1 QUALITY AND FINANCIAL PERFORMANCE OF PRIVATE EQUITY OWNED NURSING HOMES IN THE STATE OF FLORIDA By ROHIT PRADHAN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE R EQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010
2 2010 Rohit Pradhan
3 To my Mom
4 ACKNOWLEDGMENTS A work of this nature is impossible without the help and support of many peopl e. I would like to thank my committee chair, Dr. Harman, for his constant encouragement and support throughout my PhD. I acknowledge his patience in repeatedly going over the methods section with me and answering all my questions. I am indebted to Dr. Hyer for her generosity in sharing the data. To other members of my committee, Dr. Duncan, Dr. Al department chair, Dr. Duncan, for his much needed words of wisdom and encouragement. On e of the pleasures of being part of this department is the friendship I have developed and enjoyed over the years with my fellow PhD students. To all those who have traveled the same path with me t hank you for your friendship. I especially record my thank s to A lex Laberge, Michael Morris and Latarsha Chisholm ; without them this process would have been much harder. I remain grateful to Dr. Weech Maldonado for his mentorship, infinite patience and guidance throughout these last few years I have had the oppo rtunity to work with him. Above all, I thank him for his wonderful camaraderie and friendship Finally, I acknowledge my parents whose unconditional love and blind faith in me has always encouraged me to chase my own dreams, and script my own destiny.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ ........ 11 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 15 Background ................................ ................................ ................................ ............. 15 Role of Nursing Homes in the Health Care System ................................ ................ 18 Nursing Home Reimbursement ................................ ................................ ............... 20 Medicaid ................................ ................................ ................................ ........... 20 Private Pay ................................ ................................ ................................ ....... 22 Medicare ................................ ................................ ................................ ........... 22 Quality of Care in Nursing Homes ................................ ................................ .... 22 OBRA and Nursing Home Quality ................................ ................................ .... 24 Market based Approaches to Improve Quality in Nursing Homes .................... 25 Distribution of Nursing homes by Ownership/Structure ................................ .... 26 Dominance of For Profits ................................ ................................ .................. 27 Dominance of For Profit Chains ................................ ................................ ....... 28 The Boom Years in the Nursing Home Industry ................................ ............... 30 The Crisis in the Nursing Home Industry ................................ .......................... 32 Mariner Health Group ................................ ................................ ................ 36 Beverley Enterprise ................................ ................................ .................... 37 Genesis Health Care ................................ ................................ .................. 41 Defining Private Equity ................................ ................................ ..................... 43 Organization of Private Equity Market ................................ .............................. 44 Nurs ing Homes in Florida ................................ ................................ ....................... 46 Private Equity Investments in Florida Nursing Homes ................................ ............ 46 The Rationale for this Study ................................ ................................ .................... 47 2 CONCEPTUAL FRAMEWORK ................................ ................................ ............... 54 Influence of Ownership on Quality and Financial Performance ........................ 54 Ownership and Financial Performance ................................ ............................ 56 Private Equity and Financial Performance ................................ ........................ 57 Management Buyouts ................................ ................................ ...................... 58 Ownership and Quality ................................ ................................ ..................... 59 Quality of Care and Private Equity ................................ ................................ ... 60 Conceptual Frame work ................................ ................................ .................... 61
6 Agency Theory ................................ ................................ ................................ 62 Manager Accountability ................................ ................................ .................... 63 Free Cash Flo w Theory and the Role of Debt ................................ .................. 65 Other Anticipated Benefits of Private Equity Ownership ................................ ... 67 Private Equity Ownership and Quality of Care ................................ ................. 68 3 METHODS ................................ ................................ ................................ .............. 74 Datasets ................................ ................................ ................................ ........... 74 Merging Datasets ................................ ................................ ............................. 76 Population ................................ ................................ ................................ ........ 78 Quality Variables ................................ ................................ .............................. 79 Structural measures of quality ................................ ................................ .... 79 Process measures of quality ................................ ................................ ...... 82 Outcome measures of quality ................................ ................................ .... 85 Finan cial Performance Measures ................................ ................................ ..... 88 Independent Variables ................................ ................................ ..................... 91 Control Variables ................................ ................................ .............................. 92 Other Variables ................................ ................................ ................................ 93 Model ................................ ................................ ................................ ................ 93 Quality Variable Analysis ................................ ................................ .................. 95 Financial Variable Analysis ................................ ................................ ............... 97 4 RESULTS ................................ ................................ ................................ ............. 105 Hypothesis 1 ................................ ................................ ................................ ......... 105 Hypothesis 2 ................................ ................................ ................................ ......... 108 Structural Variables ................................ ................................ ........................ 109 Process Variables ................................ ................................ .......................... 110 Outcome Variables ................................ ................................ ......................... 111 5 DISCUSSION ................................ ................................ ................................ ....... 116 Quality of Care ................................ ................................ ................................ ...... 116 Fin ancial Performance ................................ ................................ .......................... 119 Managerial Implications ................................ ................................ ........................ 123 Policy Implications ................................ ................................ ................................ 126 Limitations ................................ ................................ ................................ ............. 129 Conclusions ................................ ................................ ................................ .......... 130 APPENDIX: BACKGROUND TABLES ................................ ................................ ........ 133 LIST OF REFERENCES ................................ ................................ ............................. 164 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 184
7 LIST OF TABLES Table page 1 1 Private equity investment in nursing homes ................................ ....................... 52 1 2 Florida nursing homes by organizational types ................................ ................... 53 1 3 Major private equity nursing homes acquisitions in Florida ................................ 53 3 1 RN hours per patient day distribution ................................ ............................... 100 3 2 LPN hours per patient day distribution ................................ .............................. 100 3 3 CNA hours per patient day distribution ................................ ............................. 100 3 4 Skill mix distribution ................................ ................................ .......................... 100 3 5 Pressure sore prevention distribution ................................ ............................... 100 3 6 Restorative ambulation distribution ................................ ................................ ... 101 3 7 Pressure ulcer high/low risk prevalence distribution ................................ ......... 101 3 8 Non operating revenue distribution ................................ ................................ ... 101 3 9 Non operating costs per patient day distribution ................................ ............... 101 3 10 Operating margin distribution ................................ ................................ ............ 101 3 11 Total margin distribution ................................ ................................ ................... 102 3 12 Acuity index distribution ................................ ................................ ................... 102 3 13 Dependent variables and their definitions ................................ ......................... 103 3 14 Independent variables and their definitions ................................ ...................... 104 3 15 Control variables and their definitions ................................ ............................... 104 4 1 Descriptive statistics of dependent variables ................................ .................... 112 4 2 Descriptive statistics of independent control and variables .............................. 113 4 3 Fi nancial performance of private equity nursing homes: Model 1 ..................... 113 4 4 Financial performance of private equity nursing homes: Model 2 ..................... 114 4 5 Private equity nursing homes quality indicated by structural variables: Model1 114
8 4 6 Private equity nursing homes quality indicated by structural variables: Model2 114 4 7 Private equity nursing homes quality indicated by process variables: Model1 115 4 8 Private equity nursing homes quality ind icated by process variables: Model2 115 4 9 Private equity nursing homes quality indicated by outcome variables: Model1 115 4 10 Private equity nursing homes quality indicated by outcome variables: Model2 ................................ ................................ ................................ .............. 115 5 1 Rehabilitation categories under PPS ................................ ................................ 132 A 1 T test of dependent quality variables ................................ ................................ 133 A 2 T test of dependent financial variables ................................ ............................. 134 A 3 Chi square test of L PNs on contract by Equity ................................ ................. 135 A 4 Chi square test of RNs on contract by Equity ................................ ................... 135 A 5 Chi square test of Actual harm citati ons by Equity ................................ ............ 136 A 6 Operating revenue PPD: Model 1 ................................ ................................ ..... 137 A 7 Operating revenue PPD: Model 2 ................................ ................................ ..... 137 A 8 Operating cost PPD: Model 1 ................................ ................................ ........... 138 A 9 Operating cost PPD: Model 2 ................................ ................................ ........... 138 A 10 OLS of no n operating revenues PPD: Model 1 ................................ ................. 139 A 11 OLS of non operating revenues PPD: Model 2 ................................ ................. 139 A 12 OLS of non operating costs PP D: Model 1 ................................ ....................... 140 A 13 OLS of non operating costs PPD: Model 2 ................................ ....................... 140 A 14 OLS of operating margin: Model 1 ................................ ................................ .... 141 A 15 OLS of operating margin: Model 2 ................................ ................................ .... 141 A 16 OLS of total margin: Model 1 ................................ ................................ ............ 142 A 17 OLS of total margin: Model 2 ................................ ................................ ............ 142 A 18 Logit of % Medicare: Model 1 ................................ ................................ ........... 143 A 19 Logit of % Medicare: Model 2 ................................ ................................ ........... 143
9 A 20 Logit of % Medicaid: Model 1 ................................ ................................ ............ 144 A 21 Logit of % Medicaid: Model 2 ................................ ................................ ............ 144 A 22 Logit of % Other: Model 1 ................................ ................................ ................. 145 A 23 Logit of % Other: Model 2 ................................ ................................ ................. 145 A 24 OLS of Acuity index: Model 1 ................................ ................................ ........... 146 A 25 OLS of Acuity index: Model 2 ................................ ................................ ........... 146 A 26 OLS of RN hours PPD: Model1 ................................ ................................ ........ 147 A 27 OLS of RN hours PPD: Model2 ................................ ................................ ........ 147 A 28 OLS of LPN hours PPD: Model2 ................................ ................................ ...... 148 A 29 OLS of LPN hours PPD: Model 2 ................................ ................................ ..... 148 A 30 OLS of CNA hours PPD: Model 1 ................................ ................................ ..... 149 A 31 OLS of CNA hours PPD: Model 2 ................................ ................................ ..... 149 A 32 OLS of skill mix: Model 1 ................................ ................................ .................. 150 A 33 OLS of skill mix: Model 2 ................................ ................................ .................. 150 A 34 Logistic regression of LPN s on contract: Model 1 ................................ ............. 151 A 35 Logistic regression of LPNs on contract: Model 2 ................................ ............. 151 A 36 Logistic regression of RNs on cont ract: Model 1 ................................ .............. 152 A 37 Logistic regression of RNs on contract: Model 2 ................................ .............. 152 A 38 OLS of Pressure sore prevention: Model 1 ................................ ....................... 153 A 39 OLS of Pressure sore prevention: Model 2 ................................ ....................... 153 A 40 OLS of Restorative ambulation: Model 1 ................................ .......................... 154 A 41 OLS of Restorative ambulation: Model 2 ................................ .......................... 154 A 42 Logit of use of restraints: Model 1 ................................ ................................ ..... 155 A 43 Logit of use of restraints: Model 2 ................................ ................................ ..... 155 A 44 Logit of use of catheters: Model 1 ................................ ................................ ..... 156
10 A 45 Logit of use of cath eters: Model 2 ................................ ................................ ..... 156 A 46 Logit of ADL 4 point decline: Model 1 ................................ ............................... 157 A 47 Logit of ADL 4 point decline: Model 2 ................................ ............................... 157 A 48 Logit of bowel decline: Model 1 ................................ ................................ ........ 158 A 49 Logit of bowel decline: Model 2 ................................ ................................ ........ 158 A 50 Negative binomial regression of total deficiencies: Model 1 ............................. 159 A 51 Negative binomial regression of total deficiencies: Model 2 ............................. 159 A 52 Logistic regression of actual harm citation: Model 1 ................................ ......... 160 A 53 Logistic regression of actual harm citation: Model 2 ................................ ......... 160 A 54 OLS of pressure sore h/l risk prevalence: Model 1 ................................ ........... 161 A 55 OLS of pressure sore h/l risk prevalence: Model 2 ................................ ........... 161 A 56 Correlation matrix ................................ ................................ ............................. 163
11 LIST OF FIGURES Figure page 1 1 Long term continu um of care ................................ ................................ ............. 5 0 1 2 Spending for nursing home care ................................ ................................ ........ 50 1 3 Percent nursing homes by chains 1993 20 04 ................................ .................... 51 2 1 Donabedian ................................ ................................ ................. 73 A 1 Nursing home deficiencies: scope and severity ................................ ................ 162
12 Abstract of Dissertation Presented to the Graduate School of the Unive rsity of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy QUALITY AND FINANCIAL PERFORMANCE OF PRIVATE EQUITY OWNED NURSING HOMES IN THE STATE OF FLORIDA By Rohit Pradhan December 2010 Chair: J effrey S. Harman Major: Health Services Research Purpose : P rivate equity has acquired multiple large nursing home chains within the last few years ; by 2007, it owned six of the 10 largest chains. Critics hav e alleged that private equity nursing homes comp romise on quality of care in pursuit of profits and have called for greater congressional and regulatory oversight. Quality of care in nursing homes is an area of important public concern because it concerns the well most vuln erable citizens; in addition, as the principal payer of nursing home care, government has a strong interest in nursing home performance. However, the empirical evidence on the purported impact of private equity on nursi ng home performance remains limited a nd often contradictory ; ergo this study. Methodology : Secondary data from the Medicare Cost Reports, Minimum Data Set (MDS), the On line Survey Certification of Auto mated Records (OSCAR) file, Area Resource File (ARF) term Ca re Focus dataset are combined to construct a lon gitudinal dataset for the study period 2000 2007. The initial population comprised of over 6000 observations with an average of approximately 760 nursing homes for each year of the study period. Of those 760 nursing homes,
13 approximately 200 are not for profit 300 non chain based while 40 are hospital based. All these observations are eliminated resulting in a final sample consisting of 2822 observations. D ependent quality variables co nsist of structure (RN, LPN, CNA staffing ratios skill mix, and RNs and LPNs contract ), process (restraints, catheters, pressure ulcer prevention, and restorative ambulation) and outcomes (ADL worsening, pressure sores, bowel continence decline, deficiencies and actual harm cit ation) D ependent financial variables consist of operating and non operating revenues, operating and non operating costs, operating and total margins, payer mix (census Medicare, census Medicaid, census other), and acuity index. Independent variables prima rily reflect private equity ownership. The study was analyzed using ordinary least squares (OLS), gamma distribution with log link, logit with binomial family link, negative binomial regression, and logistic regression. Results : Private equity nursing hom es have significantly worse RN staffing while they report higher CNA and LPN staffing than the control group. Skill mix is poorer in private equity nursing homes. There is no difference in process or outcome variables between private equity nursing homes a nd the control group except in case of deficiencies where they perform significantl y worse and pressure sore prevention where they report slightly better results. Private equity nursing homes have higher operating margin as well as the total margin; they also have higher operating revenues and costs. Conclusions : quality and financial performance; private equity homes are able to deliver significantly better financial performance than the control group while quality is largely similar.
14 However, cause s for concern remains particularly due significant lowering of RN staffing in private equity nursing homes and their significantly higher deficiencies Greater transparency should be ensured in private equity nursing home deals, and accountability should be fixed via an expanded and strengthened regulatory framework.
15 CHAPTER 1 INTRODUCTION Background Nursing homes are an integral part of the US healthcare system. In 2006, there were 16,269 nursi ng home facilities with over 1,760,000 certified and 52,000 uncertified beds in the U.S (Harrington, 2007) The expenditure on nursing home care totaled $115.2 billion in 2004, which represented 6.1 percent of national health expenditures (C. Smith, Cowan, Heffler, & Catlin, 2006) Experts estimate that over the next twenty years, 46% of Americans aged 65 and above will reside, at least t emporarily, in a nursing home (Spillman & Lubitz, 2002) Over the last two decades, the nursing home industry has constantly evolved in response to regulatory, financial, and environmental challenges (Stevenson & Grabowski, 2007) New regulatory requirements including minimum staffing levels were imposed following the passage of Omnibus Budget Reconci liation Act (OBRA), while the financial viability of nursing homes was threatened by reimbursement reforms introduced by the Balanced Budget Act (BBA) of 1997. As the continuum of care evolves, nursing homes face ever increasing competition from home healt h agencies, assisted living facilities and hospitals. Occupancy rates for nursing home beds fell from 92% in 1987 to 87% in 1995 (Stone., 2000) and reduced further to 85% in 2006 (Harrington, 2007) The nursing home industry has been adversely affected by the sluggish economy; the 2009 federal budget imposed a $10 billion cut in Medicare (Department of Health & Human Services, 2008) and further cuts are anticipated as the economic outlook remains poor. On similar lines, budgetary and financial constraints of states (McNichol.
16 & Lav., 2008) may further lower Medicaid reimbursement rates for nursing homes, as it has happened previously (Smith., Ramesh., Gifford., El lis., & Wachino., 2003) The nursing home industry has struggled to respond to state specific changes for instance, the rise of a litigious climate in the states of Florida and Texas (Stevenson & Studdert, 2003) leading to the strategic withdrawal of large nursing home chains from these states (Stevenson & Grabowski, 2007) Innovations introduced in the organizational structure of the nursing homes insofar as much as they affect the delivery of services, performance, and financial viability remain a salient public policy concern. Public interest in the nursing home industry is not only motivated by the fact that government is its largest payer (Agency for Healthcare Administration, 2007) but because it concerns the well being of 1.6 (Harrington, 2007) An important organizational change introduced in the nursing home industry in re cent years is private (Jensen, 2007) ; in essence, the acquired firms delist from stock exchanges and are taken private. Investments are generally m ade in underperforming publicly traded firms and private equity hopes to recoup their investment by substantially improving the financial performance of acquired firms. Private equity financing has witnessed exponential growth in the last two decades and i s currently estimated to be one sixth the size of commercial bank loan market (Fenn, Liang, & Prowse, 1995) Since 2000 the nursing home industry has witnessed large scale purchases of chains by private equity ranging from Centennia l Health bought by Hilltopper in 2000 to Manor Care acquisition by the Carlyle group in 2007
17 (Table 1). By 2007, private equity owned six of the largest chains representing 9 percent of total nursing home beds (Harrington, 2007) ments in nursing homes has sparked a vigorous public policy debate because of the significant role nursing homes play in the healthcare system and the peculiar features of private equity financing. Private equity is thought to be excessively focused on profits particularly because of the short investment ho rizon (Strmberg, 2007) ; in addition, it is considered to be the most expensive source of finance (Covitz & Liang, 2001) creating pressure to generate high amount of cash flow to service debts. Private equity owned firms are exempt from Security and Exchange Commission (SEC) regulation s and face diminished requirements for public disclosures of acquisitions and divestitures creating additional concerns over their monitoring by regulatory authorities. In October 2007, the New York Times produced an investigative report which claimed that in the pursuit of profits, private e quity owned nursing homes are compromising on quality (Duhigg, 2007) A report by Service Employees International Union (SEIU) ( 2007) has argued that private equity restructures nursing homes to maximize profit, and lowers quality of care for residents. The furor which followed these reports led to hearings by Subcommittee on Health of the House Committee on Ways and Means as well as the S before the House Committee on Ways and Means, Charlene Harrington, Professor of Nursing at University of California, San Francisco opined that equity firms do not have the expertise and experie nce to manage complex nursing home (Harrington 2007)
18 Despite its important public policy implications and the growing chorus of concern, limited independent evidence exists on quality and financial performance of private equity owned nursing homes. Two important and linked questions need to be answe red: a) Literature suggests that private equity improves the financial performance of acquired firms; is this relationship maintained in the nursing home industry? b) Does this pursuit of profits adversely affect quality of care? In other words, how do pri vate equity owned nursing homes differ in their financial performance and quality of care vis vis other investor owned homes? Role of Nursing Homes in the Health Care System facility with three or more beds that is either licensed as a nursing home by its state, certified as a nursing facility under Medicare or Medicaid, indentified as a nursing unit in (Delfosse, 1995) Nursing homes provide a place of residence, temporary or permanent, for people who require skilled nursing care and may suffer from illnesses, injuries, and functional or age related disabilities. Though most nursing home patients are over the age of 6 5, they also may cater to younger patients. The nursing home industry includes skilled nursing facilities (SNFs), Assisted living Facilities (ALFs), and retirement homes providing different levels of care to residents (Living, 2001) (Figure1). Retirement homes provide limited services, for instance, rooms assistance with personal care but not medically intensive care (Spillman, Liu, & McGilliard, 2002) SNFs provide the most complex level of medical and nursing care; Medicare pays only for care provided in facilities classified as SNFs (Giacalone, 2001)
19 In the U.S. healthcare system, nursing homes perform two important functions. First, they provide long individualized, well coordinated services that promote the maximum possible independence for people with functional limitations and that are provided over an (Shi & Singh, 2009) While longer stay in the LTC facility almost invariably results in progressive health status decline due to co mo rbidities in 2000, the number of Americans suffering from chronic conditions was 125 million (45%), a number which is stated to rise to 157 million by 2020 (Shi & Singh, 2009) status at the time of admittance, while providing maximum functional independence. The second major function of a nursing home is to provide sub acute care designed for patients who fficiently stabilized to no longer require acute care (Department of Health and Human Services, 1994) It may include wound care, intravenous therapy, ventilator support and rehabilitative therapy (Shi & Singh, 2009) The literature has treated subacute care as continuum between acute care hospitals and long term care facilities (Department of Health and Human Services, 1994) The nursing home industry in the United States is highly regulated (Agency for Healthcare Admin istration, 2007; Grabowski, 2004) At the state level, each nursing home operating within its boundaries must be licensed by the state regulatory authorities. Each state has basic standards of care which must be met by the nursing homes verifiable through inspections which are generally held once every year In the state of Florida, the Agency of Health Care Administration (AHCA) is the regulatory
20 body with oversight over nursing homes In order to accept Medicare/Medicaid patients, the nursing homes need to be certified by the Centers for Medicare and Medicaid Services (CMS). Nursing Home R eimbursement It is a highly complex system due to the multiplicity of payers Medicaid, Medicare, and private pay and because states have broad latitude in designing the ir own policies subject to federal guidelines ; Medicaid nursing home payment rates and in the specific methodology used to formulate the rates are idiosyncratic in nature. As seen in figure 2 government is the largest payer for nursing home care in 2005, Medicare and Medicaid together accounted for 60 percent of nursing home spending (Wiener, Freiman, & Brown, 2007) Therefore, public policies which affect reimbursement influence organization, financial performance and quality of care in the nursing home industry Medicaid Medicaid, the state and federal program, is the major source of funding for those with low income and assets (Department of Health and Human Services, 2009) In 2005, 44% % of nursing home revenues were derived from Medicaid (Wiener et al., 2007) while it covers approximately 70% of total available beds (Feng, 2008). In 2005, Medicai d expenditure on nursing home care totaled $53.6 billion. (Wiener et al., 2007) In their discussion of state Medicaid reimbursement policies, Swan et al. (Swan & Pickard, 2003) identify five payment methodologies: retrospective, prospective class, prospective facility specific, adjusted, and c ombination. The most common reimbursement mechanism adopted by Medicaid programs is the prospective per diem system (Feng, Grabowski, Intrator, Zinn, & Mor, 2008) In
21 contrast to retrospective payments that reflect current costs, the prospective payment rates are based on prior year costs. The main difference between prospectiv e class and prospective facility specific is that in the latter, rates are set at the facility level rather than a uniform rate across the state. Adjusted reimbursement is similar to retrospective reimbursement except for its allowance of interim rate incr eases during the financial year to reflect cost increase. Finally, the combination reimbursement methods are a mixture of prospective and retrospective reimbursement (Swan et al., 2000) As of 1997, only one state used retrospective reimbursement, 25 used prospective 21 used adjusted and 3 used combinations (Feng, Grabowski, Intrator, & Mor, 2006) Over the last two decades, states have increasingly adopted the case mix reimbursement methodology to fine tune th eir nursing home reimbursement. In 1981, merely four states used case mix payments, the number had increased to 32 in 2002 and currently 35 states have adopted this system ( Long Term Care Community Coalition 2009; Feng et al., 2006) Resource Utilization Group (RUG III) classification ( Long Term Care Community Coalit ion, 2009; Fries et al., 1994) currently in its third model, is the most commonly used model for case mix adjustment. The principle logic underlying the case mix payment system is to reimburse medical facilities according to resources consumed by particu lar group of patients, and, in turn, make it financially remunerative for them to admit high acuity patients (Arling & Daneman, 2002; Willemain, 1980) Florida follows a cost based prospective payment system for Me dicaid nursing homes. Reimbursement rates are set for each provider based on its historical cost of providing services with rate increases linked to pre determined inflation indices. In case
22 of nursing homes, ceilings are established separately for patient care costs, property costs and operating costs. There is no case mix reimbursement. Private Pay A smaller source of reimbursement for nursing homes is private pay which covers 26% of total nursing home reimbursement (Wiener et al., 2007) Nevertheless, since Me dicaid rates are only 70% of private pay rates, it is generally more lucrative for nursing homes to admit such patients (Grabowski, 2004) Medicare Medicare covers nursing home care for beneficiaries requiring skilled nursing care or rehabilitative therapy for conditions related to a minimum hospital stay of three days. All necessary services room and board, nursing care, and ancillary services such as drugs, laboratory tests, and physical therapy are covered for up to 100 days of care per spell of illness subject to a coinsurance beginning the 21 st day of the rapy (Government Accountability Office, 2002) Before 1997, Medicare paid SNFs based on retrospective reasonable cost basis. While th ere were some limits imposed on routine care (room and board), ancillary services including rehabilitation therapy were virtually unlimited (Government Accountability Office, 2002) Starting July 1 st 1998, the Medicare Prospective Payment System (PPS) established a prospectively determined per diem payment rate for Skilled Nursing Facility (SNF) patient care adjusted for case mix, ar ea wages, urban or rural status, and changes in input prices. Quality of Care in Nursing Homes Improving the quality of nursing home care continues to be an important public policy goal (Mukamel & Spector, 2000; W alshe, 2001) Regulated through a complex
23 system of federal and state regulations, nursing homes face constant regulatory and improving nursing home quality parallels it s gradual adoption of a dominant role in this industry. Before the passage of Medicare and Medicaid in 1965, no federal standards existed on regulating nursing homes (Walshe, 2001) except for the requirement for state licensure (Kumar, Norton, & Encinosa, 2006) Titles XVIII and XIX of the Social Security Act of 1965 imposed federal standards on nursing homes to ensure that they deliver an minimally acceptable standard of care (Government Accountability Office, 1999a) Subsequently, Congress enacted federal legislation in 1967 and 1972 to further strengthen the regulatory framework. H owever, nursing homes continued to face criticism for poor quality of care; a Government Accountability Office (GAO) reported ( 1971) that over half the federal certified facilities did not meet standards on physician visits and fire standards.. Similarly, Hawes ( 1996) argues that f ederal and state inspections were flawed; and were poorly implemented. In addition, the regulation was tient Process Outcome (SPO) model, structure was emphasized rather than process or outcomes (Wiener et al., 2007) The poor standard of nursing home care motivated the United States Congress to ask the non par tisan Institute of Medicine (IOM) to recommend steps to address the glaring gaps in quality of care. In a report released in 1986, IOM ( 1986) sharply
24 criticized the prevailing st andard of care in nursing homes and offered a series of recommendations to improve the quality of care. In 1987, GAO ( 1987) reported that deficiencies were widespread in patient care and recommended passage of a federal legislation to improve quality of care. Largely in response to these reports, the Congress passed the Nursing Home Reform Act as part of the Omnibus Budget Rec onciliation Act (OBRA) in 1987 which introduced sweeping reforms in nursing home regulation (Kumar et al., 2006) OBRA and Nursing Home Quality Kumar et al. ( 2006) point o ut that OBRA introduced three principal changes to the regulatory system. First, Nursing homes were obliged to provide minimum nursing hours with availability of Licensed Practioner Nurses (LPN) 24 hours a day and continuous training for care givers. It al so mandated that nursing homes provide a broader array of services including provision of rehabilitation, pharmaceutical, and dental services (Giacalone, 2001) Second, OBRA established standards of care which were much m limiting use of restraints and prohibited patient abuse, mistreatment and neglect (Wiener et al., 2007) It also ordered the Resident Assessment Instrument implemented nationwide in 1993 an important component of which was the Minimum Data Set (MDS) (Mukamel & Spector, 2003) Third, it imposed strict penalt ies on nursing homes violating government regulations (Kumar et al., 2006) It created a uniform system by merging Medicare and Medicaid standards survey and certification process (Wiener et al., 2007) Despite the sweepi ng changes introduced by OBRA, the nursing home quality of care remains a pressing public concern. Over the last few years, research continues to
25 demonstrate nursing home residents remain susceptible to poor quality and gross abuse and neglect (Government Accountability Office, 2000a, 2003; Wunderlich & Ko hler, 2001) In 2006, more than 90 percent facilities were cited for deficiencies while one fifth were cited for serious deficiencies (Wiener et al., 2007) ; importantly, this figure may be an underestimation of actual deficiencies by as much as 40% (Government Accountability Office, 2008) In its testimony before the U.S. Special Com mittee on Aging, the GAO ( 2008) expressed concern over persistent variati on in the proportion of nursing homes reporting serious deficiencies. Market based Approaches to Improve Quality in Nursing Homes In recent years, the Centers for Medicare and Medicaid Services (CMS) has moved from a purely regulatory approach in which nu rsing homes were penalized for poor quality to a broader paradigm which includes, apart from penalties, market and consumer driven approaches like quality reporting and value based purchasing. In 2001, the CMS announced the Nursing Home Quality initiative (NHQI) a four pronged approach to improve nursing home quality (Kissam et al., 2003) It included public reporting of nursing home quality measures disseminated via quality report cards (Castle & Lowe, 2005; Kissam et al., 2003) culminating in the establishment of Nursing Home Compare website (Centers for Medicare and Medicaid, 2009b) Public reporting is predicated on the idea that empowering consumers with information on quality (Mukamel & Mushlin, 1998) would incentivize competition in the nursing home market based on relative quality of care (Marshall, Shekelle, Leatherman, & Brook, 2000) Mukamel et al. ( 2008) demonstrate that publication of quality report cards resulted in improvement of quality of care in nursing homes though the effect was not similar across all quality variables.
26 The CMS Nursing Home Value Based Purchasing demonstration project seeks to reward nursing homes that can improve or deliver high quality care in four specific areas: staffing, resident outcomes, avoidable hospita lizations and reductions in deficiency citations (Centers for Medicare and Medicaid, 2009a) Participating nursing homes will be awarded points in each of the four targeted areas, and those with the highest sco res or greatest improvement will become eligible for a performance payment. Approximately 200 nursing homes in three states (New York, Wisconsin, and Arizona) are participating in the demonstration project. Because the quality of care may be affected by multiple factors of which ownership is an important factor (Davis, 1993; Hillmer, Wodchis, Gill, Anderson, & Rochon, 2005; Spector, Selden, & Cohen, 1998) ; and because of the frailties associated with nursing home residents, factors which may potentially impact nursing home quality are an important regulatory concern. Unsurprisingly, a novel form of nursing home ownership like private equity has raised concerns over quality of care and fueled demands for greater reg ulatory oversight. Distribution of Nursing home s by Ownership/Structure Unlike the hospital sector, the nursing home industry has traditionally been dominated by for profit nursing home chains. According to the National Nursing Home Survey 2004 (Figure 3), 61 % of total nursing homes are for profit; 32% not for profit, while 8% are government owned (Center for Disease Control, 2004) Similarly, 54% of nursing homes belong to a chain while 45% are independents (Center for Disease Control, 2004) Giacalone ( 2001) points out that legislative developments are for profits and
27 government; public policy has similarly promoted for profit chains as pre eminent organizational form in the nursing home industry. Dominance of For P rofits The proprietary nature of the modern nursing home industry in the United States owes its origins to the enac tment of the Social Security Act of 1935 (Social Security A dministration 1935) Previously, long term care was restricted to almshouses which delivered charitable services to the indigent. The Social Security Act introduced a federal grants in aid program to the states for means tested old age assistance (OAA) t o elderly who may not quality for old age insurance (OAI) (Bradford, 1986) Strikingly, the Act specifically prohibited aid to existing almshouses (Giacalone, 2001) while other not for profit providers were hobbled by economic deprivation of the Great Depression leading to depressed charitable donations a major source of funding. Motivated by the purchasing power of the elderly and aided by limited competition, for profit entities entered the elderly care market and the industry came to be domi profit nursing homes owned by individuals with very limited role for corporation. After the conclusion of the Second World War, the proprietary nature of the nursing home industry received further impetus from two factors: First, the vendor payments system provided federal matching grants to states for direct payments to nursing homes for care provided to OAA beneficiaries. Second, Sm all Business Administration (SBA) and Federal Housing Authority (FHA) provided government dollars for nursing home construction facilitating large scale new constructions and expansion of existing nursing homes. The advent of stable payments and constructi on loans attracted entrepreneurs and real estate speculators to the nursing home industry (Bradford, 1986)
2 8 prudent real estat industry (Bradford, 1986) A further fillip to the nursing home industry was provided by the enactment of the Social Security Amendments of 1965 which led to the establishment of Medicare and Medicaid. While nursing homes were not part of the initial legislature, the 1967 Moss Amendments provided the authority for nursing home participation in Medicaid (Giacalone, 2001) The extension of coverage allowed hitherto excluded groups to access nursing home care, and, in turn, ensured a constantly expanding market to providers. The adoption of the related reimbursement mechanism for Me dicaid subsequent to the passage of Public Law 92 603 in 1972 allowed providers virtually unlimited reimbursement for services rendered 1 (Giacalone, 2001) In addition, nursing homes were reimbursed for capital investment including the facility which in cost reimbursement, a guaranteed profit factor, and interest plus depreciation for capital (Bradford, 1986) The government, financial, and market imperatives set in motion a process favoring the proprietary model which has ensured the relative dominance of for pro fits in the nursing home industry to this day. Dominance of F or Profit Chains (Ingram & Baum, 1997) Despite some fluctuations in recent years, chains have exerted considerable influence over the 1 The public law 92 603 also introduced Supplemental Security Income (SSI) which expanded the number of people eligible for Medicaid.
29 nursing home industry (Harrington, 2007) While the proprietary mode l has been Medicare et al. (Harrington, Mullan, & Carrillo, 2004) have pointed out that before 1965, the nursing home in ( 1986) argues that increased corporatization and formation of large chains in the post Medicare e First, capital requirements inc reased because of tremendous demand for nursing home care. Due to the overtly generous reimbursement policies and favorable tax policies, nursing homes were highly profitable in the immediate post Medicare era increasing their appeal among Wall Street investors. Publicly traded chains had greater access to financ ial markets to meet their capital needs vis vis independents. Second, the stricter licensing laws and increasing complexity of reimbursement mechanisms favored chains as they had greater resources to meet regulatory and payment complexities (Baum, 1999) Third, the use of Certificate of Need (CON) limited the ability of nursing hom es to expand by constructing new nursing homes thereby leaving m ergers and acquisitions (M&A) as the easiest route for expansion for nursing home chains. The attractiveness of M&A was increased by continued strong demand for nursing home care. Publicly traded chains benefited from their ability to finance M&A because of greater access to capital markets. Finally, the deregulation mantra pursued by the Reagan administration increased nursing home consolidation by limiting anti trust
30 concerns (Bradford, 1986) term care and changes in public policy, especially reimbursement mechanisms, made the for profit chains a formidable player in the nursing home industry. The Boom Years in the Nursing Home Industry In 1983 the Pro spective Payment System (PSS) was implemented in the hospital industry. PPS replaced per diem reimbursement with fixed payments based on diagnosis groups; hospitals now had an incentive to quickly discharge their patients (Liu, Gage, Harvell, Stevenson, & Brennan, 1999) With no PPS in the nursing home industry, t he influx of high acuity patients substantially expanded post acute care opportunities for nursing homes (Fri edman & Shortell, 1988) In 1987, Congress passed the Omnibus Budget Reconciliation Act (OBRA), which removed the distinction between skilled nursing and intermediate care centers that existed for state Medicaid systems. While OBRA increased the cost of c are by mandating reforms like presence of registered nurses at all times, it also increased the reimbursement rates to alleviate the increased costs particularly for ancillary care (Liu et a l., 1999) Focusing on ancillary services rehabilitation for example virtually guaranteed higher profits. This resulted in a rapid increase in Medicare SNF payments between 1990 and 1998, Medicare payments for SNF services increased by 2.5% annually reach ing nearly $14 billion in 1998 (Government Accountability Office, 2002) driven largely by what the Office of Inspector General (OIG) termed as excessive rehabilitative therapy (Government Accountability Office, 1999b) In 1988, the Congress passed Medicare Catastrophic Coverage Act of 1988, which increased coverage for seniors and removed hospital stay requirements. Though the act was repealed the next year, nursing homes read it as a signal for increasing payments.
31 With demand for nursing home care starti ng to rise due to demographic changes, the generous reimbursement policies attracted renewed interest of Wall Street and national chains. Since avenues for new constructions were limited due to CON requirements, consolidation became the buzz word facilitat ed by creative financial mechanisms like Real Estate Investment Trusts (REITs) (Stevenson & Grabowski, 2007) The REIT structu re worked as follows: the company would sell some facilities to a REIT, which would then lease them back to the company. The sale of assets would (Harrington et al., 2004) This strategy was perhaps first adopted by Vencor, a giant nursing home chain which split into t wo companies: Vencor and Ventas. The latter owned the real state and simply leased back the property to Vencor which functioned as the operator. The acquisitions particularly in investor owned chains were largely financed by debt (Giacalone, 2001) ; overleveraged 2 chains had diminished ability to survive environmental shocks. The 1990s was also a period of consolidation. Genesis acquired Meridian Healthcare in 1995 and Multicare in 1997 while Vencor acquired Hillhaven for $1.9 billion in 1995. The second largest nursing home chain, Mariner Post Acute network was an amalgamation of Living Centers of America, Grancare, and Mariner (Giacalone, 2001) In addition, chains acquired complementar y lines of business such as assisted living, home health, and rehabilitation therapy (Walters, 1993) 2 Leverage refers to use of debt to supplement investments.
32 In a matter of decades, metamorphosized into an industry where giant chains were perhaps the most influential players. Figure 4 illustrates the rapid dominance of chains; In 1991, chains controlled 39% of the nursing home marke t, a figure which had increased to 56% by 2002 (Harrington, 20 07) Equally pertinent to note is the increased industry concentration. 3 In 1973, the three largest national chains controlled 2.2% of the market, a figure which had increased to 9.6% by 1982 (Bradford, 1986) Stevenson and Grabowski (2007) estimate that between the years 1993 and 2005, the top five and ten chains av eraged 16 percent and 24 percent of the total proportion of chain facilities The Crisis in the Nursing Home I ndustry In 2000, five of the largest nursing home chains operated under bankruptcy protection (Harrington, 2007) While this crisis was precipitated by the enactment of the Balanced Budget Act (BBA) of 19 97, its roots can be traced to developments in the nursing home industry over the previous decade. In the fall of 1997, Congress passed the Balanced Budget Act which formally ended the cost based system and marked the introduction of PPS in the nursing ho me industry. Under the new system, SNFs are paid a case mix adjusted amount for each day of care. Patients are assigned to one of 44 groups, called resource utilization groups, version III (RUG III) ( Medicare Paymen t Advisory Commission, 2003; Fries et al., 1994) which consist of seven major categories. Patients in a particular RUG group 3 An industry is said to be concentrated when it is dominated by a small number of firms who acquire increased market power. Consolidation is thought to be anti competitive..
33 are expected to require similar amount of resources. The system incentivizes efficiency as provide r cost is no longer the sole de ciding factor in deciding Medicare rates. The period following passage of BBA saw a precipitous fall in the financial fortunes of some of the largest nursing home chains. In September 1998, Sun Healthcare Group reported poor results and Genesis Health Ve ntures followed suit in November, 1999 (Keaveney, 1999) Large chains, for instance, Mariner Post Acute Network, Vencor, and Beverl y Enterprises were affected by government investigations in their Medicare billing practices resulting in large fines (Wall Street Journal, 1999) Providers complained that Medicare payments were too low and placed them in an unsustainable financial position (Government Accountability Office, 2000b; Harrington, 2007) Medicare SNF payments contrac ted by $1.3 billion between 1998 and 1999 while SNF days declined by 2% -both contrary to secular industry trends (Group, 2000) It is certainly true that nursing home payments decreased due to implementation of PPS. However, the nursing home chains overstated the adverse effects of PPS on the financial performance. A Government Acc ountability Office (GAO) ( 1999b) report released in 1999 claimed that Medicare payments remained adequate and blamed the fin ancial trouble of the large chains on rapid expansion. Similarly, testifying before the United States Senate, the Department of Health and Human Services (DHHS) ( 2000) pointed out that and expanded rapidly for several years prior to our Nursing Home Initiative and changes in payment structures. They leveraged themselves heavily, paying top dollar for their acquisitions and allowing their debt to
34 For profits had hitherto ex panded without factoring in costs. The industry was adversely affected by the cost of servicing old debts and sharply reduced Medicare payments. In addition, companies faced higher labor costs in an era of economic prosperity; staff costs typically compris e 60 70% of total nursing home costs (Feng et al., 2008) The litigation e nvironment turned increasingly hostile in Florida, for instance, the cost of litigation increased 37% annually (Department of Health and Human Services, 2006) This extraordinary confluence of factors viz. decreased Medicare rates, a tight labor market, debts accumulated due to rapid acquisitions, and a litigious environment affected the financial performance of the nursing h ome industry and prominent chains began to file for bankruptcy protection. In 1999, Vencor filed for Chapter 11 protection followed one month later by Sun Healthcare (American City Business Journals, 2000) In January 2000, Marnier followed suit, (Newswire, 2000b) a nd in March 2000, Genesis filed for bankruptcy. (Newswire, 2000a) In 2000, five of the largest nursing home chains, with over 1800 facilities, operated under Chapter 11 protection (Kitchener, O'Neill, & Harrington, 2005) Among large chains, only Beverly and Manor care remained solvent. Nursing home chains operating under Chapter 11 protection sought to emerge from bankruptcy by focusing on core operations (Stevenson & Grabowski, 2007) In sharp contrast to earlier aggressive acquisition posture, lar ge national chains began to divest facilities. Stevenson and Grabowski ( 2007) .list three factors driving divestitures: Poor Me dicaid reimbursement, geographical dissonance, and high malpractice expenses. Specifically, the nursing home chains divested from states with low Medicaid
35 rates and high malpractice liability: Florida and Texas. This led to the first round of private equit y investment in the nursing home industry; large national chains divested nursing homes in Florida (Stevenson & Grabowski, 2008) Similarly, the chains began to exit the ancillary business. As companies rationalized their portfolios and benefited from Medicare givebacks in the form of Balanced Budget Refinement Act (BBRA) of 1999 and the Benefits Improvement and Protection Act (BIPA) of 2000, they were able to improve their balance sheets. In, 2001, Vencor, Genesis Health Venture, and Sun Healthcare Gr oup emerged from bankruptcy (Business Wire, 2002) followed by Mariner Post Acute Network in 2002 (PR Newswire, 2002c) Although, the financial situation of prominent national chains has improved in recent years, raising capital remains a challenge for them. Traditionally, nursing home chains had benefited from their ability to raise capital from the financial markets (Banaszak Holl., Berta., Bowman., C., & Mitchell., 2002) However, the financial crisis in the industry made the capital markets wary of nursing homes with one analyst pointing have no growth prospects because all the traditional means ( Institutional Investor Systems, 2000) From Marc h 1998 to December 1999, the value of the public nursing home industry fell 83%, from $13.4 billion to $2.3 billion (Saphir, 2000) Reflecting these difficulties, alternative financial strategies such as REITs continue to be employed (Stevenson & Grabowski, 2007) Starved of capital, it led to the second and decidedly more massi ve entry of private equity in the nursing home industry (Stevenson & Grabowski, 2008)
36 Mariner Health Group The story of Mariner Health group is a complicated one with multiple mergers and acquisi tions and constant spinning of new companies. Mariner wa s formed by merger of two large nursing home chains: Paragon Health network and Mariner Health Group which themselves were formed by multiple mergers & acquisitions (M&A). The ARA group entered the nursing home industry in 1973 by forming ARA living cente rs America (LCA). By 1980s ARA living center was operating 290 long term facilities for the mentally impaired primarily concentrated in the South. In the mid nineties, with rapid consolidation in the long term care industry, LCA was sold to Grancare, a nur sing home operator based in Atlanta, Georgia. The merger created a new nursing home operator called Paragon Health network. On the other hand, Mariner Health Group began as a sub acute care operator. It grew rapidly in the 1990s with acquisitions of nursi ng home chains like Laurel, Pinnacle, and Convalescent Services (Stratton, 1993; The Wall Street Journal, 1995a, 1995b) In the pre PPS era, the company continued to perform well and expanded by large number of nu rsing homes purchases (The Wall Street Journal, 1997) The company was downgraded by rating agency Standard & Poor and sources of public financing died down. Left with no alternative, Mariner agre ed to be acquired by Paragon Health Network (The New York Times, 1998) effective July 31 st 1998 (PR Newswire, 1998) The new company, renamed Mariner Post Acute Network had 422 ski lled nursing, subacute and assisted living facilities in 42 states (Japsen, 1998) It was believed that the company would benefit from economies of scale. Unfo rtunately, Mariner performed poorly due to lowered reimbursements and limited access to capital
37 funds; within 6 months of the merger, the stock of the combined entity was trading at less than a dollar (Birger, 1998) Plagued by continued financial problems, Mariner filed for chapter 11 bankruptcy protection in year 2000 (Newswire, 2000b) Mariner continued to operate under bankruptcy protection for the next 2 years, emerging from bankruptcy only in 2002 after agreeing to a restructuring process. The newly formed entity was christened Mariner Health Care (PR Newswire, 2002b) and it operated 300 nursing homes, a substantial reduction from its prime (PR Newswi re, 2002c) In August 2003, Mariner Healthcare divested 19 of its Florida facilities to Formation Capital citing rapidly rising liability costs, and difficulty in obtaining malpractice insurance (Orlando Sentinel, 2003) In the same year, it further reduced its exposure in Florida by terminating the lease of seven more facilities (PR Newswire, 2003c) deteriorate in 2004. Unable to generate the necessary cash flow, Mariner agreed to be acquired by Nat ional Senior Care in a deal valued at $1 billion; the merger was completed in December 2004 (The New York Times, 2004) As is frequently the case with private equity financing, the deal was structured as a leveraged buyout (LBO). Mariner, now currently known as Seva Care currently operates approximately 180 nursing homes across United States. Beverley E nterprise Beverly enterprise was established in 1963 by Utah accountant Roy Christensen as three convalescent hospitals near Beverly Hills, California. With the growing d emand
38 home operator and witnessed rapid growth. By 1983 Beverley was the largest nursing home operator in the country. In line with the then prevailing trends in the nur sing home industry, Beverly also diversified --for instance, Pharmacy Corporation of America (PCA) was incorporated as its institutional pharmacy unit. Flush with easy Wall Street cash, Beverley continued to grow mainly by the acquisition route (Martin, 1986) investing in nursing homes as well non core business like Computran (Michael A. Anderson, 1986) However, concerns were expressed that Beverly was growing too fast (Flynn, 1986) ; in 1988, after reporting two consecutive years of losses, Business Week commented that Beverly may need (Miles, 1988) Motivated by the need to improve its financial performance, Beverley began to sell substantial number of its nursing homes (Dawson, 1989; Ihle, 1989) The restructuring (Rengers, 1992) However, the acquisition bug bit Beverley again in the mid and it sought to invest in higher margin businesses (post acute care hospitals, pharmacy services, rehab and respiratory therapy) (Tanner, 1993) As a result, Beverley made multiple acquisitions: American Transitional Hospitals Inc. of Franklin, Tenn. and Insta Care Pharmacy Services of Tampa (David. Smith, 1995) The rapid expansion negatively impacted its financial performance forcing Beverley to exit managed care and its pharmacy business. Beverly also exited Texas' punitive liability environment, selling 49 nursing homes in the state (The Wall Street Journal, 1996) The Business Week suggested that the Beverley w (Forest, 1996)
39 Beverley also faced regulatory challenges. National Labor Relations Board (NLRB) held it guilty of illegal anti union tactic (Karr, 1990) and it also set records for punitive damages in 1997 and 1998, suffering $70 million and $95 million judgments (later reduced t o $54 million and $3 million, respectively) relating to patient neglect. Beverley faced strong competitive challenges with the emergence of Sun Healthcare as a large integrated nursing home operator. As Beverly braced itself to face the challenges introdu ced by PPS, more bad news was on the way. The federal government charged Beverly with systematically falsifying records to defraud Medicare (Wall Street Journal, 1999) As part of its settlement, the company agreed to pay $200 million in fines; it was also forced to shut down ten facilities in California where Medicare fraud was held to be particularl y egregious (Hilzenrath, 2000) It was clear that only a massive restructuring process could save Beverley. In 2001 Beverly agreed to sell 49 nursing homes in Florida to a private equity e ntity called Formation Capital. The primary motivation cited was high malpractice insurance cost i n Florida (The New York Times, 2001) On similar lines, Beverly divested i ts Washington and Arizona operations, and reduced its California operations by 50% due to patient liability costs. As it continued to report poor financial reports, it decided to sell properties in Alabama, (Bassing, 2003) Mississippi (The Mi ssissippi Business Journal, 2003) and divested its home health care business in North Carolina (Knigh t Ridder Tribune Business News, 2003) The divestiture strategy continued in 2004 with company selling more than 80 nursing homes and its MATRIX Rehabilitation subsidiary, which operated outpatient
40 therapy clinics specializing in occupational health and s ports medicine. It attempted to improve its bottom line by diversifying into more profitable eldercare businesses such as hospices by acquiring Hospice USA (Business Wire, 2004) performance improved somewhat it reported operating profits though overall it was still in the red. By the end of 2004, Beverley was operating 356 nursing homes, a huge decrease in numbers from the early nineties when it was operating more than 1000 nursing homes. was reflected in its depressed stock price ; from a high of over $12, it was traded at le ss than $6 by the end of 2004. With Beverley sitting on valuable real estate, its cheap valuation meant that it was p rime target for private equity acquisition. In January 2005 Formation Capital and associates expressed interest in buying Beverley (Business Wire, 2005a) for $1.53 billion, an offer rejected by Beverly board in February (Mantone, 2005b) Formation continued to pursue Beverly offering to sw eeten the deal while soliciting support to elect its own directors to Beverly board (Business Wire, 2005b) In response, Beverly decided to offer itself for an auction (Wisenberg, 2005) leading to a frenzied bidding war between two private equity firms: National Senior Care and Formation Capital. National Senior Care offered $1 .9 billion or $12.80 per share for Beverly, a substantial premium on the $11.50 price initially offered by Formation (Loftus, 2005) Formation Capital reacted by raising its offer price to $12.90 but eventually conceded (Wall Street Journal, 2005) However, a few months later, National Senior Care
41 expressed its inability to complete the deal and was replaced by another private equity firm: Fillmore st rategic investors (Mantone, 2005a) The acquisition was completed in March 2006 (Business Wire, 2006a) Beverley continues to be owned by Fillmore Strategic investors. Genesis Health Care Genesis HealthCare was created in 2003 as a result of a spin off from Genesis Health Ventures. However, the Company's history can be traced back to the mid 1980s. Genesis Health Ventures was established in 1985 with nine nursing homes. Like other major nursing home chains of the period, Genesis grew rapidly by adopting the acquisition route (The Wall Street Journal, 1993) and by diversifying into related services such as rehabilitation therapy, diagnostic testing, respiratory therapy, and pharmacy services. It primarily focused on high acuity patients which in the pre PPS era were particularly remunerative. In early nineties it reported strong profits (Hager, 1992, 1993) eventually growing into a $2.5 billion giant by 1995. In 1997 Genesis Health Ventures and two investment funds agreed to pay $1.4 billion for nursing home operator Multicare Corp to create Genesis ElderCare Acquisition Corp in what was perhaps the first LBO seen in the LTC industry (Darby, 1997) stop shop for elder care encompass ing LTC as well as outdoor services (Telegram & Gazette, 1997) The people will pay us to keep them (Peck, 1997) he further declared. Genesis perhaps could not have chosen a worse time to indulge in a major acquisition. With the implementation of PP S, Genesis margins came under severe
42 pressure while it faced cash flow pressure due to its high debt. In 1999 the company reported a loss of over $200 million (Keaveney, 1999) Its downward spiral continued with Genesis defaulting on its loans in 2000 (Brubake, 2000) ; its bond were assigned junk status effectively killing its credit line. Left with no other option, in June 2000, Genesis filed for bankruptcy (Newswire, 2000a) It operated under Chapter 11 protection till October 2001 when it emerged as single entity having integrated Multicare which hitherto was being operated as division of Genesis (Business Wire, 2002) Genesis attempted to improve its financial performance by seeking to acquire NCS Healthcare, a large institutional pharmacy and merge it with Neighborcare, its existing institutional pharmacy arm. The combined entity would have created the second largest institutional pharmacy chain in US (PR Newswire, 2003b) However, it faced strong opposition from another large institutional pharmacy, Omnicare, and after a protracted legal battle, it terminated its agreement with NCS Healthcare (PR Newswire, 2002a) As the nursing home business continued to bleed, Genesis mulled selling its 258 nursing home to private investors (Diskin, 2002) As a first step, in 2003, Genesis divested its ten nursing homes in Florida in favor of Formation Capital (Morse, 2003) As seen in almost all nursing home divestitures in Florida, liability cost was cited as the primary motivation. Continuing its restructuring process, Genesis was split into two parts: the phar macy division Neighborcare was adopted by Genesis; it spun off the elder care business including nursing homes and rehabilitation services into a new entity Genesis HealthCare Corporation (PR Newswire, 2003a) In 2007 the company announced that it had agreed to be acquired by private equity firms Formation Capital and JER Partners for $63 per share for a total transaction
43 value of $1.65 billion (Fernandez, 2007) war with Fillmore investors, the deal was closed at a price of $69.35 a share (Business Wire, 2006b) on July13th 2007. As of 2010, Genesis continues to be owned by Formation Capital. Even this brief h istory clearly establishes that nursing home chains had expanded rapidly riding on easy money from Wall Street and innovative investment options. The crisis in the industry was indeed precipitated by the introduction of PPS, but its origins can be traced t o investment decisions made by nursing home chains with little attention paid to cash flow, debt servicing requirements, and long term profitability. Defining Private E quity Defined as the market of un 4 the private equity market has b ankruptcies in large leveraged deals (Kaplan & Str mberg, 2008) the private equity 4 Unregistered securities i.e securities not tradable on any stock exchange is a necessary but not a sufficient condition for a firm to be labeled private equity owned. Securities of a firm in which the original promoters own 100% of the sto ck would be non tradable on stock exchange but the company would be described as privately held. Whether the said privately held company is private equity owned or not is entirely dependent upon who owns the stocks. In most mature companies, additional fun ds are raised, for instance for funding an acquisition, by approaching the markets via a public offering. However, in some cases, companies may find it hard to raise funds via stock markets due to depressed stock prices or poor performance or even when pro moters may wish to exit the company entirely. In that case, they may divest their entire take in favor of private equity; only then the company can be labeled as private equity owned. In addition, private equity may invest in publicly held mature companie s and take them private by buying their entire stock and delisting it from all stock exchanges. Another avenue for private equity investment is what is known as venture capitalists who invest in start ups. Start ups typically find it difficult to raise fu nds from markets or from financial institutions as they have no proven track record of delivering decent financial performance. Venture capitalists may step in the case as they are attracted by potential high earning even though there is also the downside of frequent start up failures. The dot com bubble of early 2000s was fueled largely by venture capitalists. In recent years, venture capitalists have invested heavily in social media platforms like facebook and twitter.
44 market has grown exponentially in recent years with Morgan Stanley estimating that in activity, 50% of leverage loan volume, 33% of the h igh yield bond market, and 33% of the initial public offerings market (Jensen, 2007) Organization of Private Equity Market In its simplest form, private equity consists of three major players: issuers, private equity funds, and investors. Issuers are company which may be unable to raise public funds (Fenn et al., 1995) and have stable free cash flow to service high amount of debt ( Jensen, 1989) Private equity funds are organized as limited partnerships in which the general partners manage the fund and the limited partners provide most of the capital (Kaplan & Strmberg, 2008) Private equity funds are typically of 10 year duration after which the funds are liquidated and profits distributed among the limited and general partners (Fenn et al., 1995) Finally, there are the investors who are large institutional funds -for instance, pension funds attracted by the higher returns offered by private equity. Lim ited partners are compensated in three ways (Kapla n & Strmberg, 2008) First, they may earn an annual management fees as share of capital invested. Second, the profits earned by private equity funds are divided in 20:80 ratio between the limited and general partners. This constitutes the largest share o f income earned by limited partners and creates a strong incentive to consistently deliver superior financial performance. Third, they may charge deal and monitoring fees on an annual basis. Kaplan (2008) explains the typical private equity transaction as follows. Private equity pays 15 to 50 percent pr emium for public firms over their current stock prices. The purchase is financed heavily through the use of debt with a debt: equity ratio of
45 ranging from 6 to 1 or even 9 to 1. Debt includes a senior or secured debt arranged through a bank, institutional investors or hedge funds. It also includes a junior, subordinate to the secure debt which means that in case of defaults, the former have the first claim on the remaining resourc es of the firm. The remaining funds are the equity portion arranged from the limited partners with a small portion of equity investment coming from the new management team. The key feature of private equity deals is leverage in essence, transfer of risk from equity holders to debt providers which, in turn, increase the return on equity (Blundell Wignall, 2007) Significantly, the private equity investment is not restricted to provisi on of capital rather; it envisions provision of expertise and strategic vision to optimize firm performance ( Financial Services Authority, 2006) According to a study by Stromberg ( 2007) the median period for private equity investment is 6.8 years. In a paper published in 1989, Jensen ( 1989) argued that public corporation, the shareholders and are not listed or traded on organized exchanges. The main argument advanced by Jensen ( 1989) in favor of these new organizational forms was that they had resolved the agency conflict between owners and managers inevitable in the public vis the continued relevance of public corporations, private equity has expanded tremen dously in the last two decades: V alue and number of transactions have increased; secondary
46 buyouts have become increasingly commonplace; and industry and geographical diversity has been noticed (Jensen, 2007; Strmberg, 2007) Table 1 traces the history of private equity inves tment in nursing homes. It establishes that interest in the nursing home industry remains undiminished. In fact, the average size of deals have only increased in recent years While in 2006, only three of the top ten chains were privately held, by 2007, pri vate equity companies had purchased six of the ten largest chain s representing 9 % of total nursing home beds in the United States. Nursing Homes in Florida In 2007, reports the Agency for HealthCare Administration (ACHA) ( 2007) the Medicare, 21 are certified for Medicare but not Medicaid, six accept onl y private pay or majority of nursing homes in Florida are for profit followed by not for profits and a limited number are government owned (Table1 2). Private Equity Inve stments in Florida Nursing Homes In line with national trends Florida witnessed purchase of nursing homes by private equity firms (Table3). However, one of the primary drivers of divestment in the state of Florida has been the burden of liability insuranc e (Stevenson & Grabowski, 2008) In the State operating without liability insurance as most insurers had left the state due to high claims. In 2003, Stevenson and Studdart ( 2003) estimated that litigation costs for nursing homes h ad reached $1.1 billion annually in Florida, while Aon Risk Consultants calculated that the average risk loss per bed in Florida was over four times the national
47 average (Department of Health and Human Services, 2006) In particular, large national chains who were severely affected by tort claims took the initiative in exiting the state; by 2004, the four of the ten largest n ational chains had left the state completely (Department of Health and Human Services, 2006) One of the first national c hains to exit the state of Florida was Beverly Enterprises which operated 49 skilled nursing homes (The New York Times, 2001) facilities in Florida were bought by private equity investor Formation Capital for $165 million or about $27,000 per bed ( Irving Levin Associates 2006) Formation formed Seacrest Health Care Management to run its newly acquired facilities. In 2003, Formation acquired Genesis Florida facilities: ten nursing homes in a deal valued at $26.8 million (Morse, 2003) In August 2003, Mariner Healthcare divested 19 of its Florida facilities to Formation Capital citing rising liability costs (Stevenson & Studdert, 2003) The purchase price was $86.0 million, or a $36,000 p er bed, a substantial increase from what Formation had paid for Beverley ( Irving Levin Associates 2006) In less than two years, Formation was the largest nur sing home operator in Florida. Other smaller chains which exited Florida include Extendicare in September 2000 Kindred Healthcare in May 2003. The Rationale for t his Study Policy makers have a legitimate interest in the evolving nature of the nursing hom e industry. First, as the largest payer of the long term care in the state of Florida, for instance, 80% of nursing home revenues are derived from Medicare and Medicaid (Agency for Healthcare Administration, 2007) the government has a responsibility to ensure that nursing homes utilize its funding in a proper manner (Catherine & Phillips, 1986) In addition, ensuring the financial viability of the nursing home industry to the
48 extent they remain solvent is important: The government in the past has revised the l aw and incurred additional costs to save the industry from financial ruin ( Department of Health and Human Services 2000) Considering the expected future demand for long term care, nursing homes will continue to play an important role in the American health care industry. Second the government bears significant responsibility for ensuring that nursing homes deliver appropriate quality of care to their residents. Nursing home residents are physically and psychosocially compromised and may be entirely dependent upon their caregive rs (Agency for Healthcare Administration, 2007) According to a report b y the Kaiser Family Foundation, more than two thirds of elderly nu rsing home residents have multiple chronic conditions, 6 in 10 have multiple mental/cognitive diagnoses, and more (Wiener et al., 2007) As nursing home quality remains a matter of concern (Government Accountability Office, 2000a, 2003; Wunderlich & Kohler, 2001) analyzing the f actors which may potentially impact quality of care ownership for instance -is essential to protect nursing home residents. Third, recognizing the pressing public interest, the nursing home industry is heavily regulated by the government (Agency for Healthcare Administration, 2007) In 2001, $382 million was spent on government agencies regulating nursing homes; this figure does not include the significant costs incurred by nursing homes in complying with regulatory demands (Kumar et al., 2006) In order to ensure that the regulatory structure is not rendered obsolete by the evolving nature of the nur sing home industry, the policy makers require independent research which tracks the changes manifest in the industry:
49 in this case, the impact of a new form of ownership on quality and financial performance. The choice of Florida is in part dictated by lim itations of data availability. But it also offers one distinct advantage. In the state of Florida, nursing home divestment in favor of private equity been driven largely by high malpractice insurance. Private equity has attempted to limit the threat of lar ge malpractice claims by separating real estate and licensure, and by spinning off separate ventures for each nursing home (Duhigg, 2007) In addition, 339 of total 645 nursing homes certified to admit Medicare and Medicaid in Florida are Limited Liability Corporations (Agency for Healthcare Administration, 2007) Specifically ci ting the example of Florida, Harrington has argued that LLCs limit the ability of patients to seek redress for poor quality of care by limiting the liability of individual owners (Harrington, 2007) This study may provide some indirect evidence if tive to provide adequate quality of care. Considering the concern raised in policy circles due to private equity investments; its potential impact on both quality and financial performance; and the extremely limited and often conflicting independent resea rch available currently, (Agency for Healthcare Administration, 2007; Duhigg, 2007; Harrington, 2007; LTCQ Incorporation, 2008; Service Employees International Union, 2007; Stevenson & Grabowski, 2008) the results of this study can help policy makers better design the regulatory environment to ensure better outcomes for all the stakeholders: providers, consumers, and the government.
50 Figure 1 1 Long term continuum of care National Ce nter for Assisted Living (20 01). The Assisted Living Sourcebook Retrieved 10/12/2010 from http://www.ahcancal.org/ncal/resources/Documents/alsourcebook2001.pdf Figure 1 2 Spending for nursing home care Wiener, J. M., Fr eiman, M. P., & Brown, D. (2007). Nursing Home Care Quality: Twenty Years after the Omnibus Budget Reconciliation Act of 1987 : Kaiser Family Foundation
51 Figure 1 3. Percent nursing homes by chains 1993 2004 / Stevenson, D. G., & Grabowski, D. C. (2007). N ursing Home Divestiture and Corporate Restructuring Washington. D.C Department of Health and Human Services 44% 46% 48% 50% 52% 54% 56% 1993 1995 1997 1999 2001 2003 Series1
52 T able 1 1. Private equity investment in nursing homes Nursing Home Private investment group Year Number of fa cilities Targeted facilities type transaction Extendicare Greystone Tribeca Acquisition LLC 2000 10 Beverly Enterprises Formation Capital Properties 2002 4 9 Genesis Healthcare Formation Capital Properties 2003 8 Mariner Health Care Formation Ca pital Properties 2003 20 Entire chain transaction Centennial Healthcare Warburg Pincus 2000 9 9 Integrated Health Services Abe Briarwood 2003 241 Kindred Healthcare Senior Health Management LLC 2003 18 Mariner Health Care National S enior Care 2004 274 Meridian Formation Capital Properties 2005 7 Skilled Healthcare Group Onex Partners 2005 80 Beverly Enterprises Pearl Senior Care 2006 369 Laurel Healthcare Formation Capital Properties 2006 32 Tandem Health Care Formati on; JER Partners 2006 80 Formation Capital Affiliates General Electric 186 2006 Genesis Healthcare Formation; JER Partners 2007 220 HCR ManorCare Carlyle Group 2007 500 Haven Omega, CapitalSource Finance Nationwide Health Properties 21 2008
53 Table 1 2 Florida nursing homes by organizational types Licensee Owner Type For Profit Not for Profit Total Corporations 162 130 292 Limited Liability Companies 288 51 339 Limited Partnerships 13 0 13 Limited Liability Partnerships 9 1 10 General Partnership 3 0 3 Trusts 2 2 4 Government Owned 0 11 11 Table 1 3. Maj or private equity nursing homes acquisition s in Florida Nursing Home Private investment group Year Effective Number of units Extendicare Greystone Tribeca Acquisition LLC 2001 16 Beverely Formation 2002 49 Genesis Formation 2003 10 Mariner Formation 2003 20 Kindred Healthcare Senior Health Management LLC 2003 18 Manor Carlyle 2007 29
54 CHAPTER 2 CONCEPTU AL FRAMEWORK In this section, literature on impact of ownership and nursing home performance is reviewed. It is established that both quality and financial performance of nursing homes are influenced by ownership. Therefore, we argue that introduction of a novel ownership form private equity would influence nursing home performance, specifically, quality and financial performance. Subsequently, a conceptual framework is developed arguing that private equity nursing homes would exhibit better financial perf ormance but worse quality of care than other investor owned nursing homes Influence of Ownership on Quality and Financial Performance Three different types of ownership structures exist in the nursing home industry: private for profit (FP), private not fo r profit (NFP), and government. In 2006, 66% of nursing homes in the United States were for profit, compared to 28% not for profit and 6% government owned (Harrington, 2007) Unlike the hospital sector, the nursing home industry has historically been dominated by proprietary chains (Bradford, 1986) FPs are corporations owned by investors who are rewarded by their share of the profits generated. They are motivated by a desire to maximize share holder value and are generally considered to be more efficient and growth oriented (Bradford, 1986) NFPs are governed by section 501(1)(c) of the Internal Revenue Service (IRS) code and are subject to a non ibuting net income as dividends or above broadly defined collective purposes. In exchange for meeting these requirements, NFPs receive certain tax advantages (DiMaggio & Anheier, 1990) and may have access to non operatio nal revenues in the form of philanthropic donations.
55 The role of FPs and NFPs in healthcare sector is underpinned by differing economic philosophies. According to the competitive market model, a marketplace consists of producers with different agendas and scarce resources (Fulp, 2009) As long as the conditions of the competitive market are met viz. multiple buyers/sellers, low entry/ex it barriers, perfect information, costless transactions, and homogenous products (Bradford, 1986) prices are aligned until supply matches demand resulting in an allocation which is Pareto opti mal (Fulp, 2009) In this model, government production of services may be required mainly in case of public goods or where imperatives of social justice t rump allocative efficiencies of the market. key feature of the health care market is the presenc e of asymmetrical information between the providers and the patients. (Arrow, 2001) Hansmaan ( 1980) propounded the Trust Hypothesis which attributes the presence of NFPs in certain sectors of the economy -for instance, health care -to asy mmetrical information Accordingly, NFPs would be present in markets where consumers find it difficult to monitor quality and prefer NFPs as the non distribution constraint ostensibly circumscribes their profit motive. This paradigm essentially argues that NFPs may have less incentive to sacrifice quality in pursuit of profits as their actions are not constrained by the need for profit maximization. performance. Because of the diff ering economic and philosophical rationale, ownership may impact how nursing homes react to cost containment, market pressures and the
56 evolving regulatory environment which, may, in turn, influence their quality and financial performance. As nursing home residents are frail and may be too cognitively impaired to adequately advocate for themselves managerial behavior --dictated by type of ownership --would be particularly important factor in nursing homes (Hillmer et al., 2005) Ownership and Financial Performance Multiple studies have examined the link between ownership and financial performance in the health care industry. Primarily, the literature is focused on the hosp ital sector. An additional challenge is that different metrics of financial performance have been used by scholars. Due to the significant changes introduced in the hospital reimbursement model with the implementation of hospital PPS in 1983, this review i s confined to hospital financial p erformance in the post PPS era. Fiogone et al. ( 1996) report tha t despite their smaller average size, FP hospitals report better financial performance. In a study comparing the financial performance of FP and NFP hospitals in the state of Virginia, Shukla and Clement ( 1997) conclude that FPs have higher ope rating margins as well as Return on Assets (ROA), they attribute ( 2001) conclude that one of the determinants of the hospital financial performance was ownership; in their national sample, FPs exhibited superior financial performance to NFPs. In another study conducted on a national data set of hospitals in 2005, Younis et al. ( 2005) reported that FPs perform better than NFPs both on Return on Equity (ROE) as well as Total Profit Margin (TPM). In the state of Florida, Sear ( 199 1) conducted a study comparing hospital profitability between 50FP and 50NFP hospitals between the years 1982 and 1988 and
57 reported that FPs deliver better financial performance. However, their data included periods when PPS implementation was in process in the state of Florida (Sear, 1991) Using ROA as a measure of financial performance, Younis et al. ( 2003) reported that in Florida, FP hospitals exhibit superior economic performance compared to NFP hospitals. To summarize, despite some variations, research ha s de monstrated that FPs generally deliver better financial performance than NFPs (Baker et al., 2000) Some scholars have attributed their more ef ficient performance to greater market discipline imposed on investor owned firms (Jensen & Ruback, 1983) while others have attributed it to property rights (Clarkson, 1972) Relatively few studies have examined the effect of ownership stat us on financial performance in the nursing home industry -Weech Maldonado et al. ( 2003a) in studying the link between quality and financial performance found that FP nursing homes experienced higher operating margi ns compared to NFPs. No study however has examined financial performance of private equity owned facilities in the health care sector Therefore, the evidence from other industries is examined. P rivate Equity and Financial P erformance In a leveraged buyo investment firm using a relatively small portion of equity and a relatively large portion of (Kaplan & Strmberg, 2008) Multiple studies have examined the post facto financial performance of LBOs. In a study comparing perf ormance of 25 sample firms 2 years before and after LBOs (all firms in this sample had managers acquiring certain portion of equity), Bull ( 1989) reports that financial results improved managers. Opler ( 1992) utilizing data from 1985 1989 reported
58 that in case of 44 large LBOs, operating profit imp roved significantly 2 years after the buyouts. Utilizing the extensive Census Bureau Quarterly Finance Report (QFR) database, Long and Ravenscraft ( 1993) significant 15 percent increase in industry However, these gains were restricted to the first three years and there was a significant drop in the performance in the fourth and the fifth year (Long & Ravenscraft, 1993) Equity, Return on Investment, Margin Ratios e.t.c.), Desbrieares and Schatt ( 2002) found that acquired firms had significantly higher Return on Investment (ROI) than the industry average though they found no statistically significant difference in case of ROE. All private equity nursing home purchases in Florida were LBOs. Management Buyouts Management Buy Outs (MBOs) 5 managers acquire the firm. MBOs are frequently financed by private equity firms as banks find such acquisitions too risky. In his study of MBOs, Kaplan ( 1989) studied 76 large public firms which experienced MBOs. He reports that post buy out, the operating income increased by 24% in third year and was significantl y higher than industry average (Kaplan, 1989) Smith ( 1990) investigated 58 MBOs of public firms and concludes that o perating returns improved significantly after buy outs and was sustained subsequently. She further reported that improvement in returns was not due to layoffs or reduction in expenditure on advertising or research and development (R&D) 5 MBO is a generic term which simpl y suggests a transaction in which management of an existing firm buys the company from its owners. Since managers frequently lack the wherewithal to finance a purchase outright, they rely on outside financers. The source of funding may include FDIC insure d banks, while in other cases MBOs may be purchased with funds provided by private equity. Therefore, private equity MBOs are a subset of larger MBO market entirely predicated upon source of funding.
59 and may reflect chan ges in financial incentives for owner managers subsequent to change in ownership structure (J.A. Smith, 1990) Lichtenberg and Siegel ( 1990) repo rt that MBOs improve productivity for at least three years post management buyout. In summary, the literature provides substantial evidence that o perating performance of firms improved subsequent to private equity investments (Nikoskelainen & Wright, 2007) Ownership and Quality The relationship between ownership and quality of care in the nursing home industry has received much scholarly attention. Davis ( 1993) showed that FPs had nonpr oprietary nursing homes 1987 National Medical Expenditure Survey (NMES) database, Specter ( 1998) reports that patients with serious health conditions may prefer NFPs. Specter ( 1998) argues that in presence of asymmetrical informati on patie nts prefer NFPs as they are thought to place greater emphasis on quality due to attenuation of property right s ; an argument which has been empirically supported by Chou ( 2002) Additionally, in terms of patient outcomes as measured by mortality, infections, bedsores, hospitalizations, and functional disabilities, patients in NFPs perform significantly better (Spector et al., 1998) Harrington et al. (2002) in their national level study classify deficiencies in three categories: quality of care, quality of life, and other. They report that FPs had higher number of deficiencies across all categories. Finally, Hillmer et al. ( 2005) review the evidence concerning ownership status and quality of care in North understood in terms of Donebedian Structure (staffing) Process (Use of restraints,
60 catheters) Outcomes (development of pressure ul cers, frequency of falls) (SPO) framework. They report that in their sample of 38 studies with a total of 81 results, only 6 demonstrated that quality of care in NFPs was worse while 33 results indicate that quality was poorer in FPs (Hillmer et al., 2005) Hilmer study is particularly relevant as SPO framework is utilized in this study as well. Their findings are broadly in agreements with Rosenau and Linder ( 2003) who conducted a systematic review of performance difference between FP and NFP health care providers and report that in 59% of the 12% favored FPs. Quality in their study is multidimensional including lower adverse event rates, lower mortality rates a full continuum of health, higher Health Plan Employer Data and Information Set (HEDIS) score e.t.c (Rosenau & Linder, 2003) In summary, evidence strongly suggests that NFPs deliver better quality of care than FPs. Quality of Care and Private Equity The New York Tim es (NYT) analyzed data from Online Survey and Certificatio n and Reporting (OSCAR) and Minimum Data Set (MDS) and reported that the average private equity acquired nursing home fared worst than national average in 12 of the 14 quality indicators including b edsores, preventable infections, and use of restraints (Duhigg, 2007) However, the methodology adopted by NYT has been questioned and its findings disputed in a report prepared by LTCQ 6 ( 2007) Another report released by Service Employees International Union (SEIU) ( 2007) examined deficiencies reported at Mariner and Beverly, two large nursing home chains acquired by private equity firms. 6 LTCQ is now known as PointRight Inc
61 In both these chains, reports SEIU ( 2007) that the number of violations as well as their seriousness has increased substantially since the acquisition. Stevenson and Grabowski ( 2008) found no deterioration of quality of care i n the form of survey deficiencies or resident outcomes in private equity nursing homes. However, they do acknowledge that their data is new and that private equity deals raise (Stevenson & Grabowsk i, 2008) In an unpublished study, Harrington and Carrillo ( 2007) conducted an analysis of 105 nursing homes in California which were bought by private equity firms. They report that total nurse staffing decreased along with particularly sharp drop in the more being replaced by less expensive nursing assistants to save costs (Harrington & Carrillo, 2007) At the same time, there was an increase in deficiencies with a particularly noteworthy increase in more harmful deficiencies in these nursing homes. (AHCA) ( 2007) nursing home care suffers when a facility is owned by a privat e equity firm or an structures made understanding ownership patterns of private equity owned nursing ess these concerns (Agency for Healthcare Administration, 2007) Conceptual Framework Based on the literature review, private equity owned nursing homes would be expected to exhibit improved financial performance post acquisition. In this section,
62 utilizing Agency Theory and the Free Cash Flow Theory (FCFT) a conceptual framework is developed to support our conjecture. Agency Theory The s urvival of public corporation has been questioned since the days of Adam Smith who argued that the "joint stock company" -an analogue to the modern public corporation would find survival challenging in a competitive market because of waste and inefficiency (Daily, Dalton, & Jr, 2003) Separation of ownership and control, argued Berle and Means in 193 and of ultimate manager may, and often do, diverge, and where many of the checks (Demsetz, 1983) In essence, this paradigm assumes that interests of owners and managers were necessarily divergent and often contradictory with both acting in t heir rational self interest. In their classical 1976 paper, Jensen & Meckling ( 1976) defined agency principal engages an agent to perform a service on interest of the owner (Jensen & Meckling, 1976) The principal attempts to minimize the by either offering incentives or by incurring additional monitoring costs and bonding costs: agency costs (Eisenhardt, 1989) The principal contribution of agency theory is its elaboration of governance mechanisms to minimize the agency problem. O ne mechanism is the adoption of an outcome based approach which attempts to co align the interests of the agents with those of the principal by linking their rewards to firm performance; this, in turn, is thought to reduce the conflict of interest between the principal and the agent. Jensen
63 and Meckling ( 1976) agents develop a vested corporations may differ significantly in their ability to address agency problems. that private companies can resol corporations the conflict between owners and managers over the control and use of (Jensen, 1989) In the modern corporation, the wide dispersion of (Demsetz, 1983) Or as Berle and Means questioned: Under wide dispersion of ownership, [was there] "any justification for assuming that those in control of a modern corporation will also choose to oper (Williamson, 1984) Manager Accountability Two important mechanisms for ensuring that managers serve shareholder interests are the presence of large institutional i nvestors and board of directors While institutional investors can provide oversight by functioning as active investors, evidence suggests that their intensity of monitoring may be diminished by high private cost of monitoring (Jensen, 1989) or they may face hurdles in the form of fiduciary duties and business relations with the firm (Almazan, Hartzell, & Starks, 2005) Simil arly, legal constraints have hampered the ability of capital markets to control the management (Jensen, 1989) In public firms, the board of d irectors is an important part of corporate governance. With powers to hire, fire and compensate senior management (Baysinger & Butler, 1985)
64 (Kosnik, 1987) However, the leadership provided by board of directors has been questioned in the literature. Lorsch ( 1990) have attributed the ir ineffective performance to power gap between t he management and the directors Lorsch ( 1990) further argues that directors frequently share an incestuous relationship w ith the management they may be beholden to the management for their appointments and perks. Researchers have documented the failure of boards to properly monitor the executive management (Ingley & Walt, 2001; Sundaramurthy & Lewis, 2003) while others have argued that firms are simply a reflection of their top management (Hambrick & Mason, 1984) and they determine the strategic direction firms adopt with a limited role for board of directors. The lack of effective monitoring by either the board of directors or institutional investors vests managers w between owners and managers (Jensen, 1989) thereby exacerbating the agency probl em. Private equity placements are better positioned to resolve the agency problem and limit managerial opportunism by generating stronger incentives for the current management (Cuny & Talmor, 2007) Incentive mechanisms allow private equity to promote managerial compliance with their chosen strategies including profit maximization. A report by Financial Services Authority (FSA) ( 2006) lists some of the mechanisms adopted by private equity to control managers Managerial stock ownership : Unlike most public corporations, senior managers in private companies (Jensen, 1989) Ownership forces managers to operate the firm more efficiently as their personal
65 wealth is at risk (Fox & Marcus, 1992) and management equity is a significant predictor of LBO returns in successful buyouts (Nikoskelainen & Wright, 2007) Renneboog et al. ( 2007) private transactions total compensation are predicated on meeting performance objectives. Type of stock : Convertible preferred stock is issued to private equity firms; it is superior to common stock typically held by the management as it has the first call on resources in case of liquidation. Management employment contracts : To avoid excessive short term risk taking term perform ance. Board representation : As block shareholders, equity firms dominate company boards and are able to exercise stronger oversight (Denis, Denis, & Sarin, 1997) Private equity has the resources, staff, expertise, and most importantly sufficient incentive to constantly monitor firm performance. In contrast, board of directors in public corporations may be dominated by nominees of the professional management (Jensen, 1989) Control of access to future funds : Capital funds may be provided in stages dependent upon meeting past performance goals. If, for example, private equity has been accepted to fund future expansion, control on capital provides a sig nificant leverage to investors. Free Cash Flow Theory and the Role of D ebt At the heart of the owner manager conflict is the free cash flow flow in excess of that required to fund all investment projects with positive net present
66 values when discounted at the relevant cost of capital (Jensen, 1989) The Free Cash Flow Theory (FCFT) argues that managers should "disgorge" the cash rather than "invest it at below the cost of capital" or "waste it through organizational inefficiencies" (Jensen, 1986) Managers have an incentive to limit payouts to shareowners as the a mount of resources under their command is directly linked to their power (Jensen, 1986) As firm size is linked to manageria l remuneration and benefits (Kostiuk, 1990; Murphy, 1985) as well as prestige, managers may prefer to invest the free cash flow in inefficient mergers and acquisitions creating what Jens en has termed the agency cost of free cash flow (Jensen, 1986) FCFT predicts that the high amount of debt created in private equity transactions improves organizational performance because of the control function of debt (Bull, 1989) Debt has a salutatory effect on managerial discretion as managers are legally bound to make the interest and principal payouts limiting the wastage of free cash flow (Jensen, 1986) 7 The threat of defaulting on debt forces the management to concentrate on efficiency as the cost of failure is markedly higher than in cases of missed dividend s. Strong support for FCFT has been demonstrated by Lehn and Poulsen ( 1989) Additionally, extensive use of debt to finance buyouts limits the dispersion of equity; managers and private equity investors control the majority of stock. Concentrated ownership allows private equity investors to extensively monitor fir m strategy and performance by maintaining an active presence on the board of directors (Nikoskelainen 7 While in theo ry a similar role is played by dividends, unlike debt, dividends are not a legal requirement. While market discipline may force managers to issue dividends, managers still retain a large discretion in deciding its quantum as well as frequency.
67 & Wright, 2007) In public corporations, active investors are discou raged by insider trading and other regulations (Jensen, 2007) Extremely high level of debts may lead to bankruptcies to failure to service the 17,171 private equity sponsored buyout t ransactions occurred from January 1, 1970, to June 30, 2007. Kaplan and Strmberg ( 2008) show that 6% of deals have ended in b ankruptcy or reorganization; for all U.S. corporate bond issuers from 1980 2002. Other Anticipated Benefits of P rivat e Equity O wnership Private companies face less regulatory oversight, for example, a private equity security is exempt from registration with the Securities and Exchange Commission ( SEC ) (Fenn., Liang., & Prowse., 2005) With increased regulatory costs imposed su bsequent to the enactment of Sa rbanes Oxley Act, diminished regulatory requirements may resu lt in substantial cost savings (Strmberg, 2007) Similarly, private companies can take a more long term approach as their policies are no longer shaped b to outlook of publicly traded firms (Blundell Wignall, 2007; Robbins, Rudsenske, & Vaughan, 2008) Measures like compensation packages can be designed without Congressional or public scrutiny which helps private companies concentrate on achie ving their strategic objectives (Bull, 1989) Therefore, based on the literature review and the conceptual framework outlined above, we hypothesize, Hypothesiss1 : Private equity owned nursing homes will achieve better financial performance compared to other investor owned nursing homes in the state of Florida.
68 Private Equity Ownership and Qual ity of Care In a study in California nursing homes, O'Neill et al. ( 2003) report that among proprietary nursing homes, profits located within the highest 14% bracket were associated with significantly higher number of total as well as serious deficiencies profit (O'Neill et al., 20 03) Significantly, this relationship was not maintained for non proprietary nursing homes. A report in the Financial Times ge annual net return of 12.8 percent from their private equity investments. They also expect (Phalippou & Gottschalg, 2009) As the report suggests, private equity consistently operate under high pressure from their investors (l imited partners) for high returns; profit maximization is their raison d'tre. Jensen has argued that reputational aspect of Necessity to raise new funds makes mediocre returns a disaster (Jensen, 2007) while Ka plan points out that well performing partners are better positioned to raise future funds (Kaplan & Schoar, 2005) In order to thrive, private equity firms must consistently deliver high profits. While no literature exists on financial performance of private equity owned nursing homes, indirect evidence sugges ts that such nursing homes outperform the competition: according to the New York Times analysis, the typical investment owned nursing home earned $1700 a resident in 2005 and was 41% more profitable than the average facility (Duhigg, 2007) On similar lines, private equity firm Formation Capital sold 186 of its recently acquired nursing homes in 2006 to General Electric for $1.5 billion at a profit of
69 $ 400 million in merely 4 years (Duhigg, 2007) yielding an estimated fivefold return on equity (Investor, 2006) retical framework for identifying quality measures in ( 1988) Structure Proce ss Outcome (SPO) mod el (Figure 2 1 ). According to this framework, structural indicators are defined as the staffing patterns and organizational resources that can be associated with providing care, such as facility operating capacities (Binns, 1991) Process indicators refer to actions that are performed on or done to patients, such as medical procedures (Gustafson & Hundt, 1995) Outcome indicators are the states that result from care processes, such as increases in quality of life as well as decreased mortality rates (R. A. Kane & Kane, 1988) Good structures increase the likelihood of good processes, and good processes increase the likelihood of good outcomes. Good structure can also directly improve outcomes (Dellefield, 2000) While it is frequently possible that there is no direct interrelationship of structure and process, as well as individual patient characteristics, dictates the final outcome (Hillmer et al., 2005 p. 141 ) Weech Maldonado et al. (2004) ha ve employed the SPO framework to study the impact of nurse staffing pattern on quality of care in nursing homes A similar approach has been taken by Hilmer et al. ( 2005) in their critical review of the association between ownership status and quality of care in North American nursing homes. Nurse staffing is an important structural indicator of quality (R. L. Kane, 1998) ( 2002)
70 that nurses for recognition of adverse medica l outcomes and errors. Extant literature suggests that staffing levels as well as staffing mix has a significant influence on quality of care in nursing homes (Bowers, Esmond, & Jacobson, 2000; Cherry, 1991; Harring ton, Kovner et al., 2000; Johnson Pawlson & Infeld, 1996; Konetzka, Norton, Sloane, Kilpatrick, & Stearns, 2006; Munroe, 1990) In their systematic review of role of nurse staffing in nursing homes, Bostick et al. ( 2006 p. 366 ) There is a proven association between higher total staffing levels (especially licensed staff) and improve On similar lines, increased proportion of RN nursing is associated with better quality outcomes inc lude decreased likelihood of patient mortality and failure to rescue in hospitals (Aiken, Clarke, Sloane, Sochalski, & Silber, 2002; Estabrooks, Midodzi, Cummings, Ricker, & Giovannetti, 2005; Tourangeau et al., 200 7) Weech Maldonado et al. ( 2004) show that quality of care in nursing homes as measured by use of physical restraints, use of antipsychotic drugs, incidence or worsening of pressure ulcers, cognitive decline, and mood decline improves with higher RN staffing. Similarly, In longitudinal study of RN staffing and quality of care in nursing homes in California, Kim et al. ( 2009) show that higher level of RN staffing was associated with lower number of deficiencies in nursing homes. Anderson et al. ( 1998) in their study of 494 nursing homes conclude that RN Controlling nurse staffing costs is a mechanism nursing homes may adopt to improve profitability. Nursing homes may achieve nursing cost saving by either decreasing total nurse staffing or by substituting the more expensive Registered Nurses (RNs) for th e less expensive Licensed Practioner Nurses (LPN ) or Ce rtified Nursing
71 Assistants (CNA ). In their 2007 study on private equity nursing homes in California, Harrington and Carrillo ( 2007) found significant decrease in total nurse staffing as well as substitution of RNs by LPNs and CNAs. The need to cut costs to improve profitability may be particularly imperative in Florida nursing homes due to reimbursement issues. Medicaid, the principal payer for nursing home care, is responsible for 61% of total nursing home revenues in Florida (Agency for Healthcare Administration, 2007) Nationwide, Medicaid rates are lower than Medicare or private pay with a recent report projecting average shortfall in Medicaid nursing home reimb ursement at $12.48 per Medicaid patient day in 2008 (Eljay, 2008; LLC Eljay, 2008) Similarly, in Florida, a report by AHCA ( 2005) shows that Medicai d costs of majority of Florida nursing homes was not covered by Medicaid rates. Analysis of historical data suggests that this gap has widened sharply over time: their costs within their final per diem rate. By July 2004, only 4.67% of provide rs received 100% of their costs (Agency for Healthcare Administration, 2005) In addition, an incentive available to nursing hom e providers whose quality of care met certain standards was eliminated in 1996 to be replaced by Medicaid Adjusted Rate (MAR) (Agency for Healthcare Administration, 2005) Because it is linked to Me dicaid utilization, MAR may off If nursing homes increase Medicaid utilization in order to benefit from MAR, it limits their ability to tap lucrative private pay or Medicare patients; if, on the other hand, they keep Medicaid utilization low then the gap between costs and Medicaid reimbursement widens
72 Therefore, the nursing homes providers in Florida have limited options for increasing revenues to improve profitability leaving cost cutting as an essential profit increasing str ategy. Because proprietary nursing homes in Florida are likely to have a greater Medicaid census, (Johnson, Dobalian, Burkhard, Hedgecock, & Harman, 2004) their options are particularly limited. Therefore, based on the literature review and the conceptual framework outlined above, we hypothesize, Hypothesis 2 : Private equity nursing homes will experience poorer quality of care compared t o other investor owned nursing homes in the state of Florida.
73 Figure 2 1 Donabedian, A. (1988). The quality of care. How can it be assessed? JAMA, 260 (12), 1743 1748.
74 CHAPTER 3 METHODS Employing multiple longitud inal datasets, this study uses panel data regression to understand the effect of private equity ownership on nursing home quality and financial performance. In the following paragraphs, information about datasets is provided, followed by list of dependent and independent variables. The model is explained followed by the plan of analysis. Datasets This study combines the Online Survey Certification of Automated Records Term Care Focus dataset, Med icare Cost Reports, and Area Resource File (ARF). Online Survey Certification of Automated Records (OSCAR): It is a data network maintained by the Center s for Medicare & Medicaid Services (CMS) in cooperation with state monitoring agencies. The data is co llected as part of certification process for Medicaid and Medicare. Every nursing home is inspected at least once every fifteen months; a nursing home may be inspected earlier in case of complaints. s furnished by the operator, information on standard health and life deficiencies issues are reported by surveyors which are then compiled into a national database. Minimum Data Set (MDS): The Nursing Home R eform Act, part of the Omnibus Budget Reconcilia tion Act of 1987 (OBRA) of 1987 was the first major federal legislation directed towards improving the quality of care in nursing homes (Kumar et al., 2006) The Act included the Resident Assessment Instrument (RAI ); the foundation of MDS (Wiener et al., 2007) Each Medicare or Medicaid certified nursing home is required by
75 information, health status as well a number of services that a resident may rec eive while they reside at the nursing home. Each resident is assessed when first admitted to a nursing home, then annually with documentation each quarter of any change in status (Mor et al., 2003) Phillips et al. ( 1997) document that implementation findings based upon MDS admini strative data generally yield similar results to data obtained by research nurses trained to provide quality MDS measurement. MDS remains the most comprehensive data set available for examining quality of care in nursing homes and is primarily used to ide ntify resident needs for care (Ryan, Stone, & Raynor, 2004) MDS data was acquired using a reuse agreement from the CMS. Data is located behind a fully secure server in the Health Professions and Public Health building operated by the IT department at the University of Florida. It was originally acquired by the University of South Flori da. Long Term Care Focus dataset (LTC Focus dataset): This dataset was created status of nursing home residents, characteristics of care facilities, and state policies rel evant to long t erm care services and financing (Long Term Care Focus, 2010) The dataset has been built using Medicare enrollm ent dataset; Medicare claims data, and MDS. Both f acility and county level datasets for the state of Florida (2000 2007) were downloaded from the Long Term Care (LTC) Focus website. To maintain consistency, quality variables were constructed using the LTC Focus dataset to the extent possible. Medicare Cost Reports : A public access dataset, it was acquired through the CMS website. All CMS certified nursing homes accepting Medicare beds are required to
76 submit their cost reports annually. The Medicare Paymen t and Advisory Commission (MedPac) has used the Medicare cost reports in its reports to the Congress. The Medicare Cost Reports was used to calculate the financial performance variables. Area Resource Files ( ARF ): It contains data on socioeconomic and dem ographic characteristics of markets where nursing homes are located collected by more than fifty sources including (CMS) and United States Census Bureau. It includes over 7000 variables including geographic codes and classifications; health professions su pply and detailed demographics; health facility numbers and types. ARF is released ann ually and historical data is available. Merging D atasets As a first step, yearly datasets consisting of four of the datasets viz. OSCAR, MDS, LTC Focus dataset, and Medic are cost reports was created. The Medicare cost reports proved particularly challenging as many observations had multiple cost reports. This discrepancy may be attributed to updated cost reports filed by a facility in a particular financial year or new co st reports filed in the case of change in ownership. exported in a separate dataset. Arithmetic mean was calculated and for each observation, a single Medicare cost report was created. These observations were merged back with the yearly datasets. Subsequently, all the yearly datasets were merged to create a master dataset pertaining to the years of study: 2000 2007. This was subsequently merged with market variables derived from ARF 2008 and LTC Focus county level dataset. Identifying private equity nursing homes is challenging due to complex organizational structures erected by private equity to limit liability claims. As a first step,
77 a search was carried out in Lexus Nexus us chains, prevailing market conditions, and the deta ils of private equity nursing home acquisition: Number of facilities involved, year of acquisitions, and cost. This the Security & Exchange Commission (SEC) ( 2010) Edgar is a useful tool as all reports, and oth ( 2010) and this information is publicly available. To identify individual nursing homes acquire d by private equity firms, OSCAR data was supplemented with by accessing websites of private equity nursing home chains. Many such websites have a detailed list of individual facilities owned by that particular nursing home chain which can be matched with the OSCAR dataset. Since many of these websites were no longer online, WayBack machine was used to access them (WayBack Machine, 2 010) WayBack machine is an internet tool which Finally, a unique dataset with a list of Florida nursing homes acquired by private equity was provided by the University of South Florida (USF) and was used to identify private equity nursing homes in the master dataset. The use of multiple sources, and in particular, the USF dataset provides reasonable confidence that private equity nursing homes were correctly identified i n our dataset, and is one of the unique strengths of this study.
78 Population This s tudy includes all for profit, chain, free standing nursing homes in the state of Florida included in the OSCAR, MDS, Long Term Focus datasets, and which filed Medicare cost reports between the years 2000 2007. The initial population comprised of over 6000 observations with an average of approximately 760 nursing homes for each year of the study period. Of those 760 nursing homes, approximately 200 are NFPs, 300 non chain bas ed while 40 are hospital based. 8 For reasons discussed below, all these observations are eliminated resulting in the removal of over 3200 observations --NFP (1500), non chain (2300), and hospital based facilities (300). The final sample has 2822 observat ions spread over the 8 year study period. There are several reasons for focusing on for profit, chain, and free standing facilities. The underlying premise is to ensure that nursing homes included in the study are similar to each other in their organizati onal structure, profit motive, and response to regulatory, environmental, and market conditions. Hospital based facilities differ significantly from free standing facilities (Weech Maldonado, Neff, & Mor, 2003b) as they have to frequently manage sub acute patients from their own hospitals. On similar lines, chain based facilities may differ significantly from their free standing counterparts. Chain affiliation has been promoted as a mechanism that can improve financial performance by increasing access to capital, introducing economies of scale (Harrington, 2007) greater labor and management resources larger patient base, m ore referrals from within the system, market power, brand identity, increased clout with both regulatory agencies and third party payers, community effectiveness (i.e., greater 8 These cat egories are not mutually exclusive, and some nursing homes may be classified in multiple groups --for instance, a NFP facility may be non chain or hospital based.
79 service scope and quality), heightened social status and visibility, legitimacy in the institutional environment, and enhanced information systems and technologies (Banaszak Holl. et al., 2002; Bazzoli & Chan, 2000; Tennyson & Fottler, 2000) NFPs are governed by section 501(1) (c) of the Int ernal Revenue Service code and are subject to a non distribution constrain t which proscribes distributing net income as dividend s or above they must serve one of several broadly defined collective purposes In exchange f or meeting these requirements, NFPs r eceive certain tax advantages (DiMaggio & Anheier, 1990) Naturally, the FPs and NFPs may react in different ways to cost containment, market pressures and the evolving regulatory environment. Finally, some nursing homes included in OSCAR or MDS m ay not file Medicare cost reports and were omitted from the study. Quality Variables In this study, nu rsing home quality measures encompass all three dimensions of (Donabedian, 1988) SPO framework for quality assessment. In the SPO framework, s tructur e refers to the professional and organizational resources that can be associated with providing care ; p rocess re fer s to actions that are performed on or done to patients ; and o utcome s are the states that result from care processes (R. A. Kane & Kane, 1988) Good structures increase the likelihood of good processes, and good processes increase the likelihood of good outcomes. Structural measures of quality Nurse staffing is the most frequent measure of structure in nursing home literature (Hillmer et al., 2005) A large number of studies in the nursing home literature ha ve established that staffing intensity as well as staffing mix has a significant influence on quality of care (Bowers et al., 2000; Cherry, 1991; Harrington, Kovner et al., 2000;
80 Johnson Pawlson & Infeld, 1996; Kone tzka, Norton, Sloane et al., 2006; Munroe, 1990) Existing academic literature on private equity nursing homes has noted significant changes in nurse staffing patterns and expressed concern over its impact on quality (Harrington, 2007; Stevenson & Grabowski, 2008) Finally, nurse staffing is one of the most significant costs in nursing home operations. A nursing home striving for better profitability would ordinarily be expected to look towards nurse staffing for significant cost savings. Nursing homes typically employ three different types of nursing service personnel: Registered nurses (RNs), Licensed Practi cal Nurses (LPNs) and Certified Nursing Assistants (CNAs). The literature indicates that the nurse staffin g as well as proportions of the different types of nursing staff has a significant effect on patient mortality and quality of care. Nurse s taffing is represented by nursing intensity variables and skill mix. In addition, this study incorporates two variab les measuring the percentage of RN and LPN contract staffing in a nursing home. Literature suggests that use of contract staffing may increase nursing home costs and negatively impact quality of care (Bourbonniere et al., 2006) The following nurse staffing variables are employed in this study. Registered nurse hours per patient day (RNHRPPD) : Derived from the LTC focus dataset, it is a measure of RN intensity. Literature suggests that nurse qualification has an independent effect on quality; higher qualified nurses ensure lower morbidity and mortality (Aiken, Clarke, & Sloane, 2002; Aiken, Clarke, Sloane et al., 2002; Estabrooks et al., 2005; Weech Maldonado et al., 2004) In their study in post operative complications in hospitals, Kovner & Gregen ( 1998) demonstrate an inverse correlation between RN nurse intensity and occurrence of adverse events including
81 urinary tract infections (UTI). However, compared to LPNs and CNAs, RNs are more expensive. Therefore, this variable is employed to capture if private equity nursing homes have shifted their nurse staffing patterns to save costs in order to improve profits. Licensed p ractical nurse hours per patient day (LPNHRPPD): Derived from the LTC focus dataset, it is one of the two measures of non RN staffing. The role of adequate nurse staffing in ensuring nursing home quality has been widely acknowledged in the nursing home qua lity literature (Bowers et al., 2000; Harrington, Kovner et al., 2000) Though less qualified than RNs, with the rise of cost as a major concern, LPNs have increasingly been used in nursing homes. Based on our conc eptual framework, in order to increase profits, private equity nursing homes would be expected to substitute RNs with LPNs. Certified nurse assistant hours per patient day (CNAHRPPD): Derived from the LTC focus dataset, it is the second measure of non RN staffing. It is conceptualized on similar lines to LPNs, with private equity nursing homes expected to substitute RNs with CNAs in order to improve profitability. Skill mix : It is defined as the composition of the nursing staff by licensure or educationa l status (Kim et al., 2009) In this study skill mix was operational i zed as the r atio of number of RN FTEs divided by number of RN FTEs plus LPN FTE s. This variable as has been derived from the LTC focus dataset and is intended to capture shifts in nurse staffing patterns. For instance, if a facility lowered its RN FTEs, and substituted them with LPN FTEs, the skill mix would be negatively affected. Literature suggests that decrease in skill mix may lead to harmful effects for patients if substitute
82 LPNs and CNAs are made to work beyond their technical expertise. Lower skill mix may decrease overall nursing home quality, and result in higher workloads for RNs as LPNs and CNAs are less autonomo us in their functioning (Buchan & Poz, 2002) Registered Nurses contract : An OSCAR variable it indicates the per centage of total RN FTEs worked by contract RNs. In literature RN contract staffing has been coded as a dichotomous level with 5% threshold; nursing homes with more than 5% RN contract staffing are coded as one while others are coded as zero (Bourbonniere et al., 2006) However, considering the employment of RNs as contract nurses is relatively rare in Florida, in our study it was coded as a dichotomous variable indicating whether a facility employed RNs on contract or not. Increased hiring of RNs on contract increases costs and has a negative effect on quality (Bourbonniere et al., 2006) Licensed Nurses Contract : An OSCAR variable it indicates the percentage of total LPN FTEs worked by contract LPNs. In literature a 5% threshold level has been used to code LPN contract staffing as a dichotomous variable (Bourbonniere et a l., 2006) However, due to the relative rarity with which LPNs are engaged on contract in Florida, in our study it was coded as a dichotomous variable simply indicating whether a nursing home employed LPNs on contract or not. Similar to the case with RNs increased number of LPNs on contract increases costs and adversely affects quality (Bourbonniere et al., 20 06) Process measures of quality Process measures of quality reflect what is done to the patient (Hillmer et al., 2005) and whether it is done with adequate skill (R. L. Kane, 1998) Process measures are widely employed in the healthcare sector. Since 2005, the CMS has publicly reported process measures in hospitals. Unlike outcome measures like mortality,
83 process measures have proved more useful in distinguishing between high quality and low quality hospitals (Shih & Schoenbaum, 2007) On simila r lines it has been argued that process measures can be measured in the short term; are more amenable to clinical intervention than outcome measures; and set measureable markers for improvement (Werner, Bradlow, & Asch, 2008) Recognizing the role of process measures in understanding nursing home quality, CMS has incorporated process measures like use of restraints and use of catheters in its Nursing Home Compare database (Centers for Medicare and Medicaid Services, 2008a) .The following process measures of quality have been used in this study. Pressure sore prevention : This variable is construct ed as a facility composite score (0 4) of pressure sore prevention processes derived from four MDS dichotomous (yes/no) item s : turning/repositioning program, pressure relieving seat, pressure relieving mattress and ointment application. These four variable s were selected based on factor analysis with varimax rotation of all skin care processes captured by the MDS The pressure sore prevention composite had adequate internal consistency showing a Cronbach alpha of 0.82. Pressure sore prevention is an import ant variable as nursing home patients are highly susceptible to developing pressure sores because of their often limited mobility. Ensuring a regular turning schedule reduces the likelihood of pressure sores. Similarly, provision of pressure relieving dev ice distributes pressure over a greater body surface area lowering the risk of pressure sores (Centers for Medicare and Medicaid Services, 2008b) Restorative ambulation : A facility level continuous variable that measures the faci
84 nursing aides. It is generated by dividing the resident level restorative variable that represents the number of days of ambulation provided in the 7 days before the assess ment date by the total number of residents in the facility Nursing home residents on a restorative program are more likely to maintain their functional mobility because they walk on a regular basis. While walking residents may require only CNAs, the order s must be issued by a Physical Therapist (PT) and the entire program may be overseen by a licensed nurse. Use of restraints : It is a continuous variable defined as the proportion of residents who are physically restrained daily. The use of physical restr aints in nursing homes has been actively discouraged in the last two decades. OBRA (1987) mandated nursing ed for the purposes of (Castle & Mor, 1998 p.122 ) The literature has linked use of restraints with higher morbidity, cognitive decline, as well as an overall negative influenc e on resident quality of life (Castle & Mor, 1998) Restraints is one of the 8 (Abt Associates, 2004) Use of catheters : It is a continuous variable that represents the proportions of residents who had a catheter inserted and left in their bladder Use of catheters has been associated with higher rates of nosocomial ur inary tract infections in nursing home patients (Warren, 2001) while Kunin et al. (1 992) have shown that use of catheters is correlated with higher morbidity and morta lity among elderly nursing home patients. Use of catheters is a CMS quality measure used in the Nursing Home Compare website.
85 Outcome measures of quality The risk adjusted Quality Indicator (QI) score is a facility level QI score adjusted for the specif ic risk for that QI in the nursing facility. The risk adjusted QI score can be thought of as an estimate of what the nursing facility's QI rate would be if the facility had residents with average risk (Abt Associates, 2004) These outcome measures of nursing home quality have been validated by Abt Associates (2004) and are part of the CMS National Nursing Hom e Quality Measures and are currently used in the CMS Nursing Home Compare website ADL and bowel continence w orsening were risk adjusted using covariate (multivariate regression) models, while pressure ulcer prevalence was risk adjusted according to the s tratification method. Risk adjustment in the stratification method consisted of two steps: first, a weighted average per quarter was created for the high and low risk measures; second, an average was obtained across quarters The following risk adjusted MDS variables are used. ADL decline in function : is defined as referring to a set of common, everyday tasks, performance of which is required for personal self care and independent living (Wiener, Hanley, Clark, & Van Nostrand, 1990 p. 230 ) ADL decline in function is a risk adjusted ratio of the proportion of the residents who experienced a 4 point ADL decline (out of 16 points) relative to the total number of residents in the nursing homes who had a valid prior assessment. Residents excluded from the denominator include those identified as being comatose, having end stage disease, or receiving hospice care. Four Measures are included in ADL: bed mobility, transfer, eating, and toileting with each measure assigned 4 points for a total of 16 me quality reporting (Abt Associates, 2004)
86 Pressure ulcer high/low risk prevalence : It is a continuous variable that measures the percentage of residents who have pressure sore s (stage 1 4). Per Abt quality measures which should be reported together for high risk as well as low risk patients (Abt Associates, 2004) Residents considered high risk for pressure ulcers if they met any of the following criteria: impaired in mobility or transfer, comatose, suffer malnutrition, and end stage disease. The variable is risk adjusted u sing facility admission profile taking into account the prevalence of pressure sores among admissions over the past 12 months (Abt Associates, 2004) Bowel decline in continence : B owel decline in continence is a continuous variable that measures the proportion of residents whose bowel functions declined. It is a measure of the proportion of residents who have become incontinent. It is another ed for national reporting. The numerator for this (Abt Associates, 2004) Three different types of QMs are calculated depending upon the profile of the denominator: high and low risk which includes all residents regardless of risk, high risk with only high risk residents and low risk with only low risk residents. In our study we use the first type of QM which includes all residents irrespective of their risk. Residents excluded from the denominator include those identified as being comatose, having end stage disease, re ceiving hospice care, or having an ostomy present. Covariates include having short term memory problem, dressing problem, or bladder incontinence in the prior assessment.
87 Our study also uses two non risk adjusted outcome variables to measure quality. They are as follows. Deficiencies : Recommended as a quality measure by IOM (Institute of Medicine, 1996) in its report released in 1996, deficiencies are an o verall measure of nursing home quality and are part of the federal survey process. Deficiencies are monitored by state survey agencies operating under federal contract which evaluate nursing homes on a periodic basis (Harrington, Zimmerman, Karon, Robinson, & Beutel, 2000) If a facility fails to meet a particular standard, a citation is issued and the d eficiency is entered in the OSCAR system (Harrington, Zimmerman et al., 2000) Harrington et al. argue ( 2000) that deficiencies are an accurate reflection of quality of a nursing home as they are subject to extensive guidelines, and nursing homes have the right to challenge their citations. Federal regulations have 179 d ifferent standard of nursing home care classified into 17 major categories. Some of the categories under which deficiencies may be issued include dietary and rehabilitative services, pharmacy, physician services, falls, and pressure sores (Harrington, Zimmerman et al., 2000) Deficienci es have been frequently used in literature to measure nursing home quality (Grabowski, 2004; Kim et al., 2009; D B. Smith, Feng, Fennell, Zinn, & Mor, 2007) The total number of deficiencies in this study is a coun t variable which is the sum of all state and federal deficiencies and is located in the OSCAR dataset. Actual harm citation : It is a dichotomous variable measuring whether a facility was cited due to actual harm occurring to residents on a state survey It is organized according to a Scope and Severity system followed by all state health agencies under federal guidelines (Figure A 1). Under this system, each deficiency is classified
88 according to the level of harm to the residents and its scope within the facility There are (Colorado Department of Public Health & Environment, 2010) In ou r study, following Stevenson & Grabowski ( 2008) and 0 otherwise. (Figure A 1). Financial Perform ance Measures Operating revenue per patient day (OPRPPD): Operating revenue is defined as time charges such as sale of distressed properties to increase cash flow, it provides insight actual performance. In case of nursing homes, it is mainly income accruing from patient services. Nursing homes seeking to improve profitability can either seek to increase revenues, cut costs, or mixture of both. It is incorporated in this s tudy to understand what avenue private equity nursing homes adopt to improve profitability. OPRPPD is calculated by dividing operating revenues by total patient days. Operating cost per patient day (OC PPD): Operating expense is defined as cost incurred by a firm in discharging its normal business operations. Firms generally would seek to minimize their operating costs without adversely affecting their ability to compete in the market. In case of nursing homes, operating cost is primarily expense incurred o n providing patient services. In nursing home, staffing costs constitute over two third of total costs (Grabowski, Feng, Intrator, & Mor, 2004; Konetzka, Stearns, &
89 Park, 2008) ; it is included in this study to unde rstand the strategies adopted by private equity nursing homes to maximize profits. OCPPD is calculated by dividing operating expense by total patient days. Non operating revenue per patient day (NORPPD): It refers to income derived from operations which are not part of main business of a firm. It can include interest income and gain on sale of assets. For instance, a nursing home chain may show large non operating revenue by selling off a few nursing homes if the sale yielded a capital gain. In case of N FPs, philanthropic contributions usually form a significant portion of the non operating revenue. NORPPD is calculated by dividing non operating revenue by total patient days. Non operating cost per patient day (NOCPPD): it is defined as expense incurred loss on sale of assets, interest payments, and depreciation. NOCPPD is calculated by dividing non operating cost by total patient days. Census Medicare, Census Medicaid & Census O ther : While Medicare is responsible only for a relatively small portion of total nursing home revenues, it has been well recognized that Medicare patients are more financially attractive (Konetzka, Norton, & Stearns, 2006) than Medicaid patients. While Medicaid is the principal payer for nursing home care nationwide, recent reports suggest that cost of Medicaid patients are not covered by Medicaid payments (Agency for Healthcare Administration, 2005; Eljay, 2008) Census other --principally, private pay patients --supports a relatively small portion of nursing home beds. However, they are usually considered most financially remunerative. Therefore, nursing homes striving to maximize financial performance, a
90 shift in payer mix may be noticed shift in patient census in favor of Medicare and private pay patients at the cost of Medicaid patients. These variables would capture if a similar strategy is being purs ued by private equity nursing homes. Operating margin : It focuses on core business functions and excludes the influence of non operating income like endowments and non operating expenses such as interest income. It is calculated as following: Operating m argin: operating revenue operating expenses Operating revenue Total profit margin : It is the overall measure of financial performance. It includes all revenues (operating and non operating revenues) and all expenses (ope rating and non operating expe nses It is important to include this measure as private equity investments typically involve large amount of debt which may impact total profit margin (Atrill, 2008) 9 Total profit margin: t otal revenue total expenses Total revenue Operating margin and total profi t margin have been used previously in the literature to measure nursing home financial performance. Dependent variables are summarized in Tables 3 13. Acuity index : Derived from the LTC focus dataset, it is a facility level measure of resident acuity. The Acuity index is derived from the LTC focus dataset and is used as a measure of resident acuity at the facility level. Residents of a facility with higher acuity index would in principle require more intensive care. Acuity index can impact financial perfor mance as facilities with higher resident acuity may have higher revenues, but, at the same time, is likely to face higher costs than facilities with lower acuity. 9 Care is valuable real estate held by the latter which may help in financing the debt.
91 Independent V ariables Private equity ownership : A dichotomous variable (0 or 1) for nursing home ownership: whether the said nursing home is owned by private equity or not. Nursing homes are code d as 1 post acquisition; before their acquisition they are treated as part of the control group. The year of acquisition is counted a part of post acqui sition and nursing homes coded as 1. Year prior equity acquisition : A dichotomous variable coded as 1 in the year prior to private equity acquisition of a particular nursing home and 0 otherwise. This variable would indicate if nursing home behavior chang es before acquisition. For instance, nursing homes may focus on improving financial performance to extract the maximum valuation possible. The emphasis on financial performance, if present, may have a detrimental effect on nursing home quality. Years post acquisition : This indicates the number of years a nursing home has been owned by private equity. This variable is useful in understanding the impact of private equity ownership as literature indicates that number of years of private equity ownership has a n independent effect on the performance of acquired firms. This variable is coded as 0 prior to acquisition and in the year of acquisition; it is coded as 1 in subsequent years. Years post acquisition squared : Private equity LBO literature suggests that i n firms acquired by private equity, performance peaks in the first two years post acquisition, after which the impact of private equity ownership is minimal. This variable is designed to capture if the effect of private equity ownership is linear or quadra tic. Independent variables are summarized in Table 3 14.
92 Control Variables Control variables include organizational and market variables that may be associated with financial performance and quality: size, payer mix, occupancy rate, acuity index, per cap ita income county level, metropolitan, excess capacity, and Herfindahl index. Nur sing home size is measured by number of residents Larger facilities benefit from economies of scale, which are expected to result in lower resident cost per day (Banaszak Holl. et al., 2002; Bazzoli & Chan, 2000) Payer mix variables are the percentage s of Medicare and M edicaid residents. Nursing homes with a greater proportion of Medicare and private pay residents are expected to hav e greater revenues and hence may experience better financial performance. For example, it has been estimated that private pay residents pay on average 40% more than the Medicaid rate (Grabowski, 2004) The occupancy rate is the percentage of nursing home beds occupied by residents. Facilities with higher occupa ncy rates are more likely to experience better financial performance as they increase their revenues and make better use of their capacity. The Acuity index is derived from the LTC focus dataset and is used as a measure of resident acuity at the facility l evel, which is based on resident mobility and nursing factors, such as proportion of residents that are bedfast, requiring assist with ambulation or transfers, and receiving suctioning or intravenous therapy. Metropolitan represents a dichotomous variable that identifies whether a nursing home is located in a metropolitan area (1=yes, 0= no). Since facilities in metropolitan and non metropolitan areas are subject to different environmental factors, they may have different financial performance. The proport ion of the county population that is 65 years and older is included as another market variable. Since the majority of residents in nursing homes
93 are over the age of 65, it would be expected that counties with a younger population (less than 65 years) will be more competitive and less profitable. We include the nursing across markets. We use two measures of market competition: Herfindahl Index and excess capacity. The Herfinda hl index is a measure of market concentration and is derived from the LTC focus dataset. It is calculated by squaring each facility's total beds and the sum for all facilities in the county is calculated This sum is subsequently divided by sum of all co unty beds squared. The Herfindahl index represents perfect competition when it registers a score of 0, while a score of 1 represents a monopolistic market. Excess capacity is the average number of empty beds per facility in the county. Markets with highe r excess capacity are more competitive. Control variables are summarized in Table 3 15. Other V ariables Provider number : The unique provider used to identify individual nursing facilities, and is common across different datasets. Time : It is used to iden tify the number of years of this study and is coded as 2000:1, 2001:2, 2003:3 and so on. M odel Our study is a longitudinal study (2000 2007) and is organized as a panel data regression. Kim et al. ( 2009) point that using large administrative datasets may cause omitted variable bias; they are not collected specifically for a particular research project and may therefore exclude measure which impact variables of interest. To address omitted va riable bias, this study uses random effects. Random effects model assumes
94 that the model heterogeneity is due to time invariant traits located within individual nursing homes. T he general model is as follows t+1 it +T it + Tit 2 +it Where Y: Dependent Variable (quality and financial performance measures) P.E: A dichotomous variable (0=control group, 1=Private equity ownership) C: Control variables ( Size, p ayer mix occupancy rate, acuity index, per capita income county level, metropolitan, Herfindahl Index and excess capacity ) i: Individual facility t: Year Coded as 1..2..3 for each year individual year in the dataset. Year prior equity acquisition : A dichotomous variable coded a s 1 if the nursing home ownership changes in t+1. T: It reflects the y ears post acquisition: It will be modeled as T=0 before a upon the years subsequent to the acquisiti on. For instance, a nursing home acquired in 2003 wi ll have T=0 for 2000, 2001, 2002 and 2003 while T=1 in 2004, 2 in 2005 and so on. T 2 : Years post acquisition squared : Error term For ratios, Ordinary Least Square (OLS) is used. To satisfy the normal ity assumption, it is ensured that skewness and kurtosis are as close to 0 and 3 as possible. To improve the distribution, outliers are dropped while in other cases transformations may be necessary. Log, square root, and cube transformations are
95 attempted. In case of log transformations, if kurtosis is less than 4, gamma distribution with log link is used otherwise OLS is used. Logit is used for proportions and Odds ratio is calculated. Logistic regression is used for dichotomous variables and Odds ratio is calculated. In case of count variables, negative binomial regression is the preferred approach. Quality V ariable Analysis Registered nurse hours per patient day (RNHRPPD): The distribution of RNHRPPD can be seen in Table 3 1. In order to use Ordinary L east Square (OLS) the normality assumption must be satisfied; the goal is to reduce kurtosis as close to 0 as possible and skewness to 3. As seen in Table 3 1, with log transformation, the skewness and kurtosis approximates the desired values of 0 and 3 r espectively. As kurtosis is less than 4, a gamma distribution with log link is used to avoid the known issues of heteroskedasticity Licensed nurse hours per patient day (LPNHRPPD): The distribution of LPNHRPPD can be seen in Table 3 2. Log transformation achieved the best results with skewness and kurtosis closest to 0 and 3. As kurtosis is less than 4, a gamma distribution with log link is used to avoid the known issues of heteroskedasticity Certified nurse assistant hours per patient day (CNAHRPPD): T he distribution of LPNHRPPD can be seen in Table3 3. D ropping observations + 5 standard deviations of mean achieved the best results with skewness and kurtosis closest to 0 and 3 respectively. Therefore OLS is used to analyze CNAHRPPD Skill mix: The distr ibution of skill mix can be seen in T able 3 4. T he untransformed variables achieved the best results with skewness and kurtosis closest to 0 and 3 respectively. Therefore OLS is used for skill mix.
96 RN contract staffing : It is a dichotomous variable modele d as 1 in presence of RN contract staffing and zero otherwise OLS is not appropriate in this case as estimates would not be efficient. Therefore, logistic regression is used for RN contract staffing. LPN contract staffing: It is a dichotomous variable m odeled as 1 in presence of RN contract staffing and zero otherwise OLS is not appropriate in this case as estimates would not be efficient. Therefore, logistic regression is used. Pressure sore prevention : The distribution of pressure sore prevention can be seen in Ta ble 3 5. c ube root transformation achieved the best results with skewness and kurtosis closest to 0 and 3 respectively. OLS is used for pressure sore prevention. Restorative ambulation: The distribution of restorative ambulation can be seen in Ta ble 3 6. D ropping outliers achieved the best results with skewness and kurtosis closest to 0 and 3 respectively. Therefore, OLS is used for restorative ambulation. Use of restraints : It is the proportion of residents on whom restraints are used dai ly. Since it is a proportion, OLS is not appropriate. Therefore for restraints, binomial family with logit link is used Use of catheters : It is the proportion of residents on who se bladders are catheterized. Since it is a proportion, OLS is not appropriat e. Therefore for catheters, binomial family with logit link is used. ADL decline in function : It is the proportion of the residents who experienced a 4 point ADL decline (out of 16 points) relative to the total number of residents in the nursing homes. Sin ce it is a proportion, OLS is not appropriate. Therefore for ADL decline in function, binomial family with logit link is used
97 Pressure ulcer high/low risk prevalence : The distribution of pressure ulcer high/low risk prevalence can be seen in Table 3 7. l o g transformation achieved the best results with skewness and kurtosis closest to 0 and 3 respectively. As kurtosis is less than 4, a gamma distribution with log link is used. Bowel decline in continence : I t is the proportion of residents whose bowel funct ions declined As it is a proportion, OLS is not appropriate. Therefore for bowel decline in continence, binomial family with logit link is used Number of deficiencies: It is a count variable, and therefore OLS is not appropriate as the distribution is n ot normal. Therefore is analyzed using negative binomial regression. Actual harm citation : It is a dichotomous variable modeled as 1 in presence of harm and zero otherwise OLS is not appropriate in this case as estimates would not be efficient. Therefor e, logistic regression is used for LPN contract staffing. Financial Variable A nalysis Operating revenues per patient day (OPRPPD): The distribution of OPPPD tends to be skewed to the right as few facilities may have significantly higher values than the me an. Therefore, OLS is not appropriate in this case and a gamma distribution with log link is used (Fleishman, Cohen, Manning, & Kosinski, 2006; Pagano et al., 2009) Facilities with OPRPPD greater than 5 standard d eviations of the mean were dropped. Non operating revenues per patient day (NORPPD) : The distributions of NORPPD can be seen in Ta ble 3 8. L og transformation achieved the best distribution. As kurtosis is less than 4, a gamma distribution with log link is used to avoid the known issues of heteroskedasticity
98 Operating costs per patient day (OCPPD) : The literature suggests that distribution of OCPPD tends to be skewed as some nursing homes may have much higher values than the mean. Therefore, OLS may not be appropriate in this case and a gamma distribution with log link is used. Facilities with OCPPD greater than 5 standard deviations of the mean were dropped. Non operating costs per patient day (NOCPPD) : The distribution of NOCPPD can be seen in Table 3 9. c ube root transformation achieved the best distribution. With skewness and kurtosis approaching the desired values of 0 and 3 respectively, OLS is used for NOCPPD. Operating margin : OLS with random effects is used to analyze operating margin. In order to use OLS, the normality assumption must be satisfied; the goal is to reduce kurtosis as close to 0 as possible and skewnes s to 3. D ropping +/ 5 standard deviations achieved the best results with skewness and kurtosis closest to 0 and 3 respectively. Ther efore, OLS is used for operating margin Total margin : The distribution of total margin can be seen in Tabl e 3 11. d ropping +/ 3 standard deviations achieved the best results with skewness and kurtosis closest to 0 and 3 respectively. Therefore, OLS is us ed for total margin. Census Medicare : It is the proportion of facility residents who are supported primarily by Medicare. Since it is a proportion, OLS is not appropriate. Therefore for census Medicare, binomial family with logit link is used Census Medi caid : It is the proportion of the facility residents who are supported primarily by Medicaid. Since it is a proportion, OLS is not appropriate. Therefore for census Medicaid, binomial family with logit link is used
99 Census Other : It is the proportion of fa cility residents who are supported by modes of support other than Medicare and Medicaid --primarily private pay. Since it is a proportion, OLS is not appropriate. Therefore for census other, binomial family with logit link is used. Acuity Index : Acuity ind ex is a continuous variable, and its distribution can be seen in Tabl e 3 12. T he untransformed variable achieved the best distribution with skewness and kurtosis closest to 0 and 3 respectively. Therefore, OLS is used for acuity index.
100 Tabl e 3 1. RN hours per patient day distribution Transformations Mean Skewness Kurtosis Original .29 14.5 268.3 Dropping outliers .27 14.5 268.3 Log 1.4 .04 3 Square root .5 .5 1.23 Cube root .5 .13 .58 *Observations with RNHRPPD=0 were dropped as nur sing homes are mandated by law to have RNs. A total of 13 observations dropped. Table 3 2. LPN hours per patient day distribution Transformations Mean Skewness Kurtosis Original .91 19.2 404.3 Dropping outliers .8 1.8 14.2 Log .16 .3 2.5 Square ro ot .92 .09 7.08 Cube root .94 1.16 16.9 Observations + 5 standard deviations dropped. A total of 9 observations dropped. Table 3 3. CNA hours per patient day distribution Transformations Mean Skewness Kurtosis Original 2.5 2.8 33.9 Dropping outlier s 2.5 .32 1.6 Log .8 .67 1.47 Square root 1.57 .18 .99 Cube root 1.35 .34 1.03 Observations + 5 standard deviations dropped. A total of 6 observations dropped. Table 3 4. Skill mix distribution Transformations Mean Skewness Kurtosis Original .2 3 1.1 2.7 Dropping outliers .23 1.1 2.7 Log 1.5 .77 1.38 Square root .47 .20 .29 Cube root .59 .09 .21 Observations + 3 standard deviations dropped. No observations dropped. Table 3 5. Pressure sore prevention distribution Transformations Mean Skewness Kurtosis Original 1.7 .03 .29 Dropping outliers Log .48 .1.71 5.76 Square root 1.3 .64 .43 Cube root 1.89 .91 1.20 Dropping observations +/ 5 standard deviations drops all observations therefore was not attempted.
101 Table 3 6. Rest orative ambulation distribution Transformations Mean Skewness Kurtosis Original .63 1.62 4.05 Dropping outliers .62 1.38 2.56 Square root .71 0 .07 Cube root .77 .81 1.04 Observations + 5 standard deviations d ropped. 30 observations dropped ** Lo g transformation not attempted as a lot of zero values and natural log of zero is undefined. Table 3 7. Pressure ulcer high/low risk prevalence distribution Transformations Mean Skewness Kurtosis Original .08 1.19 4.78 Dropping outliers .08 1.57 2.54 Log 2.5 .74 1.75 Square root .29 .19 1.08 Cube root .43 .1 .88 Observations + 5 standard deviations dropped. 27 observations dropped Table 3 8. Non operating revenue distribution Transformations Mean Skewness Kurtosis Original 8.3 25.1 763.2 Dropping outliers 2.3 10.5 128.8 Log .83 .06 2.2 Square root .95 5.5 40.1 Cube root .88 3.4 17.7 Observations + 3 standard deviations dropped. Values more than $200 dropped. 22 observations dropped Table 3 9 Non o perating costs per patient da y distribution Transformations Mean Skewness Kurtosis Original 22.1 7.4 82.1 Dropping outliers 19.2 1.4 4.9 Log 2.8 4.4 72.1 Square root 4.2 .2 1.0 Cube root 2.6 .2 1.23 Observ ations with negative values dropped. Observations +/ 3 standard devi ations dropped. 118 observations dropped Table 3 10 Operating margin distribution Transformations Mean Skewness Kurtosis Original .07 11.3 247.9 Dropping outliers .09 .9 2.7 Log/square root/cube root** Observations +/ 3 standard deviation s dropped. 29 observations dropped ** Log, square root and cube root transformation was not attempted as operating margin has large number of negative values and these values would be undefined
102 Table 3 11 Total margin distribution Transformations Mean S kewness Kurtosis Original .01 9.3 157.9 Dropping outliers .001 1.1 4.8 Log/square root/cube root** Observations +/ 3 standard deviations dropped. 38 observations dropped ** Log, square root and cube root transformation was not att empted as total margin has large number of negative values and these values would be undefined Table 3 12. Acuity index distribution Transformations Mean Skewness Kurtosis Original 11.5 .16 .24 Dropping outliers 11.5 .16 .24 Log 2.4 .49 .53 Square root 3 .3 .3 .34 Cube root 2.2 .38 .39 *+ 3 standard deviations dropped. No observations
103 Table 3 13. Dependent variables and their definitions Dependent variable Definition RN hours per patient day Registered nurses hours/total patient days LPN hours per patient day Licensed nurse hours/total patient days CNA hours per patient day Certified nurse assistant hours/total patient days Skill Mix RN FTEs/RN FTEs+LPN FTEs RN contract staffing P ercentage of total RN FTEs worked by contract RNs LP N contract staffing Percentage of total LPN FTEs worked by contract RNs Pressure sore prevention A facility composite score (0 4) of pressure sore prevention processes derived from four MDS dichotomous (yes/no) item s Restorative ambulation A verage numbe r of days in a week that residents are walked using restorative nursing aides. Use of restraints Proportion of residents who are restrained daily Use of catheters Proportion of residents who had a catheter inserted and left in their bladder Number of de ficiencies Total number of deficiencies Actual harm citations Measures whether actual harm was caused to resident in a state survey QI ADL decline Percent of residents who had 4 pt decline QI h/l Pressure sore Percent of residents who acquired a pressu re sore QI Bowel Worsening Percent of residents whose bowel function declined Operating revenue per patient day Operating revenues/total patient days Non operating revenues per patient day Non operating revenues/total patient days Operating cost per pa tient day Operating cost/total patient days Non operating cost per patient day Non operating cost/total patient days Operating Margin Operating revenue operating cost/patient revenue Total Margin Total Revenue total cost / total Revenue Census Medic are Proportion of facility residents supported primarily by Medicare Census Medicaid Proportion of facility residents supported primarily by Medicaid Census Other Proportion of facility residents not supported by Medicare and Medicaid Acuity index The a verage Resource Utilization Group Nursing Case Mix Index
104 Table 3 14. Independent variables and their definitions Independent variables Definition Equity owned Indicates whether a nursing home is owned by private equity (0/1) Year prior equity ownersh ip A dichotomous variable coded as 1 year prior to private equity acquisition of a nursing home Years post acquisition Indicates numbers of years of private equity ownership of a particular nursing home Years squared Years post acquisition squared. Indic ates whether the effect is linear or quadratic Table 3 15 Control variables and their definitions Control Variables D efinition Ownership Dichotomous Total residents # of residents Percent Medicare Percent beds that are designated Medicare Percent Me dicaid Percent beds that are designated Medicaid Percent Other Percent beds that are designated Other Payer Occupancy Rate Number of occupied beds in facility/total number of beds Acuindex The average Resource Utilization Group Nursing Case Mix Index Metropolitan Dichotomous Variable (0,1) Per capita income Per capita income of the country in which the nursing home is located Percentage over 65 Proportion of patients in the nursing home county aged 65 and over Herfindahl index Measure of nursing ho me concentrati on/competition in the county ran ging from 0 to 1 Excess capacity Average number of empty beds in a county
105 CHAPTER 4 RESULTS Table 4 1 provides the descriptive statistics of all dependent variables while Table 4 2 provides descriptiv e statistics for independent and control variables. Most dependent variables were skewed and non kurtotic, and therefore required either dropping of outliers, transformations, or a combination of both. Bivariate statistics for all the dependent variables a re available on Tables A 1 to A 5. Two models are run for each dependent variable: A fully specified model which includes all the independent variables as well as all control variables capturing the quality and financial performance of private equity nurs ing homes over time. Squared year variable is used in the model to understand if the shift in nursing home performance as years of private equity ownership progress is quadratic or linear. It is interpreted as linear if the squared year variable is non si gnificant. If in the initial model the squared year variable is non significant, it is dropped from model 1 for that particular dependent variable. If the squared year variable is significant, then the inflexion point is calculated. The inflexion point can be understood as the exact point of time when there is a shift in the dependent variable as years of equity ownership progress. The second model includes all the dependent and contr ol variables but the only independent variable incorporated in this model is equity ownership. The second model is designed to capture the overall effect of equity ownership on nursing home quality and financial performance Hypothesis 1 Hypothesis#1 : Equity owned nursing homes deliver better financial performance than other i nvestor owned nursing homes. In this study, financial performance is
106 visualized in terms of operating margin and total margin. The results suggest that hypot hesis #1 is supported (Table 4 3 and Table 4 4 ). Private equity nursing homes report 1% lower opera ting margin than the control group. However this result is significant only at 10% level. Prior to their acquisition, these nursing homes had 8% lower operating margin (p=.05). This shift is explained by positive correlation of operating margin with years of equity ownership (p=.01) with operating margin increasing by 3% for each year of equity ownership. In terms of total margin, there is no statistically significant difference between private equity nursing homes and the control group; however, total ma rgin is positively correlated with years of private equity ownership (p=.001). In the year prior to their acquisition, these nursing homes had 3% higher total margin than the control group (p=.001). Chi square test conducted to capture the overall effect of private equity ownership is significant for both operating margin (p=.001) and total margin (p=.01). In model 2 (Table 4 4 ), priva te equity nursing homes report 2 % higher operating margin as well as 1 % higher total margin (p=.001).Overall it appears th at private equity nursing homes report better financial performance than the control group. This difference is highly statistically significant We include revenues and costs in our study to understand the dominant template adopted by private equity nursing homes to improve financial performance. For instance, some private equity nursing homes may choose to focus on maximizing revenues, while others may attempt to improve financial performance by cutting costs. In model 1 (Table 4 3 ) private equity nursing homes have higher operating revenues PPD than the control group. (p=.001); this improvement in operating revenues
107 PPD is positively correlated with increasing years of private equity ownership; for every year of private equity ownership, operating revenue PPD increases by 3% (p=.001). In the year prior to private equity acquisition, these nursing homes had 3% lower operating revenue PPD. However, this result is only significant at 10% level. Private equity nursing homes also report 11% higher operating co st PPD (p=.001) than the control group; the operating cost PPD is negatively correlated with years of private equity ownership (p=.001). The results suggest that private equity nursing homes have higher operating revenues as well as higher operating costs than the control group. Chi square test is significant in the case of both operating revenue PPD and operating cost PPD (p=.001) indicating an overall impact of private equity ownership. Private equity nursing homes have 120% lower non operating revenues PPD than the control group (p=.001). However, in somewhat surprising results, before their acquisitions, these nursing homes had 130% higher non operating revenues (p=.001). In terms of non operating costs PPD, although there is no statistically significa nt difference between private equity nursing homes and the control group, it is positively correlated with years of private equity ownership (.001). Chi square is highly significant in case of both non operating revenue PPD and non operating cost PPD (P=.0 01) indicating an overall effect of private equity ownership. There is no statistically significant difference between private equity nursing homes and the control group on census Medicare, census Medicaid, and Census other. Chi square test in case of all three payer mix variables is non significant indicating no overall effect of private equity ownership on these variables. Similarly, there is no statistically significant difference
108 between private equity nursing homes and the control group on residents ac uity. However, for every year of private equity ownership, acuity increases by 8% (p=.001). In Model 2 (Table 4 4) private equity nursing homes have 17% higher operating revenues PPD (p=.001) and 15 % higher operating costs PPD than the control group (p=. 001). The non operating revenue PPD is 86 % lower (p=.001) while th eir non operating cost PPD is 27 % higher (p=.001). Private equity nursing homes report 27% higher acuity index than the control group (p=.001). However, there is no statistically signific ant difference between private equity nursing homes and the control group in their payer mix viz. census Medicare, census Medicaid and census other (primarily private pay). Year squared variable is sig nificant for operating margin, operating cost PPD, non operating cost PPD and non operating revenue PPD The inflexion point for operating margin is 6 years; operating margin increases for the first 6 years post acquisition, and then declines T he inflexion poi nt for operating cost PPD is 5 years; it declines for the first 5 years post acquisition and then increases. The inflexion point for non operating cost PPD and non operating revenue PPD is 8 and 6.8 years respectively. Full results for financial variables are available on T ables A 6 to A 25 Hypothesis 2 Hypothesis #2 states that private equity owned nursing homes are likely to experience poorer quality of care compared to other investor owned nursing homes. For the resu lts partially support hypothesis #2 but differ significantly across the different d imensions of the SPO model (Table 4 5 and Table 4 6).
109 Structural V ariables In terms of staff ing variables Model 1 (Table 4 5 ) results suggest that private equity nursing h omes have 11% lower RN hours PPD compared to the control group (p=.05) while years of private equity ownership is negatively correlated with RN hours PPD (p=.05). Private equity owned nursing homes have 10% higher LPN hours PPD (p=.001) with positive corr elation with years of private equity ownership (p=.05). Private equity nursing homes report 32% higher CNA hours PPD compared to the control group (p=.001). CNA hours PPD is also positively correlated with years of equity ownership (p=.05). The difference in CNA hours PPD between private equity owned nursing homes and the control group is particularly striking as before acquisition, they had 13% less CNA hours per patient day (p=.05). Chi square test was highly significant in case of all three variables viz RN, LPN and CNA hours PPD indicating an overall statistically significant impact of private equity ownership on nurse staffing. The shift in nurse staffing patterns is reflected in the skill mix where private equity nursing homes have 4% lower skill mix than the control group (p=.001) and this difference is sustained as nursing homes spend more years under private equity ownership (p=.01). It appears that as predicted in the conceptual framework, private equity nursing homes have substituted RNs with LPN s and CNAs. In terms of nurse staffing on contract, the odds of equity owned nursing homes hiring RNs on contract are 2.1 times higher (p=.05). However, as years of private equity ownership progress, the odds of hiring RNs on contract decline (p=.05). On similar lines, the odds of private equity nursing homes hiring LPNs on contract are 6 times higher (p=.001) as compared to the control group though it is important to note that even prior to their acquisition, the odds of these particular nursing homes hir ing LPNs on contract
110 was 2.2 times higher (p=.01). As noted in the case of RNs, the odds of hiring LPNs on contract decline with progressive years of private equity ownership (p=.001). It indicates that private equity nursing homes may have initially hired more RNs/LPNs on contract but the strategy shifts with the passage of time with these nursing homes attempting to minimize hiring RNs and LPNs on contract Chi square test was significant in case of LPNs on contract (p=.001) but not on RNs on contract. In Model 2 (Table 4 6), private equity nursing homes report 12% lower RN hours PPD compared to the control group (p=.001) while they have 12% higher LPN hours PPD (p=.001) and 42% higher CNA hours PPD (p=.001). The skill mix in private equity nursi ng homes is 4% lower (p=.001) compared to the control group. The odds of private equity nursing homes hiring RNs on contract is 2.1 times higher (p=.01) while the odds of these nursing homes hiring LPNs on contract is 2.3 times higher (p=.001) compared to the con trol group. Process V ariables In model 1 (Table 4 7), p rivate equity nursing homes report 2% lower pressure sore prevention than the control group. However, this result is statistically significant only at the 10% level. However, for every year of priv ate equity ownership, nursing homes report 1% higher pressure sore prevention (p=.001) while restorative ambulation declines by 3% for every year of private equity ownership (p= .001). Chi square test is highly significant in case of both pressure sore pre vention as well as restorative ambulation (p=.001). There is no statistically significant difference between private equity nursing homes and the control group in case of use of restraints and catheters. The chi square test is non significant in case of bo th these variables indicating that there is no overall impact of private equity ownership.
111 Outcome Variables In model 1 (Table 4 9), Private equity nursing homes have 10% higher deficiencies as compared to the control group (p=.01) (Table4 6) while thes e nursing homes had 18 % higher total deficiencies before their acquisition. (p=.001). This shift may be explained by 3% decline in deficiencies for every year of private equity ownership (p=.05). There is an overall impact of private equity ownership on d eficiencies as Chi square test is highly significant ( p=.001). Private equity nursing homes report better results on actual harm citation variable (OR=.53) though this result is significant only at 10% significance level. Chi square test is non significan t. In case of all other outcome variables viz. ADL 4point decline, bowel worsening and pressure ulcer high/low risk prevalence, there is no statistically significant difference between private equity nursing homes and the control group. In Model 2 (Table 4 10 ), private equity nursing homes have 13% higher deficiencies as compared to the control group (p=.001). All other outcome variables are non significant. Full results are available on Tables A 26 to A 5 5.
112 Table 4 1. Descriptive statistics of depend ent variables Dependent variables N Mean Minimum Maximum Skewness Kurtosis RN hours PPD 2737 .27 .01 2.0 1.9 9.8 LPN hours PPD 2737 .87 0.0 3.3 1.4 10.3 CNA hours PPD 2737 2.5 .8 5.2 .32 1.6 Skill mix 2737 .23 .01 1.0 .89 1.45 Pressure sore preventio n 2737 1.8 .03 3.5 .03 .3 Restorative ambulation 2737 .62 0.0 3.2 1.38 2.57 % of restraints 2723 .08 0.0 .39 .85 .71 % of catheters 1310 .09 .0 .66 1.9 8.9 ADL decline 4 point 2729 .12 .009 .34 .73 1.08 Pressure ulcer h/l risk prevalence 2723 .08 .4 .009 1.2 4.79 # of deficiencies 2737 7.83 0 33 1.1 1.38 Operating revenue PPD 2645 189.0 25.3 458.1 .64 1.8 Non operating revenue PPD 2539 2.4 0.0 192.6 10.5 130.8 Operating cost PPD 2645 170.1 22.9 479.0 .6 2.9 Non operating cost PPD 2645 19.9 0.0 8 0.6 1.3 4.3 Operating margin 2645 .09 .33 .44 .8 2.7 Total margin 2539 .001 .54 .44 1.1 4.9 Census Medicare 2645 .17 0.0 1.0 1.5 5.4 Census Medicaid 2645 .62 0.0 .98 .8 1.2 Census Other 2645 .20 0.0 1.0 1.3 3.6 Acuity index 2645 11.6 7.9 15.6 .16 .25
113 Table 4 2. Descriptive statistics of in dependent control and variables Independe nt variables N Mean Minimum Maximum Standard deviation Equity ownership 2737 .15 0 1 .15 % Medicare 2737 .17 0.0 .9 .1 % Medicaid 2737 .61 0.0 .98 .17 % Other 2737 .20 0.0 1.0 .13 Acuity index prevention 2737 11.5 7.8 18.6 1.1 Occupanc y rate 2736 .88 .27 1.0 .1 Per capita income 2595 37217 15971 57446 9620 Metro 2595 .89 .0 1 .3 County population + 65 2595 .19 .08 .34 .06 Herfindahl index 2732 1 .01 1 .16 Excess capacity 2732 14.7 2 87 5.7 Table 4 3. Financial performance of private equity nursing homes: Model 1 Dependent variable Equity owned Year prior equity acquisition Years post acquisition Years post acquisition squared Operating reven ue PPD 1 08 ***(.01) .03 + (. 02) .03***(.00) Non operating revenue PPD .1.2***(.15) 1.3***(.17) .55***(.12) .08***(.02) Operating cost PPD .11***(.01 ) .007 (.02) .003(.005) .006*(.00) Non operating cost PPD .04(.04) .004(.04) .16***(.03) .02***(.00) Operating margin .01 + (.00) .08*(.01) .03***(.00) .005***(.00) Total margin 1 .008(.00) .03***(.02) .007***(.00) Census Medicare^ 1 1.04(.21) 1.02(.27) 1.05(.05) Census Medicaid^ 1 1.09(.15) 1.02(.18) .94(.03) Census Other^ 1 .78(.17) .88(.23) 1.03(. 06) Acuity index 1 .06(.09) .06(.12) .08***(.02) Significance levels: *.05 **.01***.001 + .1 ^: Odds ratio 1 Squared year variable dropped from the model due to non significance.
114 Table 4 4. Financial performance of private equity nursing homes: Mode l 2 Dependent variable Equity owned Operating revenue PPD .17***(.01) Non operating revenue PPD .86***(.1) Operating cost PPD .15***(.01) Non operating cost PPD .27***(.02) Operating margin .02***(.00) Total margin .01*(.00) Census Medicare^ 1.17(. 17) Census Medicaid^ .95(.1) Census Other^ .87(.13) Acuity index .25***(.06) Significance levels: *.05 **.01***.001 ^: Odds ratio Table 4 5. Private equity nursing homes quality indicated by structural variables: Model1 Dependent variable Equity owne d Year prior equity acquisition Years post acquisition RN hours per patient day .11* (.05) .07(.05) .02*(.04) LPN hours per patient day .10***(.02) .01(.02) .01*(.01) CNA hours per patient day .32***(.04) .13*(.05) .03*(.01) Skill mix .04***(.00) .01(.01) .006**(.00) RN contract staffing^ 2.1*(1.0) 1.0(.57) .74*(.28) LPN contract staffing^ 6.0***(1.6) 2.2**(.72) .67***(.27) Significance levels: *.05 **.01***.001 ^: Odds ratio Squared year variable dropped due to non significance Table 4 6. P rivate equity nursing homes quality indicated by structural variables: Model2 Dependent variable Equity owned RN hours per patient day .12***(.03) LPN hours per patient day .12***(.01) CNA hours per patient day .42***(.03) Skill mix .04***(.00) RN c ontract staffing^ 2.1*(.28) LPN contract staffing^ 2.3***(.45) Significance level: *=.05 **=.01 ***=.001 + =.1 ^Odds ratio
115 Table 4 7 Private equity nursing homes quality indicated by process variables: Model1 Dependent variable Equity owned Year prior equity acquisition Years post acquisition Pressure sore prevention .02(.01) + .003(.01) .01***(.00) Restorative ambulation .04(.04) .06(.04) .03***(.01) Use of restraints^ 1.0(.26) .94(.29) .97(.06) Use of catheters^ 1.3(.5) 1.0(.22) .92(.09) Sig nificance level: *=.05 **=.01 ***=.001 + =.1 ^Odds ratio Squared year variable dropped from models as It was statistically non significant. Table 4 8. Private equity nursing homes quality indicated by process variables: Model2 Dependent variable Equity own ed Pressure sore prevention .02*(.00) Restorative ambulation .005(.02) Use of restraints^ 1.0(.18) Use of catheters^ 1.0(.28) Significance level: *=.05 **=.01 ***=.001 +=.1 ^Odds ratio Table 4 9. Private equity nursing homes quality indicated by ou tcome variables: Model1 Dependent variable Equity owned Year prior equity acquisition Years post acquisition ADL 4 point decline^ 1.0(.24) 1.1(.36) .97(.25) Bowel worsening^ 1.0(.21) 1.1(.27) 1.0(.06 ) Pressure ulcer high/low risk prevalence .01(.03) 01(.03) .002(.00) Total deficiencies .10*(.05) .18***(.06) .03*(.01) Actual harm citation^ .53 + (.19) 1.7 + (.49) 1.4(.45) Significance level: *=.05 **=.01 ***=.001 + =.1 ^Odds ratio Squared year variable dropped from model as It was statistically non signi ficant. Table 4 10. Private equity nursing homes quality indicated by outcome variables: Model2 Dependent variable Equity owned ADL 4 point decline^ .96(.16) Bowel worsening^ .95(.14) Pressure ulcer high/low risk prevalence .01(.02) Total deficien cies .13***(.03) Actual harm citation^ .7(.16) Significance level: *=.05 **=.01 ***=.001 + =.1 ^Odds ratio
116 CHAPTER 5 DISCUSSION wariness and concern in policy and academic circ les. The critics have argued that the organization of private equity financing and its different motivations --in particular, the quest for short term profitability --substantially alters the industry structure, and potentially has a negative impact on nu rsing home quality. However, absence of independent empirical research has hampered this important policy debate; ergo this study with its stated goal of understanding the impact of private equity ownership on nursing home performance. By addressing this gap in the extant literature, this study furthers our understanding of drivers of nursing home performance in particular, the impact of private equity ownership. Quality of Care We hypothesized that private equity nursing homes are likely to deliver poore r quality of care to their residents compared to other investor owned nursing homes. The results are mixed and only partially support our hypothesis. As anticipated, private equity nursing homes have lower RN intensity, and higher LPN and CNA intensity com pared to the control group. The shift is particularly sharp in the case of CNAs. One possible explanation may be Senate Bill (SB) 1202 signed into law in the year 2001. Aimed at improving nursing home quality, SB 1202 imposed minimum CNA staffing levels in Florida and was to be fully implemented by the year 2003 (Polivka, Salmon, Hyer, Johnson, & Hedgecock, 2003) As private equity nursing homes had lower CNAs than the control group pre acquisition, and th e law was implemented over 2001 2003 --the
117 same period as private equity acquisitions --it may explain why private equity nursing homes so sharply increased their CNA staffing. The change in nurse staffing pattern is reflected in the lower skill mix of pr ivate equity nursing homes. As argued in the conceptual framework, it appears that private equity nursing homes are replacing higher cost RNs with LPNs and nursing aides. Interestingly, (SB) 1202 had provided incentives for Florida nursing homes to increas e their licensed nursing ratios (RNs and LPNs). It appears that private equity nursing homes have complied with the law by disproportionately increasing the number of LPN staffing instead of the more expensive RN staffing. In case of RN staffing, the resu lts support those reported by Harrington & Carrillo ( 2007) in their study of pr ivate equity nursing homes in California. Ordinarily, decrease in RN staffing would have a negative impact on nursing home quality (Aiken, Clarke, & Sloane, 2002; Weech Maldonado et al., 2004) The results are mor e ambiguous in our study. Despite the troubling decline in RN staffing intensity, private equity nursing homes perform similar to the control group in all process and outcome quality variables except pressure sore prevention and deficiencies. Private equit y nursing homes perform slightly better in pressure sore prevention, and significantly worse in number of total deficiencies. performance in respect to pressure sore prevention may reflect their higher CNA ratios while deficiencies are more sensitive to RN staffing which is significantly lower in these nursing homes as compared to the control group. These results are broadly similar to those reported by Stevenson and Grabowski ( 2008) in their national level study of
118 p rivate equity nursing homes: Decline in RN staffing but little significant difference in quality in private equity and other nursing homes. Stevenson and Grabowski ( 2008) argue that though RN staffing is lower in private equity nursing homes, it may no t be a cause of immediate concern as in their sample these nursing homes had lower RN staffing before their acquisition. While our findings are similar to Stevenson and Grabowski ( 2008) it is pertinent to note that we also report a decline in RN staff ing with every progressive year of private equity ownership. Therefore, we remain less sanguine about future quality in private equity nursing home; it is possible that more recent data shows a more negative impact of lower RN staffing on quality. Despite the sharp increase in CNA staffing intensity, private equity nursing homes perform no better in process variables like restorative ambulation and; both these variables may be positively impacted by CNA nurse staffing. For instance, improving patient mobil (Thomas, Hyer, Andel, & Weech Maldonado, 2009) Studies have show n that increasing CNA staffing without adequate RN supervision may bring little additional benefit (Buchan & Poz, 2 002) Another possible explanation may be that as reported by Thomas et al. ( 2009) one of the uni ntended consequence of SB 1202 is decline in indirect care staff (housekeeping, therapists, physicians, dietary and laundry staff) as nursing homes may be using the newly mandated CNAs for indirect care rather than direct patient care. While we cannot offe r definitive evidence as we do not measure indirect care staff in our study, a similar shift in private equity nursing homes would circumscribe the potential quality gains anticipated from increased CNA staffing and may help explain our results.
119 Deficienc ies remain a cause of concern as it is one significant quality differentiator between private equity nursing home and the control group. Not only private equity nursing homes had significantly higher deficiencies, what is particularly troubling is the posi tive association of deficiencies with every progressive year of equity ownership. Deficiencies are important as they a re the only indicator of a facility is adhering to federal regulations (Harrington, Zimmerman et al., 2000) Literature suggests that nurse staffing has a strong infl uence on deficiencies with Harrington et al. (2000) reporting that decreased RN intensity was associated with increase in total deficiencies. Similarly Kim et al. report that skill mix as measured by ratio of RNs to total licensed nursing staff is negatively correlated with total numb er of deficiencies (Kim et al., 2009) Therefore, it appears that decrease in RN staffing intensity and skill mix has had a negative impact on deficiencies. Financial Performance We hypothes ized that private equity nursing homes deliver better financial performance, as measured by operating and total margins, than the control group. The results fully support our hypothesis. Private equity nursing homes exhibit higher operating margin as well as total margin. Interestingly, both the margins are positively associated with progressive years of private equity ownership suggesting continued potential for improved financial performance. These findings are somewhat contrary to those reported in the p rivate equity literature where some studies suggest that improvement in private equity acquired firms is confined to first two or three years post acquisition (Long & Ravenscraft, 1993) We had anticipated that private equity nursing homes would improve their financial performance either by increasing revenues, lowering costs, or a combination thereof. As
120 an ticipated, private equity nursing homes indeed have higher operating revenues. One common route adopted in the nursing home industry to improve operating revenues is shift in patient census in favor of Medicare and private pay patients. No such shift in pa yer mix is noted in our study. Private equity nursing homes have higher acuity index than the control group. In presence of case mix adjustment, it would ordinarily result in higher operating revenues as this system reimburse s medical facilities according to resources consumed by particular group of patients (Arling & Daneman, 2002) ; facilities with higher acuity would be expected to have more resource intensive patients, and therefore enjoy higher reimbursement rates. However, Florida follows a cost based prospe ctive payment sy stem for Medicaid nursing homes with no case mix adjustment. Though there is no shift in Medicare census in private equity nursing homes, they may still be able to increase Medicare revenues by adopting two different strategies: increasing length of stay and increasing intensity of rehabilitation therapy. Increasing l ength of stay : In a departure from hospital PPS, which uses admissions as unit of payment, SNF PPS use the day as the unit of payment (White, 2003a) Ther efore, while PPS creates an incentive for SNFs to reduce their cost of pro viding a day of care, it includes no incentive for reducing the length of stay. In its 2002 re port to the Congress, MedPac (2002) expressed concern that providers may attempt to increase their profitability by increasing length of stay in fact, average length of stay in nursing homes increased by one day between 1999 and 200 0 (Medicare Payment Advisory Commission, 2003) Increasing i ntensity of rehabilitation t herapy : Under RUG III patients are initially classified into seven major categories one of which is rehabili tation. During the
121 initial RUG development process, patients in the rehabilitation category were deemed most resource intensive. (Fries et al., 1994) The reh abilitation category is furt her divided into five sub categories dependent upon the number of minutes of therapy received. With varying payment rates (Table5 1). Payment for rehabilitation therapy services under the current SNF PPS is based on the number of minutes of therapy which are predicated upon provider expectations rather than clinical characteristics of the patients (Centers for Medicare and Medicaid Services, 2007) fee schedule certain advantages, it also creates incentives for providers to classify patients in rehabilitation groups with the most favorable payment rates ( Centers for Medicare and Medicaid Services, 2007) In a report released in 2002, the GAO ( 2002) found that more patients were classified rehabilitation payment groups whi ch are the most profitable groups on a payment per minute of therapy bas is (Centers for Medicare and Medicaid Services, 2007) while in 2007 MedPac (2007) pointed out that due to skewed payment over time the number of beneficiaries receiving therapy has increased, as has the amount of therapy each be neficiary has received Similar results were reported by White in his study of the effect of PPS on rehabilitation thera py (White, 2003b) While we cannot provide conclusive evidence as we do not measure the relevant variables in this study, it can be speculated that private equity nursing homes may have improved their operating revenues by adjusting the length of stay and by providing increased rehabilitation therapy Private equity nursing homes also report significantly higher operating costs than the control group. One possible explanation is simply their higher acuity; if a facility has
122 more resource intensive patients, its patients cost would normally increase. On similar lines, if private equity nursing homes are providing higher amount of rehabilitative therapy, it would also lead to higher costs. If true, an interesting strategic approach as it would suggest that private equity nursing homes are willing to incur higher pa tient costs if it leads to a disproportionate increase in patient revenues. An interesting finding is that in the year prior to their acquisition, private equity nursing homes had significantly poorer operating margin while their total margin was superior to the control group. Operating margin reflects actual operational efficiency of the firm while total margins include relatively fixed costs such as depreciation, capital costs, and interest payments. Therefore, assuming that the financial performance of nursing homes would be reflected in their selling price, and considering the moribund status of the nursing home industry in that time period, private equity would potentially be able to acquire these nursing homes for a lower price. As private equity oper ates on the principle of making operational improvements, and specializes in financial distressed firms, it would be better placed to improve nursing home operating performance and thereby increase operating margins. Better total margins in these nursing h omes are an additional benefit as depreciation or interest payments are usually fixed costs. Additionally, as private equity deals are highly leveraged, and real estate was utilized to raise debt, a highly depreciated property --reflected in better total m argins --would be a particularly attractive target. It appears that private equity acquired nursing homes with potential for best financial results. Whether this selection is a product of a deliberative process or just providence is open to conjecture, but it would be reasonable to conclude that
123 private equity would be well positioned to understand balance sheets. Interestingly, contrary to general trends, private equity nursing homes had sharply higher non operating revenues in the year prior to their acqu isition. It may be possible that these nursing homes were selling assets, and registering capital gains in order to make their balance sheet more attractive. However, as the mean of non operating revenues is low, it should be interpreted conservatively. O verall it appears that there necessarily a tradeoff between quality and financial performance. Private equity nursing homes are able to deliver better financial performance without a significant decline in quality. While it is important to remain ca utious, these findings are certainly encouraging. Managerial Implications Management of private equity nursing homes may believe that quality is of secondary importance as long as financial goals are met and they are able to earn an adequate return on the ir investments. However, regulators are increasingly moving towards a regime which actively rewards good quality while penalizing healthcare facilities delivering poor quality of care. For instance, in 2007, CMS announced that it --avoidable conditions resulting from physician error or hospital deficiencies (Rosenthal, 2007) NHQI launched by CMS in 2001 aims to improve nursing home quality through a mixture of public reporting, penalties, and incentives for higher quality (Kissam et al., 2003) CMS value based purchasing demonstration project seeks to reward nursing homes delivering higher quality of care in four specific areas: staffing, resid ent outcomes, avoidable hospitalizations and reductions in deficiency citations (Centers for Medicare and
124 Medicaid, 2009a) Incidentally, deficiencies and staffing are two variables where private equity nursing homes differ most significantly from the control group in this study. Public reporting may yet pose a more formidable challenge to private equity nursing homes. Public reporting is p redic ated on the idea that consumers armed with information (Mukamel & Mushlin, 1998) would spur competition in the nursing home industry based on quality of care (Marshall et al., 2000) The need to address the vital challenge of dissemination of quality information has led to the development of nursing home compare website; information once hard to find is now available at the click of a mouse. In Florida the Office of Health Quality Assurance inspects each Florida nursing home and ass igns them ratings of Superior, Standard or C onditional indicating their co mpliance with minimum quality standards (Troyer & Thompson, 2004) With their less than savory reputation for quality, private equity nursing homes may be adversely affected as literature suggests that perception s of quality affect firm profitability. McGuire et al. ( 1990) reported that there was a strong correlation between firm quality as reported in the Fortune magazine annual survey of corporat e reputations and f uture profitability while Roberts & Dowling ( 2002) report that reputation may provide firms with sustained competitive advantage (Barney, 1991) leading to long term profitability. Moreover, Nelson et al (1992) has may explain 17 27% variation in hospital financial performance. Therefore, private equity nursing homes may face a financial imperative to invest in quality. However, addressing quality on ly partially solves the reputation conundrum for private equity nursing homes. In his seminal paper, Arrow ( 2001) argues that a key feature of the health care market is the presence of asymmetrical information between
125 ( 1980) propounded the Trust Hypothesis which argues that the presence of NFPs in certain sectors of the economy (e.g. health care) is explained by greater trust placed in them by consumers due to the attenuation of profit motive. Correspondin gly, in presence of asymmetrical information, consumers are less likely to trust profit driven enterprises. Consumer concerns are likely to be exacerbated in the case of private equity nursing homes due to their byzantine organizational structures and a d eliberate policy of obfuscation and non transparency. It has proved challenging for even regulators to navigate their maze of complex ownership structures (Agency for Healthcare Administration, 2007) The primary justification offered is huge liability costs. Or as a private equity executive put it: over almost any th and that the barriers erected by private equity were necessary for the survival of nursing home industry (Duhigg, 2007) While it may help them lower liability costs, non transparency damages the reputation and credibility of private equity nursing homes. Private equity has paid for its damaged reputation in the form of critical media reports, Congressional hearings, and public opprobrium. Therefore, it is advisable that management take s steps to assuage the concerns of different stakeholders. Even with reference to liability costs, literature suggests in many cases health care facilities admitt ing to medical errors have noticed a decline in lawsuits (Berlin, 2006; Sack, 2008; Wojcieszak, Banja, & Houk, 2006) and have benefited from consequent cost savings. Many states have issued laws prohibiting courts from admitting apologies as evidence in malpractice lawsuits (Berlin, 2006) ; a similar bill was introduced in United States Se nate in 2005 by (then) Senators Obama and Clinton
126 (Glendinning, 2005) A less conformational attitude towards patients and patient advocates, nursing home employees, and other stakeholders will serve private equity well. Finally, priva te equity nursing homes perform better on financial margins in our study but their operating costs are also significantly higher. Quality management literature posits a positive relationship between quality and lower costs. According to the Deming Chain, q uality improvement lowers waste (less rework, fewer mistakes, snags and delays) and thereby reduc es costs (Deming, 1986) In this scenario, quality improvement entails managing the production process with the goal of imple menting clinically proven processes of care. For instance, Frantz et al. ( 1995) showed that the implementation of a skin care protocol in a long term care facility reduced treatment costs. Some researchers have argued a business case for nursing home quality (Weech Maldonado et al., 2003a) but even a less ambitious approach linking higher quality to lower cost of production would prove financially beneficial to private equity nursing homes. Policy Implications Federal and state governments are concerned with nursing home performance as they are the principle payers of nursing home care and because it affects the well being of highly vulnerable citizens. The introduction of a novel form of ownership in the nursing home industry would justifiably raise regulatory concerns. Our results suggest that quality of care delivered at private equity nursing homes is largely similar to other investor owned nursing homes while financial performance is significantly better. However, there is still cause for concern.
127 Perhaps the most pressing issue is of transparency. The complicated ownership and operating structures limit an important incentive to improve quality. In their study of malpractice suits in Florida nursing homes, Troyer & Thompson ( 2004) report that chain level legal claims are associated with higher measurable quality of fir ms associated with that chain. It has been reported that in many cases lawyers simply gave up attempting to sue the private equity owners (Duhigg, 2007) Weisbord ( 1988) divides nursing homes quality measures in to Type I and Type II measures. Type I measures are easily quantifiable and are frequently the focus of nursing home inspections. Type II measures are less tangible, harder to measure, and are frequently unobserved by regulator s. Examples would include relationship with staff and neglect (Troyer & Thompson, 2004) Nursing homes may focus their resources on a Type I measures to avoid citations while Type II measures may be under resourced. As Type II measures are important for patient satisfaction, and yet unobserved by r egulators, litigation may reflect particularly blatant examples of problems with Type II measures (Troyer & Thompson, 2004) Correspondingly, nursing homes facing less threat of lawsuits may feel emboldened to divert more resources from Type II measures negati vely impacting patient satisfaction and quality though this deterioration would not be captured in administrative datasets. Stevenson and Grabowski ( 2008) argue CMS regulation of nursing home quality is not based on ownership, but reflects facility lev el outcomes and deficiencies. If the behavior of a facility is guided by its ownership, a facility based approach may be ineffectual. While it is true that this paradigm is applicable to non private equity chains as well, it becomes especially important in case of private equity as the complicated
128 ownership structures for instance, separation of ownership and licensure --creates legal walls between facilities, and the identification of a common owner may be challenging. It also protects them from other poss ible chain wide sanctions including termination from Medicare and Medicaid programs for criminal activities, and prosecution for false Medicare or Medicaid claims (Stevenson & Grabowski, 2008) Just like in the case of residents, the complicated ownersh ip structures handicap regulators and limit their policing tools. Therefore, regulators must ensure greater transparency in private equity nursing home deals particularly in relation to ownership. Serious thought needs to be given to an expanded regulatory (Stevenson & Grabowski, 2008) from facility to the chain level recognizing that chain is not merely an amalgamation of disparate units but represents a common governing and operational philosophy. Anoth er worrying aspect of private equity nursing homes is the short investment period typically not extending beyond 5 7 years. Private equity may adopt a different strategic outlook than long term investors as investments maturing over a longer time horizon w ould make little financial sense. We had stressed the importance of reputation to private equity nursing homes and how it may nudge them towards improving quality. An important caveat is in order here. Assuming a reasonable priori that improving quality of nursing homes may incur substantial initial costs, private equity may conclude that the investment yields little benefits as they are too far in the horizon. Even data driven efforts like Nursing Home Compare website are blunt tools and most nursing homes would be expected to congregate near the mean. Therefore, an excessive reliance on public reporting may be overly optimistic and the anticipated gains may not
129 materialize. Development of more granular measures of quality and sharper linkages between quali ty and reimbursement is required. The importance of a strong regulatory framework cannot be stressed enough. Two other concerns remain. First, private equity nursing homes deliver better financial performance in our study but they also carry a large amoun t of debt created to finance the acquisitions. Their future financial viability remains a matter of regulatory concern especially in an era of bleak government finances. Second, as noted, private equity nursing homes have introduced changes in nurse staffi ng prima facie structured to save costs. Nurse staffing is a primary target for cost savings if these nursing homes face future financial pressures. The role of minimum nurse staffing in improving nursing homes quality has been emphasized in the literature Hyer et al. ( 2009) recently reported that Florida nursing homes quality has substantially improved since the imposition of minimum nurse staffing regulations in the state. Imposition of federal minimum nurse staffing levels may be helpful and something CMS should seriously consider. Limitations This study presents several limitations. First, staffing data are based on OSCAR data, which is self reported and it is not subject to regular audits. However, a recent study by Grabowski et al ( 2004) found a strong inter survey agreement between OSCAR and their own survey with respect to RN, LPN, and CNA full time equivalents (FTEs) data. Second, this study is limited to facilities with Medicare residents. This results in the exclusion of facilities that are exclusively private pay or Medicaid. However, the number of such nursing homes is limited. Third, our explanations for cost and revenues are speculative in nature in absence of relevant variables. For instance, we speculate that
130 the increase in private equity operating revenues may be explained by increase in length of stay and more re habilitative therapy. Since we do not measure either the length of stay or the rehab minutes, our argument may or may not reflect actual strategic choices adopted by private equity nursing homes. In turn, that may color the managerial and policy implicatio ns of our study. Fourth, nursing home deals are recent in nature; the full effects of private equity ownership may only be fully manifest in more recent data. Therefore, conclusions, particularly in reference to quality, should be drawn conservatively from our study. Finally, this study is restricted only to Florida which may make generalizations across states and at the national level difficult Conclusions Should private equity be allowed to invest in nursing homes? Any proposal to limit private equity t of deliberate malfeasance especially in an industry dominated by proprietary chains. And as we have noted, there is no significant difference in quality between private equity and other investor o wned nursing homes in our study. The focus therefore has to be on the twin issues of transparency and accountability. Transparency a nd accountability are inextricably linked. Without ensuring that nursing homes ownership is more transparent, accountabili ty cannot be fixed. And without a strengthened regulatory framework which attempts to move beyond facility level monitoring system, accountability would be limited. These are challenges which may be particularly relevant to private equity ownership but are by no means limited to them. A ddressing them would facilitate achieving what engages policymakers, regula tors, and patients alike: N ursing homes ensure quality care to their residents and
131 in instances where this expectation is belied effective tools are available to punish the guilty and to compensate the victims.
132 Table 5 1. Rehabilitation categories under PPS Rehabilitation Category Number of minutes of therapy Ultra High 720 Very high 500 High 325 Medium 150 Low 45
133 APPE NDIX B ACKGROUND TABLES Table A 1. T test of dependent quality variables Dependent variable N Mean P value RN hours PPD Equity 614 .25 .003 Non equity 2123 .27 LPN hours PPD Equity 614 .84 .002 Non equity 2123 .88 CNA hours PPD Equity 614 2.5 .5 Non equity 2123 2.5 Skill mix Equity 614 .23 .62 Non equity 2123 .23 Pressure sore prevention Equity 614 1.8 .06 Non equity 2123 1.73 Restorative ambulation Equity 614 .61 .48 Non equity 2123 .63 % Catheter Equity 33 2 .09 .05 Non equity 978 .09 % restraints Equity 614 .08 .18 Non equity 2109 .08 ADL 4 point decline Equity 614 .13 .06 Non equity 2115 .12 Bowel worsening Equity 610 .15 .009 Non equity 2083 .16 Pressure ulcer hi low prevalence Equity 614 .08 .88 Non equity 2109 .08 Total deficiencies Equity 614 9.1 .001 Non equity 2123 7.4
134 Table A 2. T test of dependent financial variables Dependent variable N Mean P value Operating revenue PPD Equity 614 191.2 .17 Non equ ity 2031 188.4 Non operating revenue PPD Equity 573 3.2 .06 Non equity 1966 2.1 Operating cost PPD Equity 614 169.7 .7 Non equity 2031 170.3 Non operating cost PPD Equity 614 18.3 .01 Non equity 2031 19.4 Operating margin Equity 614 .1 .001 Non equity 2031 .09 Total margin Equity 573 .02 .001 Non equity 1966 .008 Census Medicare Equity 614 .16 .001 Non equity 2031 .17 Census Medicaid Equity 614 .65 .001 Non equity 2031 .61 Census Other Equity 614 .17 001 Non equity 2031 .2 Acuity index Equity 614 11.6 .1 Non equity 2031 11.6
135 Table A 3. Chi square test of LPNs on contract by Equity Total 2123 614 2737 77.57 22.43 100.00 (Chi square test P value=.001) Table A 4. Chi square test of RNs on contract by Equity Percent Total 2123 614 2737 77.57 22.43 100.00 ( Chi square test: P value= .04 )
136 Table A 5. Chi square test of Actual harm citations by Equity Total 2086 612 2698 77.32 22.68 100.00 Frequency Missing = 39 ( Chi square test: P value=.13)
13 7 Table A 6. Operating revenue PPD : Model 1 Operating revenue PPD Coefficient Standard Error P value Equity owned .08 .01 .001 Year prior equity acquisition .02 .02 .2 Years post acquisition .03 .00 .001 % Medicare .8 .05 .001 % Medicaid .2 .04 .001 Acuity index .03 .00 .001 O ccupancy rate .6 .05 .01 Total residents .00 .00 .001 County per capita income .67 .85 .4 Metro .02 .03 .3 County population over 65 .08 .11 .4 Herfindahl index .06 .06 .2 Excess capacity .00 .00 .001 Table A 7. Operating revenue PPD: Model 2 Operating revenue PPD Coefficient Standard Error P value Equity owned .17 .01 .001 % Medicare .8 .05 .001 % Medicaid .2 .04 .001 Acuity index .03 .00 .001 Occupancy rate .6 .05 .001 Total residents .00 .00 .001 County per capita income .7 .8 .3 M etro .03 .03 .2 County population over 65 .08 .11 .4 Herfindahl index .06 .06 .2 Excess capacity .00 .00 .001
138 Table A 8 Operating cost PPD: Model 1 Operating cost PPD Coefficient Standard Error P value Equity owned .1 .02 .001 Year prior e quity acquisition .02 .02 .2 Years post acquisition .00 .01 .8 Squared year .00 .00 .03 % Medicare .8 .05 .001 % Medicaid .2 .04 .001 Acuity index .02 .00 .001 Occupancy rate .7 .05 .001 Total residents .00 .00 .001 County per capita income .15 .8 .001 Metro .00 .03 .2 County population over 65 .14 .11 .2 Herfindahl index .08 .06 .1 Excess capacity .00 .00 .001 Table A 9 Operating cost PPD: Model 2 Operating cost PPD Coefficient Standard Error P value Equity owned .14 .01 .001 % Medi care .8 .05 .001 % Medicaid .2 .04 .001 Acuity index .03 .00 .001 Occupancy rate .7 .05 .001 Total residents .00 .00 .001 County per capita income .15 .8 .07 Metro .03 .03 .2 County population over 65 .14 .11 .2 Herfindahl index .09 .00 .001 Excess capacity .00 .00 .001
139 Table A 10. OLS of non operating revenues PPD: Model 1 Non operating revenues PPD Coefficient Standard Error P value Equity owned .1 .15 .001 Year prior equity acquisition 1.3 .17 .001 Years post acquisition .55 .1 .001 Squared year .08 .02 .001 % Medicare .12 .4 .7 % Medicaid .16 .3 .001 Acuity index .00 .02 .9 Occupancy rate 1.2 .33 .001 Total residents .00 .00 .01 County per capita income .00 .64 .06 Metro .35 .23 .1 County population over 65 .5 .8 .4 Herfindahl index .3 .4 .4 Excess capacity .00 .6 .009 Table A 11 OLS of non operating revenues PPD: Model 2 Non operating revenues PPD Coefficient Standard Error P value Equity owned .8 .1 .001 % Medicare .17 .4 .6 % Medicaid 1.7 .3 .001 Acuity index .00 .02 .8 Occupancy rate 1.2 .34 .001 Total residents .00 .00 .02 County per capita income .00 .6 .07 Metro .3 .2 .1 County population over 65 .6 .8 .4 Herfindahl index .3 .4 .3 Excess capacity .00 .00 .7
140 Table A 12. OL S of non operating costs PPD: Model 1 Non operating costs PPD Coefficient Standard Error P value Equity owned .02 .04 .5 Year prior equity acquisition .004 .04 .9 Years post acquisition .16 .02 .001 Squared year .02 .00 .001 % Medicare .05 .11 .6 % Medicaid .35 .08 .001 Acuity index .01 .00 .041 Occupancy rate .8 .1 .001 Total residents .00 .00 .2 County per capita income .3 .2 .09 Metro .00 .07 .9 County population over 65 .2 .2 .4 Herfindahl index .009 .1 .9 Excess capacity .00 .00 01 Table A 13 OLS of non operating costs PPD: Model 2 Non operating costs PPD Coefficient Standard Error P value Equity owned .23 .02 .001 % Medicare .08 .11 .4 % Medicaid .4 .08 .001 Acuity index .01 .00 .01 Occupancy rate .8 .1 .001 Total resi dents .00 .00 .1 County per capita income .3 .2 .1 Metro .02 .07 .7 County population over 65 .19 .2 .5 Herfindahl index .001 .15 .9 Excess capacity .00 .00 .06
141 Table A 14. OLS of operating margin : Model 1 Operating margin Coefficient Sta ndard Error P value Equity owned .01 .00 .1 Year prior equity acquisition .4 .01 .001 Years post acquisition .03 .00 .001 Squared year .00 .00 .001 % Medicare .00 .02 .9 % Medicaid .01 .01 .3 Acuity index .00 .00 .3 Occupancy rate .1 .02 .001 Total residents .00 .00 .2 County per capita income .6 .3 .05 Metro .00 .01 .9 County population over 65 .05 .04 .2 Herfindahl index .01 .00 .01 Excess capacity .04 .03 .2 Table A 15 OLS of operating margin : Model 2 Operating margin Coefficient Standard Error P value Equity owned .03 .00 .001 % Medicare .00 .02 .9 % Medicaid .02 .01 .1 Acuity index .00 .00 .1 Occupancy rate .15 .01 .001 Total residents .00 .00 .1 County per capita income .6 .3 .07 Metro .00 .01 .8 County population over 65 .04 .04 .3 Herfindahl index .01 .02 .4 Excess capacity .00 .00 .002
142 Table A 16. OLS of total margin : Model 1 Total margin Coefficient Standard Error P value Equity owned .008 .00 .3 Year prior equity acquisition .03 .01 .001 Years post acquisition .007 .00 .01 % Medicare .05 .02 .02 % Medicaid .03 .01 .04 Acuity index .00 .00 .06 Occupancy rate .16 .02 .001 Total residents .0002 .00 .01 County per capita income .4 .3 .2 Metro .008 .01 .5 County population over 65 .009 .0 4 .8 Herfindahl index .009 .00 .001 Excess capacity .001 .00 .001 Table A 17 OLS of total margin : Model 2 Total margin Coefficient Standard Error P value Equity owned .01 .00 .005 % Medicare .06 .02 .02 % Medicaid .03 .01 .04 Acuity index .003 00 .06 Occupancy rate .16 .02 .001 Total residents .00 .00 .001 County per capita income .4 .3 .1 Metro .008 .01 .5 County population over 65 .009 .04 .8 Herfindahl index .01 .02 .6 Excess capacity .001 .00 .001
143 Table A 18 Logit of % Medicare : Model 1 % Medicare^ Coefficient Standard Error P value Equity owned 1.0 .2 .8 Year prior equity acquisition 1.0 .2 .9 Years post acquisition 1.0 .05 .2 Acuity index 1.0 .04 .03 Occupancy rate 1.2 .79 .7 Total residents 1.0 .02 .9 County per capita income 1.0 .00 .9 Metro 1.0 .43 .9 County population over 65 3.1 4.9 .4 Herfindahl index .7 .6 .7 Excess capacity .98 .01 .2 ^ Odds ratio Table A 19 Logit of % Medicare : Model 2 % Medicare^ Coefficient Standard Error P value Equity owne d 1.1 .17 .27 Acuity index 1.4 .04 .3 Occupancy rate 1.28 .8 .6 Total residents 1.0 .00 .9 County per capita income 1.0 .00 .9 Metro 1.0 .4 .9 County population over 65 2.1 4.9 .4 Herfindahl index .74 .65 .7 Excess capacity .98 .01 .2 ^ Odds ratio
144 Table A 20 Logit of % Medicaid : Model 1 % Medicaid^ Coefficient Standard Error P value Equity owned 1.0 .1 .5 Year prior equity acquisition 1.0 .18 .8 Years post acquisition .94 .03 .1 Acuity index .98 .03 .6 Occupancy rate .5 .2 .2 T otal residents 1.0 .00 .7 County per capita income .99 .00 .3 Metro .93 .3 .8 County population over 65 .1 .2 .1 Herfindahl index 2.1 1.5 .3 Excess capacity 1.0 .00 .7 ^ Odds ratio Table A 21 Logit of % Medicaid : Model 2 % Medicaid^ Coefficient Sta ndard Error P value Equity owned .95 .1 .6 Acuity index .98 .03 .5 Occupancy rate .5 .2 .1 Total residents 1.0 .00 .7 County per capita income .99 .09 .3 Metro .9 .3 .9 County population over 65 .17 .23 .1 Herfindahl index 2.1 1.5 .2 Excess capaci ty 1.0 .00 .6 ^ Odds ratio
145 Table A 22 Logit of % Other : Model 1 % Other^ Coefficient Standard Error P value Equity owned .78 .1 .2 Year prior equity acquisition .88 .23 .6 Years post acquisition 1.0 .06 .5 Acuity index .98 .04 .7 Occupancy rate 2.7 1.6 .09 Total residents .99 .00 .2 County per capita income 1.0 .00 .2 Metro 1.1 .48 .7 County population over 65 4.4 6.3 .2 Herfindahl index .4 .4 .3 Excess capacity 1.0 .00 .4 ^ Odds ratio Table A 23 Logit of % Other : Model 2 % Other^ Coefficient Standard Error P value Equity owned .8 .13 .3 Acuity index .9 .04 .8 Occupancy rate 2.7 1.6 .09 Total residents .99 .00 .2 County per capita income 1.0 .00 .2 Metro 1.1 .4 .7 County population over 65 4.4 6.3 .2 Herfindahl index .4 .39 .3 Excess capacity 1.0 .00 .4 ^ Odds ratio
146 Table A 24 OLS of Acuity index : Model 1 Total margin Coefficient Standard Error P value Equity owned .06 .09 .5 Year prior equity acquisition .02 .11 .8 Years post acquisition .07 .02 .002 % Medicare .8 .29 .005 % Medicaid .1 .2 .6 Occupancy rate .21 .2 .4 Total residents .001 .00 .3 County per capita income .00 .04 .01 Metro .01 .16 .9 County population over 65 .08 .6 .8 Herfindahl index .77 .3 .01 Excess capacity .003 .00 .3 Tab le A 25 OLS of Acuity index : Model 2 Total margin Coefficient Standard Error P value Equity owned .24 .06 .001 % Medicare .8 .29 .004 % Medicaid .06 .2 .7 Occupancy rate .22 .26 .4 Total residents .00 .00 .2 County per capita income .00 .04 .02 Me tro .001 .16 .9 County population over 65 .07 .6 .9 Herfindahl index .7 .3 .01 Excess capacity .00 .00 .2
147 Table A 26. OLS of RN hours PPD: Model1 RN hours PPD Coefficient Standard Error P value Equity owned .09 .04 .05 Year prior equity acq uisition .07 .05 .23 Years post acquisition .02 .01 .04 % Medicare .49 .15 .001 % Medicaid .49 .11 .001 Acuity index .01 .01 .319 Occupancy rate .15 .14 .28 Total residents .00 .00 .8 County per capita income .211 .26 .4 Metro .09 .09 .3 C ounty population over 65 .61 .35 .08 Herfindahl index .11 .18 .5 Excess capacity .004 .00 .06 Table A 27. OLS of RN hours PPD: Model2 RN hours PPD Coefficient Standard Error P value Equity owned .12 .03 .001 % Medicare 51 .15 .001 % Medicaid .49 .11 .001 Acuity index .011 .01 .26 Occupancy rate .15 .14 .28 Total residents .00 .00 .8 County per capita income .28 .26 .4 Metro .09 .09 .3 County population over 65 .62 .35 .07 Herfindahl index .11 .18 .5 Excess capacity .004 .00 .04
148 Tab le A 28 OLS of LPN hours PPD: Model2 LPN hours PPD Coefficient Standard Error P value Equity owned .1 .02 .001 Year prior equity acquisition .01 .01 .4 Years post acquisition .01 .00 .06 % Medicare .46 .06 .001 % Medicaid .01 .04 .8 Acuity index .01 .00 .001 Occupancy rate .45 .01 .001 Total residents .00 .00 .912 County per capita income .28 .1 .8 Metro .00 .03 .9 County population over 65 .18 .15 .18 Herfindahl index .07 .07 .3 Excess capacity .00 .00 .001 Table A 29 OLS of LPN hours PPD: Model 2 LPN hours PPD Coefficient Standard Error P value Equity owned .12 .01 .001 % Medicare .46 .06 .001 % Medicaid .009 .04 .84 Acuity index .01 .00 .001 Occupancy rate .45 .06 .001 Total residents .00 .00 .874 County per capita income .31 .1 .7 Metro .00 .03 8 County population over 65 .18 .14 .18 Herfindahl index .07 .07 .29 Excess capacity .00 .00 .001
149 Table A 30 OLS of CNA hours PPD: Model 1 CNA hours PPD Coefficient Standard Error P value Equity owned .32 .04 .001 Year pri or equity acquisition .13 .05 .02 Years post acquisition .03 .01 .01 % Medicare 1.0 .15 .001 % Medicaid .02 .1 .8 Acuity index .04 .01 .001 Occupancy rate .45 .13 .001 Total residents .00 .00 .001 County per capita income .38 .17 .03 Metro .04 .06 .4 County population over 65 .5 .2 .01 Herfindahl index .08 .12 .5 Excess capacity .01 .00 .001 Table A 31 OLS of CNA hours PPD: Model 2 CNA hours PPD Coefficient Standard Error P value Equity owned .42 .03 .001 % Medicare 1.0 .15 .001 % Me dicaid .004 .1 .9 Acuity index .04 .01 .001 Occupancy rate .46 .13 .001 Total residents .00 .00 .001 County per capita income .39 .17 ;02 Metro .05 .06 .4 County population over 65 .55 .22 .01 Herfindahl index .08 .12 .5 Excess capacity .01 .0 0 .001
150 Table A 32 OLS of skill mix : Model 1 Skill mix Coefficient Standard Error P value Equity owned ..04 .00 .001 Year prior equity acquisition .01 .01 11 Years post acquisition .00 .00 .001 % Medicare .13 .02 .001 % Medicaid .02 .001 .001 Acuity index .004 .00 .02 Total residents .00 .00 .31 County per capita income .12 .54 .8 Metro .01 .01 .3 County population over 65 .1 .07 .1 Herfindahl index .01 .03 .6 Excess capacity .001 .00 .001 Table A 33 OLS of skill mix : Model 2 Ski ll mix Coefficient Standard Error P value Equity owned .04 .00 .001 % Medicare .13 .02 .001 % Medicaid .05 .02 .01 Acuity index .00 .00 .01 Total residents .00 .00 .2 County per capita income .1 .5 .8 Metro .01 .01 .3 County population over 65 .01 .03 .1 Herfindahl index .01 .03 .6 Excess capacity .00 .05 .001
151 Table A 34 Logistic regression of LPNs on contract: Model 1 LPNs on contract^ Coefficient Standard Error P value Equity owned 6.0 1.6 .001 Year prior equity acquisition 2.2 .72 .01 Years post acquisition .67 .05 .001 % Medicare .79 .83 .8 % Medicaid 2.7 2.0 .1 Acuity index .98 .06 .7 Occupancy rate 1.2 1.08 .8 Total residents 1.0 .00 .2 County per capita income .99 .00 .6 Metro 1.2 .55 .6 County population over 65 34 52 .02 Herfindahl index .66 .5 .6 Excess capacity 1.02 .01 .1 ^ Odds ratio Table A 35 Logistic regression of LPNs on contract: Model 2 LPNs on contract^ Coefficient Standard Error P value Equity owned 2.3 .45 .001 % Medicare .79 .83 .8 % Medicaid 3.4 2.5 .09 Acuity index .9 .06 .5 Occupancy rate 1.1 1.0 .8 Total residents 1.0 .00 .2 County per capita income .99 .00 .5 Metro 1.3 .59 .5 County population over 65 35.5 54.0 .02 Herfindahl index .66 .58 .6 Excess capacity 1.0 .01 .04 ^ Odds ratio
152 Table A 36 Logistic regression of RNs on contract: Model 1 RNs on contract^ Coefficient Standard Error P value Equity owned 2.1 .84 .05 Year prior equity acquisition 1.0 .5 .9 Years post acquisition .7 .1 .04 % Medicare .4 .5 .5 % Medicaid .6 .6 6 Acuity index 1.1 .11 .1 Occupancy rate 3.0 3.9 .28 Total residents .00 .00 .8 County per capita income .211 .26 .4 Metro .09 .09 .3 County population over 65 .61 .35 .08 Herfindahl index .11 .18 .5 Excess capacity .004 .00 .06 ^ Odds ratio Ta ble A 37 Logistic regression of RNs on contract: Model 2 RNs on contract^ Coefficient Standard Error P value Equity owned 1.1 .35 .5 % Medicare .4 .6 .5 % Medicaid .72 .74 .75 Acuity index 1.1 .11 .18 Occupancy rate 3.0 3.9 .38 Total residents 1.00 .00 .23 County per capita income 1.00 .00 .23 Metro .62 .36 .4 County population over 65 4.2 .89 .4 Herfindahl index 1.57 1.76 .6 Excess capacity 1.04 .01 .007 ^ Odds ratio
153 Table A 38. OLS of Pressure sore prevention: Model 1 Pressure sore prevent ion Coefficient Standard Error P value Equity owned .02 .01 .09 Year prior equity acquisition .003 .01 .8 Years post acquisition .01 .00 .001 % Medicare .23 .04 .001 % Medicaid .0 .3 .001 Acuity index .03 .00 .001 Occupancy rate .06 .03 .07 Tot al residents .00 .00 .5 County per capita income .4 .6 .9 Metro .03 .02 .1 County population over 65 .1 .08 .04 Herfindahl index .03 .04 .4 Excess capacity .00 .00 .05 Table A 39. OLS of Pressure sore prevention: Model 2 RNs on contract^ Coeffi cient Standard Error P value Equity owned .02 .00 .02 % Medicare .2 .0 .001 % Medicaid .1 .03 .001 Acuity index .03 .00 .001 Occupancy rate .06 .03 .07 Total residents .00 .00 .5 County per capita income .7 .6 .9 Metro .03 .02 .1 County popula tion over 65 .1 .08 .04 Herfindahl index .03 .04 .4 Excess capacity .00 .00 .02
154 Table A 40. OLS of Restorative ambulation: Model 1 Restorative ambulation Coefficient Standard Error P value Equity owned .04 .04 .2 Year prior equity acquisition .06 .04 .1 Years post acquisition .3 .01 .002 % Medicare .1 .1 .3 % Medicaid .09 .09 .2 Acuity index .02 .00 .01 Occupancy rate .6 .1 .001 Total residents .00 .00 .01 County per capita income .5 .2 .01 Metro .07 .07 .3 County population over 65 .2 .2 .4 Herfindahl index .1 .1 .3 Excess capacity .00 .00 .8 Table A 41. OLS of Restorative ambulation: Model 2 RNs on contract^ Coefficient Standard Error P value Equity owned .00 .02 .8 % Medicare .1 .1 .3 % Medicaid .08 .09 .3 Acuity index .02 .00 .01 Occupancy rate .5 .1 .001 Total residents .00 .00 .01 County per capita income .5 .2 .01 Metro .08 .07 .2 County population over 65 .2 .2 .3 Herfindahl index .12 12 .3 Excess capacity .00 .00 .7
155 Table A 42. Logit of use of restra ints: Model 1 Use of restraints^ Coefficient Standard Error P value Equity owned 1.08 .2 .7 Year prior equity acquisition .9 .3 .8 Years post acquisition .9 .07 .6 % Medicare 1.0 .97 .9 % Medicaid 1.2 .83 .7 Acuity index 1.0 .06 .2 Occupancy rate .9 8 .8 .9 Total residents 1.0 .00 .3 County per capita income .99 .99 .5 Metro 1.1 .5 .8 County population over 65 2.1 4.2 .7 Herfindahl index .93 .99 .9 Excess capacity .99 .01 .9 ^ Odds ratio Table A 43. Logit of use of restraints: Model 2 Use of re straints^ Coefficient Standard Error P value Equity owned 1.0 .2 .8 % Medicare 1.0 .9 .9 % Medicaid 1.2 .8 .7 Acuity index 1.0 .06 .3 Occupancy rate .9 .8 .9 Total residents 1.0 .00 .3 County per capita income .99 .00 .5 Metro 1.1 .5 .8 County pop ulation over 65 2.1 4.2 .7 Herfindahl index .9 .9 .9 Excess capacity 1.0 .01 .99 Odds ratio
156 Table A 44. Logit of use of catheters: Model 1 Use of catheters^ Coefficient Standard Error P value Equity owned 1.3 .5 .4 Year prior equity acquisition 1 .0 .3 .4 Years post acquisition .9 .1 .4 % Medicare 1.8 2.9 .7 % Medicaid .63 .7 .6 Acuity index 1.0 .1 .7 Occupancy rate .7 1.0 .8 Total residents 1.0 .00 .4 County per capita income .99 .00 .4 Metro 1.0 .6 .8 County population over 65 1.8 3.5 .7 Herfindahl index .8 .9 .8 Excess capacity 1.0 .02 .8 ^ Odds ratio Table A 45 Logit of use of catheters: Model 2 Use of catheters^ Coefficient Standard Error P value Equity owned 1.0 .3 .7 % Medicare 1.8 2.8 .7 % Medicaid .6 .7 .7 Acuity index 1.0 .1 .7 Occupancy rate .7 1.0 .8 Total residents 1.0 .00 .4 County per capita income .99 .00 .3 Metro 1.0 .6 .8 County population over 65 1.9 3.7 .7 Herfindahl index .8 .9 .8 Excess capacity 1.0 .02 .8 ^Odds ratio
157 Table A 46 Logit of ADL 4 point decline: Model 1 ADL 4 point decline^ Coefficient Standard Error P value Equity owned 1.0 .2 .8 Year prior equity acquisition 1.1 .3 .7 Years post acquisition .9 .07 .7 % Medicare 1.8 1.5 .4 % Medicaid .65 .38 .4 Acuity index 1.0 .05 .9 Occupancy ra te .7 1.0 .8 Total residents 1.0 .00 .9 County per capita income .99 .00 .7 Metro 1.0 .4 .9 County population over 65 1.0 1.4 .9 Herfindahl index .9 .8 .9 Excess capacity 1.0 .01 .8 ^Odds ratio Table A 47 Logit of ADL 4 point decline: Model 2 ADL 4 point decline^ Coefficient Standard Error P value Equity owned .9 .2 .8 % Medicare 1.8 1.5 .4 % Medicaid .6 .3 .4 Acuity index 1.0 .05 .9 Occupancy rate .9 .7 .9 Total residents .99 .00 .9 County per capita income .99 .00 .7 Metro 1.0 .4 .9 Coun ty population over 65 1.0 1.5 .9 Herfindahl index .9 .7 .9 Excess capacity 1.0 .01 .8 ^Odds ratio
158 Table A 48. Logit of bowel decline: Model 1 Bowel decline^ Coefficient Standard Error P value Equity owned 1.0 .2 .8 Year prior equity acquisition 1.1 .3 .6 Years post acquisition .9 .07 .7 % Medicare 1.2 .9 .8 % Medicaid .7 .4 .5 Acuity index 1.0 .05 .4 Occupancy rate .9 .7 .9 Total residents 1.0 .00 .9 County per capita income .99 .00 .7 Metro 1.0 .4 .9 County population over 65 1.0 1.4 9 Herfindahl index .9 .7 .9 Excess capacity 1.0 .01 .6 ^ Odds ratio Table A 49. Logit of bowel decline: Model 2 Bowel decline^ Coefficient Standard Error P value Equity owned .9 .2 .7 % Medicare 1.2 .9 .8 % Medicaid .7 .4 .5 Acuity index 1.0 .05 .4 Occupancy rate .9 .7 .9 Total residents 1.0 .00 .7 County per capita income 1.0 .00 .7 Metro 1.0 .4 .8 County population over 65 1.1 1.5 .9 Herfindahl index .9 .7 .9 Excess capacity 1.0 .01 .6 ^ Odds ratio
159 Table A 50. Negative binomial regression of total deficiencies: Model 1 Total deficiencies Coefficient Standard Error P value Equity owned .1 .05 .04 Year prior equity acquisition .18 .06 .004 Years post acquisition .03 .01 .04 % Medicare .005 .18 .9 % Medicaid .11 .12 .3 Acuity index .0 0 .01 .4 Occupancy rate .55 .15 .001 Total residents .002 .00 .001 County per capita income .00 .00 .001 Metro .07 .08 .3 County population over 65 .3 .2 .2 Herfindahl index .03 .15 .8 Excess capacity .00 .00 .7 Table A 51. Negative binomial regression of total deficiencies: Model 2 Total deficiencies Coefficient Standard Error P value Equity owned .13 .03 .001 % Medicare .01 .18 .9 % Medicaid .12 .12 .3 Acuity index .01 .01 .3 Occupancy rate .5 .15 .01 Total residents .00 .00 .001 Cou nty per capita income .00 .00 .001 Metro .07 .08 .3 County population over 65 .3 .28 .2 Herfindahl index .03 .15 .8 Excess capacity .00 .00 .7
160 Table A 52. Logistic regression of actual harm citation: Model 1 Total deficiencies^ Coefficient Stand ard Error P value Equity owned ..6 .2 .1 Year prior equity acquisition 1.6 .4 .08 Years post acquisition 1.1 .1 .1 % Medicare .66 .55 .6 % Medicaid 1.9 1.0 .2 Acuity index .9 .05 .2 Occupancy rate .93 .6 .9 Total residents .99 .00 .1 County per ca pita income 1.0 .00 .001 Metro .82 .2 .5 County population over 65 5.5 5.8 .1 Herfindahl index 1.4 .8 .5 Excess capacity 1.0 .01 .4 ^Odds ratio Table A 53. Logistic regression of actual harm citation: Model 2 Total deficiencies^ Coefficient Standard Error P value Equity owned .8 .1 .001 % Medicare .7 .6 .9 % Medicaid 2.0 1.1 .3 Acuity index .93 .05 .3 Occupancy rate .96 .6 .01 Total residents .99 .00 .001 County per capita income 1.0 .00 .001 Metro .83 .25 .3 County population over 65 5.6 5.9 .2 Herfindahl index 1.4 .8 .8 Excess capacity 1.0 .01 .7 ^ Odds ratio
161 Table A 54. OLS of pressure sore h/l risk prevalence : Model 1 Total deficiencies^ Coefficient Standard Error P value Equity owned .01 .03 .6 Year prior equity acquisition 01 .03 .6 Years post acquisition .00 .00 .8 % Medicare .4 .1 .001 % Medicaid .28 .07 .001 Acuity index .02 .00 .001 Occupancy rate .03 .09 .6 Total residents .00 .00 .7 County per capita income .15 .14 .9 Metro .11 .05 .03 County population o ver 65 .08 .18 .6 Herfindahl index .4 .1 .001 Excess capacity .00 .00 .001 Table A 55. OLS of pressure sore h/l risk prevalence : Model 2 Total deficiencies^ Coefficient Standard Error P value Equity owned .01 .02 .5 % Medicare .4 .1 .001 % Medica id .2 .07 .001 Acuity index .02 .00 .001 Occupancy rate .03 .09 .6 Total residents .00 .00 .7 County per capita income .16 .14 .9 Metro .11 .05 .03 County population over 65 .08 .18 .6 Herfindahl index .4 .1 .001 Excess capacity .00 .00 .01
162 Figure A 1 Nursing home deficiencies: scope and severity
163 Table A 56. Correlation matrix % Medicare % Medicaid MMdicaid Acuity Index Occupancy rate Total residents Per c apita Income M etro Population 65 + Excess capacity Herf in dahl % Medicare 1.000 % Medicaid 0.673 1 0 0 0 Acuity Index 0.058 0.049 1 .0 0 Occupancy rate 0.212 0.152 0 074 1 0 0 0 Total res idents 0.042 0.181 0.044 0 .238 1 0 0 0 Per capita Income 0.057 0.259 0 0 58 0 073 0 054 1 0 0 0 Metro 0.043 0 169 0 0 39 0 0 02 0 056 0 360 1 0 0 0 Population 65+ 0 1 2 3 0 226 0 006 0 093 0 07 0 0 309 0 0 31 1 0 0 0 Excess capacity 0 0 79 0 010 0 027 0 320 0 0 58 0 099 0 077 0 068 1 0 0 0 Herfindahl Index 0 1 0 4 0 193 0 064 0 0 12 0 058 0 398 0 689 0 1 43 0 0 72 1 0 0 0
164 LIST OF REFERENCES 1. Abt Associates. (2004). National nursing home quality measures. Users manual (v1.2), Cambridge: A bt Associates. 2. Agency for Healthcare Administration. (2005). Nursing Home Cost and Payment Rates : Agency for Healthcare A dministration. 3. Age ncy for Healthcare Administration. (2007). Long Term Care Review: Florida Nursing Homes Regulation, Quality, Ownership, and Reimbursement : Agency for Healthcare A dministration. 4. Aiken, L. H., Clarke, S. P., & Sloane, D. M. (2002). Hospital staffing, org anization, and quality of care: Cross national findings. Nurs Outlook, 50 (5), 187 194. 5. Aiken, L. H., Clarke, S. P., Sloane, D. M., Sochalski, J., & Silber, J. H. (2002). Hospital nurse staffing and patient mortality, nurse burnout, and job dissatisfact ion. JAMA, 288 (16), 1987 1993. 6. Almazan, A., Hartzell, J. C., & Starks, L. T. (2005). Active Institutional Shareholders and Costs of Monitoring: Evidence from Executive Compensation. Financial Management, 34 (4), 5 34. 7. American City Business Journals (2000). Tough times for nursing home operator Atlanta: American City Busi ness Journals. 8. Anderson, M. A. (1986). Nursing Home Colossus Buys Most of Computran Business Journal 9. Anderson, R. A., Hsieh, P. C., & Su, H. F. (1998). Resource allocati on and resident outcomes in nursing homes: comparisons between the best and worst. Res Nurs Health, 21 (4), 297 313. 9 Arling, G., & Daneman, B. (2002). Nursing home case mix reimbursement in Mississippi and South Dakota. Health Serv Res, 37 (2), 377 395. 10 Arrow, K. J. (2001). Uncertainty and the Welfare Economics of Medical Care. Journal of Health Politics, Policy & Law, 26 (5), 33p. 12 Atrill, P. (2008). Private equity industry. Retrieved 04/27/2008, from http://www.acca.co.uk/students/publications/finance_matters/archive/2007/73/30 25240 13 Baker, C. M., Messmer, P. L., Gyurko, C. C., Domagala, S. E., Conly, F. M., Eads, T. S., et al. (2000). Hospital ownership, performance, and outcomes: assessing the state of the science. J Nurs Adm, 30 (5), 227 240.
165 14 Banaszak Holl., J., Berta., W. B., Bowman., D. M., C., J. A., & Mitchell., B. W. (2002). The rise of human service chains: antecedents to acquisitio ns and their effects on the quality of care in US nursing homes. Managerial and Decision Economics, 23 (4 5), 261 282. 15 Barney, J. (1991). Firm Resources and Sustained Competitive Advantage. Journal of Management, 17 (1), 99 120. 16 Bassing, T. (2003). Beverly Alabama selling 7 of 21 nursing homes. Birmingham Business Journal, 20 17 Baum, J. A. C. (1999). The rise of chain nursing homes in Ontario. Social Forces 543 584. 19. Baysinger, B. D., & Butler, H. N. (1985). Corporate Governance and the Boa rd of Directors: Performance Effects of Changes in Board Composition. J Law Econ Organ, 1 (1), 101 124. 18 Bazzoli, G. J., & Chan, B. (2000). The Financial Performance of Hospitals Belonging to Health Networks and Systems. Inquiry (00469580), 37 (3), 234. 1 9 Berlin, L. (2006). Will Saying "I'm Sorry" Prevent a Malpractice Lawsuit? AJR Am J Roentgenol, 187 (1), 10 15. 20 Binns, G. S. (1991). The relationship between quality, cost, and market share in hospitals. Topics in Healthcare financing 21 Birger, J. (1998, 07/15). Health care gets hot on Wall St Crain's New York Business, p. 1, 22 Blundell Wignall, A. (2007). The Private Equity Boom: Causes and Policy Issues : Organization of Economic Cooperation and Deve lopment. 23 Bostick, J. E., Rantz, M. J., Flesner, M. K., & Riggs, C. J. (2006). Systematic review of studies of staffing and quality in nursing homes. J Am Med Dir Assoc, 7 (6), 366 376. 24 Bourbonniere, M., Feng, Z., Intrator, O., Angelelli, J., Mor, V., & Zinn, J. S. (2006). The use of con tract licensed nursing staff in U.S. nursing homes. Med Care Res Rev, 63 (1), 88 109. 25 Bowers, B. J., Esmond, S., & Jacobson, N. (2000). The relationship between staffing and quality in long term care facilities: exploring the views of nurse aides. J Nu rs Care Qual, 14 (4), 55 64; quiz 73 55. 26 Bradford, H. B. (Ed.). (1986). For Profit Enterprise in Health Care Washinton D.C National Academy Press.
166 27 Brubake, B. (2000, 03/24). Md. Nursing Home Operator Genesis to Revamp Finances Washington Post 2 8 Buchan, J., & Poz, D. R. M. (2002). Skill mix in the health care workforce: reviewing the evidence (No. 80). Geneva: World Health Organization. 29 Bull, I. (1989). Financial Performance of Leveraged Buyouts: An Empirical Analysis. Journal of Busine ss Venturing, 4 263 279. 30 Business Wire. (2002). Genesis Health Ventures Emerges From Bankruptcy Business Wire 31 Business Wire. (2004). Beverly Subsidiary Completes Hospice USA Acquisition Business Wire 32 Business Wire. (2005a). Beverly Enter prises Responds to 13 D Filing Business Wire 33 Business Wire. (2005b). Formation Capital Mails Proxy Statement to Beverly Enterprises Shareholders to Begin Solicitin g Votes for Its Board Nominees. Business Wire 34 Business Wire. (2006a). Beverly Ente rprises Completes Merger With Pearl Senior Care. Business Wire 35 Business Wire. (2006b). Genesis HealthCare Announces Amended Merger Agreement with Formation Capital and JER Partners Business Wire 36 Castle, N. G., & Lowe, T. J. (2005). Report cards and nursing homes. Gerontologist, 45 (1), 48 67. 37 Castle, N. G., & Mor, V. (1998). Physical restraints in nursing homes: a review of the literature since the Nursing Home Reform Act of 1987. Med Care Res Rev, 55 (2), 139 170; discussion 171 136. 38 Ca therine, H., & Phillips, C. D. (1986). The Changing Structure of the Nursing Home Industry and the Impact of Ownership on Quality, Cost, and Access : Institut e of Medicine. 39 Center for Disease Control. (2004). The National Home Survey 2004: Overview Wa s hington D.C. Center for Disease Control 40 Centers for Medicare and Medicaid. (2009a). Medicare Demonstration Show Paying for Quality Health Care Pays off Retrieved 11/16/2009. from http://www2.cms.hhs.gov/apps/media/press/release.asp?Counte r=3495&intNum PerPage=10&checkDate=&checkKey=&srchType=1&numDays=3500&srchOpt= 0&srchData=&keywordType=All&chkNewsType=1%2C+2%2C+3%2C+4%2C+5 &intPage=&showAll=&pYear=&year=&desc=false&cboOrder=date
167 41 Centers for Medicare and Medicaid. (2009b). Nursing Ho me Compare (Publication. Retrieved 11/20/2009, from Centers for Medicare and Medicaid: http://www.medicare.gov/nhcompare/ 42 Centers for Medicare and Medicaid Services. (2008a). Nursing Home Quality Initia tive [Electronic Version]. Retrieved 08/08/2010, from http://www.cms.gov/NursingHomeQualityInits/10_NHQIQualityMeasures.asp 43 Centers for Medicare and Medicaid Services. (2008b). RAI MDS Users Manual Washington D.C Centers for Medicare and Medicaid Serv ices 44 Centers for Medicare and Medicaid Services (2007). Final Report to CMS: Options for Improving Medicare Payment for SNFs Washington D.C Centers for Medicare and Medicaid Services. 45 Cherry, R. L. (1991). Agents of nursing home quality of care: ombudsmen and staff ratios revisited. Gerontologist, 31 (3), 302 308. 46 Chou, S. Y. (2002). Asymmetric Information, Ownership and Quality of Care: An Empirical Analysi s of Nursing Homes. Journal of Health Economics, 21 2 293 311. 47 Clarkson, K. W. (1972). Some Implications of Property Rights in Hospital Management. Journal of Law & Economics, 15 (2), 363 384. 48 Colorado Department of Public Health & Environment. ( 2010). The Survey Process -Scope and Severity [Electronic Version]. Retrieved 08/16/2010, from http://www.cdphe.state.co.us/hf/ncf/ncfscopeseverity.htm 49 Medicare Payment Advisory Commission, (2003). Assessing payment adequacy and updating payments fo r skilled nursing facility services Washington D.C Medicare Payment Advisory Commission 50 Covitz, D., & Liang, N. (2001). Recent Developments in the Private Equity Market and the Role of Preferred Returns Washington D.C Board of Governor s of the Feder al Reserve System 51 Cuny, C. J., & Talmor, E. (2007). A Theory of Private Equity Turnarounds. Journal of Corporate Finance, 13 4 629 646. 52 Daily, C. M., Dalton, D. R., & Jr, A. A. C. (2003). Corporate Governance: Decades of Dialogue and Data. The Academy of Management Review, 28 (3), 371 382. 53 Darby, R. (1997). Genesis climbs in bed with a pair of LBO firms to get Multicare cos. The Investment Dealers' Digest, 63 (42).
168 54 Davis, M. A. (1993). Nursing home ownership revisited: market, cost and q uality relationships. Med Care, 31 (11), 1062 1068. 55 Dawson, D. (1989). Nursing Home Giant Selling 370 Facilities Pacific Business News, p. 2, 56 Delfosse, R. (1995). Urban and rural classification of National Health Provider Inventory Providers: Un ited States 1991. Washington D.C National Center for Healt h Statistics. 57 Dellefield, M. E. (2000). The relationship between nurse staffing in nursing homes and quality indicators. J Gerontol Nurs, 26 (6), 14 28. 58 Deming, W. E. (1986). Out of Cris is Cambridge: Center for Advanced Engineering Study. 59 Demsetz, H. (1983). The Structure of ownership and the theory of the firm. Journal of Law and Economics, XXVI 60 Denis, D. J., Denis, D. K., & Sarin, A. (1997). Ownership structure and top execu tive turnover. Journal of Financial Economics, 45 (2), 193 222. 61 Department of Health and Human Services (2000) Testimony on Nursing Home Bankruptcies United States Senate 62 Department of Health and Human Services. (1994). Subacute care: review of the literature Wa shington D.C. 63 Department of Health and Human Services. (2006). The Nursing Home Liablity Insurance Market: A Case Study of Florida Washington D.C Department of Health and Human Services. 64 Department of Health and Human Services (2009). Nursing Homes: Paying for Care [Electronic Version]. Retrieved 10/02/2009, from http://www.medicare.gov/Nursing/Payment.asp 65 Department of Health and Human Services (2008). HHS Propose s $737 Billion Budget for Fiscal Year 2009 Retrieved 05/22. from http://www.hhs.gov/news/press/2008pres/02/20080204a.html 66 Desbrieres, P., & Schatt, A. (2002). The impacts of LBOs on the performance of acquired firms:The French Case. Journal of Business Finance & Accounting, 29 (5&6). 67 DiMaggio, P. J., & Anheier, H. K. (1990). The Sociology of Nonprofit Organizations and Sectors. Annual Review of Sociology, 16 (1), 23p.
169 68 Disk in, C. (2002). Pennsylvania Based Genesis Health Ventures Looks to Sell 258 Nursing Homes Knight Ridder Tribune Business 69 Donabedian, A. (1988). The quality of care. How can it be assessed? JAMA, 260 (12), 1743 1748. 70 Duhigg, C. (2007, 10/24/2007) Inquries at investor owned nursing homes The New York Times Retrieved 01/15/1008, from http://www.nytimes.com/2007/10/24/business/24retirement.html?_r=1&oref=slo gi n 71 Eisenhardt, K. M. (1989). Agency Theory: An Assessment and Review. The Academy of Management Review, 14 (1), 57 74. 72 Eljay. (2008). A Report on Shortfalls in Medicaid Funding for Nursing Home Care : American Health Care Associationo. Document N umber) 73 Eljay, L. (2008). A Report on Shortfalls in Medicaid Funding for Nursing Home Care : American Health Care Associationo. Document Number) 74 Estabrooks, C. A., Midodzi, W. K., Cummings, G. G., Ricker, K. L., & Giovannetti, P. (2005). The impact of hospital nursing characteristics on 30 day mortality. Nurs Res, 54 (2), 74 84. 75 Feng, Z., Grabowski, D. C., Intrator, O., & Mor, V. (2006). The effect of state medicaid case mix payment on nursing home resident acuity. Health Serv Res, 41 (4 Pt 1), 1 317 1336. 76 Feng, Z., Grabowski, D. C., Intrator, O., Zinn, J., & Mor, V. (2008). Medicaid payment rates, case mix reimbursement, and nursing home staffing -1996 2004. Med Care, 46 (1), 33 40. 77 Fenn, G. W., Liang, N., & Prowse, S. (1995). The Economi cs of the Private Equity Market Washington D.C. : Board of Governors of the Federal R eserve System. 78 Fenn., G. W., Liang., N., & Prowse., S. (2005). The Economics of Private Equity Market Washington D.C: Board of Governors of the Federal R eserve Syst em. 79 Fernandez, B. (2007). $1.25 billion for Genesis HealthCare: Two private equity firms are buying the Kennett Square operator of nursing homes in what one expert termed "a financial play Knight Ridder Tribune Business News 80 Financial Services Authority (2006). Private equity: A discussion of risk and regulatory engagement London: Financial Services Authority
170 81. Fleishman, J. A., Cohen, J. W., Manning, W. G., & Kosinski, M. (2006). Using the SF 12 Health Status Measure to Improve Prediction s of Medical Expenditures. Medical Care, 44 (5) 82. Flynn, J. (1986). A Nursing Home Company Faces Up to Its Midlife Crisis. Business Week 83. Forest, S. A. (1996). Might a new doctor cure Beverly. Business Week 84. Forgione, A., Schiff, D. A., & Crumb performance: An inventory of financial and nonfinancial Metrics. International Association of Management J Forum on Research in Healthcare Financial Management, 8 (2), 65 83. 85. Fox, I., & Marcus, A. (1992). The Caus es and Consequences of Leveraged Management Buyouts. Academy of Management Review, 17 (1), 62 85. 86. Frantz, R. A., Bergquist, S., & Specht, J. (1995). The cost of treating pressure ulcers following implementation of a research based skin care protocol in a long term care facility. Adv Wound Care, 8 (1), 36 45. 87. Friedman, B., & Shortell, S. (1988). The financial performance of selected investor owned and not for profit system hospitals before and after Medicare prospective payment. Health Serv Res, 23 (2 ), 237 267. 88. Fries, B. E., Schneider, D. P., Foley, W. J., Gavazzi, M., Burke, R., & Cornelius, E. (1994). Refining a case mix measure for nursing homes: Resource Utilization Groups (RUG III). Med Care, 32 (7), 668 685. 89. Fulp, E. W. (2009). Competit ive Market Model [Electronic Version]. Retrieved 11/20/2009, from http://www.cs.wfu.edu/~fulp/Papers/ewfchp5.pdf 90. Giacalone, A. J. (2001). The US Nursing Home Industry New York: M.E Sharpe. 9 1. Glendinning, D. (2005). New bill promotes apology, negotiations to handle errors [Electronic Version]. American Medical News Retrieved 08/27/2010, from http://www.ama assn.org/amedn ews/2005/10/24/gvsa1024.htm 92. Government Accountability Office. (1971). Problems in Providing Proper Care to Medicaid and Medicare Patients in Skilled Nursing Homes Washington D.C General Accounta bility Office. 93. Government Accountability Office. (1987). Medicare & Medicaid: stronger enforcement of nursing home requirements needed Washington D.C Government Accountability Officeo. Document Number)
171 94. Government Accountability Office. (1999a). Additional Steps Needed to Strengthen Enforcement of F ederal Quality Standards Washington D.C Government Accounta bility Office. 95. Go vernment Accountability Office. (1999b). Medicare Post Acute Care: Better Information Needed Before Modifying BBA Reforms Washington D.C Government Accountab ility Office 9 6. Government Accountability Office. (2000a). Nursing Homes: Sustained Efforts Are Essential to Realize Potential of the Quality Initiatives -Report to the Special Committee on Aging, U.S. Senate Washington D.C Government Accountab ility Office 97. Go vernment Accountability Office. (2000b). Nursing Homes: Aggregate Medicare Payments Are Adequate Despite Bankruptcies Washington D.C Government Accountab ility Office. 98. Government Accountability Office. (2002). Providers have reacted to Medicare payme nt system by changing practices Government Accountability Office Wa shington D.C. 99. Government Accountability Office. (2003). Nursing Home Quality: Prevalence of Serious Problems, While Declining Reinforces Importance of Enhanced Oversight : Governmen t Accounta bility Office. 100. Government Accountability Office. (2008). Nursing Homes: Federal monitoring surveys demonstrate continued understatement of serious care problems and CMS oversight weaknesses Washington D.C Government Accounta bility Office. 101. Grabowski, D. C. (2004). A longitudinal study of Medicaid payment, private pay price and nursing home quality. Int J Health Care Finance Econ, 4 (1), 5 26. 102. Grabowski, D. C., Feng, Z., Intrator, O., & Mor, V. (2004). Recent trends in state nurs ing home payment policies. Health Aff (Millwood), Suppl Web Exclusives W4 363 373. 103 Gustafson, D. H., & Hundt, A. S. (1995). Findings of innovation research applied to quality management principles for health care. Health Care Manage Rev, 20 (2), 16 3 3. 104 Hager, G. V. (1992). Genesis Health Ventures Reports Record Fiscal 1992 Results 105 Hager, G. V. ( 1993). Genesis Health Ventures Reports Record F iscal 1993 results
172 106 Hambrick, D. C., & Mason, P. A. (1984). Upper Echelons: The Organization as a Reflection of Its Top Managers. The Academy of Management Review, 9 (2), 193 206. 107 Hansmann, H. (1980). The role of nonprofit enterprise. Yale Law Journal, 89 (3), 835 901. 108 Statement of Charlene Harrington, Ph.D., Professor of Sociology and N ursing, Department of Social and Behavioral Sciences, University of California, San Francisco, California House of Representiaves (2007). 109 Harrington, C., & Carrillo, H. (2007). Analysis of On Line Survey Certification and Reporting data from the Cen ters for Medicare and Medicaid Services. University of California. 111. Harrington, C., Kovner, C., Mezey, M., Kayser Jones, J., Burger, S., Mohler, M., et al. (2000). Experts recommend minimum nurse staffing standards for nursing facilities in the United States. Gerontologist, 40 (1), 5 16. 111 Harrington, C., Mullan, J. T., & Carrillo, H. (2004). State nursing home enforcement systems. J Health Polit Policy Law, 29 (1), 43 73. 112 Harrington, C., Woolhandler, S., Mullan, J., Carrillo, H., & Himmelstein D. U. (2001). Does investor ownership of nursing homes compromise the quality of care? Am J Public Health, 91 (9), 1452 1455. 113 Harrington, C., Woolhandler, S., Mullan, J., Carrillo, H., & Himmelstein, D. U. (2002). Does investor ownership of nursing homes compromise the quality of care? Int J Health Serv, 32 (2), 315 325. 114 Harrington, C., Zimmerman, D., Karon, S. L., Robinson, J., & Beutel, P. (2000). Nursing home staffing and its relationship to deficiencies. J Gerontol B Psychol Sci Soc Sci, 55 ( 5), S278 287. 115 Hawes, C. (1996). Assuring Nursing Home Quality: The History and Impact of Federal Standards in OBRA 87 New York: Commonwealth Fund 116 Hillmer, M. P., Wodchis, W. P., Gill, S. S., Anderson, G. M., & Rochon, P. A. (2005). Nursing ho me profit status and quality of care: is there any evidence of an association? Med Care Res Rev, 62 (2), 139 166. 117 Hilzenrath, D. S. (2000, 02/04). Nursing Home Firm Settles Fraud Case; Beverly Enterprises to Pay $175 Million in Medicare Complaint Was hington Post, p. E3, 118 Hyer, K., Thomas, K., Mehra, S., Johnson, C., & Harman, J. (2009). Analyses on Outcomes of Increased Nurse Staffing Policies in Florida Nursing Homes: Staffing Levels, Quality and Costs Tampa: University of South Florida
173 119 Ihle, B., Whitehead, Bruce. (1989). Beverly Enterprises Completes Sale of All 44 Nursing Homes, Retirement Villas in Iowa Business Wire 120 Ingley, C. B., & Walt, N. T. V. d. (2001). The Strategic Board: the changing role of directors in developing a nd maintaining corporate capability. Corporate Governance An International Review, 9 (3), 174 185. 121 Ingram, P., & Baum, J. A. C. (1997). Chain Affiliation and the Failure of Manhattan Hotels, 1898 1980. Administrative Science Quarterly, 42 (1), 68 102. 122 Institute of Medicine. (1996). Nursing staff in hospital and nursing homes: Is it adequate? Washington D.C: Institute of Medicine. 123 Institute of Medicine (1986). Improving the Quality of Care in Nursing Homes Washington D.C National Academy Pr ess 124 Irving Levin Associates (2006). Formation Capital Deals: Largest private sale of skilled nursing facilities goes to G.E Irving Levin Associates 125 Japsen, B. (1998, 04/20). New Medicare PPS sparks Mariner Paragon merger Modern Healthcare 126 Jensen, M. C. (1986). Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers. American Economic Review, 76 (2), 323. 127 Jensen, M. C. (1989). Eclipse of the Public Corporation. Harvard Business Review, 67 (5), 61 74. 128 Jensen, M. C. (2 007). The Economic Case For Private Equity (and Some Concerns) Paper presented at the Harvard Business School Centennial Conference on Private Equity. Retrieved 10/02/2009, from http://ssrn.com/abstract=963530 129 Jensen, M. C., & Meckling, W. H. (1976). Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure. Journal of Financial Economics, 3 (4), 305 360. 130 Jensen, M. C., & Ruback, R. S. (1983). The market for corporate control: The scientific evidence. J Finan Econ 11(1 4). 5 50 131 Johnson Pawlson, J., & Infeld, D. L. (1996). Nurse staffing and quality of care in nursing facilities. J Gerontol Nurs, 22 (8), 36 45. 132 Johnson, C. E., Dobalian, A., Burkhard, J., Hedgecock, D. K., & Harman, J. (2004). Factors predicting lawsuits against nursing homes in Florida 1997 2001. Gerontologist, 44 (3), 339 347.
174 133 Kane, R. A., & Kane, R. L. (1988). Long term care: variations on a quality assurance theme. Inquiry, 25 (1), 132 146. 134 Ka ne, R. L. (1998). Assuring quality in nursing home care. J Am Geriatr Soc, 46 (2), 232 237. 135 Kaplan, S. N. (1989). The effects of management buyouts on operating performance and value. Journal of Financial Economics, 24 217 254. 136 Kaplan, S. N., & Schoar, A. (2005). Private equity performance: Returns, persistence, and capital Flows. The Journal of Finance, LX (4). 137 Kaplan, S. N., & Strmberg, P. (2008). Leveraged Buyouts and Private Equity Cambridge: National Bureau Econ omic Research. 138 K arr, A. R. (1990). NLRB Judge Finds Beverly Enterprises Guilty of Illegal Anti Union Tactics. Wall Street Journal, p. 1, 139 Keaveney, B. (1999, 12/03). Genesis Health Loses $290M for Year But Expects Profitability Ahead. The Daily Record 140 Kim, H., Harrington, C., & Greene, W. H. (2009). Registered nurse staffing mix and quality of care in nursing homes: a longitudinal analysis. Gerontologist, 49 (1), 81 90. 141 Kissam, S., Gifford, D., Parks, P., Patry, G., Palmer, L., Wilkes, L., et al. (2003 ). Approaches to quality improvement in nursing homes: lessons learned from the six state pilot of CMS's Nursing Home Quality Initiative. BMC Geriatr, 3 2. 142 Kitchener, M., O'Neill, C., & Harrington, C. (2005). Chain reaction: an exploratory study of nursing home bankruptcy in California. J Aging Soc Policy, 17 (4), 19 35. 143 Knight Ridder Tribune Business News. (2003, 07/02). Home Care Agency Firm Sells North Carolina Holdings Knight Ridder Tribune Business News 144 Konetzka, R. T., Norton, E. C., Sloane, P. D., Kilpatrick, K. E., & Stearns, S. C. (2006). Medicare prospective payment and quality of care for long stay nursing facility residents. Med Care, 44 (3), 270 276. 145 Konetzka, R. T., Norton, E. C., & Stearns, S. C. (2006). Medicare paym ent changes and nursing home quality: effects on long stay residents. Int J Health Care Finance Econ, 6 (3), 173 189. 146 Konetzka, R. T., Stearns, S. C., & Park, J. (2008). The staffing outcomes relationship in nursing homes. Health Serv Res, 43 (3), 1025 1042. 147 Kosnik, R. D. (1987). Greenmail: A Study of Board Performance in Corporate Governance. Administrative Science Quarterly, 32 (2), 163 185.
175 148 Kostiuk, F. P. (1990). Firm size and executive compensation. The Journal of Human Resources 25 (1). 1 49 Kovner, C., & Gergen, P. J. (1998). Nurse staffing levels and adverse events following surgery in U.S. hospitals. Image J Nurs Sch, 30 (4), 315 321. 150 Kumar, V., Norton, E. C., & Encinosa, W. E. (2006). OBRA 1987 and the quality of nursing home car e. Int J Health Care Finance Econ, 6 (1), 49 81. 151 Kunin, C. M., Douthitt, S., Dancing, J., Anderson, J., & Moeschberger, M. (1992). The association between the use of urinary catheters and morbidity and mortality among elderly patients in nursing homes Am J Epidemiol, 135 (3), 291 301. 152 Lehn, K., & Poulsen, A. (1989). Free cash flow and stockholder gains in going private transactions. The Journal of Finance XLIV (3). 153 Lichtenberg, F. R., & Siegel, D. (1990). The effects of leveraged buyouts on productivity and related aspects of firm behavior. Journal of Financial Economics, 27 (1), 165 194. 154 Liu, K., Gage, B., ., Harvell, J., Stevenson, D., & Brennan, N. (1999). Medicare's Post Acute Care Benefit: Background, Trends, and Issues to Be Faced. Washinton D.C Urban Instituteo. Document Number) 155 Loftus, P. (2005, 08/18). Beverly Enterprises to Be Acquired By Investor Group for $1.63 Billion Wall Street Journal 156 Long Term Care Community Coalition (2009) Modifying the Case Mix Medicai d Nursing Home System to Encourage Quality, Access and Efficiency New York: Long Term Care Commun ity Coalition 157 Long Term Care Focus. (2010). Long Term Care Focus (Publication. Retrieved 08/15/2010, from Brown University: http://ltcfocus.org/ 158 Long, F. W., & Ravenscraft, J. D. (1993). The Financial Performance of Whole Company LBOs (Discusison Paper No. CES 93). Washington D.C,: US Bureau of Census 159 Lorsch, J., & Young, J. (1990). Pawns or Potentates: The Reality of America's Corporate Boards. The Executive, 4 (4), 85 87. 160 LTCQ. (2007). Acquisition of Nursing Home Chains by Private Investment Groups: Abbreviated Analysis of the New York Times Article : LTCQ 161 LTCQ Incorporation. (2008). Acquisition of Nursing Home Chains by Private Investment Groups: Abbreviated Analysis of the New York Times Article Boston: LTCQ Incorporation
176 162 Mantone, J. (2005a). A new suitor for Beverly. Modern Healthcare, 35 163 Mantone, J. (2005b). Unwelcome advance. M odern Healthcare, 35 164 Marshall, M. N., Shekelle, P. G., Leatherman, S., & Brook, R. H. (2000). The public release of performance data: what do we expect to gain? A review of the evidence. JAMA, 283 (14), 1866 1874. 165 Martin, P. (1986). Three Nursi ng Homes Slated by Beverly The Business Journal 166 McGuire, J. B., Scheeweis, T., & Branch, B. (1990). Perceptions of Firm Quality: A Cause or Result of Firm Performance. Journal of Management, 16 (1), 167. 167 McNichol., E. C., & Lav., I. (2008). 2 7 states faced Total Budget Shortfall of at least $47 billion in2009; 4 other expect budget problems 16 8 Medicare Payment Advisory Commission. (2002). Report to the Congress: Medicare Payment Policy, Washington, DC. Medicare Payment Advisory Commission 169 Medicare Payment Advisory Commission. (2003). Report to the Congress: Medicare Payment Policy Washington D.C Medicare Payment Advisory Commission 170 Medicare Payment Advisory Commission. (2007). Report to the Congress: Medicare Payment Policy Washington D.C Medicare Payment Advisory Commission 171 Miles, G. L. (1988). This Nursing Home Giant May Need Intensive Care. Business Week 172 Mor, V., Berg, K., Angelelli, J., Gifford, D., Morris, J., & Moore, T. (2003). The quality of quality mea surement in U.S. nursing homes. Gerontologist, 43 Spec No 2 37 46. 173 Morse, A. (2003). Genesis unloads Florida nursing homes (Publication. Retrieved 07/15: 174 Mukamel, D. B., & Mushlin, A. I. (1998). Quality of care information makes a difference: an analysis of market share and price changes after publication of the New York State Cardiac Surgery Mortality Reports. Med Care, 36 (7), 945 954. 175 Mukamel, D. B., & Spector, W. D. (2000). Nursing home costs and risk adjusted outcome measures of qual ity. Med Care, 38 (1), 78 89. 176 Mukamel, D. B., & Spector, W. D. (2003). Quality report cards and nursing home quality. Gerontologist, 43 Spec No 2 58 66.
177 177 Mukamel, D. B., Weimer, D. L., Spector, W. D., Ladd, H., & Zinn, J. S. (2008). Publication of quality report cards and trends in reported quality measures in nursing homes. Health Serv Res, 43 (4), 1244 1262. 178 Munroe, D. J. (1990). The influence of registered nurse staffing on the quality of nursing home care. Res Nurs Health, 13 (4), 263 270 179 Murphy, J. K. (1985). Corporate performance and managerial renumeration: An empeirical evidence. Journal of Accounting and Finance, 7 11 42. 180. National Center for Assisted Living (20 01). The Assisted Living Sourcebook Retrieved 10/12/2010 from http://www.ahcancal.org/ncal/resources/Documents/alsourcebook2001.pdf 181 Nelson, C. E., Rust, T. R., Zahorik, A., Rose, L. R., Batalden, P ., & Siemanski, A. B. (1992). Do patient perceptions of quality relate to hospital financial performance Journal o f Healthcare Marketing, December 182 Newswire, P. (2000a, 07/11/2008). Genesis Health Ventures Files Ch. 11 PR Newswire 183 Newswire, P. (2000b, 01/18). Mariner Post Acute Network Announces Filing for Chapter 11 PR Newswire, p. 1, 184 Nikoskelai nen, E., & Wright, M. (2007). The impact of corporate governance mechanisms on value increase in leveraged buyouts. Journal of Corporate Finance, 13 (4), 511 537. 185 Institutional Investor Systems (2000) Nursing home cos. seek to go private Institution al Inv estor Systems 186 O'Neill, C., Harrington, C., Kitchener, M., & Saliba, D. (2003). Quality of care in nursing homes: an analysis of relationships among profit, quality, and ownership. Med Care, 41 (12), 1318 1330. 187 Opler, C. T. (1992). Operatin g performance in leveraged buyouts: Evidence from 1985 1989. Financial Management, 21 (1), 27 34. 188 Orlando Sentinel. (2003). Mariner Sells off Nursing Homes Orlando Sentinel 189 Pagano, E., Bo, S., Petrinco, M., Rosato, R., Merletti, F., & Gregori D. (2009). Factors affecting hospitalization costs in Type 2 diabetic patients. J Diabetes Complications, 23 (1), 1 6. 190 Peck, R. L. (1997). "People will pay us to keep them home...". Nursing Homes, 46 (7), 39 42.
178 191 Phalippou, L., & Gottschalg, O. (2009). The performance of private equity funds. The Review of Financial Studies, 22 (4). 192 Phillips, C. D., Morris, J. N., Hawes, C., Fries, B. E., Mor, V., Nennstiel, M., et al. (1997). Association of the Resident Assessment Instrument (RAI) with chan ges in function, cognition, and psychosocial status. J Am Geriatr Soc, 45 (8), 986 993. 193 Polivka, L., Salmon, J. R., Hyer, K., Johnson, C., & Hedgecock, D. (2003). The nursing home problem in Florida. Gerontologist, 43 Spec No 2 7 18. 194 PR Newswir e. (1998). Paragon Health Network and Mariner Health Group Merger Declared Effective. PR Newswire 195 PR Newswire. (2002a). Genesis Health Ventures to Terminat e Agreement with NCS Healthcare. PR Newswire 196 PR Newswire. (2002b). Mariner Post Acute Net work(R) and Mariner Health Group, Inc. Announce The Confirmation of Their Joint Plan of Reorgan ization by the Bankruptcy Court. PR Newswire 197 PR Newswire. (2002c, 05/14). Mariner Post Acute Network, Inc. and Mariner Health Group, Inc. Emerge From Chapt er 11 PR Newswire, p. 1, 198 PR Newswire. (2003a). Genesis Health Ventures Announces Filing of Registration Statement in Connection with the Spin Off of Genesis HealthCare Corporation. PR Newswire 199 PR Newswire. (2003b). Genesis Health Ventures to Acquire NCS HealthCare Transaction Creates Second Largest In stitutional Pharmacy in the U.S. PR Newswire 200 PR Newswire. (2003c). Mariner Health Care Announces Agreement to Terminate Leases Relat ing to Seven Florida Facilities. PR Newswire 201 Rengers C. (1992). Beverly's Long Term Recuperation: Net Income Is Up as a Fort Smith Company Continues Its Comeback. Arkansas Business, 202 Renneboog, L., Simons, T., & Wright, M. (2007). Why do public firms go private in the UK? The impact of private equit y investors, incentive realignment and undervaluation. Journal of Corporate Finance 591 628. 203 Robbins, C. J., Rudsenske, T., & Vaughan, J. S. (2008). Private equity investment in health care services. Health Aff (Millwood), 27 (5), 1389 1398. 204 Ro berts, P. W., & Dowling, G. R. (2002). Corporate reputation and sustained superior financial performance. Strategic Management Journal, 23 (12), 1077 1093.
179 205 Rosenau, P. V., & Linder, S. H. (2003). Two Decades of Research Comparing For Profit and Nonpro fit Health Provider Performance in the United States Social Science Quarterly (Blackwell Publishing Limited), 84 (2), 219 241. 206 Rosenthal, M. B. (2007). Nonpayment for performance? Medicare's new reimbursement rule. N Engl J Med, 357 (16), 1573 1575. 2 07 Ryan, J., Stone, R. I., & Raynor, C. R. (2004). Using large data sets in long term care to measure and improve quality. Nurs Outlook, 52 (1), 38 44. 208 Sack, K. (2008) Doctors say 'I'm Sorry before 'See you in Court The New York Times Retrieved 0 8/25/2010, from http://www.nytimes.com/2008/05/18/us/18apology.html?pagewanted=1 209 Saphir, A. (2000). Rebuilding from the rubble : Crain Communications, Incorporated 210 Sear, M. A. (1991). Comparison of efficiency and profitability of investor owned multihospital systems with not for profit hospitals. Health Care Management Review, 16 (2), 31 37. 211 Service Employees International Union. (2007). Equity and inequity: How private equity buyouts hurt nursing home residents Washington D.C Servic e Emp loyees International Union. 212 Shi, L., & Singh, A. D. (2009). Essentials of the US Health Care System (2nd ed.). Boston: Jones & Bartlett 213 Shih, A., & Schoenbaum, C. S. (2007). Measuring Hospital Performance: The Importance of Process Measures N ew York: Commonwealth Fund 214 Shukla, R. K., & Clement, J. (1997). A comparative analysis of revenue and cost management strategies of not for profit and for profit. Hospital & Health Services Administration, 42 (1), 117. 215 Smith, C., Cowan, C., Hef fler, S., & Catlin, A. (2006). National health spending in 2004: recent slowdown led by prescription drug spending. Health Aff (Millwood), 25 (1), 186 196. 216 Smith, D. (1995). Beverly Enterprises grows with acquisitions Arkansas Business, 217 Smith D. B., Feng, Z., Fennell, M. L., Zinn, J. S., & Mor, V. (2007). Separate and unequal: racial segregation and disparities in quality across U.S. nursing homes. Health Aff (Millwood), 26 (5), 1448 1458. 218 Smith, J. A. (1990). Corporate ownership structu re and performance: The case of management buyouts. Journal of Financial Economics, 27 143 164.
180 219 Smith., V., Ramesh., R., Gifford., K., Ellis., E., & Wachino., V. (2003). States respond to fiscal pressure: State Medicaid spending growth and cost cont aintme nt in fiscal years 2003 & 2004 Washington D.C Kaiser Commission on Medicaid & Uninsured 220 Spector, W. D., Selden, T. M., & Cohen, J. W. (1998). The impact of ownership type on nursing home outcomes. Health Econ, 7 (7), 639 653. 221 Spillman, B C., Liu, K., & McGilliard, C. (2002). Trends in Residential Long Term Care: Use of Nursing Homes and Assisted Living and Characteristics of Facilities and Residents Washington D.C: Urban Institute 222 Spillman, B. C., & Lubitz, J. (2002). New estimat es of lifetime nursing home use: have patterns of use changed? Med Care, 40 (10), 965 975. 223 Social Security Administration (1935) Social Security Act Retrieved 12/01/2010 from http://www.ssa.gov/OP_Home/ssact/comp ssa.htm 224 Stevenson, D. G., & Gra bowski, D. C. (2007). Nursing Home Divestiture and Corporate Restructuring Washington. D.C Department of Health and Human Services 225 Stevenson, D. G., & Grabowski, D. C. (2008). Private equity investment and nursing home care: is it a big deal? Healt h Aff (Millwood), 27 (5), 1399 1408. 226 Stevenson, D. G., & Studdert, D. M. (2003). The rise of nursing home litigation: findings from a national survey of attorneys. Health Aff (Millwood), 22 (2), 219 229. 227 Stone., R. (2000). Long term care for the disabled elderly: Current policy, emerging trends, and imp lications for the 21st century. Milbank Memorial Fund 228 Stratton, W. A. (1993). Mariner Health Group announces facility acquisition Business Wire: 229 Strmberg, P. (2007). The new demograph y of private equity : Swedish Institute for Financial Research 230 Sundaramurthy, C., & Lewis, M. (2003). Control and Collaboration: Paradoxes of Governance. The Academy of Management Review, 28 (3), 397 415. 231 Swan, J. H., Harrington, C., Clemena, W. Pickard, R. B., Studer, L., & DeWit, S. K. (2000). Medicaid nursing facility reimbursement methods: 1979 1997. Med Care Res Rev, 57 (3), 361 378. 232 Swan, J. H., & Pickard, R. B. (2003). Medicaid case mix nursing home reimbursement in three states. J H ealth Soc Policy, 16 (4), 27 53.
181 233 Tanner, L. (1993). Nursing home giant tackles growing specialty care marke Dallas Business Journal 234 Telegram & Gazette. (1997, 06/17). Multicare sold to Genesis ; $1.4B for nursing homes Telegram & Gazette 2 35 Tennyson, D. H., & Fottler, M. D. (2000). Does System Membership Enhance Financial Performance in Hospitals? Med Care Res Rev 57 (1), 22p. 236 The Lewin Group (2000). Briefing Chartbook on the Effect of the Balanced Budget Act of 1997 and the Balanc ed Budget Refinement Act of 1999 on Medicare Payments to Skilled Nursing Facilities Falls Church: Th e Lewin Group. 237 The Mississippi Business Journal. (2003). Beverly Enterprises sells facilities in Mississippi. The Mississippi Business Journal, 25 2 38 The New York Times. (1998, 04/14). Paragon in $581 Million Deal for Mariner The New York Times, p. 1, 239 The New York Times. (2001, 07/17). Nursing Home Owner to Sell Some Operations in Florida The New York Times, p. C6, 240 The New York Time s. (2004, 05/30). National Senior Care Buys Mariner Health Care The New York Times, p. 1, 241 The Wall Street Journal. (1993, 09/21). Genesis Health Ventures Inc The Wall Street Journal 242 The Wall Street Journal. (1995a, 01/10). Mariner Health G roup agrees to buy operator of nursing facilities The Wall Street Journal, p. 1, 243 The Wall Street Journal. (1995b, 01/07). Mariner Health set to acquire Pinnacle through stock swap The Wall Street Journal, p. 1, 244 The Wall Street Journal. (199 6, 10/02). Beverly Enterprises is considering selling its 43 Texas facilities Wall Street Journal 245 The Wall Street Journal. (1997, 07/10). Mariner Health plans purchase The Wall Street Journal, p. 1, 246 Wall Street Journal 92005) Beverly E nter prises Accepts Buyout Offer Wall Street Journal 247 Thomas, K. S., Hyer, K., Andel, R., & Weech Maldonado, R. (2009). The Unintended Consequences of Staffing Mandates in Florida Nursing Homes: Impacts on Indirect Care Staff. Med Care Res Rev
182 248 To urangeau, A. E., Doran, D. M., McGillis Hall, L., O'Brien Pallas, L., Pringle, D., Tu, J. V., et al. (2007). Impact of hospital nursing care on 30 day mortality for acute medical patients. J Adv Nurs, 57 (1), 32 44. 249 Troyer, J. L., & Thompson, H. G., J r. (2004). The Impact of Litigation on Nursing Home Quality. Journal of Health Politics Policy and Law, 29 (1), 11 42. 250 US Security & Exchange Commission. (2010). Filings & Forms (Publication. Retrieved 08/15/2010, from US Security & Exchange Commissio n: 251 Wall Street Journal. (1999, 07/28). Beverly Enterprises, U.S. Reach Medicare Settlement Wall Street Journal 252 Walshe, K. (2001). Regulating U.S. nursing homes: are we learning from experience? Health Aff (Millwood), 20 (6), 128 144. 253 W alters, D. (1993). Nursing homes' in transition: Health care reform, lifestyle trends drive internal, aesthetic changes in health care facilities Arkansas Business, 25 4 Warren, J. W. (2001). Catheter associated urinary tract infections. International Journal of Antimicrobial Agents, 17 (4), 299 303. 255 WayBack Machine. (2010). WayBack machine (Publication. Retrieved 08/08/2010, from WayBack Machine: http://www.archive.org/web/web.php 256 Weech Maldon ado, R., Meret Hanke, L., Neff, M. C., & Mor, V. (2004). Nurse staffing patterns and quality of care in nursing homes. Health Care Manage Rev, 29 (2), 107 116. 257 Weech Maldonado, R., Neff, G., & Mor, V. (2003a). Does quality of care lead to better finan cial performance?: the case of the nursing home industry. Health Care Manage Rev, 28 (3), 201 216. 258 Weech Maldonado, R., Neff, G., & Mor, V. (2003b). The relationship between quality of care and financial performance in nursing homes. J Health Care Fin ance, 29 (3), 48 60. 259 Weisbrod, B. A. (1988). The Nonprofit Economy Cambridge: Harvard University Press. 260 Werner, R. M., Bradlow, E. T., & Asch, D. A. (2008). Does Hospital Performance on Process Measures Directly Measure High Quality Care or Is I t a Marker of Unmeasured Care? Health Services Research, 43 (5p1), 1464 1484. 261 White, C. (2003a). The New Medicare Prospective Payment System for Skilled Nursing Facilities: Impacts on Payments, Services and Quality of Care. Harvard University, Boston.
183 262 White, C. (2003b). Rehabilitation therapy in skilled nursing facilities: effects of Medicare's new prospective payment system. Health Aff (Millwood), 22 (3), 214 223. 26 3 Wiener, J. M., Freiman, M. P., & Brown, D. (2007). Nursing Home Care Quality: Twenty Years After the Omnibus Budget Reconciliation Act of 1987 : Kaiser Family Foundation o. Document Number) 264 Wiener, J. M., Hanley, R. J., Clark, R., & Van Nostrand, J. F. (1990). Measuring the activities of daily living: comparisons across nation al surveys. J Gerontol, 45 (6), S229 237. 265 Willemain, T. R. (1980). A comparison of patient centered and case mix reimbursement for nursing home care. Health Serv Res, 15 (4), 365 377. 266 Williamson, O. (1984). Corporate Governance. The Yale Law Jour nal, 93 (7), 1197 1230. 267 Wisenberg, D. B. (2005, 03/23). Beverly Decides To Auction Itself Amid Proxy Fight; Nursing Home Operator Spurned Offer of Suitors Who Now Seek New Board. Wall Street Journal 268 Wojcieszak, D., Banja, J., & Houk, C. (2006) The Sorry Works! Coalition: Making the Case for Full Disclosure. Joint Commission Journal on Quality and Patient Safety, 32 344 350. 269 Wunderlich, G. S., & Kohler, O. P. (2001). Improving the Quality of Long term Care Washington D.C: Institut e of M edicine. 270 Younis, M., & Forgione, A. (2005). Using Return on Equity and Total Profit Margin to Evaluate Hospital Performance in the US: A Piecewise Regression Analysis. Journal of Health Care Finance, 31 (3), 82 88. 271 Younis, M., Forgione, D. A., K han, M., & Barkoulas, J. (2003). Hospital Profitability in Florida: A Revisitation. Research in Healthcare Financial Management, 8 (1), 95. 272 Younis, M., Rice, J., & Barkoulas, J. (2001). An Empirical Investigation of Hospital Profitability in the Post PPS Era. Journal of Health Care Finance, 28 (2), 65.
184 BIOGRAPHICAL SKETCH Rohit Pradhan completed his undergraduate degree in Medicine from India in 2002. Subsequently he worked as a physician in India for more than two years. He has done graduate work at State University of New Jersey, Rutgers in health policy and management. He completed his PhD in health services research, manageme nt, and policy in December 2010 from the University of Florida. His areas of interest include long term care especially nurs ing homes, and organizational theory. Lately, he has also developed an interest in global health especially with rural health care delivery systems.