<%BANNER%>

Clinical Outcomes, Healthcare Utilization and Costs in Patients with Neck Pain Utilizing Physical Therapy

MISSING IMAGE

Material Information

Title:
Clinical Outcomes, Healthcare Utilization and Costs in Patients with Neck Pain Utilizing Physical Therapy
Physical Description:
1 online resource (132 p.)
Language:
english
Creator:
Horn, Maggie E
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Rehabilitation Science
Committee Chair:
Bishop, Mark Donald
Committee Members:
Peoples-Sheps, Mary D
George, Steven
Harman, Jeffrey Scott
Brennan, Gerard

Subjects

Subjects / Keywords:
cost -- neck -- pain -- physical -- therapy -- utilization
Rehabilitation Science -- Dissertations, Academic -- UF
Genre:
Rehabilitation Science thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Approximately 35% of people with neck pain seek care from physical therapists (PT). Clinical practice guidelines were published for neck pain. These guidelines recommend sub-grouping patients into treatment groups and provide recommendations for interventions. Past research in the low back indicates that receiving guideline adherent care can improve clinical outcomes and decrease physical therapy utilization and receiving physical therapy intervention early in an episode can decrease healthcare utilization and costs.Therefore we wished to examine if certain subgroups of patients have improved clinical outcomes; the effect of receiving guideline adherent care on clinical outcomes, utilization and costs; and the effect of receiving guideline adherent care in combination with timing of intervention on clinical outcomes,utilization and costs.   The sample in this study included 3485 people with neck pain.From this sample, three separate sets of analyses were performed to examine differences in clinical outcomes, utilization and costs in the sample. The results of this study found that patients sub-grouped in acute neck pain group had the most favorable outcomes compared to all other groups and the patients in the chronic neck pain group had the least favorable outcomes. Patients who received guideline adherent care did not significantly differ in their clinical outcomes but had had approximately half the number visits to physical therapy and healthcare providers, fewer diagnostic images and prescription medication during the year of physical therapy and less costs for physical therapy. When looking at the combination of guideline adherent care and timing, those patients who received early intervention with guideline adherent care experienced improved clinical outcomes, highest efficiency and best value for physical therapy, less costs for visits to healthcare providers and fewer prescription medications during the year of physical therapy. The primary limitations in this study were timing of physical therapy was reliant on patient report and adherent care was determined from billing codes. These findings support that providing early intervention with guideline adherent care can improve clinical outcomes and potentially decrease healthcare utilization and costs. Further research with a more rigorous methodology is needed to validate these findings.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Maggie E Horn.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: Bishop, Mark Donald.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0045680:00001

MISSING IMAGE

Material Information

Title:
Clinical Outcomes, Healthcare Utilization and Costs in Patients with Neck Pain Utilizing Physical Therapy
Physical Description:
1 online resource (132 p.)
Language:
english
Creator:
Horn, Maggie E
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Rehabilitation Science
Committee Chair:
Bishop, Mark Donald
Committee Members:
Peoples-Sheps, Mary D
George, Steven
Harman, Jeffrey Scott
Brennan, Gerard

Subjects

Subjects / Keywords:
cost -- neck -- pain -- physical -- therapy -- utilization
Rehabilitation Science -- Dissertations, Academic -- UF
Genre:
Rehabilitation Science thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Approximately 35% of people with neck pain seek care from physical therapists (PT). Clinical practice guidelines were published for neck pain. These guidelines recommend sub-grouping patients into treatment groups and provide recommendations for interventions. Past research in the low back indicates that receiving guideline adherent care can improve clinical outcomes and decrease physical therapy utilization and receiving physical therapy intervention early in an episode can decrease healthcare utilization and costs.Therefore we wished to examine if certain subgroups of patients have improved clinical outcomes; the effect of receiving guideline adherent care on clinical outcomes, utilization and costs; and the effect of receiving guideline adherent care in combination with timing of intervention on clinical outcomes,utilization and costs.   The sample in this study included 3485 people with neck pain.From this sample, three separate sets of analyses were performed to examine differences in clinical outcomes, utilization and costs in the sample. The results of this study found that patients sub-grouped in acute neck pain group had the most favorable outcomes compared to all other groups and the patients in the chronic neck pain group had the least favorable outcomes. Patients who received guideline adherent care did not significantly differ in their clinical outcomes but had had approximately half the number visits to physical therapy and healthcare providers, fewer diagnostic images and prescription medication during the year of physical therapy and less costs for physical therapy. When looking at the combination of guideline adherent care and timing, those patients who received early intervention with guideline adherent care experienced improved clinical outcomes, highest efficiency and best value for physical therapy, less costs for visits to healthcare providers and fewer prescription medications during the year of physical therapy. The primary limitations in this study were timing of physical therapy was reliant on patient report and adherent care was determined from billing codes. These findings support that providing early intervention with guideline adherent care can improve clinical outcomes and potentially decrease healthcare utilization and costs. Further research with a more rigorous methodology is needed to validate these findings.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Maggie E Horn.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: Bishop, Mark Donald.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0045680:00001


This item has the following downloads:


Full Text

PAGE 1

1 CLINICAL OUTCOMES, HEALTHCARE UTILIZATION AND COSTS IN PATIENTS WITH NECK PAIN UTILIZING PHYSICAL THERAPY By MAGGIE ELIZABETH HORN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FU LFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 Maggie Elizabeth Horn

PAGE 3

3 To my wonderful husband Tod d, my beautiful daughter Laney and my parents Nina a nd Bruce Horn; it is through your support and love I was able to achieve this accomplishment

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank and acknowledge the people who have been instrumental in my success. First and foremost, I would like to thank Dr. Mark Bishop. Without his mentorship and support none o f this would have been possible For the last 6 years he has not only been a mentor, but a friend. I will be forever indebted to him for my achievements Second, I would like to thank Dr. Gerard Brennan and Intermountain Healthcare for offering to take me as a student for my public health internship, being a member on my dissertation committee and providing me with invaluable clinical and research insight for my thesis. Third, I would like to sincerely thank Dr. Mary Peoples Sheps. She was instrumental in my success in the public health program and has provided me with the much needed insight into the public health impacts of my dissertation. Without Dr. Jeff Harman, I would not have made it through what I th ought at the time was endless data cleaning and analysis. I would like to thank Dr. Steve George for his mentorship with this dissertation and his guidance and nurturing of my future research career. I would also like to thank all those people who have su pported me along the way, Joel Bialosky, Claudia Senesac, Meryl Alappattu, Jason Beneciuk, Chaz Gay, Roy Coronado, Cory Simon and Cally House. Last but not least, I owe sincere thanks to my husband Todd Robinson, a k a Mr. Mom. I could not ask for more fr om a husband and a father; without him I would not be the person I am t oday and my su ccess is a direct result of his support

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 12 ABSTRACT ................................ ................................ ................................ ................... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 15 2 LITERATURE REVIEW ................................ ................................ .......................... 17 Magnitude of Neck Pain as a Public Health Problem ................................ .............. 17 Societal Impact of Pain ................................ ................................ ..................... 17 Economic Impacts of Neck Pain ................................ ................................ ....... 18 Epidemiology of Neck Pain: Incidence and Prevalence Rates of Neck Pain Reported in the Literature ................................ ................................ ............. 21 Management of Neck Pain in the Healthcare System ................................ ............. 22 Management of Neck Pain by Physical Therapists ................................ ................. 23 Public Health Significance of Neck Pain and Relationship to Rehabilitation ........... 24 3 SPECIFIC AIMS AND HYPOTHESES ................................ ................................ .... 26 Specific Aim 1: Description of Clinical Characteristics, Outcomes and Physical Therapy Utilization in Patients with Neck Pain ................................ ..................... 26 Specific Aim 2: Does Receiving Guideline Adherent Ph ysical Therapy Intervention Improve Clinical Outcomes, Healthcare Utilization and Costs in Patients with Neck Pain? ................................ ................................ ..................... 27 Specific Aim 3: Does the Timing of Guideline Adherent Care for Patients w ith Neck Pain Improve Clinical Outcomes, Healthcare Utilization and Costs? .......... 28 4 DESCRIPTION OF CLINICAL CHARACTERISTICS, OUTCOMES AND PHYSICAL THERAPY UTILIZATION IN PATIENTS WITH NECK PAIN ................ 30 Background ................................ ................................ ................................ ............. 30 Purpose ................................ ................................ ................................ .................. 31 Methods ................................ ................................ ................................ .................. 32 Database ................................ ................................ ................................ .......... 32 Sample ................................ ................................ ................................ ....... 32 Patient clinical sub grouping ................................ ................................ ...... 33 Clinical outcomes ................................ ................................ ....................... 33 Health related quality of life ................................ ................................ ........ 34

PAGE 6

6 Utilization ................................ ................................ ................................ ... 35 Data Analysis ................................ ................................ ................................ ... 35 Description and comparison of patient characteristics, clinical outcomes and physical therapy utilization ................................ ............................... 35 Clinical outcomes ................................ ................................ ....................... 36 Examination of clinical predictors of value and efficiency .......................... 36 Results ................................ ................................ ................................ .................... 36 Description of patient characteristics, clinical outcomes and physical therapy utilization ................................ ................................ .......................... 37 Comparison of patient characteristic s, clinical outcomes and physical therapy utilization by clinical sub group ................................ ......................... 37 Disability ................................ ................................ ................................ ..... 39 Pain ................................ ................................ ................................ ............ 40 Health related quality of life (HRQL) ................................ .......................... 40 Efficiency and value ................................ ................................ ................... 40 Summary of Findings ................................ ................................ .............................. 41 5 DOES RECEIVING GUIDELINE ADHERENT PHYSICAL THERAPY INTERVENTION IMPROVE CLINICAL OUTCOMES, HEALTHCARE UTILIZATION AND COSTS IN PERSONS WITH NECK PAIN? ............................. 47 Background ................................ ................................ ................................ ............. 47 Purpose ................................ ................................ ................................ .................. 48 Methods ................................ ................................ ................................ .................. 49 Dat abase ................................ ................................ ................................ .......... 49 Sample ................................ ................................ ................................ ............. 50 Determining Adherent Care ................................ ................................ .............. 50 Clinical Outcome Measures ................................ ................................ .............. 52 Health Related Quality of Life ................................ ................................ ........... 53 Financial Outcomes and Healthcare Utilization ................................ ................ 53 Physical therapy utilization and costs ................................ ........................ 53 Healthcare utilization and costs ................................ ................................ 54 Data Analysis ................................ ................................ ................................ ... 54 Comparison of Clinical Outcomes ................................ ................................ .... 54 Comparison of Healthcare Utilization ................................ ............................... 55 Results ................................ ................................ ................................ .................... 57 Clinical Outcomes ................................ ................................ ............................ 59 Successful outcome in disability ................................ ................................ 59 Successful pain management ................................ ................................ .... 59 HRQL ................................ ................................ ................................ ......... 60 Healthcare Utilization during the Year of Admission to Physical Therapy ........ 60 Healthcare Utilization the Year after Admission to Physical Therapy ............... 60 Costs during the Year of Admission to Physical Therapy ................................ 61 Costs during the Year after admission to Physical Therapy ............................. 61 Summary of Findings ................................ ................................ .............................. 61

PAGE 7

7 6 DOES THE TIMING OF GUIDELINE ADHERENT CARE FOR PATIENTS WITH NECK PAIN IMPROVE CLINICAL OUTCOMES, HEALTHCARE UTILIZATION AND COSTS? ................................ ................................ .................. 67 Background ................................ ................................ ................................ ............. 67 Purpose ................................ ................................ ................................ .................. 68 Methods ................................ ................................ ................................ .................. 68 Database ................................ ................................ ................................ .......... 68 Sample ................................ ................................ ................................ ............. 69 Analyses #1:Clinical outcomes: ................................ ................................ 69 Analyses#2: Healthcare utilization and costs ................................ ............. 70 Clinical Care Groups ................................ ................................ ........................ 70 Clinical Outcome Measures ................................ ................................ .............. 72 Examination of Clinical Predictors of Value and Efficiency ............................... 73 Health Related Quality of Life (HRQL) ................................ ............................. 73 Physical Therapy Utilization and Costs ................................ ............................ 74 Healthcare Utilization and Costs ................................ ................................ ...... 74 Data Analysis ................................ ................................ ................................ ... 74 Analyses #1: Clinical outcom es ................................ ................................ 74 Analysis #2: Comparison of healthcare utilization and costs during the year of admission to physical therapy ................................ ..................... 75 Results ................................ ................................ ................................ .................... 77 Analyses #1: Clinical Outcomes ................................ ................................ ....... 77 Disability ................................ ................................ ................................ ..... 78 Pain ................................ ................................ ................................ ............ 78 Health related quality of life ................................ ................................ ........ 79 Efficiency and Value ................................ ................................ .................. 79 Analyses #2: Healthcare Utilization and Costs ................................ ................. 81 Physical therapy utilization and costs ................................ ............................... 81 Visits ................................ ................................ ................................ .......... 81 LOS ................................ ................................ ................................ ............ 82 Costs ................................ ................................ ................................ .......... 82 Healthcare Providers Utilization and Costs ................................ ...................... 82 Diagnostic Imaging ................................ ................................ ........................... 83 Prescription Medication ................................ ................................ .................... 83 Summary of Findings ................................ ................................ .............................. 84 7 DISCUSSION AND CONCLUSIONS ................................ ................................ ...... 88 Discussion ................................ ................................ ................................ .............. 88 Statement of Principle Findings ................................ ................................ ........ 89 Strengths ................................ ................................ ................................ .......... 90 Limitations ................................ ................................ ................................ ........ 90 Measures of value and efficiency ................................ ............................... 90 Sub grouping approaches ................................ ................................ .......... 91 Methodological limitations and considerations ................................ ........... 93 Poten tial Implications ................................ ................................ ....................... 95

PAGE 8

8 Unanswered Questions and Future Research Directions ................................ 95 Conclusion ................................ ................................ ................................ .............. 97 LIST OF REFERENCES ................................ ................................ ............................. 123 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 132

PAGE 9

9 LIST OF TABLES Table page 1 1 Description of Clinical Characteristics and Outcomes from Physical Therapy. ... 99 1 2 Odds of Achieving MCID for NDI Score (19 points). ................................ ......... 100 1 3 Odds of Achieving MCID for NPRS (1.3 points). ................................ .............. 100 1 4 Predictors of Efficiency for an Episode of Physical Therapy for Neck Pain. ..... 101 1 5 Predictors of Value for an Episode of Physical Therapy for Neck Pain. ........... 101 2 1 CPT Codes Occurring in each Phase of Treatment. ................................ ......... 102 2 2 Baseline Clinical Characteristics of Sample by Adherent Care Group. ............. 103 2 3 Clinical Outcomes and Course of Physical Therapy by Adherent Care Gr oup. 103 2 4 Healthcare Utilization and Costs by Adherent Care Group for the Year of Admission to Physical Therapy. ................................ ................................ ........ 104 2 5 Heal thcare Utilization and Expenditures the Year after Admission to Physical Therapy. ................................ ................................ ................................ ........... 105 2 6 Adjusted Odds Ratio (aOR) for Successful Disability Outcome from Physical Therapy (50% change in NDI). ................................ ................................ ......... 105 2 7 Adjusted Odds Ratio (aOR) for Successful Pain Management from Physical Therapy (50% change in NPRS). ................................ ................................ ..... 106 2 8 Healthcare Utilization the Year of Admission to Physical Therapy. .................. 107 2 9 Adjusted Odds Ratio (aOR) for Prescription Medication during Year of Admission to Physical Therapy. ................................ ................................ ........ 108 2 10 Healthcare Utilization the Year after Admission to Physical Therapy. .............. 110 2 11 Costs during the Year of Admission to Physical Therapy. ................................ 111 2 12 Costs during the Year after admission to Physical Therapy. ............................ 112 3 1 Comparison of Baseline Characteristics of Clinical Care Groups. .................... 112 3 2 Comparison of Clinical Outcomes and Process Variables by Clinical Care Group. ................................ ................................ ................................ .............. 113

PAGE 10

10 3 4 The Adjusted Odds Ratio (aOR) of Ac hieving Successful Pain Management (50% Improvement in NPRS). ................................ ................................ .......... 115 3 5 Model for Prediction of Efficiency of Physical Therapy Intervention. ................ 116 3 6 Model for Prediction of Value of Physical Therapy Intervention. ...................... 117 3 7 Comparison of Unadjusted Healthcare Utilization and Costs by Clinical Group. ................................ ................................ ................................ .............. 118 3 8 Comparison of Physical Therapy Utilization and Costs by Clinical Care Group. ................................ ................................ ................................ .............. 119 3 9 Comparison of Health Care Utilization and Costs by Clinical Care Group the Year of Admission to Physical Therapy. ................................ ........................... 120 3 10 Number of Diagnostic Images the Year of Admission to Physical Therapy. ..... 121 3 11 Number of Prescriptions the Year of Admission to Physical Therapy. .............. 122

PAGE 11

11 LIST OF FIGURES Figure page 1 1 Description of Clinical Sub g roups ................................ ................................ ...... 45 1 2 Derivation of Sample for Specific Aim 1 ................................ ............................. 46 2 1 Derivation of the Sample for Specific Aim 2 ................................ ....................... 66 3 1 Derivation of Samples for Analyses #1 and #2 for Specific Aim 3 ...................... 87

PAGE 12

12 LIST OF ABBREVIATIONS APTA American Physical Therapy Association CMS Centers for Medicare and Medicaid Serv ices CPT Current Procedural Terminology HRQL Health Related Quality of Life ICD 9 International Classification of Diseases, Ninth Revision IHC Intermountain Healthcare Inc IASP International Association for the Study of Pain LBP Low back pain LOS Length of Stay MCID Minimal Clinically Important Difference MEPS Medical Expenditure Panel Survey MPPR Multiple Procedure Payment R eduction NDI Neck Disability Index NHIS National Health Interview Survey NPRS Numerical Pain Rating Scale PT Physical Therapy RVU Rel ative Value Unit SF 6 D Short form version 6D TBC Treatment Classification System WHO World Health Organization

PAGE 13

13 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requiremen ts for the Degree of Doctor of Philosophy CLINICAL OUTCOMES, HEALTHCARE UTIL I ZATION AND COSTS IN PATIENTS WITH NECK PAIN UTILIZING PHYSICAL THERAPY By Maggie Elizabeth Horn August 2013 Chair: Mark Bishop Major: Rehabilitation Science Approximately 35% o f people with neck pain seek care from physical therapists (PT). C linical practice guidelines were published for neck pain. These guidelines recommend subgrouping patients int o treatment groups and provide recommendations for intervent ions Past research i n the low back indicates that receiving guideline adherent care can improve clinical outcomes and decrease physical therapy utilization and receiving physical therapy intervention early in an episode can decrease healthcare utilization and costs. Therefore we wished to examine if certain subgroups of patients have improved clinical outcomes; the effect of receiving guideline adherent care on clinical outcomes, utiliz ation and costs ; and the effect of receiving guideline adherent care in combination with tim ing of intervention on clinical outcomes, utilization and costs. The sample in this study included 3485 people with neck pain. From this sample, three separate sets of analyses were performed to examine differences in clinical outcomes, utilization and c osts in the sample. The results of this study found that patients sub grouped in acute neck pain group had the most favorable outcomes

PAGE 14

14 compared to all other groups and the patients in the chronic neck pain group had the least favorable outcomes. Patients w ho received guideline adherent care did not significantly differ in their clinical outcomes but had had approximately half the number visits to physical therap y and healthcare providers, fewer di agnostic images and prescription medication during the y ear o f physical therapy and less costs for physical therapy When looking at the combination of guideline adherent care and timing, those patients who received early intervention with guideline adherent care experienced i mproved clinical outcomes h ighest effic iency and b est value for physical therapy less costs for visits to healthcare providers and fewer prescription medication s during the year of physical ther a py. The primary limitations in this study were timing of physical therapy was reliant on patient re port and adherent care was determined from billing codes. The se findings support that providing early intervention with guideline adherent care can improve clinical outcome s and potentially decrease healthcare utilization and costs. Further research with a more rigorous methodology is needed to validate these findings.

PAGE 15

15 CHAPTER 1 INTRODUCTION The magnitude of social and economic burden of neck pain is immense and this burden has led to neck pain being regarded as a major public health problem 1 Approximately 22 70% of the population will have neck pain at some point in their life 2, 3 The point prevalence of neck pain varies from 10 20% and approximately 54% of people repor t having neck pain in the last 6 months 4 and there is reason to believe the prevalence of neck pain is increasing 5 Additionally, spine pain is one of the top physical s ability to work and perform activities of daily living 6 A large portion of people experiencing painful musculoskeletal conditions wil l utilize outpatient physical therapy services, which will undoubtedly account for a significant percentage of health care expenditures 7 Given these statistics, it is evident that neck pain is a musculoskeletal pain condition that is common, affects a significant proportion of the population and presents as a pu blic health a nd economic problem In order to address neck pain as a public health problem, the condition must be addressed from both a clinical and healthcare utilization perspective. Approximately 35% of persons with neck pain reported being seen by a physical thera pist; the most frequently visited provider 8 Physical therapists provide interventions to patients with neck pain 9 11 but variation exists in the inventions provided to patients. In 2008, the Orthopaedic Section of the American Physical Therapy Association (APTA) created evidence based practice guidelines for orthopaedic physical therapy management of patients with the musculoskeletal impairment of neck pain 4 The purpose of this guideline was to describe evidence based physical therapy diagnosis, prognosis, intervention and assessment of outcomes for

PAGE 16

16 neck pain. Specifically, this guideline made recommendations based on best available evidence and expert opinion for interventio ns for neck pain. The guideline advocate d for the use of primarily active interventions in conjunction with manipulation/mobilization when indicated to promote optimal outcomes. Prior research in the low back found that ph ysical therapist intervention that was guideline adherent produced improved clinical outcomes and decreased utilization and costs for physical therapy 12 but this has not been investigated in neck pain. Additionall y, l iterature supports that a relationship has also been indicated between the timing of physical therapy initiation for acute LBP and subsequent healthcare utilization 13 Gellhorn et al (2012) found that patients who received physical therapy early after an episode of acute LBP were at lower risk for subsequent LBP related healthcare utilization over the following year compared to those who received physical therapy at later times 13 It is reasonable to extend this rationale to neck pain; that is, early interven tion and evidence based and guideline adherent care improve clinical outcomes and decrease healthcare utilization and costs. Therefore the primary goals o f this dissertation were 1) T o d escribe the clinical characteristics, physical therapy utilization and costs in persons seeking care for neck pain 2) To determine whether receiving guideline adherent physical therapy intervention improves clinical outcomes, decreases healthcare utilization and costs and 3) To determine the extent to which timing of physic al therapy in conjunction with receiving guideline adherent care improves clinical outcomes and decreases healthcare utilization and costs.

PAGE 17

17 CHAPTER 2 LITERATURE REVIEW Magnitude of Neck Pain as a Public Health Problem Societal Impact of P ain The societ al impact of pain is immense with pain conditions affecting at least 116 million U.S. adults 14 The International Association for the Study of Pain (IASP), defines pain as an unpleasant sensory and emotional experience associated with actual or potential tissue damage 15 Additionally, IASP appreciates pain as not purely physical, but views the pain experience as a constellation of biological, psychological and social factors. One of the most frequent reported types of pain is musculoskeletal pain. It is one of the most common reasons for entry into the healthcare system and its profound societal impact is now recognized by multiple international governments and by the World Hea lth Organization (WHO) 16 From 1996 2006, the number of persons reporting a musculoskeletal disease increased nearly 14 million from the 76 million reported in 1996 17 Musculoskeletal pain is responsible for approximately 25% of the total expense of illness in developed nations 16, 18 The prevalence rates of general musculoskeletal pain vary depending on the source. According to the National Health Interview Survey (NHIS), approximately 49% of US adults, reported having a chronic musculoskeletal condition i n 2005 6 Other sources report a musculoskeletal pain prevalence rate of up to 65% 19 The term musculoskeletal pain can encompass many disorders including neck and low back pain. Healthcare utilization and medical costs for spine conditions have increased steadily from 1997 2005 and these increases in utilization did not correlate

PAGE 18

18 with improvement s in health 20 Moreover, musculoskeletal pain can be conceptu alized as a public health challenge due to increasing prevalence and proportion of the population affected and it is considered an epidemic 6, 21 Neck pain and its related disability have an enormous impact on individuals and their families, communities, healthcare systems and businesses 22 25 The impact of neck pain is likely t o vary significantly between and within populations depending on socioeconomic status, general access to health services, occupational distribution, pain perception and other factors that are associated with the onset and prognosis of neck pain 26 Individuals with neck pain have limitations of varying degrees ranging from difficulties with mobility during activities of daily living to the reduced ability to participate in work, and social obligations; further increasing the burde n associated with neck pain 24 Economic Impacts of Neck Pain Neck pain has major economic consequences through the cost of health care, work absenteeism, insurance, and pressure on health care systems. A study in the Netherlands found that direct costs, such as health care, amounted to just 23% of this figure while indirect costs, such as work absenteeism and disability, amounted to 77% of the total costs 27 The annual cost of pain was greater than the annual costs in 2010 US dollars of heart disease ($309 billion) cancer ($243 billion) and diabetes ($188 billion) and nearly 30 percent higher than the combined cost of cancer and diabetes 14 Healthcare utilization and medical costs f or spine conditions have increased steadily from 1997 2005 and these increases in utilization and cost did not correlate with improvements in health 20 For the years 2004 2006, the annual average direct cost for musculoskeletal health care as a direct result of patients with a musculoskeletal

PAGE 19

19 disease in addition to o ther health issues was estimated to be $576 billion (4.5% GDP). Indirect cost, expressed primarily as wage losses was estimated to be $373.1 billion (2.9% of the GDP). The total cost for all persons with a musculoskeletal disease, either treated a primary condition or in addition to another condition costs for persons with a musculoskeletal disease in addition to the cost to society in the form of decreased wages was estimated to be nearly $950 billion per year (7.4% of GDP) 17 Of interest here are costs related specifically to neck pain, a type of spine pain. Recent studies, calculated sp ine related healthcare trends using the Medical Expenditure Panel Survey (MEPS). MEPS, maintained by the Agency for Healthcare Research and Quality, is a nationally representative annual cross sectional survey of household medical utilization that is suppl emented by provider and employer records 28, 29 Using this survey allows researchers produce unbiased national estimates of expenditures, utilization, and health status. This survey is widely used in research on na tional utilization trends and policy 29 One MEPS study examined the health care utilization and expenditures related to the spine in outpatient, inpatient, pharmacy and emergency categories from 1997 2006 29 This study found that for inpatient visits, the annual per user expenditure for hospitalizations increased an average of 3.7% per year ($13,040 in 1997 (95% CI: 12,370 13,710) to $17,909 in 2006). In 2006, inpatient expen ditures accounted for 29.3% of total national spine related costs. The proportion of respondents with spine problems did not increase nor did the mean number of hospitalizations. Outpatient visits increased 4.7% over the study period in respondents with sp ine problems. The proportion of patients with spine problems seeking outpatient services

PAGE 20

20 did not change, but the mean number of visits increased in this time period contributing to a 6.7% average annual increase in total national outpatient expenditures($1 0.4 billion in 1997 (95% CI: $9.6 billion $11.2 billion) to $18.4 billion in 2006 (95% CI: $16.6 billion $20.1 billion). This category exhibited the largest absolute increase among all the categories, and is responsible for 53.4% of all spine related healt hcare costs 29 The average annual per user expenditures for prescription medication increased an average of 10.2% per year ($166 per patient in 1997 (95%CI: $164 $167) to $397 in 2006 (95% CI: $393 $402). There was also an increase in the expenditure for prescriptions (139%) and increase in the number of users (39.9%). Additionally, there was a 660% increase in the expenditures for opioid medication 29 In another study whic h used the MEPS data to examine health care expenditures for spine pain, researchers examined the health care expenditures for spine pain for ambulatory care services including physical therapy, chiropractic care, medical physicians and primary care physic ians 30 Trends in number of visits and cost s were examined. F or medical physicians the mean number of visit were between 2.9 3.7, for chiropractic care between 7.2 9.3 visits, and for physical therapy between 6.8 11.4 visits. Visits to specialists increased from 6.6 visits in 2003 to 8.9 visits in 2006. Among adults who reported any expenditure on medical care for a primary diagnosis of a spine condition, the mean inflation adjusted expenditure increased by 95% (from $487 in 1999 to $950 in 2008). Chiropractic care costs, varied little, but physical therapy costs pea ked in 2002, and then contracted. The specialty care costs increased significantly and primary care physician costs varied slightly 30

PAGE 21

21 Epidemiology of Neck Pain : Incidence and Prevalence Rates of Neck Pain Reported in the Literature Accurate prevalence estimates are desirable to serve as a ba sis for etiological studie s, health care evaluation and to assess the effect of neck pain in general populations 1, 27, 31 Unfortunately there is a great deal of variability in the reporting of r ates of neck pain prevalence. Prevalence rat es can be reported as point or period prevalence. Often when prevalence rates are being compared across or between studies, consideration for the type of prevalence rate are not reported or compared with studies with a different method of reporting prevale nce Hoy et al. (2010) examined the epidemiology of neck pain; twenty seven studies reported period prevalence of one, two and three week rates; one, three and six month rates; and lifetime rates 25 Considerations for the definition are important when examining and reporting incidence rates for neck pai n from the literature. Spine pain, and neck pain specifically differs from other musculoskeletal pain disorders because the course of neck pain is often recurrent rather than isolated to one episode. A population based estimate reports that 80% of all acut e neck pain resolves within days to weeks 32 But of those persons that report current neck pain, 50 75% r eport having neck pain again in 1 to 5 years later 33 Additionally, episodes of neck pain often occur in childhood or adolescence. One study specifically examining neck pain in adolescents found that at twelve mont hs, in an initially pain free population of adolescents, 21% reported experiencing neck pain 34 Therefore, most studies may not be accurately reporting incidence rates for neck pain. Incidence and prevalence rates of neck pain vary across populations, countries and study designs. This variation is evident in the reported literature. Incidence rates in the US are reported as 21.3% 34 17.9% in the United Kingdom 35 21.3 % in Finland 34

PAGE 22

22 and a range from 10.4 13.3% in Canada 36, 37 The reported prevalence of neck pain in the US ranges from 30 50% 38, 39 22.7% in Canada 37 31.7% in Denmark 40 and as high as 66.3% 73% in Sweden 7, 41 Management of Neck Pain in the Healthcare System Neck pain is second only to low back pain as the most common musculoskeletal disorder in population surveys 42, 43 Neck pain is a musculoskeletal pain condition that can encompa ss a wide variety of causes, symptoms, physiological and psychological contributions. In the US healthcare system, neck pain is treated by multiple professionals. In a telephones based survey on persons with neck pain, 71.9% reported seeing a primary care physician, 31.6% an orthopedic surgeon, 29.1% a neurologist, 3.9% a psychologist or psychiatrist, 40.4% a chiropractor, and 35.25% by a physical therapist 8 Number of visits to these providers in a year ranged from 3 17.2 visi ts, with physical therapists being the most frequent provider visited. Moreover, 77.6% of the sample reported seeing greater than 3 provider types for their neck pain 8 US studies on healthcare utilization in 2001 and 2002 reve aled 10.2 million visits to physician offices and outpatient departments for spine pain 44 Interventions for neck pain are often categorized by invasive and non invasive interventions. The Bone and Joint Task Force published a recent series on the treatment of neck pain which examined the literature from 1980 2006 and reported the best available evidence for both surgical and nonsurgical interventions for neck pain. Some common non invasive interventions cited for n eck pain were exercise, manual therapy, physical modalities, cervical collars, low level laser therapy (LLLT), combined approaches and medication. The results of the literature review found that for neck pain other than whiplash associated disorders, manua l therapy, supervised exercises and

PAGE 23

23 LLLT were more effective than sham or alternative interventions, but no intervention was found to be superior to another 44 Surgical procedures for neck pain can be costly and a re often not found to be effective. The Bone and Joint Taskforce examined the use of surgical and injection interventions in a population of neck pain patients that surgery is being performed in the absence of cervical surgery for stabilization of fractur es, dislocation, control of tumor, hemorrhage, infection, threatening neurologic loss or progressive spinal compression exists concerning surgical interventions in persons with neck pain and possibly radiculopathy, when only common aging and degenerative processes are found upon 45 The results of this literature review found that, evidence does not support intra articular steroid injections or radiof requency neurotomy for treatment of neck pain, nor is anterior cervical fusion or cervical disc arthoplasty or cervical disc arthroplsty for radicular symptoms without radiculopathy or underlying pathology indicat ed. Short term improvements with surgical intervention for radiculopathy with pain were found, but there no long term studies to date validate this finding. Lastly, there is support for short term pain relief in cervical radicular symptoms with epidural or selective root injections with corticoste roids, but these findings are not validated in populations examining using this technique to prevent surgical intervention 45 Management of Neck Pain by Physical Therapists Neck pain is a common condition that is often treated by physical thera pists 9 11 Recent evidence suggests that there are a plethora of interventions for neck pain, many of which are not found to be effective or have small effect sizes 10, 46 In order to improve clinical decision making, decrease variability and streamline interventions, treatment

PAGE 24

24 based classification (TBC) systems have been published. 4, 47, 48 By grouping patients into homogenous trea tment groups, interventions should be based on most relevant impairments and the patients should be given interventions for these impairments Published studies examining TBC systems for neck pain have identified possible subgroups for treatment and demo nstrated improved patient outcomes with using a classification system; but were unable evaluate the effect of classification on treatment outcomes in a controlled design or able to identify the healthcare utilization and costs associated with using a TBC a pproach 48, 49 In 2008, the Orthopaedic Section of the American Physical Therapy Association (APTA) created evidence based practice guidelines for orthopaedic physical therapy management of patients with the musc uloskeletal impairment of neck pain 4 The purpose of this guideline was to describe evidence based physical therapy diagnosis, prognosis, intervention and assessment of outcomes for neck pain. Specifically, this gui deline made recommendations based on best available evidence and expert opinion for interventions for neck pain. Overall, the guidelines advocate s for the use of primarily active interventions to promote optimal physical therapy outcomes. A st udy in patien ts with acute LBP examined the effect of guideline adherenc e of physical therapy intervention for low back pain and found that adherence rate to guidelines was 40.4%; those who received guideline adherent care had improved disability and pain outcomes and required fewer visits and subsequently less expenditures for physical therapy 12 Public Health Significance of Neck Pain and Relationship to Rehabilitation A population based estimate reports that 80% of acute neck pain resolves within days to weeks 32 but recurrence rates are high 50 Literature in the low back supports that with early physical therapy intervention, there is a decreased od ds of advance

PAGE 25

25 imaging, physician visits, surgery, injections, opioid medication and subsequent care 51 The literature in neck pain supporting early intervention is less forthcoming. It is generally accepted that neck pain has a favorable course of care 10 ; but recurr ence rates are high 10 Evidence suggests that physical therapy can be an effective secondary prevention strategy for neck pain 52 The totality of these findings, suggest treati ng neck pain in the early may decrease the percentage of the population experiencing chronic or recurrent pain in the future, and thereby reducing unnecessary costs and healthcare utilization, but this has yet to be studied. The overarching g oals of this project are to examine healthcare utilization, expenditures and clinical outcomes in persons with neck pain seeking interventions provided by physical therapists. These issues are paramount to understanding present and future health care costs and expenditures, value of specific services such as physical therapy for treatment of neck pain and improving healthcare delivery for patients with neck pain. Addressing these issues will provide a comprehensive understanding of neck pain and provide nee ded insight into the relationship of healthcare utilization, expenditures and clinical outcomes from physical therapy intervention.

PAGE 26

26 CHAPTER 3 SPECIFIC AIMS AND HYPOTHESES The primary goals of this dissertatio n were to 1) D escribe the clinical characteris tics, physical therapy utilization and costs in persons seeking care for neck pain 2) D etermine whether receiving guideline adherent physical therapy intervention improves clinical outcomes, decreases healt hcare utilization and costs 3) D etermine the exten t to which the timing of receiving guideline adherent care (i.e. early versus later in an episode of pain) improves clinical outcomes and decreases healthcare utilization and costs. The goals were addressed through three inter related specific aims. The fo llowing sections provide initial hypotheses and support for each specific aim Specific Aim 1 : Description of Clinical C haracteristics, Outcomes and Physical Therapy Utilization in Patients with Neck Pain This aim was an exploratory aim. First, we wished to describe the clinical characte ristics, clinical outcomes and physical therapy utilization of patients receiving outpatient physical therapy for neck pain using clinical sub groups. This sample contained clinical sub gr oups of patients with neck pain. Pati ents were categorized into five clinical sub groups : acute neck pain, chronic neck pain, neck pain with arm pain, neck pain with headache and whiplash. Second, we wished to compare clinical outcomes of pain (NPRS) and disability (NDI) among clinical sub g roups and determine if membership in a speci fic clinical sub group improved clinical outcomes The current research on sub groupin g methods in neck pain is not as well developed as these methods are in the low back 5 3, 54 Subsequently, the current literature does not support the effectiveness of using TBCs to improve clinical outcomes. Therefore this aim is focused examining factors that improve outcomes that can be readily disseminated to the current clinical pract ice setting

PAGE 27

27 independent of adoption of a specific classification system and treatment approach. Existing evidence shows age, gender and psychological factors are prognostic of the course of neck pain 33 but less is known about symptom specific factors for outcomes from physical therapy. Currently, we do not operate in a value based healthcare system, rather the current system rewards volume 55 In 2011 the Centers for Medicare and Medicaid Services (CMS) adopted a multiple procedure payment reduction (MPPR) which introduced a 20 25% reduction in reimbursement of the relative value unit (RVU) of any procedure bil led after the first practice expense components of the RVU. Under this reduction, effective April 1 st 2013, a 50% reduction in reimbursement will replace the 20 25% reduction in reimbursement in outpatient settings 56 This is just one example the evolution of healthcare system and the changes to reimbursement practices which necessitates achievin g greater improvements with fewer visits and or billed units. For our third and last point, we wished to examine if membership in a specific clinical sub group is predictive of increased efficiency and value for an episode of physical therapy. Specific Aim 2 : Does Receiving Gui deline Adherent Physical Therapy Intervention Improve Clinical Outcomes, Healthcare Utilization and Costs in Patient s with Neck Pain? The current literature concerning effective interventional strategies for treatment of neck pain are limited; studies have investigated the use of clinical predication rules 57 63 combined interventional strategies 64 but have yet to address how employing these interventional strategies can impr ove not only clinical outcomes but healthcare utilization and costs in patients with neck pain. The Orthopaedic Section of the American Physical Therapy Association (APTA) published clinical practice guidelines for orthopaedic physical therapy management o f patients with the musculoskeletal

PAGE 28

28 impairment of neck pain 4 This guideline made recommendations based on best available evidence and expert opinion for interventions for neck pain. Overall, the guidelines advocate the use of primarily active interventions in conjunction with manipulation/mobilization when indicated to promote optimal outcomes. 65 Fritz et al. (2007) examined the effect of guideline adherence in physical therapy intervention for low back pain and found that adherence rate to guidelines was moderate, but those who received guideline adherent care had improved disability and pain outcomes and required less visits and subsequently less expenditures for physical therapy 12 Therefore in this aim we wished to determine the extent to which patients who receive g uideline adherent interventions have i mproved clinical outcomes d ecreased healthcare utilization and costs for healthcare services compared to patients receiving non adherent guideline interventions We hypothesized that those receiving gui deline adherent care will have improved clinical outcomes from physical therapy, decreased healthcare utilization and costs for healthcare services the year of receiving physical therapy and the following year compared to those who do not receive guideline adherent care. Specific Aim 3 : Does the Timing of Guideline Adherent Care for Patients with Neck Pain Improve Clinical Outcomes, Healthcare Utilization and Costs ? The clinical practice guidelines for neck pain advocate use of specific interventions matche d to clinical presentation in effort to improve clinical outcomes 4 Additionally, the literature supports that a relationship has also been indicated between the timing of physical therapy initiation for acute LBP and subsequent healthcare utilization. Gellhorn et al. (2012) found that patients who received physical therapy early after an episode of acute LBP were at lower risk for subsequent LBP related healthcare

PAGE 29

29 utilization over the following year compared to tho se who received physical therapy at later times 13 It is reasonable to extend the rationale to neck pain; that is, early intervention and evidence based and guideline adherent care can improve clinical outcomes and decrease healthcare utilization and costs. Th is aim synthesizes the principles from specific aims 1 and 2. We wished to determine the extent to which patients categorized into clinical care groups based on timing of physical therapy intervention and receiving guideline adherent care differ in their c linical outcomes, healthcare utilization and costs during the year of receiving physical therapy intervention. Our first hypothesis tests whether patients who receive early guideline adherent intervention have improved clinical outcomes compared to those patients who receive late physical therapy intervention with and without guideline adherent care and patients who receive early non adherent guideline care. Additionally, we hypothesized that patients who receive early guideline adherent intervention would have decreased healthcare utilization and costs for physical therapy and healthcare providers compared to all other groups.

PAGE 30

30 CHAPTER 4 DESCRIPTION OF CLINICAL CHARACTERISTICS, OUTCOMES AND PHYSICAL THERAPY UTILIZATION IN PATIENTS WITH NECK PAIN Background Neck pain is second only to low back pain as the most common musculoskeletal disorder in population surveys and is associated with high injury and disability claims 42, 43 The point prevalence of neck pain varies from 10 20% [1, 12 14] and there is reason to believe the prevalence of neck pain is increasing 16 Approximately 35% of persons with neck pain reported being seen by a physical therapist; the most frequently visi ted provider 8 Physical therapists provide interventions to patients with neck pain 9 11 but variation exists in the inventions provided to patients. To improve clinical decis ion maki ng and provide guidance for physical therapy intervention, Treatment Based Classification (TBC) systems have been published. Evidence supports the use of treatment based classification systems for patients with LBP 5 3, 54 but the evidence for implementation and wide spread adoption of TBCs for neck pain is still limited. Wang et al. (2003), Childs et al. (2004) and Childs et al. (2008) each proposed classification systems for neck pain. 4, 47, 48 The purposes of these TBC systems were to create homogenous groups of patients based on clinical findings and existing evidence to guide treatment by reducing inappropriate interventions for neck pain with the end goal of improving p atient outcomes. The published classification systems for neck pain were beneficial in identifying possible homogenous sub groups to guide physical therapy intervention, but published studies were not able to fully evaluate the effectiveness of using TBCs to improve clinical outcomes. Fritz and Brennan (2007) examined the preliminary use the TBC proposed by Childs et al (2004). The authors

PAGE 31

31 classification groups there were a ssociated improvements in NDI and NPRS scores ; however, the authors were unable to draw conclusions about the effectiveness of the TBC or predictive validity of the TBC used 48 Therefore the implementation of a specific treatment based classification system for neck pain with the purpose of improving clinica l outcomes has been limited at best given the current evidence, lack of validation of a specific classification system and subsequent effectiveness studies. In this study we were interested in taking a different approach f ocused on determining clinical fa ctors that improve outcomes which can be readily disseminate d to the current clinical setting independent of adoption of a sp ecific classification system and treatment a pproach. Existing evidence supports age, gender and psychological factors are prognosti c of the course of neck pain 33 but less is known about symptom specific factors for outcomes from physical therapy Specifically we are interested in examining the differences in clinical outcomes among groups o f patients with a primary clinical presentation of acute neck pain, chronic neck pain, neck pain with arm pain, neck pain with headache and neck pain from whiplash in an outpatient clinical setting. Purpose The purposes of this study were to 1) Describe t he clinical character istics, clinical outcomes, and physical therapy utilization of patients receiving outpatient physical therapy for neck p ain using clinical sub groups; 2) C ompare clinical outcomes of pain (NPRS) and disability (NDI) among clinical sub groups and determine if membership in a specific clinical sub group increases the odds of achieving minimal clinically import ant

PAGE 32

32 difference NPRS or NDI and 3) E xamine if membership in a specific clinical sub group is predictive of increased efficiency and value for an episode of physical therapy. Methods Database Data for this study was collected from 11 outpat ient physical therapy clinics at In termountain Healthcare Inc Intermountain Healthcare is a private, non profit integrated healthcare system locate d in the Salt Lake City UT and surrounding regions. Data for patients receiving physical therapy are entered into an electronic database that tracks clinical outcomes and process data related to the delivery of physical therapy services. These variables i clinical sub grouping related to impairments and symptoms reported at initial evaluation. Utilization variables included number of visits, length of stay (LOS) and total c harges for physical therapy. C linical outcomes that were collected in the sample in this study were the Neck Disability Index (NDI) and Numerical Pain Rating Scale (NPRS). Clinical outcomes have been tracked since 2002 and previous studies utilizing this database report that data for 9 4.5% of patients receiving physical therapy were collected 66 Sample The sample in this study included patients who underwent an episode of physical therapy care for treatment of neck pain. An episode of care was defined as the date from initial evaluation to the date of the last visit. Data for this study w as retrospectively retrieved from the electronic clinical outcomes database from the dates January 1, 2008 to December 31 st 2012. Patients included in the samp le must have met the additional inclusion criteria of: non surgical patient, NDI score of 10 or greater and NPRS of 2 or greater at initial evaluation, 2 or more visits and a LOS between 2 180 days.

PAGE 33

33 Patient clinical s ub grouping Patients included in the a nalysis received treatment by a physical therapist for a primary compliant of neck pain. Patients were categorized in clinical sub group s These sub groups included patients with a presentation of: acute neck pain, chronic neck pain, neck pain with arm pai n, neck pain with headache and neck pain with whiplash. Descriptions of these clinical sub groups can be found in Figure 1 1 Clinical o utcomes During the episode of care patients completed the NDI and NPRS. Scores from the initial visit and final visit we re recorded. The NDI is a condition specific outcome measure comprised of 10 items; each item is scored from 0 to 5. The total score is expressed as a percentage and is reflective of a level of disability related to neck pain where high percentages are re lated to higher disability. This questionnaire asks the patient to rate how their neck pain affects their tolerance for activities of daily living. The NDI is a commonly used outcome measure for people with neck pain is found to be reliable and valid in th e neck pain population 67 71 Both the NDI and NPRS exhibit fair to moderate test retest reliability in patients with mechanical neck pain and both instruments also show adequate responsiveness in this patient population 72, 73 Change in NPRS and change i n NDI scores were calculated for each patient in the sample. This was calculated by subtracting the initial score from the score at the last visit for NPRS and NDI. Patients were categorized as achieving Minimal Clinically Important Difference (MCID) if th e amount of improvement from initial evaluation to

PAGE 34

34 discharge was equal to or greater than the MCID for each of the measures. For the NDI, MCID was defined as a 19 point change and for the NPRS the MCID was1.3 points 72 If patients initial NDI score was less than the MCID for the NDI, they were categorized as improved if they had a last visit score of 0 out of 100, indicating no disability. A variable for value of physical therapy was created. This variable was calculated by dividing the change in NDI during an episode of care by the total charges for the physical therapy divided by 100 during an episode of care (Change in NDI/ (Charges for PT/100). Smaller values indicate lower value and larger number higher value. Additionall y a variable for efficiency of physical therapy was created by dividing the change in NDI score by the to tal number of visits during an episode of care (Change in NDI/ # of visits to PT). Smaller values represent lower efficiency and larger values higher efficiency. Health related quality of l ife We planned to measure the differences in HRQL between clinical sub groups. In order to do this, we needed to convert the NDI into a SF 6D score. The SF 6D is a widely used classification for describing health status. The SF 6D provides a score for general health from 0 1, where 0 equals death and 1 equals perfect heal th 74 Using a standard gamble technique, admission, dischar ge and change scores for NDI were converted to SF 6 scores using the equation (SF 6D= 0.0115(NDI Score) + 0.8383). This equation was published by Richardson and Berven (2012) who found that correlations between the NDI and SF 6D are strong; and high corre latio ns between SF 6D and NDI score permit models to be used to calculate utilities, change in utilities and quality adjusted life years 75

PAGE 35

35 Utilization For each episode of care included in the sample, the length of stay (LOS) was defined as the number of days between the initial evaluation and the last visit. The number of visits was defined as the number of visits that patients attended during an episode of care. Total cost s for physical therapy were defined as the total charge amount for the episode of care All cost s were adjusted t o 2008 costs by adjusting cost s from subsequent years by a rate of 0.968. The rate used was published in th e Consumer Price Index by the US bureau of labor and statistics for medical ca re services 76 Data Analysis Data analysis was performed using SPSS statistical software (version 21.0). Significance level for all analyses was set at 0.05. Description and comparison of p atient characteristics, clinical outcomes and physical therapy u tilization Descriptive statistics were calculated for all persons with neck pain in this sample. Means and standard deviations were reported for continuous vari ables and percentages were reported for categorical variables. Baseline patient characteristics, clinical outcomes, HRQL and physical therapy utilization were compared among clinical sub groups using chi squared analyses for categorical variables and one way ANOVAs for continuous variables. For variables with skewed distributions (LOS, visits and total c ost s for PT) Kruskal Wallis test was used to compare clinical sub groups and medians with interquartile range were reported.

PAGE 36

36 Clinical o utcomes The proporti on of patients in each clinical sub group achieving MCID for the NPRS and NDI were compared using separate binary logistic regression models. Clinical sub group was the main predictor of interest. Covariates in the models included age, gender, LOS and numb er of visits. Separate repeated measures ANOVA models (Within Time; Between Clinical sub group) were performed to examine differences in change in clinical outcome scores over time. Dependent variables in se parate models were NDI score, NPRS and HRQL (SF 6D) score We were specifically interested in the effect of clin ical sub group on NDI score, NPRS and HRQL (SF 6D) score Covariates included in the models were age, gender, LOS and number of visits Examination of c l inical predictors of value and e fficie ncy W e examined predictors of value and efficiency using separate linear regression models. The model for predicting value (change in NDI/charges for PT /100 ) included baseline NDI score, age LOS and number of visits as covariates The model for predicting efficiency (change in NDI/number of visits) included baseline NDI score, age and LOS as covariates. Clinical sub groups and gender were included as factors in each model, with the chronic neck pain group as the reference group and males as the reference g ender. Results The initial sample included 3484 patients who received physical therapy intervention for neck pain during this time period. After inclusion criteria were met and missing observations removed, 2732 episodes of care remained in the sample. See Figure 1 2 for derivation of the sample.

PAGE 37

37 Description of patient characteristics, clinical outcomes and physical therapy utilization The mean age of patients in the sample was 49.39(SD15.91) years old. Females comprised 68.8% of the total sample. Patient s were seen for a median of 5 visits and had a median LOS of 25 days. The median total cost s for physical th erapy in this sample were 744.83 US dollars. At the initial visit in the sample, patients in the sample reported an average baseline NDI score of 36 .53(SD16.16) and NPRS of 5.55(SD2.19). At the last visit, patients reported an average NDI score of 25.66(SD17.02) and NPRS of 3.53(SD2.44), with a mean change score of 10.86(SD14 .40) points for the NDI and 2.02(SD2.46 ) points for NPRS. Approximately twenty nine percent of the sample achieved MCID for the NDI and 55.4% for the NPRS. When examining value and efficiency, on average for every 100 dollars charged for physical therapy, patients reported approximately a 1.6 point change in NDI score; addit ionally physical therapy efficiency was 2.26 point change in NDI per visit on average in the sample. For HRQL, t he average admission SF 6D score in the sample was 0.42(SD0.19), the average discharge score was 0.54(SD0.42) and the average change score was 0.12(SD0.17). Comparison of patient characteristics, clinical outcomes and physical therapy utilization by clinical sub group The most common clinical sub group was the chronic neck pain group comprising 39.5% of the total sample followed by the neck pa in with arm pain group 22.5%, acute neck pain 16.5%, headache with neck pain 12.4% and whiplash 8.0%. When comparing the clinical sub groups, those patients in the whiplash group were the youngest with a mean age of 42.84(SD15.36) years and the oldest gro up was

PAGE 38

38 the patients in the chronic neck pain group aged 52.45(SD16.77) years. The majority of the sample were female in each clinical sub group with proportions ranging from 64.1 79.0%, with the highest proportion of females in the headache group and the lowest proportion in the acute neck pain group 64.1%. Clinical sub groups varied in their median LOS with those in patients in the acute neck pain group having a median LOS of 21 days versus the whiplash group with the longest LOS at 35 days. There was l ess variation in the number of visits with all groups reporting a median number of visits of 5 except whiplash reporting 6 visits. When examining baseline NDI score and NPRS, patients in the chronic neck pain group entered with the lowest reported disabili ty on the NDI 34.31(SD15.22) and NPRS 5.28(SD2.14). The patients in the acute neck pain group 40.18(SD17.54) and whiplash group 40.70(SD17.27) reported the highest baseline NDI scores. The patients in the acute neck pain group reported the highest base line NPRS 5.96(SD2.23). The greatest change in NDI 16.31(SD16.15) and NPRS 2.76(SD2.51) was demonstrated by the patients in the acute neck pain group. The proportion of patients in each group achieving MCID in NDI scores were the highest in patients in the whiplash group (40.6%), followed by patients in the acute neck pain group (38.8%), then (29.3%) in the neck pain with arm pain, (26.3%) in the headache group and (23.6%) the chronic group. Overall a greater proportion of patients in each group achieved MCID in NPRS with 68% of the acute neck pain group achieving MCID, followed by the whiplash group (62.1%), neck pain with arm pain group (56.4%), chronic group (50.9%) and the headache group (46.4%).

PAGE 39

39 The acute neck pain group demonstrated the highest effi ciency and value c ompared to the other groups where on average for every 100 dollars charged for PT patients reported a 2.42 point change in NDI (value) and 3.65 point change in NDI per visit (efficiency). The other clinical sub groups ranged from 1.27 1.6 6 reported change in NDI per 100 dollars spent (value) and 1.84 2.39 change in NDI per visit (efficiency). The patients in the chronic neck pain clinical sub group had the highest admission HRQL (SF 6D) score (0.44) and those in the acute and whiplash gro ups had the lowest scores at approximately 0.37 each. At discharge those in the acute neck pain group had the highest HRQL (SF 6D) (0.57) followed by the chronic group (0.54), headache group (0.54), neck pain with arm pain group (0.53) and whiplash group ( 0.53). The greatest changes in HRQL were reported by patients in the acute neck pain group with a 0.19 change. The least change was reported by patients the chronic neck pain group (0.099). Clinical sub groups significantly differed in all clinical charac teristics with the exception of discharge NDI score (p=0.114) and discharge SF 6D score (p=0.109). See Table 1 1. Disability A significant interaction of time and clinical sub group was found for NDI score after controlling for age, gender, LOS and numbe r of visits (F 4,2673 = 23.02, p<0.0001). When compared to the chronic group, patients in the acute neck pain group (p<0.0001) were 2 times more likely, patients in the whiplash group (p<0.0001) were 2.2 times more likely and patients in the neck pain with arm pain group (P=0.03) were 1.29 times more likely to achieve MCID for NDI. The headache group did not significantly

PAGE 40

40 differ in odds of achieving MCID compared to th e chronic group (p=0.311). See Table 1 2. Pain There was a significant interaction of tim e and clinical sub group for NPRS (F 4,2671 = 16.47, p<0.0001), after controlling for age, gender, LOS and number of visits. Patients in the acute group therapy were 2.07 times more likely (p<0.0001), and patients in the whiplash group 1.58 times more like ly (p=0.004) than patients in the chronic group achieve MCID in NPRS. The odds of achieving MCID for NPRS did not differ for the patients in the headache (p=0.241) or neck pain with arm pain group (p=0.105) when compared to the chronic group. See table 1.3 Health related quality of l ife (HRQL) There was a significant interaction of time and clini cal sub group for HRQL (SF 6D) (F 4,2676 = 23.03, p<0.001 ), after controlling for age, gender, LOS and number of visits Efficiency and v alue When examinin g the eff iciency and value of physical therapy for patie nts during an episode of care, we were specifically interested in determining if clinical sub group was a significant predictor of thes e dependent variables. Table s 1 4 and 1 5 present models predicting effici ency and value. The contribution of the specific predictors is expressed by the regression coefficient. Positive coefficients indicate the change in efficiency was greater in the presence of the predictor and a negative coefficient attenuates the effect of the predictor on the dependent variable. Furthermore the magnitude of the coefficient indicates the strength of the association of the predictor with dependent variable. Additionally, for the clinical sub groups, the coefficient can be

PAGE 41

41 interpreted as the increase in the dependent variable when a patient is in that clinical sub group compared to the chronic clinical sub group when all other predictors are held constant. When examining predictors of efficiency, after adjusting for age, gender, LOS, baseline NDI and number of visits, compared to the chronic group, the acute neck pain group ( B = 1.29, p<0.001) was the only clinical sub group that was a significant predictor of efficiency in the model but the whiplash sub group approached significance (B =0.48, p=0.06) Patients in the acute sub group had the greatest change in NDI per visit with 3.63 point change and the chronic subgroup was the least efficient with only a 1.84 point change in NDI per visit. Although modest, this model accounted for 13.1% of the total variance in efficiency. When examining predictors of value, after adjusting for age, gender, LOS and baseline NDI, compared to the chronic group, the acute group is the only significant clinical sub group predicting value in the model( B=0.79, p<0. 001) ; again whiplash approached significance (B=0.35, p=0.06) The acute sub group demonstrated with highest value with a 2.42 point change in NDI per 100 dollars spent, where the chronic subgroup demonstrated the least amount of value with only a 1.27 poi nt change in NDI per 100 dollars spent. This model accounts for 11.4% of the variance in value. Summary of Findings Evidence shows that there is variation in outcomes in persons with neck pain and outcomes are not cons istently favorable, some neck pain resolves where others continue to experience persistent pain and disability 77 We know that certain factors are prognostic for the course of neck pain and what interventions are likely to

PAGE 42

42 produce a favorable outcome 58, 59, 61, 78, 79 but what we currently know little about symptom specific prognostic factors of clinical outcomes from physical therapy 4 This study describes the cl inical characteristics, clinical outcomes and predictors of efficiency and value in patients receiving physical therapy intervention for neck pain and examined differences in these variables among clinical sub groups. Past research has examined the use of TBC systems, but these classification systems have not been validated, found effective in the clinical setting or been accepted as readily as it has been in for the low back. This study was designed to augment the existing knowledge on TBC system literatur e by examining the prognostic value of symptom specific clinical presentations of neck pain These findings may be used by clinicians in conjunction with current classification systems to improve clinical decision making by physical therapists by adding a temporal component to the classification systems. O verall in this study patients with acute neck pain demonstrated the greatest improvements in pain ( NPRS ) disability ( NDI ) had the highest proportions of patients achieving MCID in NDI and the highest pr oportions achieving MCID in NPRS. Also patients with acute neck pain had greatest change in NDI per dollar spent (value) and required fewer visits for greater change in NDI (efficiency). Conversely, overall patients with chronic neck pain had poorer outco mes with the least change in NDI and NPRS, smaller proportions of patients achieving MCID for NDI, and smallest change in NDI per dollar spent and smallest change in NDI per visit. This study provides information about the prognostic value of clinical sub grouping. Specifically, those patients with acute neck pain tend to do better overall on most metrics where patients with chronic neck pain tend to do worse. Patients with the

PAGE 43

43 clinical sub grouping of whiplash behave similar to patients with acute neck pa in when looking at clinical outcomes, but have longer LOS and number of visits during an episode of physical therapy. There are several limitations in the current study. First, the clinical care groups were defined by temporal parameters for some groups a nd primary clinical presentation for others. Some patients may present with both temporal and a clinical presentation consistent with one of the other groups. By default, patients would be categorized by their clinical presentation (headache, neck pain wit h arm pain or whiplash). Second, there was no attempt to standardize interventions provided by physical therapists. Therefore patients may have received differential treatment based on their clinical sub group. Lastly, although the authors made attempts to control for known confounders such as age and gender, we were unable to control for psychological status or comorbid conditions which may affect outcomes 80 Understanding factors which can influence the clinical co urse of care for neck pain is paramount to improving not only clinical outcomes but decreasing the burden neck pain places on the individual and the healthcare system. The current study found that those patients with presenting with acute neck pain have th e most favorable clinical outcomes compared to patients with neck pain with arm pain, neck pain with headache, neck pain with whiplash and chronic neck pain. Furthermore, patients with chronic neck pain have the least favorable clinical outcomes. The resul ts demonstrate that when intervention for neck pain is provided early, more favorable outcomes are achieved. Although the authors cannot say definitively why the patients in the acute neck pain group performed better due to the study design, we may surmise that patients with

PAGE 44

44 acute neck pain have a greater capacity for improvement (i.e., higher pain at initial visit) when compared to other groups such as the chronic group, where symptoms are less likely to change great amounts due to the nature of chronic pa in. Future policy and efforts should be directed a treating patients in the earlier stages of neck pain to increase the likelihood of a successful outcome verses treatment in the chronic stages where outcomes tend to be less favorable. Future studies are n eeded to examine how membership in different clinical sub groups, specifically acute and chronic neck pain groups affects healthcare utilization and costs.

PAGE 45

45 Group Description Acute Neck Pain <4 weeks duration, not presenting with primary complaint of hea dache or neck pain with arm pain or whiplash Chronic Neck Pain >4 weeks duration, presents with or without ROM restrictions, not presenting with primary complaint of headache or neck pain with arm pain or whiplash. Neck Pain with Arm Pain Presents with p rimary complaint of neck pain with arm pain, can be of varying duration or ROM restrictions, does not have a primary complaint of neck pain with headache or whiplash Neck Pain with Headache Presents with primary complaint of neck pain with a headache, can be of varying duration or ROM restrictions, does not have primary complaint of neck pain with arm pain or whiplash Whiplash Presents with primary complaint of whiplash injury with neck pain, can be of varying duration or ROM restrictions, does not have p rimary complaint of neck pain with arm pain or headache Figure 1 1 Description of Clinical Sub groups

PAGE 46

46 Figure 1 2. Derivation of Sample for Specific Aim 1

PAGE 47

47 CHAPTER 5 DOES RECEIVING GUIDELINE ADHERENT PHYSICAL THERAPY INTERVENTION IMPROVE CLINICAL OU TCOMES, HEALTHCARE UTILIZATION AND COSTS IN PERSONS WITH NECK PAIN? Background Healthcare utilization and medical costs for spine conditions have increased steadily from 1997 2005 and these increases in utilization did not correlate with improvements in h ealth 20 Neck pain and its related disability have an enormous impact on individuals and their families, communities, healthcare systems and businesses 22 25 Approximately 22 70% of the population will have neck pain at some point in their life 2, 3 and of those persons that report current neck pain, 50 75% report having neck pain again in 1 to 5 years later 33 Because of the economic impact and burden of the disease, neck pain may be thought o f as a significant public problem with impacts on the individual and the healthcare system 1 In efforts to decrease inappropriate interventions and practice variation, clinical guidelines have been published for a wide variety of conditions 81 83 The intents and purposes of these guidelines are to decrease utilization of ineffective treatments and therapies and increase the use of evidence based interventions to improve pati ent outcomes and decrease costs 65 Fritz et al. (2007) examined the effect of guideli ne adherence in physical therapy intervention for low back pain and found that adherence rate to guidelines was moderate, but those who received guideline adherent care had improved disability and pain outcomes and required fewer visits and subsequently le ss expenditures for physical therapy 12 In 2008, the Orthopaedic Section of the American Physical Therapy Association (APTA) created evidence based practice guidelines for orthopaedic physical therapy

PAGE 48

48 management of patients with the musculoskeletal impairment of neck pain 4 The purpose of this guideline was to describe evidence based physical therapy diagnosis, prognosis, intervention and assessment of outcomes for neck pain. Specifically, this guideline made recommendations based on best available evidence and expert opinion for interventions for neck pain. Intervention recommendations included stretching and strengthening exercises, centralization exercises, upper quarter an d nerve mobilization procedures, mechanical traction and patient education in conjunction with exercises. Overall, the guidelines advocate the use of primarily active interventions in conjunction with manipulation/mobilization when indicated to promote opt imal outcomes. To date no studies have been identified examining the relationship between receiving physical therapy intervention adherent with the clinical practice guideline for neck pain and clinical outcomes, healthcare utilization and costs related t o neck pain. Purpose The purposes of this study were to determine the extent to which patients who rece ive gui deline adherent care differ in 1) C linical outcomes 2) H ealthcare utilization and 3) C osts for healthcare services compared to those patients who receive guideline non adherent care We hypothesize that those patients receiving guideline adherent care will have improved clinical outcomes from physical therapy, decreased healthcare utilization and decreased costs for healthcare services during the year of receiving physical therapy and the following year compared to those who do not receive guideline adherent care.

PAGE 49

49 Methods Database Data for this study was collected from 11 outpatient physical therapy clinics of In termountain Healthcare Inc. Intermou ntain Healthcare is a private, non profit integrated healthcare system located in the Salt Lake City, UT and surrounding regions. All therapists working these clin ics were salaried employees of Intermountain Healthcare and did not receive financial incenti ves based on number or type of billing codes used per patient visit. Reimbursement for physical therapy was based on fee for service for all patients in this analysis 12 Data for patients receiving interventions pr ovided by a physical therapist are entered into an electronic clinical outcomes database that tracks outcomes and process data related to the delivery of services by physical therapists. These variables include nd subgrouping related to impairments and symptoms reported at initial evaluation. Physical therapy utilization variables in this database include number of visits, length of stay (LOS) and total cost of physical therapy. Clinical outcomes that were collec ted in the sample in this study were the Neck Disability Index (NDI) and Numerical Pain Rating Scale (NPRS). Clinical outcomes have been tracked since 2002 and previous studies utilizing this database report that data for 94.5% of patients receiving physic al therapy were collected 66 Records from the clinical outcomes database were linked to a billing database using an enterprise master patient index number to capture healthcare utilization and costs in the sample. This database contains current procedural terminology (CPT) codes billed by the physical therapist providing interventions at each visit. Healthcare utilization and costs for services provided by healthcare providers related to an

PAGE 50

50 International Classific ation of Diseases code ( ICD 9 ) pertaining to the neck region were extracted for patients in the sample. The total number of visits to healthcare providers, number of prescriptions, number of diagnostic images and costs for all visits to healthcare provider s and cost of prescriptions were extracte d for the year of and year following being admitted to physical therapy for an episode of care for all patients in the sample. Sample The sample in this stud y included patients who underwent an episode of physical t herapy for treatment of neck pain. An episode of care was defined as the time from initial evaluation to the time of the last visit. Data for this study w as retrospectively retrieved from the electronic clinical outcomes database from the dates January 1 2008 to December 31, 2012. A subgroup of patients with private insurance was selected for this analysis. Patients included in the sample met the additional inclusion criteria of: NDI score of 10 or greater at initial evaluation and 2 or more visits and a LOS gr eater than 2 days. See Figure 2 1 for derivation of sample. Determining Adherent Care To determine a dherent care groups the episode of care was divided into 2 phases. The first phase consisted of physical therapy interventions received from 0 14 day s. Phase II treatment included interventions received from day 15 and forward. The determination of proportion of active interventions received by patients was determined separately for each phase. If patients did not enter phase II, the proportion of acti ve interventions received in phase I was used to determine adherent care group for patients during an episode of care.

PAGE 51

51 CPT codes billed for physical therapy visits occurring in each phase of treatment were examined and categorized as a ctive, passive or all owed. See Table 2 1. Codes classified as active indicated procedures adherent to clinical guidelines for neck pain for physical therapists 4 Passive codes were those that indicated treatment inconsistent with gui deline recommendation or there was weak or inconsistent evidence supporting the use of the intervention for neck pain. Allowed codes were codes that could not be adequately categorized into active or passive codes or were evaluation, testing or equipment c odes. Similar methodology had been previously used by Fritz et al (2007) to determine the effect of guideline adherence in patients with low back pain 12 Our study used similar methodology to Fritz et al (2007) f or determining adherent and non adherent care In this study manual therapy was included in phase I as an active code and an allowed code in phase II. The clinical guidelines recommend manipulation (thrust) and mobilization (non thrust manipulation) as an intervention for mechanical neck disorders and should be combined with exercise for maximum benefit 64 Also, Fritz et al (2007) recommend ed that manipulation and mobilization be received early in an episode of care 48 Moreover, the guidelines for neck pain advocate t hat every patient visit to a physical therapist should include primarily active treatments, such as strengthening, coordination and endurance 84, 85 and stretching 86 interventions. The number of active and passive codes billed at each visits were recorded. For each phase, the percentage of active codes was calculated as [(number of active codes/ (number of active codes + number of passive codes)) x100=%Active Care]. Guideline adherent care was defined as occurring when the percentage of active codes within the phase was at least 75%, and each visit must have included an active code.

PAGE 52

52 A dherent care was calculated two ways in this sample. First, adherent care w as calculated a s continuous variable where the % adherence for each patient refers to the ratio of active care to passive care received by phase and for the entire episode of care. Second, a dichotomous variable to represent receiving adherent care was calculated. Two groups were point of receiving at least 75% active care in both phases of treatment. This algorithm has been previously published 12 Clinical Outcome Measures Patients completed the Neck Disability Index (NDI) and the Numerical Pain Rating Scale (NPRS) at the initial visit and final visit. The Neck Disability Index is a condition specific outcome measure c omprised of 10 items; each item is scored from 0 to 5. The total score is expressed as a percentage and is reflective of a level of disability related to neck pain where high percentages are related to higher disability. This questionnaire asks the patient to rate how their neck pain affects their tolerance for activities of daily living. The NDI is a commonly used outcome measure for people with neck pain is found to be reliable and valid in the neck pain population 67 71 Both the NDI and NPRS exhibit fair to moderate test retest reliability in p atients with mechanical neck pain and both instruments also show adequate responsiveness in this patient population 72, 73 Change in pain and change in disability scores were calculated for each patient in the sampl e. This was calculated by subtracting the discharge score from the score at the initial visit for NPRS and NDI. Change score was then divided by the score at the

PAGE 53

53 initial visit then multiplied by 100 to create a percentage improvement. A dichotomous variabl e to indicate a successful outcome from physical therapy was created. A 50% therapy as successful and less than 50% as unsuccessful. Additionally, a dichotomous variable t o indicate successful pain management was created with a 50% improvement in NPRS or greater was determined to be successful pain management and less than 50% was determined to be unsuccessful pain management. This threshold has been used in previous studie s 12, 87, 88 Health Related Quality of Life We wished to measure the differences in HRQL between adherent care groups. In order to do this, we needed to convert the NDI into a SF 6D score. The SF 6D is a widely us ed classification for describing health status. The SF 6D provides a score for general health from 0 1, where 0 equals death and 1 equals perfect health 74 Using a standard gamble technique, admission, discharge and change scores for NDI were converted to SF 6 scores using the equation (SF 6D= 0.0115(NDI Score) + 0.8383). This equation was published by Richardson and Berven (2012) who found that correlations between the NDI and SF 6D are strong; and high correlatio ns between SF 6D and NDI score permit models to be used to calculate utilities, change in utilities a nd quality adjusted life years 75 Financial Outcomes and Healthcare Utilization Physical therapy utilization and c osts For each episode of care included in the sample, the length of stay (LOS) was defined as the num ber of days between the initial evaluation and the last visit. The number of visits was defined as the number of visits that patients attended during an

PAGE 54

54 episode of care. Total costs for physical therapy were defined as the total cost for the episode of car e. All costs were adjusted to 2008 costs by adjusting costs from subsequent years by a rate of 0.968. The rate used was from the Consumer Price Index published by the US bureau of labor and statistics for medical care services 76 Healthcare utilization and c osts In addition to physical therapy, data was retrieved from patients in the sample who received services from a healthcare provider with an ICD 9 diagnos is of neck pain or neck pain related diagnosis. Utilization variables included number of visits to healthcare providers, number and type of prescriptions and number of diagnostic images. Cost variables included costs for services from healthcare providers and costs for prescriptions and were adjusted to 2008 costs 76 These variables were examined for the year of admission to physical therapy and the year follo wing admission to physical therapy Data Analysis In unadjusted analyses, patient chara cteristics, clinical outcomes and utilization variables where compared between patients receiving adherent care verses non adherent care using independent samples t tes 2 tests for categorical variables. When comparing utilization and cost variables between groups, Mann Whitney U nonparametric tests were used due to violations of assumptions of normality for these variables. Comparison of Clinical Outcomes Separate binary logistic regression models were built to determine if receiving adherent care improved the odds of achieving a successful outcome or successful pain management with physical therapy intervention. Achieving 50% change in N DI

PAGE 55

55 (successful physical therapy outcome) and 50% change in pain (successful pain management) were the dependent variables of interest in these models. In addition to adherent care group, covariates in the models incl uded age, gender, LOS for PT, number of PT visits and comorbid LBP R epeated measures ANOVA models (Within Time; Between Adherent care group ) were performed to examine differences in change in HRQL (SF 6D) scores over time. Dependent variable in the models was HRQL (SF 6D) We were specifical ly interested in the effect of adherent care group on HRQL (SF 6D) score C ovariates included in the model were age, gender, LOS and number of visits Comparison of Healthcare Utilization We examined predictors of healthcare utilization using separate Pois son regression models. Incident rate ratios (IRR) with confidence intervals were reported. We were interested in examining predictors of healthcare utilization the y ear patients were admitted to PT and the year after admission to PT In separate regressio n models, we examined predictors of: (1) number of visits to PT (2) number of visits to healthcare providers the year of PT, (3) number of prescriptions the year of PT (4) number of diagnostic imaging the year of PT, (5) number of visits to healthcare pro viders the year after PT, (6) number of prescriptions the year after PT (7) number of diagnostic imaging the year after PT. Predictors in each the models varied slightly depending on the nature of the dependent variable. For the model examining number of visits to PT, predictors in the model included: receiving adherent care, age, gender, admission NPRS and NDI score. To examine the number of prescriptions the year of receiving physical therapy predictors included: receiving adherent care, age, gender and admission NPRS The

PAGE 56

56 model predicting number of visits to healthcare providers the year of PT, predictors in the model included: receiving adherent care, age, gender, admission NDI and NPRS and number of PT visits. The model for diagnostic imaging included the predictors of: receiving adherent care, age, gender, admission NDI and NPRS. The models predicting utilization the year after PT used discharge pain and disability as covariates where admission NDI or admission NPRS were included in the models. To exa mine prescription drug use during the year of receiving physical therapy, separate binary logistic regression models were built to assess the odds of being prescribed medication for pain the year of being admitted to PT. We examined the odds of being presc ribed (1) non narcotic analgesics, (2) opioids, (3) anti inflammatory medication and (4) corticosteroids. Predictors in the models included receiving adherent care, age, gender, PT LOS, number of PT visits, admit NPR and presence of comorbid LBP. Lastly to examine costs for an episode of physical therapy care, costs for healthcare visits and prescriptions, gamma regression with log link function was performed. This type of regression was used because the cost variables were positively skewed. Outcome variab les of interest in these models were (1) total costs for an episode of PT (2) total costs for healthcare visits the year of admission to PT (3) total costs for prescriptions the year of admission to PT (4) total costs for healthcare visits the year afte r admission to PT and (5) total costs for prescriptions the year after admission to PT Exponentiated betas with 95%CI were reported to improve interpretability of findings.

PAGE 57

57 Predictors in the model for costs for an episode of PT included: receiving adheren t care, age, gender, number of visits and LOS, admission NPRS, admission NDI score and comorbid low back pain. Predictors in the model for costs for healthcare visits included: receiving adherent care, age, gender, number of PT visits, admission NPRS, admi ssion NDI score and comorbid LBP. Predictors in the model examining costs for prescriptions during the year of PT included: receiving adherent care, age, gender, admission NPRS and comorbid LBP. Predictors in the models for costs for healthcare visits and prescriptions were the same as the previous models with exception of using discharge scores in the models for clinical outcomes. IBM SPSS 21.0 and Intercooled STATA 12.1 were used to perform analyses. Significance level was set at 0.05 for all analyses. Re sults The initial sample in this study included 3485 episodes of care. After inclusion criteria were met and missing observations removed, 298 episodes of care were included in the analyses. See Figure 2 1 for derivation of the sample. Eleven percent of th e patients in the sample were categorized as receiving adherent care (75% or greater active codes) during an episode of physical therapy. When examining each phase individually, 13.7% of the sample received adherent care in phase I and 1.7% received adhere nt care in phase II. When examining adherent care a s a continuous variable the mean percentage of active care received by patients was 40.47% in phase I and was 20.83% in phase II. Approximately 27% of sample did not have an episode of care that last gre ater than 14 days. In unadjusted analyses patients in the adherent care groups did not significantly differ in baseline characteristics of age (p=0.18), gender (p=0.70), admission NPRS

PAGE 58

58 (p=0.79), admission NDI scores (p=0.42) or SF 6D scores (p=0.50), but did differ in proportions of patients with comorbid LBP (p=0.02), with those patients in the adherent care group having a higher proportion of p atients with comorbid LBP. See Table 2 2 Nor did patients in adherent care groups differ in discharge NDI score (p=0.84), NDI change score (p=0.54), discharge NPRS (p=0.37), NPRS change (p=0.55), or proportions in adherent care groups achieving 50% change in NPRS (p=0.21), discharge SF 6D score (p=0.26) or change in SF 6D score (p=0.51). But patients did differ on the proportion of patients achieving a 50% change in disability score (p=0.05), where the non adherent care group had higher proportions of patients achievi ng 50% change in NDI. Adherent c are groups also differed in number of visits for PT (p<0.001), LOS ( p<0.001) and costs for physical therapy (p<0.001) where the adherent care group attended an average of 3 fewer visits, had an average of 34 fewer days for an episode of care and had an average of 587.5 2 less total costs for PT. See Table 2 3. When comparin g healthcare utilization and costs for visits to healthcare providers, prescription medication and diagnostic imaging for the year of being admitted to PT in unadjusted analyses the adherent care groups only significantly differed in the number of visits to healthcare providers the year of PT (p<0.001), were those in the adherent care group had approximately half the visits of the non adherent care group. Patients did not differ in the costs for visits to healthcare providers (p=0.07), number of prescripti on medications (p=0.59) or costs of prescription medications (p=0.99), number of diagnostic images related to the neck (p=0.09) or the proportion of non narcotic analgesics(p=0.55), opioids (p=0.55), anti inflammatory (p=0.57), musculoskeletal

PAGE 59

59 therapeutic agents (p=0.51) or corticosteroids (p=0.98) prescribed the year of PT based on a dherent care group. See Table 2 4 Furthermore, patients in adherent care groups did not differ in the number of visits (p=0.31) or costs (p=0.23) to healthcare providers, numbe r of (p=0.24) or costs (p=0.24) for prescription medications or the number of diagnostic images (p=0.52) the year af ter being admitted to PT. See Table 2 5. Clinical Outcomes Successful ou tcome in d isability We were interested in examining the odds of ac hieving a successful outcome and successful pain management with physical therapy intervention between adherent care groups. When controlling for age, gender, PT LOS, number of PT visits and comorbid LBP the odds of achieving a successful outcome were not different if patients received adherent care or non adherent care (aOR=0.29, p=0.058). But the presence of comorbid LBP decreased the odds of achieving a successful outcome by approximately 50% (aOR= 0.49, p=0.04). Successful pain m anagement When examining successful pain management from physical therapy intervention, there were no differences in the odds of achieving successful pain outcome from PT between groups (aOR= 0.75, p=0.56) nor did comorbid LBP change the odds of successful pain management (aOR=0. 70, p=0.24) See T ables 2 6 and 2 7 for models.

PAGE 60

60 HRQL There were no main effects of adherent care group (F 1,290 = 2.40, p=0.12 ) or two way interaction s for changes in HRQL (SF 6D) scores over time between adherent care groups (F 1,290 =0.59, p=0.48 ) Healthc are U tilization during the Year of Admission to Physical T herapy Those rec eiving adherent care, attended 5 4% fewer PT visits ( Adjusted Mean Difference= 3.63 visits ) during an episode of care (IRR= 0.44, p=<0.001 ), had 2 5% fewer prescription medication (Adjus ted Mean Difference=1.00 prescription ) ( (IRR= 0.75, p=0.02), atten ded 46 % fewer visits to healthcare providers (Adjusted Mean Difference=7.26 visits ) (IRR=0.54, p<0.001) and had 43 % fewer number of diagnostic images the year of PT (Adjusted Mean Difference = 0.43 images ) (IRR=0.57,p=0.02) compared to those receiving non adherent care, while holding all other variables constant in each model. See table 2.8. During the year of admission to physical therapy, receiving adherent care did not change the odds of bein g prescribed non narcotic analgesics (p=0.84), opioids (p=0.43), anti inflammatory medication (p=0.39), musculoskeletal therapeutic agents (p=0.27) or corticosteroids (p=0.54) during the year of being admitted to physical therapy. Although receiving adhere nt care did not increase the odds of prescription medication use, having comorbid LBP increased the odds of being prescribed non narcotic analgesics (p=0.05), anti inflammatory medication (p=0.01), musculoskeletal therapeutic agents (p=0.05) and corticoste roids (p=0.05). See Table 2 9. Healthcare U tilization the Y ear after Admission to Physical T herapy When examining the utilization variables of number of visits to healthcare providers, number of prescriptions and number of diagnostics images the year after

PAGE 61

61 admission to PT we found that those receiving adherent care when holding all other variables constant in the models, had a 70% increased rate of prescription utilization (Adjusted Mean Difference=0.9 prescriptions) ( IRR= 1.70, p< 0.001) and a 53 % less visit s to healthcare providers (Adjusted Mean Difference=1.23 visits) ( IRR=0.47, p=0.031). There was not a statistical difference in the rate of imaging the year after PT be tween the groups ( IRR= 1.91, p=0.20 ). See Table 2 10. Costs during the Year of Admission to Physical Therapy During the year that patients were admitted to physical therapy, while holding all other variables co nstant in the models, the adherent care group had 22% less costs for physical therapy compar ed to those receiving non adherent care (M ean Difference 172.55 USD) (e B =0.78, 95%CI 0.69, 0.89, p<0.001), but did not differ in their costs to healthcare providers ( e B =0.79, 95%CI 0.26, 2.24, p=0.68) or costs f or prescription medication (e B =0.74 95%CI 0 .21, 2.51, p=0.62). See Table 2 11. Costs dur ing the Year after admission to Physical Therapy The year after admission to physical therapy, while holding other variabl es in the models constant, adh e rent care groups did not significantly differ in their costs for health care providers (e B =0.24, 95%CI 0.03, 2.02, p=0.19) or costs for prescription medication ( e B =2.61, 95%CI 0 .07, 9.97, p=0.16). See table 2 12. Summary of Findings The current healthcare system is evolving and there is a need for streamlining interventions by healthcare providers to incre ase favorable clinical outcomes and decrease unnecessary healthcare utilization and expenditures. Clinical practice guidelines have been published for numerous conditions and much research has been devoted to measuring the effects of guideline disseminatio n 65 Recently published

PAGE 62

62 literature supports that adherence to the guideline recommend ations for back pain improved clinical outcomes and decreased physical therapy utilization and costs 12 Where the current literature is currently lacking is the dissemination of clinical guidelines for neck pain an d the effect of providing care adherent with the guidelines on clinical outcomes, utilization and costs for physical therapy and healthcare providers. This study specifically addresses the need for literature related to clinical outcomes, healthcare utiliz ation and costs of care for neck pain. The adherent rates in this study were lower than prior studies on adherence in the lo w back literature. In our study, the percentage of patients receiving adherent care (75% or greater active care) was approximately 1 1% for the entire sample and prior studies have reported adherence rates of low back guidelines of approximately 40% 89 90 We had hypothesized that those receiving active c are would have improved clinical outcomes from physical therapy. Surprisingly, the results of our findings were that guideline adherent care which advocates primarily active interventions did not improve the odds of successful physical therapy or successfu l pain management as defined in our study. But when examining a successful disability outcome the effect of group almost reached statistically significance ( p=0.058), where those patients who received adherent care actually were less likely to achieve a su ccessful outcome which is contradictory to our hypothesis and past findings in low back pain. In addition, the presence of comorbid low back pain, the odds of achieving a successful outcome was cut in half. Possibl e reasons for this null finding are that groups were compared based on the treatments they received rather their clinical presentation on clinical outcomes. Therefore the findings may not be a true reflection of the potential for difference in

PAGE 63

63 outcomes between groups. Further analyses would benef it from examining adherent care in a more homogenous sample of patients. A lthough receiving guideline adherent care did not have a significant effect on clinical outcomes. Receiving a dherent care did have a positive effect on immediate and downstream healt hcare costs and utilization, where utilization of physi cal therapy was decreased by approximately 4 visits the number of healthcare visits by approximately 7 visits, patients had 1 fewer prescription and approximately half the number of diagnostic images compared to the non adherent group Additionally, patients in the adherent care group had approximately 20% less costs for an episode of physical therapy compared to those receiving non adherent care. It is beyond the scope of this study to infer the exact reason for this decrease in utilization because we did not control who received guideline adherent care. Additionally, although we controlled for some patient factors during analyses, we were unable to control for non measured factors such as psychosocial factors. However, the implication of these findings is that when receiving guideline adherent care, patients appear to utilize less healthcare resources. Lastly, we hypothesized that the benefits of receiving guideline adherent care would continue to have a positive effect on healthcare utilization and costs the year after physical therapy. Those in the adherent care group continued to have approximately half the number of visits to healthcare providers, but actually had increases in the number of prescrip tion medications. The savings seen during the year of admission to physical therapy in the adherent care group were not translated for the following year for costs of visits to healthcare providers or prescription medications. The results of this

PAGE 64

64 would sug gest that receiving a dherent care decreases utilization and costs the year of physical therapy but the benefits are seen f or the following year. There are several limitations in the current study. First and foremost, the determination of active care was b ased on CPT codes; billed codes may encompass a large variety of interventions and there is no way to specify exactly what treatments were provided. Second, clinical guidelines are provided as a recommendation only; and do not take into account the variabi lity in patient presentation that may affect a sample size in these analyses was relatively small and there were disproportionate group sizes, therefore the generaliza bility of these results may be limited. Although the current study has noted limitations, it demonstrates the value of providing guideline adherent care on healthcare utilization and costs in patients with neck pain. To date, there are not any published st udies that have examined a clinical guideline recommended active care approach on clinical outcomes, healthcare utilization and costs in patients with neck pain. Although the findings from this study did not find statistically significant differences in cl inical outcomes between a dherent care groups as past literature in the low back has, we still feel the findings of this study are valuable. Future research is needed to elucidate the reasons behind this null finding. In conclusion, this study provides new, clinically relevant information about the benefits of a guideline adherent care approach to physical therapy intervention for neck pain. Receiving guideline adherent physical therapy intervention decreases the number of visits and costs of physical therap y and decreases the healthcare utilization and costs the year of physical therapy, but the effects of a guideline adherent approach did not

PAGE 65

65 lead to improved clinical outcomes compared to non adherent interventions or did this approach decrease healthcare u tilization and costs the year after physical therapy.

PAGE 66

66 Figure 2 1 Deri v ation of the Sample for Specific Aim 2

PAGE 67

67 CHAPTER 6 DOES THE TIMING OF GUIDELINE ADHERENT CARE FOR PATIENTS WITH NECK PAIN IMPROVE CLINICAL OUTCOMES, HEALTHCARE UTILIZATION AND COSTS ? Background Neck pain is a common musculoskeletal condition 42 The course of neck pain is variable 33 and while often neck pain resolves without intervention, neck pain also has a high rate of recurren ce 91 and development of chronic symptoms 37, 92 Non specific chronic neck pain represents the most frequently reported type of neck pain and can contribute to significant healthcare costs and burdens to the individual 93, 94 Treatments for neck pain can range from surgical intervention to more conservative interventions such as physical therapy. Approximately 25% of people with neck pain seek care in outpatient physical therapy 9 11, 95 Clinical guidelines for neck pain based on best available evidence and expert opinion advocate for providing specific interventions which are matched to clinical presentation 4 These guidelines support the use of active interventions such as therapeutic exercise often in conjunction with thrust and non thrust manipulation for treatment of neck pain and advocate an overall active approach to physical therapy intervention that minimizes the use of passive non evid ence based interventions 64 Prior research in the low back found that physical therapist intervention that was guideline adherent produced improved clinical outcomes and decreased utilization and costs for physical therapy 12 but this has not been investigated in neck pain. Additionally, l iterature supports that a relationship has also been indicated between the timing of physical therapy initiation for acute LBP and subsequ ent healthcare utilization Gellhorn et al (2012) found that patients who received physical therapy early after an episode of acute LBP were at lower risk for subsequent LBP

PAGE 68

68 related healthcare utilization over the following year compared to those who rece ived physical therapy at later times 13 It is reasonable to extend the rationale to neck pain; that is, early intervention and evidence based and guideline adherent care can improve clinical outcomes and decrease healthcare utilization and costs. Purpose The p urpose of this study was to examine how patients categorized into clinical care groups based on acuity of symptoms a proxy measure for timing and receiving guideline adherent care differ in their clinical outcomes, healthcare utilization and costs during the year of receiving physical therapy intervention. Specifically, we hypothesized that patients categorized with acute neck pain who receive a guideline adherent, active approach to physical therapy intervention would have improved clinical outcomes compa red to those patients who are categorized as chronic neck pain with and without guideline adherent care and acute non adherent guideline care. Additionally, we hypothesized that patients in the acute adherent care group would have decreased healthcare util ization and costs for physical therapy and healthcare providers compared to all other groups. Methods Database Data for this study was collected from 11 outpatient physical therapy clinics of Intermountain Healthcare Inc (IHC). Intermountain Healthcare is a private, non profit integrated healthcare system located in Salt Lake City, UT and surrounding regions. Data for patients receiving physical therapy are entered into an electronic database that tracks clinical outcomes and process data related to the del ivery of physical therapy

PAGE 69

69 subgrouping related to impairments and symptoms reported at initial evaluation. Utilization variables included number of visits, length of stay (LOS ) and total cost of physical therapy. Clinical outcomes that were collected in the sample in this study were the Neck Disability Index (NDI) and Numerical Pain Rating Scale (NPRS). Clinical outcomes have been tracked since 2002 and previous studies utilizi ng this database report that data for 94.5% of patients receiving physical therapy were collected 12 Sample For this study two related data sets, both drawn from afore m entioned clinical databases, was used. In the initial sample of patients, we wished to examine the differences between groups based on the acuity of their symptoms and whether they received guideline adherent interventions on their clinical outcomes from physical therapy. In a subset of these patient s where the data was available, we compared healthcare utilization and costs for the episode of physical therapy to other healthcare providers. Analyses #1:Clinical o utcomes: The sample in these analyses included patients who underwent an episode of physi cal therapy for treatment of neck pain. An episode of care was defined as the date from initial evaluation to the date of the last visit. Data for this study w as retrospectively retrieved from the electronic clinical outcomes database from the dates Janu ary 1, 2008 to December 31 st 2012. Patients were chosen for this sample if they were categorized in the clinical sub groups of acute, whiplash and chronic neck pain described in specific aim 1. The patients included in the sample must have met the addition al inclusion criteria of: non surgical patient, NDI score of 10 or greater and NPRS of 2 or greater at initial evaluation, 2 or more visits and LOS between 2 180 days.

PAGE 70

70 Analyses #2: Healthcare utilization and c osts The sample in these analyses included patie nts who underwent an episode of physical therapy care for treatment of neck pain from the dates January 1, 2008 to December 31 st 2012 and had a clinical sub grouping of acute, whiplash or chronic neck pain. Patients were included in this analysis if both billing and clinica l data w as available. Additionally patients in the analysis must have met the criteria of NDI score of 10 or greater, two or more visits, LOS between 2 180 days. Records from the clinical outcomes database were linked to a billing databa se using an enterprise master patient index number to capture healthcare utilization and c ost s in the sample. Healthcare utilization and cost s for services provided by healthcare providers including physical therapists related to an ICD 9 code pertaining to the neck region were extracted for patients in the sample. The total number of visits to physical therapy during an episode of care, number of visits to healthcare providers, number of prescriptions, number of diagnos tic images and costs for vi sits to h ealthcare providers for the year of physical therapy were examined for all patients in the sample. See Figure 3.1 for derivation of the samples for analyses 1 and 2. Clinical Care Groups Patients included in the se analyses received interventions provided by a physical therapist for a primary compliant of neck pain. Patients were categorized into clinical sub groups by the licensed physical therapist at the initial evaluation. These categories included patients with a primary complaint of: acute neck pain, chronic neck pain, neck pain with arm pain, neck pain with headache and whiplash. Patients who were sub grouped into the categories acute neck pain, whiplash and chronic neck pain were chose n for the s e current analyse s. Based on the findings from prior ana lyses in this

PAGE 71

71 dissertation, the patients in the acute neck pain and whiplash groups were combined to form one clinical group; acute neck pain to indicate early intervention Both groups of patients reported a duration of s ymptoms less than 4 weeks and thes e groups exhibited similar baseline clinical characteristics and outcomes. Patients categorized as having chronic neck pain remained in the chronic group to indicate late intervention. Next, patients within these groups were further categorized into receiv ing guideline adherent care or non guideline adherent care. To determine adherent care, CPT codes billed during an episode of care for physical therapy were examined and categorized as active, passive or allowed. The number of active and passive codes bill ed at each visits were recorded for two phases of treatment: Phase 1 (0 14days), Phase II (15 days and greater). The percentage of active codes was calculated as [(number of active codes/ (number of active codes + number of passive codes)) x100=%Adherent C are]. Adherent care was defined as occurring when the percentage of active codes within the phase was at least 75%, and each visit must have included an active code. Adherent care was calculated as a continuous variable for the entire sample by phase. Two adherent treatment. This algorithm has been previously published 12 and is deta iled in specific aim 2 Clinical care groups based on acuity of symptoms ( proxy measure for timing early and late intervention ) and receiving adher ent or non adherent care, four c linical care groups were formed for comparison in these analyses: Chronic No n adherent, Chronic Adherent, Acute Non adherent and Acute Adherent.

PAGE 72

72 Clinical Outcome Measures Patients completed the Neck Disability Index (NDI) and the Numerical Pain Rating Scale (NPRS) at the initial visit and final visit. The Neck Disability Index is a condition specific outcome measure comprised of 10 items; each item is scored from 0 to 5. The total score is expressed as a percentage and is reflective of a level of disability related to neck pain where high percentages are related to higher disabili ty. This questionnaire asks the patient to rate how their neck pain affects their tolerance for activities of daily living. The NDI is a commonly used outcome measure for people with neck pain is found to be reliable and valid in the neck pain population 67 71 Both the NDI and NPRS exhibit fair t o moderate test retest reliability in patients with mechanical neck pain and both instruments also show adequate responsiveness in this patient population 72, 73 Change in pain and change in disability scores were calculated for each patient in the sample. This was calculated by subtracting the discharge score from the score at the initial visit for NPRS and NDI. Change score was then divided by the score at the initial visit then multiplied by 100 to create a perc entage improvement. A dichotomous variable to indicate a successful outcome from physical therapy was created. A 50% therapy as successful and less than 50% as unsuccessfu l. A dichotomous variable to indicate successful pain management was created with a 50% improvement in NPRS or greater was determined to be successful pain management and less than 50% was

PAGE 73

73 determined to be unsuccessful pain management. This threshold has b een used in previous studies 12, 87, 88 Examination of Clinical Predictors of Value and Efficiency We examined potential predictors of value and efficiency using linear regressi on models. The model predicting valu e ((change in NDI/charges for PT ) x100 ) included the variables baseline NDI score, age, gender, LOS and number of visits as covariates. Clinical care groups were included as factors, with the chronic non adherent care group as the reference group. The mode l for predicting efficiency (change in NDI/number of visits) included baseline NDI score, age and LOS as covariates. Clinical care groups and gender were included as factors, with the chronic non adherent care group as the reference group for clinical care groups and male for the reference gender Health Related Quality of Life (HRQL) We wished to measure the differences in HRQL (SF 6D) between clinical care groups. In order to do this, we needed to convert the NDI into a SF 6D score. The SF 6D is a widely used classification for describing health status. The SF 6D provides a score for general health from 0 1, where 0 equals death and 1 equals perfect health 74 Using a standard gamble technique, admission, discharge and change scores for NDI were converted to SF 6 scores using the equation (SF 6D= 0.0115(NDI Score) + 0.8383). This equation was published by Richardson and Berven (2012) who found that correlations between the NDI and SF 6D are strong; and high correlations between SF 6D and NDI score permit models to be used to calculate utilities, change in utilities and quality adjusted life years 75

PAGE 74

74 Physical Therapy Utilization and Costs For each episode of care included in the sample, the length of stay (LOS) was defined as the number of days between the initial evaluation a nd the last visit. The number of visits was defined as the number of visits that patients attended during an episode of care. Total costs for physical therapy were defined as the total charge amount for the episode of care. Healthcare Utilization and Cost s In addition to physical therapy, data was retrieved from patients in the sample who received healthcare with an ICD 9 diagnosis of neck pain or neck pain related diagnosis. Utilization variables included: number and cost of visits to healthcare providers number of prescriptions and number of diagnostic images. T hese variables were examined during the year of admission to physical therapy. All cost s were adjusted to 2008 costs by adjusting c ost s from subsequent years by a rate of 0.968. The rate used was from the Consumer Price Index published by the US bureau of labor and statistics for medical care services 76 Data Analysis Analyse s #1: Clinical o utcomes P atient characteristics and clinical outcomes were compared between the clinical care groups. In unadjusted analyses patients with chronic neck pain receiving non adherent care, patients with chronic neck pain receiving adherent care and patients with acut e neck pain receiving non adherent care, patients with acute neck pain receiving adherent care were compared using one way ANOVAs for continuous variables and chi squared tests for categorical variables.

PAGE 75

75 Separate repeated measures ANOVA models (Within Ti me; Between C linical care group) were performed to examine differences in change in clinical outcome scores over time. Dependent variables in separate models we re NDI score, NPRS and HRQL Covariates included in the models were age, gender, LOS and numbe r of visits. We were specifically interested in the effect of clin ical care group on NDI score, NPRS and HRQL (SF 6D) score The proportion of patients with episodes of care achieving a successful outcome from physical therapy (50% change in NDI score) and successful pain management (50% change in NPRS) were compared using separate binary logistic regression models. We were specifically interested in the effect of clinical care group on odds of achieving success. Covariates included in the analyses in inclu ded age, gender, LOS and visits. We examined potential predictors of value and efficiency using linear regression models. The model for predicting value (change in NDI/ ( charges for PT x100) ) included baseline NDI score, age, gender, LOS and number of vis its as covariates. Clinical care groups were included as factors with the chronic non adherent care group as the reference group. The model for predicting efficiency (change in NDI/number of visits) covariates included baseline NDI score, age, gender and L OS. Analysis #2: Comparison of healthcare utilization and c ost s during the y ear of admission to physical t herapy In unadjusted analyses, healthcare utilization and costs were compared among clinical care groups using Kruskal Wallis nonparametric tests due to violations of assumptions of normality for these variables.

PAGE 76

76 We examined predictors of healthcare utilization using separate Poisson regression models. Incident rate ratios (IRR) with confidence intervals were reported. We were interested in examining predictors of healthcare utilization the year patients were admitted to PT. In separate regression models, we examined predictors of: (1) number of visits to physical therapy, (2) number of visits to healthcare providers the year of PT, (3) number of pres criptions the year of PT (4) number of diagnostic imaging the year of PT (5) LOS for physical therapy Predictors in each the models varied slightly depending on the nature of the dependent variable. For the model examining number of visits to PT, predict ors in the model included clini cal care group, age, gender, admission NPRS and NDI score. The model predicting number of visits to healthcare providers the year of PT, predictors in the model included: clinical care group, age, gender, admission NDI and NP RS and number of PT visits. To examine the number of prescriptions the year of receiving physical therapy predictors incl uded: clinical care group, age, gender and admission NPRS The model for diagnostic imaging included the predictors of: patient group, age, gender, admission NDI and NPRS. For the model examining PT LOS, predictors included clinical care groups, age, gender, admission NDI score and admission NPRS. Lastly to examine costs for an episode of physical therapy care and costs for healthcare vis its the year of admission to PT, separate gamma regression with log link function was performed Dependent variables i n these models were total cost s for p hysical therapy and total cost s for healthcare visits the year of admission to physical therapy. Expo nentiated B with 95% confidence intervals was reported for the independent variables.

PAGE 77

77 Pr edictors in the model for cost s for an episode of physi cal therapy included: clinical care group, age, gender, number of visits admission NPRS, admission NDI score and comorbid low back pain. Pr edictors in the model for cost s for healthcare visits included: clinical care group, age, gender, number of visits, admission NPRS, admission NDI score and comorbid LBP. IBM SPSS 21.0 and Intercooled STATA 12.1 were used to perf orm analyses. Significance level was set at 0.05 for all analyses. Results Analyses #1: Clinical Outcomes The first sample included 3485 episodes of care. After the inclusion criteria were met and missing observations removed, 1750 episodes of care remaine d. Clinical groups were compared for differences in baseline characteristics. In unadjusted analyses, groups differed in their age (p<0.001), admission NDI score (p<0.001) and admission NPRS (p<0.001) but not gender (p=0.24). See Table 3 1 When comparing unadjusted clinical outcomes, groups differed in change in NDI score ( p <0.001), change in pain score (p<0.001), discharge pain score (p=0.02) and proportion achieving a successful outcome( p <0.001) and proportion achieving successful pain management (p<0.0 01), but did not differ in discharge NDI score (p<0.08).Groups differed in the value (change in NDI/ ( charges for PT /100 ) (p<0.001) and efficiency (charges for PT/ PT visits) (p<0.001) of the episode of PT. Additionally, groups differed in admission HRQL (S F 6D) score (p<0.001), change in HRQL (SF 6D) score (p<0.001), but not discharge HRQL (SF 6D) score(p=0.08). See Table 3 2.

PAGE 78

78 Disability D ifferences among clinical care groups for change in NDI over an episode of care were examined. There was statistically significant interaction of time and clinical care group ( F 3, 1715 =26.739, p<0.0001) where the acute adherent experienced the greatest changes in pain Patients in the acute adherent care group were approximately 2 times more likely to achieve a successful outcome from physical therapy intervention compared to those patients in the chronic non adherent care group (p=0.01). Additionally, those patients in the acute non adherent care group were1.6 times more likely to achieve a successful outcome (p<0.001) co mpared to patients in the chronic non adherent group but those sub grouped into the chronic neck adherent care group were no more likely to a successful outcome from physical therapy intervention (p=0.10) than patients in the c hronic non adherent group Fu rthermore when comparing the odds of achieving a successful outco me, the acute adherent group had 0.8 times greater odds of achieving success compared to the acute non adherent group; and when compared to the chronic adherent gro up, the acute adherent grou p had 3.07 times greater odds of achieving a successful outcome See Table 3 3 Pain The difference among clinical care groups for the change in NPRS over an episode of care was examined. There was a significant interaction of time and clinical sub group ( F 3,1711 =17.88, p<0.001). Patients in the acute adherent care group were 2.4 times more likely to achieve successful pain management with physical therapy intervention compared to those

PAGE 79

79 patients in the chronic non adherent care group (p=0.002). Additiona lly, those patients in the acute non adherent care group were1.6 times more likely to achieve successful pain management (p<0.001), but those classified in the chronic adherent care group were no more likely to achieve pain management with physical therapy intervention than those patients in the chronic non adherent care group (p=0.27). Furthermore when examining the odds of achieving successful pain management in the acute adherent care group compared to other groups the acute adherent group is 1.5 times more likely to achieve successful pain management compared to the acute non adherent group and when compared the chronic adherent group, the acute adherent group is 1.84 times more likely to achieve successful pain management. See Table 3 4. Health relate d quality of l ife Differences among clinic al care groups for change in HRQL (SF 6D) score over an episode of care were examined. There was statistically significant interaction of time a nd clinical care group ( F 3, 1715 =26.06 p<0.0001). When looking at spe cific group comparisons, only the acute non adherent care group significantly differed from the chronic non adherent group (. F 3, 1715 =2 .68, p=0.045 ) There were no differences statistically significant differences found between any other groups Efficiency and Value In these analyses we were specifically interested in determining if clinical care group was a significant predictor o f value or efficiency. Tables 3 5 and 3 6 present models predicting efficiency and value of physical therapy intervention. The co ntribution of the specific predictors is expressed by the regression coefficient. Positive coefficients indicate the change in efficiency was greater in the presence of the predictor and a

PAGE 80

80 negative coefficient attenuates the effect of the predictor on the dependent variable. Furthermore the magnitude of the coefficient indicates the strength of the association of the predictor with dependent variable. Additionally, for the clinical care groups, the coefficient can be interpreted as the increase in the depen dent variable when a patient in that clinical care group compared to the chronic non adherent care group when all other predictors are held constant. In the model predicting efficiency of physical therapy, when controlling for age, gender and LOS patients in both the acute with adherent care and acute with non adherent care showed increased efficiency compared to the chronic with non adherent care group; where membership in the acute non adherent care group (p<0.001) incre ased efficiency by a rate of 0.92 and membership in the acut e adherent care group (p<0.001) increased efficiency by a rate of 2 .49 Membership in the chronic adherent care group compared to the chronic non adherent care group did not improve efficienc y. In other words, the acute adherent care group experienced 5.3 point change in NDI and the acute non adherent care group only experienced a 3 point change in NDI compared to the chronic non adherent group who experienced only a 1.79 change in NDI per visit. This model accounted for 18% of th e variance in efficiency scores. When examining value, after controlling for age, gender and LOS, again having membership both the acute adherent care (p<0.001) and the acute non adherent care (p<0.001) groups increased value by a rate of 1 .26 and 0.64 res pectively compared to the chronic non adherent group. Membership in the chronic adherent care group did not increase value (p=0.629).

PAGE 81

81 In other words for every 100 dollars spent, the acute adherent care group had a 3.2 point improvement in NDI and the acut e non adherent group had 2 point change in NDI compared to only a 1.2 point change in the in the chronic non adherent group. This model accounted for 16% of the variance in efficiency scores. Overall the ac ut e adherent care group had the greatest impr ovements in NDI per 100 dollars spent. Analyses #2: Healthcare Utilization and Costs Table 3 7 displays the un adjusted values for clinical care groups. Patients in clinical care groups significantly differed in their number of visits to physical therapy ( p<0.001), PT LOS (p<0.001) and c ost s for physical therapy (p<0.001), but did not differ in their number of visits to healthcare providers (p=0.31), number of diagnostic images (p=0.34) or prescriptions (p=0.87). Nor did they differ in their cost s to health care providers (p=0.20). Physical therapy utilization and costs Visits P atients in the chronic adherent group differed in the number of visits for physical therapy compared to the chronic non adherent group, where they averaged approximately 3 fewer visits (p<0.001) and the patients in the acute adherent group (p=0.01) had approximately 2.5 fewer visits compared to the chronic non adherent group There was not a statistically signifi cant difference between the acute non adherent group (p=0.15 ) when compared to the chronic non adher ent group on number of visits. W hen comparing the acute groups only non adherent group attended 3 more visits than the acute adherent group.

PAGE 82

82 LOS Both the chronic adherent (p<0.001) and acute adherent (p<0.001) groups had signific antly shorter LOS compared to the chronic non adherent group, where the chronic adherent group had LOS 30 days shorter and the acute adherent group had a 28 day shorter LOS. The LOS of the acute non adherent group was approximately 10 more day s than the c h ronic non adherent group (p<0.001 ). Additionally t he LOS of the acute non adherent group was 38 days longer than the acute adherent care group. C osts When examining differences in c ost s for physical therapy between groups holding all other variables in th e model constant, only those patients in the chronic adherent ca re group differed in their cost s for phys ical therapy, where their cost s were approximately 20% less than those in the chronic non adherent group (p=0.02). Those in the acute non adherent (p=0 .71) and acute adherent (p=0.19) did not significantly differ from those in the chronic non adherent group in the c osts for physical therapy. See Table 3 8. Healthcare Providers Utilization and Costs The number of visits to healthcare providers significant ly differed in all groups from the patients in the chronic non adherent gr oup (14 visits) The chronic adherent (7.2 visits) acute adherent (8.5 visits) and acute non adherent (7.5 visits) all had approximately half the visits (p<0.001 ) compared to the c hroni c non adherent care group. See Table 3 9. Furthermore, when examining differences in cost s for visits to healthcare providers the year of admission to physical therapy, those patients in the acute non adherent group (p<0.001) and the patien ts in the a cute adherent group (p=0.02) differed

PAGE 83

83 in their cost s where the acute non adherent group had 68% of the cost for visits and the acute adherent had 81% of the cost to healthcare providers compared to the chronic non adherent care group. Patients in the chro nic adherent care group did not significantly differ from chronic non adherent group (p=0.27) in costs to healthcare providers during the year of admission to physical therapy Overall the acute adherent group had the lowest cost for visits to healthcare providers. Diagnostic I maging When examining the differences in the number of diagnostic images per year both the chronic adherent and acute non adherent groups significantly differed from the chronic non adherent care group where the patients in the chr onic adherent care group (p=0.005) had 73% fewer diagnostic images and the patients in the acute non adherent care group (p=0.004) had 46% fewer diagnostic images compared to the chronic non adherent care group. The acute adherent care group did not differ from the chronic non adherent care group in number of diagnostic images during the year of physical therapy (p=0.54). See Table 3 10 Prescription M edication After adjusting for variables in the model, only patients in the acute adherent care group had a lower rate of prescription medication usage compared to the chronic non adherent care group (p=0.04), where the patients in the acute adherent group had approximately 40% fewer prescriptions the year of physical therapy. The patients in the chronic adhere nt care (p=0.38) and the acute non adherent group (p=0.21) did not differ from the chronic non adherent care group in numbe r of prescriptions. See Table 3 11

PAGE 84

84 Summary of Findings Recent research has advocated ad hering to clinical guidelines to improve pat ient outcomes and decrease costs related to physical therapy 12 but there is lacking evidence to support that providing adherent care in neck pain can improve clinical outcomes. Furthermore, it has been suggested a nd supported by current research that receiving early intervention for spine pain can improve clinical outcome and decrease future cost. The purpose of this study was to examine how patients with neck pain differ in their clinical outcomes, healthcare util ization and costs when sub grouped by timing of physical therapy and receiving adherent care. We hypothesized that patients with acute neck pain who receive adherent care would have improved clinical outcomes compared to other patients in acute non adheren t care, chronic adherent care and chronic non adherent care. Our primary hypothesis was supported; those patients with acute neck pain who received adherent care had the greatest increase odds of achieving a successful outcome in disability and successful pain management, achieved greater change in NDI over fewer visits and per dollar spent had greater improvements in NDI Our secondary hypothesis examined the differences in healthcare utilization and costs between these clinical care groups. Findings were less consistent with respect to both acuity and adherent care; where no one group performed better across the board, rather adherent care dictated decreases in utilization. Both the chronic and acute patients receiving adhe rent care had fewer visits to PT For PT LOS, both the acute and chronic adherent care groups had significantly shorter LOS The only group that had significantly lower costs for physical therapy was the chronic adherent group; where their c ost s were approximately 20% l ower than the pati ents in the chronic non adherent

PAGE 85

85 group. Furthermore, the chronic adherent group actually had the fewest number of visits (2.6 visits ) and shortest LOS (8.0 days ) Combined, these results suggest that when looking at physical therapy outcomes specifically, receiving adherent care early in an episode of neck pain improves clinical outcomes, but when looking at the big picture, early intervention paired with adherent care does not necessarily decrease physical therapy utilization and costs but rather receivin g adherent c are appears to be more important When considering other healthcare utilization and costs the findings did not follow a specific pattern For diagnostic imaging the chronic adherent and acute non adherent groups had less imaging compared to the chronic non adherent group. Although it is beyond the scope of these data and analyses, there may be psychological comorbidities of the patients in the non adherent groups that may contribute the differences in rates of imaging between these gr oups. When examining prescription medication, only the acute adher ent care group had statistically fewer prescription medications than the chronic non adherent group suggesting that adherent physical therapy might decrease the rate of prescription use. Lastly, when looking at total healthcare utilization and costs, although all groups had fewer visits to healthcare providers than did the chr onic non adherent group. When translating these findings to cost s both the acute non adherent and acute adherent groups had sta tisticall y significantly lower costs for visits to healthcare providers. This supports recommendations that providing early intervention during an episode of acute pain can decrease future healthcare and utilization 51 and that physical therapy can be an effective secondary prevention strategy for neck pai n 52

PAGE 86

86 There are numerous strengths to this study. This is the first study that has examined clinical presentation combined with treatment approach, utilization and expenditures in patients with neck pain. This study adds to the current literature to support that those with acute symptoms who receive an active approach consistent with guideline adherent care have improved clinical outcomes in multiple domains. Moreover, when looking at overall costs and uti lization, receiving adherent care decreases utilization and costs not only with physical therapy but with visits and costs to health care providers. But there are noted limitations to this study as well. These findings should be interpreted with caution; adherent care in this study was defined with CPT codes, therefore there is a wide variation in the procedures billed under these codes. Additionally, we did not control the specific interventions provided nor were intervening therapists trained in providin g care specific to the clinical guidelines for neck pain Secondly, for the cost and utilization analyses, there were small numbers in the adherent care groups which increased the confidence intervals about the point estimates and may narrow the generaliza bility of these findings. Receiving guideline adherent care in the presence of acute neck pain improves clinical outcomes of pain, disability, and improves value and efficiency for physical therapy. Receiving adherent care in the presence of acute neck pa in does not necessarily reduce costs or utilizations, r ather receiving adherent care may be the driving force behind decreased costs and utilizations in this sample.

PAGE 87

87 Figure 3 1 Derivation of Samples for Analyses #1 and #2 for Specific Aim 3

PAGE 88

88 CHAPTER 7 DISCUSSION AND CONCLUSIONS Discussion A large portion of people experiencing painful musculoskeletal conditions utilize outpatient physical therapy services 7 and approximately 35% of persons with neck pain reported being seen by a physical therapist 8 Physical therapists provide interventions t o patients with neck pain 9 11 but variation exists in the inventions provided to patients. In effort to decrease this variation and potentially improve clinical outcomes, clinical practice guidelines have been pu blished for neck pain 4 The guidelines advocate the use of primarily active interventions. Prior research in the low back found that physical therapist intervention that was guideline adherent produced improved cli nical outcomes and decreased utilization and costs for physical therapy 12 but this has not been investigated in neck pain. In addition to receiving guideline adherent care, the timing of treatment may be just as i mportant. Gellhorn et al (2012) found that patients who received physical therapy early after an episode of acute LBP were at lower risk for subsequent LBP related healthcare utilization over the following year compared to those who received physical ther apy at later times 13 It is reasonable to extend this rationale to neck pain; that is, early intervention and evidence based and guideline adherent care improve clinical outcomes and decrease healthcare utilization and costs. Therefore the primary goals of thi s dissertat ion were to 1) D escribe the clinical characteristics, physical therapy utilization and costs in persons seeking care for neck pain 2) D etermine the extent to which receiving guideline adherent physical therapy intervention improves clinical outc omes, decreases healt hcare utilization and costs 3) D etermine the extent to which timing of physical therapy intervention in conjunction with

PAGE 89

89 receiving guideline adherent care improves clinical outcomes and decreases healthcare utilization and costs. This was addressed through three interrelated specific aims. Statement of Principle F indings Clinical sub groups of acute neck pain, chronic neck pain, neck pain with arm pain, neck pain with headache and whiplash differ in their clinical characteristics, healt hcare utilization and costs. Specifically, the acute neck pain and whiplash groups are more likely to have successful disability and pain outcomes with physical therapy than the chronic group. Moreover, the patients in the acute neck pain group have the hi ghest efficiency and value for physical therapy; that is, they require fewer visits for greater changes in disability and have greater changes in disability for dollar spent than patients with chronic pain who have the lowest efficiency and value. T hese fi ndings are not unexpected ; previous work has supported that in general neck pain has a favorable course of care and patients with chronic low back pain incur the majority of the healthcare costs associated with managing that condition 4, 96, 97 Receiving guideline adherent care had an immediate positive downstream effect on healthcare utilization costs where the patients receiving adherent care had approximately half the number visits to physical therapy and healthc are providers, less diagnostic imaging and prescription medication during the year of physical therapy and l ower c ost s for physical therapy. Unfortunately these decreases in healthcare utilization and costs did not continue the year after receiving physica l therapy The positive effect of receiving adherent care may be limited to the year in which physical therapy is provided due to the influence of other health conditions that may arise after one year. Furthermore when looking at the combination of timing and adherent care, w hen patients receive early intervention with guideline adherent care, they have increased

PAGE 90

90 odds of achieving a successful outcome from physical therapy, successful pain management and have the greatest value and efficiency with physical therapy, reduces costs to healthcare providers and decreases prescription medication use, but does not necessarily reduce other utilizations Strengths This study has numerous strengths. First, this study is the first study examining the differences in cl inical outcomes, healthcare utilization and costs when receiving guideline adherent care. Second, we examined the clinically important c oncepts of efficiency and value in physical therapy; which have been yet to be documented in patients with neck pain but are critical Third, this study was able to examine healthcare utilization and costs in patients with neck pain beginning with the physical therapist to how these patients are using the healthcare system. Fourth, the findings of this study addressed how t he combination of early intervention paired with adherent physical therapy affect clinical outcomes, healthcare utilization and costs. Lastly, although these findings are not yet validated they can easily be translated in the clinical setting; treating pat ients early with guideline adherent care has the potential to improve clinical outcome s and adherent care independent of timing can decrease healthcare utilization and costs. Limitations Measure s of value and e fficiency In this project we wished to create a variable to define the value of physical therapy. Value has been previously defined similarly by Porter et al. (2010) as the health outcome achieved per dollar spent 98 We defined value as the change in NDI divided by the total charges for physical therapy divided by 100. We chose to define

PAGE 91

91 value this way to be able to interpret the change in disability per 100 dollars spent for phys ical therapy. Value is an important concept that has yet to be studied in patients utilizing physical therapy for neck pain. The limitations of this measurement are that value can only be directly compared among clinical subgroups rather than the overall v alue of physical therapy care to the healthcare system. Furthermore, we wished to create variable to define efficiency of physical therapy. In this sample we defined efficiency as the change in NDI divided by number of visits to physical therapy. In this p roject efficiency is another metric to determine the change in outcome per utilization. This measure is useful to be able to examine changes in NDI per visit (utilization) rather than over costs. The benefit of this approach in addition to reporting value we are able to minimize the variability associated with charges per physical therapy visit and examine the change in NDI per visit independent of billing practices Sub grouping approaches In specific aim 1, we were interested in examining the differences between clinical sub groups on clinical outcomes. Past research on neck pain has proposed m ultiple sub grouping approaches using treatment based classification systems, but these sub grouping approaches have not been validated as an effective way to impro ve clinical outcomes. The sample in this study included patients categorized into sub groups used at Intermountain Healthcare that have not been previously described in the literature. These subgroups were used in these analyses to compare clinical outcome s. The purpose of analyzing these sub groups was to determine if certain subgroups are predictive of improved clinical outcomes; independent of controlling for the type of intervention received. Furthermore, we used these sub groups instead of previously

PAGE 92

92 published treatment based classification system groups because our data was retrieved retrospectively; we were unable based on the available data to group patients into previously published subgroups. Additionally, examining the sample of patients with nec k pain using these sub groups added a temporal component to examining sub grouping approaches in neck pain. In specific aim 2, we were interested in comparing groups of patients based on the type of physical therapy intervention they received using billing codes for physical therapy In this study, we defined receiving guideline adherent care when the patients were billed for codes that are representative of receiving interventions consistent with guideline recommendations 75% of the time during an episode of care We chose to use this cut point for two reasons. First, when looking at codes in each category, codes in the passive category primarily contain modalities; codes in the active category contain codes billed for exercise and manual therapy. The resu lts of meta analyses reported in the clinical practice guidelines recommend manipulation (thrust) and mobilization(non thrust manipulation) as an intervention for mechanical neck disorders and should be combined with exercise for maximum benefit 64 Second, using this cut point of 75% has been previously published in the low back literature which served as the basis for these analyses 12 Using a 75% cut point is rigorous; bu t the definition of adherent care (75%) would indicate that, assuming patients were billed for four units, three of those units should be consistent with evidence and one unit would be a modality.

PAGE 93

93 In specific aim 3, we were interested in examining the ef fects of both receiving guideline adherent care and the effect of timing of PT services. Past research has supported in the low back that receiving guideline adherent care improves clinical outcomes and physical therapy utilization, furthermore receiving e arly physical therapy intervention can decrease healthcare utilization and costs. A noted limitation of this aim was to date there i s no evidence that providing early intervention with guideline adherent care can improve clinical outcomes or decrease utili zati ons in patients with neck pain. W e wished fill the gap in the literature with this study and felt it was appropriate b ased on previous evidence in the low back clinical knowledge of neck pain and the findings from the first two aims to answer the prop osed hypotheses. Methodological limitations and c onsiderations Numerous methodological considerations were taken into account in the analyses in this project. First, the data was retrospectively collected from a large outcomes database. The quality and val idity of the clinical data is dependent upon those collecting and inputting the data. Furthermore, when examining cost and utilization data there were limitations; data was analyzed at the level of the year due to privacy measures in place at Intermountain Healthcare Therefore we lose some specificity in reporting the temporality of healthcare costs and utilizations associated with physical therapy during the year of admission to PT and the following year When determining adherent care, we used billing c odes associated with physical therapy interventions. There were three categories of codes, active, passive and allowed. These categories represent codes that are commonly billed from certain interventions. There are a few noted limitations to this approach First, using this approach does not allow us to know exactly what interventions were received. Second,

PAGE 94

94 there are small effect sizes for interventions for physical therapy 10, 46 ; therefore the results of our find ings need to be interpreted with caution. Third, the equation used to calculate percent adherent care for each patient was active codes/ (passive + active codes) x100=%adherent care. This calculation does not account to for allowed codes nor do we have ava ilable in our dataset the proportion of allowed codes billed. Although this is a limitation, based on the codes categorized in the active, passive and allowed categories we have confidence that we are measuring adherence appropriately. When creating the su bgroups for comparison for aims one and three, we utilized data that was patient reported to the physical therapist; the physical therapist then categorized patients into clinical subgroups. Furthermore, when looking at the temporal subgroups, the acute ne ck pain group was categorized as having neck pain of duration less than four weeks and chronic neck pain was categorized as having duration of symptoms greater than 4 weeks. The duration of symptoms required for membership in the chronic group in this stud y is lower than traditional definitions of chronic neck pain which are can range in duration of symptoms of 3 to 6 months or longer 99 Because of this limitation the author acknowledges there may be some miscla ssification of patients who may not have the traditional definition of chronic neck pain, but rather they have sub acute neck pain. But because of this limitation, the authors use the terminology early verses late physical therapy in specific aim 3. Based on the methodological limitations, the author acknowledges that there are inherent biases in the analyses used in this study. First, there is sampling bias in this project. Our sample may not representative of all persons with neck pain who receive physica l therapy. Second, the subgroups used in the analyses in specific aim 1 were not

PAGE 95

95 formed a priori; the author was confined to the subgroups and definitions used by the healthcare system where the sample was obtained. Due to this, there may be inflation in t he error in the statistical modeling due to misclassification bias. Furthermore, when performing the statistical analyses for the utilization and cost data, the sample was relatively small and groups for comparison were uneven. Although the authors found s tatistically significant findings, the confidence intervals were large. Therefore these findings should be interpreted with caution. Potential I mplications At some point in their life, 22 70% of the population will have neck pain 2, 3 and approximately 25% of people with neck pain seek care in outpatient physical therapy 9 11, 95 A large portion of people will utilize outpatient physical therapy services, which will undoubted ly account for a significant percentage of health care expenditures 7 T herefore it is paramount that the interventions for neck pain be evidenced based and cost effective. Although this study did not perform a formal cost effective analysis, our findings support that providing guideline adherent care can decrease healthcare u tilization and costs in patients with neck pain Our study did find that when the considering the effect of adherent care, early intervention with adherent care is superior to non adherent care in patients with both acute and chronic neck pain. The results of this study can inform physical therapists and healthcare providers about the benef its of treating neck pain early with guideline adherent care on both clinical outcomes and healthcare utilization and costs. Unanswered Questions and Future Research Dire ctions Although the results of this study are promising, there is much research that needs to be performed to better understand neck pain from a clinical and healthcare

PAGE 96

96 system. First, there is a need to validate a subgrouping approach to physical therapy i ntervention. The most recent classification system was published in t he clinical practice guidelines; but this has yet to be validated in a clinical setting and does not consider the acuity of symptoms in its classification. The findings our study support that indeed the acuity of symptoms are important and e ffect clinical outcomes. Therefore future research should aim to validate a classification system that considers acuity of symptoms in categorizing patients in addition to other clinical information. Fu rthermore, our study did not find statistical differences in clinical outcomes between groups receiving adherent and non adherent care alone Future studies would benefit from investigating this concept further Lastly, future studies should extend the fi ndings of our study by creating recommendations for parameters of care in patients with neck pain to decrease unnecessary utilization and costs and promote evidence based care to improve clinical outcomes. From this project, the author is able to extend th ese findings to shape future research in persons with neck pain. W e found that providing care early can improve clinical outcomes and providing guideline adherent care can decrease healthcare utilization and cost, together early intervention with guideline adherent is superior. In future projects the author would like to examine the concepts of early physical therapy intervention as defined by time from referral and evidenced based care as defined by specific interventions received in a more methodologicall y rigorous design. It is necessary to perform these studies in order to validate these preliminary findings. Furthermore, the author would like to investigate the cost effectiveness of early intervention and receiving evidenced based care in a formal analy sis. These issues are

PAGE 97

97 paramount to support the use of physical therapy as a clinically and cost effective intervention for neck pain. Conclusion This study examined a sample of patients with neck pain and examined the effect of adherent care and early int ervention on clinical outcomes, healthcare utilization and costs. We found that patients who receive early intervention have the most favorable clinical outcomes, the highest efficiency and the best value. Specifically, they require fewer visits to achiev e greater changes in pain and have greater changes in pain per dollar spent. The implications of this finding support that getting patients to physical therapy earlier in the course of neck pain can improve clinical outcomes on multiple levels and these pa tients receive the greatest change in disability with fewer visits and fewer dollars spent. When looking at how providing guideline adherent care affects clinical outcomes, we were surprised to find that, in general, simply providing adherent care did not improve clinical outcomes compared to providing non adherent care in this sample Patients receiving adherent care had approximately half the number visits to physical therapy and healthcare providers, less diagnostic imaging and prescription medication du ring the year of physical therapy and l ower costs for physical therapy These findings support that when patients receive guideline adherent care, their utilization of the healthcare system on multiple levels is decreased. The findings reveal that in this sample, t he patients receiving adherent care used less resources, time and money to achieve similar outcomes as the non adherent group. W hen a broader approach is taken to examine the combination of these findings, when patients received early interventio n with gu ideline adherent care they

PAGE 98

98 experienced, i mproved clinical outcomes the h ighest efficiency the best value, l ess costs for visits to healthcare providers and lowest prescription medication use. W e can now see that in this sample providing adherent care early in an episode of neck pain can improve clinical outcomes more th an just providing adherent care. Taking the summation of these findings into consid eration, providing timely physical therapy with evidenced based intervention has the potential t o optimize both clinical outcomes and decrease the burden of neck pain on the healthcare system.

PAGE 99

99 Table 1 1 Description of Clinical Characteristics and Outcomes from Physical Therapy Variable Clinical Sub groups Acute Chronic Neck Pain w/ AP Heada che Whiplash Total Sample p value Age (Years) 46.23 (15.41) 52.45 (16.77) 51.02 (12.31) 45.18 (17.05) 42.84 (15.36) 49.39 (15.91) <0.00 1 % Female 64.1% (289) 70.5% (760) 61.6% (378) 79.0% (267) 72.6% (159) 68 .8% (1853) <0.00 1 PT LOS (days) 21 (2,180) 26 ( 2,175) 23 (2,175) 24 (2,177) 35 (3,173) 25 (2,180) <0.00 1 Visits 5 (2,54) 5 (2,30) 5 (2,40) 5 (2,30) 6 (2,32) 5 (2,180) <0.00 1 A DM NDI 40.18 (17.54) 34.31 (15.22) 36.55 (16.02) 35.97 (15.34) 40.70 (17.27) 36.53 (16.16) <0.00 1 D C NDI 23.86 (17.34) 25.6 4 (16.30) 26.31 (17.41) 26.10 ( 16.79) 26.97 (18.82) 25.66 (17.02) 0.114 CHG NDI 16.31 (16.15) 8.68 (13.31) 10.18 (13.66) 9.87 (13.39) 13.73 (15.93) 10.86 (14.40) <0.00 1 A DM NPRS 5.96 (2.23) 5.28 (2.14) 5.79 (2.20) 5.40 (2.25) 5.60 (2.04) 5.55 (2.19 ) <0.0 0 1 D C NPRS 3.20 (2.50) 3.57 (2.37) 3.67 (2.50) 3.76 (2.48) 3.32 (2.35) 3.53 ( 2.44) 0.004 CHG NPRS 2.76 (2.51) 1.71 (2.26) 2.12 (2.50) 1.64 (2.72) 2.28 (2.37) 2.02 (2.46) <0.00 1 Cost 733.75 (184.79, 8266.16 ) 703.29 (95.79,7959.48) 769.32 (111.47,10598.67) 769.31 (161.67,5754.88) 913.28 (194.97, 5957.42) 744.83 (95.79, 10598.67 ) <0.00 1 Value 2.42(2.82) 1.27(2.48) 1.49(2.69) 1.38(2.55) 1.66(2.41) 1.56(2.60) <0.00 1 Efficiency 3.65(4.15) 1.84(3.40) 2.04(3.11) 2.11(3.76) 2.39 (3.33) 2.26(3.57) <0.00 1 MCID NDI 38.8%(175) 23.6%(254) 29.3(180) 26.3%(89) 40.6%(89) 29.1%(787) <0.00 1 MCID NPRS 68.1%(307) 50.9%(549) 56.4%(346) 46.4%(157) 62.1%(136) 55.4%(1495) <0.00 1 ADM SF 6D 0.37(0.20) 0.44(0.18) 0.42(0.184) 0.42(0.18) 0.37(0.19) 0.42(0.19) <0.00 1 D C SF 6D 0.56( 0.38) 0.54(0.44) 0.54(0.42) 0.53(0.42) 0.53(0.37) 0.54(0.42) 0.109 CHG SF 6D 0.19(0.19) 0.09 (0.15) 0.12(0.1570) 0.11(0.15) 0.16(0.18) 0.12(0.17) <0.00 1 All values represent Mean s ( standard deviations ), Percentages ( Frequency Count)

PAGE 100

100 T able 1 2. Odds of A chieving MCID for NDI Score (19 points) Variable Adjusted Odds Ratio (aOR) p value 95% Confidence Interval for aOR Lower Upper Clinical Sub group Chronic Ref Acute 2.00 <0.001 1.57 2.55 Neck P ain w/AP 1.29 0.03 1.03 1.61 Headache 1.16 0 .31 0.87 1.54 Whiplash 2.21 <0.001 1.62 3.02 LOS (days) 0.99 0.002 0.99 0.998 Visits 1.06 <0.001 1.03 1.09 Gender Male Female 1.11 0.27 0.92 1.34 Age 1.00 0.47 0.99 1.01 Table 1 3. Odds of Achieving MCID for NPRS (1.3 points) Variable Adju sted Odds Ratio (aOR) p value 95% Confidence Interval for aOR Lower Upper Clinical Sub group Chronic Ref Acute 2.07 <0.001 1.63 2.62 Neck pain w / AP 1.18 0.1 1 0.97 1.45 Headache 0.86 0.24 0.67 1.11 Whiplash 1.58 0.004 1.16 2.15 LOS (da ys) 0.99 0.002 0.99 1.00 Visits 1.09 <0.001 1.06 1.12 Gender Male Female 0.98 0.79 0.83 1.16 Age 1.01 0.03 1.00 1.01

PAGE 101

101 Table 1 4. Predictors of Efficiency for an Episode of Physical Therapy for Neck Pain Variable B Confidence Interval fo r B t p value Lower Upper Clinical Sub group Chronic Ref Acute 1.29 0.93 1.67 6.85 <0.001 Neck Pain w/AP 0.01 0.32 0.34 0.08 0.94 Headache 0.16 0.25 0.57 0.76 0.45 Whiplash 0.48 0.01 0.97 1.91 0.06 Age 0.002 0.10 0.01 0.58 0 .56 Gender Male Ref Female 0.04 0.23 0.31 0.28 0.78 LOS (days) 0.03 0.03 0.02 10.97 <0.001 Admit NDI Score 0.06 0.05 0.07 14.23 <0.001 R 2 =0.131 Table 1 5. Predictors of Value for an Episode of Physical Therapy for Neck Pain Variabl e B Confidence Interval for B t p value Lower Upper Clinical Sub group Chronic Ref Acute 0.79 0.52 1.07 5.69 <0.001 Neck Pain/ Arm Pain 0.09 0.14 0.34 0.77 0.44 Headache 0.05 0.24 0.36 0.36 0.72 Whiplash 0.35 0.01 0.72 1.89 0.06 Age 0.00 0.01 0.01 0.12 0.91 Gender 0.02 0.18 0.22 0.19 0.85 LOS (days) 0.02 0.02 0.01 10.14 <0.001 Baseline NDI Score 0.04 0.04 0.05 13.6 6 <0.001 R 2 =0.114

PAGE 102

102 Table 2 1 CPT C odes Occurring in each P hase of T reatment Phase I(0 14 days) Active P assive Allowed Code Procedure Code Procedure Code Procedure 97110 Therapeutic exercise 97035 Ultrasound 97001 PT eval 97350 Therapeutic activity 97010 Hot or Cold Pack 97002 PT re eval 97535 Self Care management training G0283, 970032 Electrical Stimul ation 99070 Misc Supplies 97112 Neuromuscular Re ed 97012 Mechanical Traction 97750 Physical performance/meas ure 97150 Group Therapeutic Procedures 97124 Massage Therapy 97113 Aquatic Therapy with exercise 97140 Manual Therapy 97024 diathermy 97116 Gait training 97026 Infrared heat Phase II(15days till end of care) Active Passive Allowed Code Procedure Code Procedure Code Procedure Code Procedure Code Procedure Code Procedure 97110 Therapeutic exercise 97035 Ultrasound 97001 PT eval 97350 Ther apeutic activity 97010 Hot or Cold Pack 97002 PT re eval 97535 Self Care management training G0283, 970032 Electrical Stimulation 99070 Misc Supplies 97112 Neuromuscular Re ed 97012 Mechanical Traction 97750 Physical performance/meas ure 97150 Group Ther apeutic Procedures 97124 Massage Therapy 97113 Aquatic Therapy with exercise 97024 Diathermy 97116 Gait training 97026 Infrared heat 97140 Manual therapy

PAGE 103

103 Table 2 2 Baseline Clinical Characteristics of Sample by A dherent Care Group Variable Non A dherent Care (<75%) Adherent Care (>75%) p value Age 46.78(13.42) 43.48(13.01) 0.18 Gender Female 72.8% 69.7% 0.70 Comorbid LBP 31.1% 53.3% 0.02 Admit NDI 35.45(15.06) 37.36(17.28) 0.42 Admit Pain 5.40(2.26) 5.46(2.82) 0.79 Admission SF 6D 0.43(0.17 ) 0.41(0.20 ) 0.50 Reported as Mean (standard deviation) or Percentage (%) Table 2 3 Clinical Outcomes and C ourse of Physical Therapy by A dherent Care Group Variable Non Adherent Care (<75%) Adherent Care (>75%) p value Disability (NDI) Discharge S core 25.62(16.02) 28.97(16.10) 0.84 Change Score 9.89(12.47) 8.39(11.56) 0.54 Successful Outcome 34.2% 15.4% 0.05 Pain (NPRS) Discharge pain 3.61(2.43) 3.91(2.13) 0.37 Change in Pain 1.79(2.33) 0.596 0.55 Successful Pain Management 47.6 0 % 34.6% 0.21 SF 6D Discharge SF 6D 0.54(0.18) 0.5051(0.18) 0.26 Change in SF 6D 0.11(0.14) 0.0965(0.13) 0.51 Process Variables LOS (days) 45.70(50.81) 11.54(8.71) <0.001 Visits 6.52(4.58) 3.12(1.36) <0.001 Reported as mean(standard deviation)

PAGE 104

104 Table 2 4 H ealthcare Utilization and Costs by Adherent Care G roup for the Year of A dmission to Physical Therapy Group Non Adherent Care (<75%) Adherent Care (>75%) Sample p value Physical Therapy Costs (USD) 1048.69(768.87) 461.17(203.42) 984.33(749.81) <0. 00 1 Healthcare Providers Costs (USD) 3199.00(9631.72) 1757.48(4884.49) 3017.90(9201.90) 0.07 Number of Visits 13.76(17.89) 6.48(9.27) 12.86(17.27) <0.00 1 Prescriptions Number of Prescriptions 4.14(6.28) 3.39(5.30) 4.033(6.18) 0.59 Non narcot ic Analgesics 5.6% 8.7% 6.2% 0.55 Opioids 63.1% 69.6% 62.4% 0.55 Anti I nflamm atory 62.6% 56.5% 61.57% 0.57 Musculosketal therapy agents 58.1% 65.2% 55.37% 0.51 Corticosteroid s 30.7% 30.4% 30.99% 0.98 Costs (USD) 319.59(848.63) 306.39(690.02) 315.69(830 .53) 0.99 Imaging Number of Diagnostic Images 1.04(1.80) 0.63(1.36) 0.98(1.75) 0.09 Reported as mean(standard deviation)

PAGE 105

105 Table 2 5 Healthcare Utilization and Expenditures the Y ear a fter Admission to Physical T herapy Group Non Adherent Care (<75% ) Adherent Care (<75%) Sample p value Healthcare Providers C ost s for Visits (USD) 860.56(4807.31) 282.22(1008.04) 796.08(4546.27) 0.23 Number of Visits 2.46(7.94) 1.21(4.14) 2.32(7.61) 0.31 Prescriptions Number of Prescriptions 1.48(4.09) 2.47(8.88) 1.57(4.74) 0.24 Cost s for Prescriptions (USD) 75.20(227.87) 469.93(1985.40) 113.03(650.64) 0.24 Imaging Number of Diagnostic Images 0.07 (7.94) 0.15(0.62) 0.08 (0.45) 0.52 Reported as mean (standard deviation) Table 2 6 Adjusted Odds Ra tio (aOR) for Successful Disability Outcome from Physical Therapy (50% change in NDI) aOR P 95% CI for OR Lower Upper Group Non Adherent Ref Adherent 0.29 0.058 0.08 1.05 LOS 0.99 0.02 0.97 1.00 Visits 1.14 0.02 1.02 1.28 Age 0.99 0.4 1 0.97 1.01 Gender Male Ref Female 0.61 0.16 0.31 1.21 Comorbid LBP 0.49 0.04 0.25 0.98

PAGE 106

106 Table 2 7 Adjusted Odds Ratio (aOR) for Successful Pain Management from Physical Therapy (50% change in NPRS ) aOR P 95% CI for OR Lower Upper G roup Non Adherent Ref Adherent 0.75 0.56 0.29 1.956 LOS 0.99 0.09 0.98 1.00 Visits 1.12 0.01 1.02 1.23 Age 1.01 0.65 0.98 1.03 Gender Male Ref Female 0.93 0.81 0.50 1.71 Comorbid LBP 0.70 0.24 0.39 1.26

PAGE 107

107 Table 2 8 Health c are Utilization the Year of Admission to P hysical Therapy Variable B 95%CI for IRR IRR P Lower Upper PT visits Non Adherent Ref Adherent 0.81 0.36 0.55 0.44 <0.001 Age 0.01 1.00 1.004 1.00 <0.001 Gender Male Female 0.07 0. 97 1.19 1.07 0.19 Admit NDI 0.01 1.00 1.01 1.01 0.003 Admit NPR 0.01 0.99 1.04 1.01 0.32 Prescriptions the Year of PT Non Adherent Ref Adherent 0.29 0.59 0.95 0.75 0.02 Age 0.003 0.99 1.01 1.00 0.26 Gender Male Female 0.3 7 0.61 0.79 0.69 <0.001 Admit NPR 0.08 1.05 1.11 1.05 <0.001 Healthcare Visits the Year of PT Non Adherent Ref Adherent 0.61 0.47 0.62 0.54 <0.001 Age 0.01 0.99 1.00 0.99 <0.001 Gender Male Female 0.32 0.68 0.77 0.72 <0.0 01 Admit NDI 0.004 0.99 1.00 0.996 0.01 Admit NPR 0.03 1.02 1.05 1.03 <0.001 Number of PT visits 0.04 1.04 1.05 1.04 <0.001 Number of Diagnostic Imaging the Year of PT Non Adherent Ref Adherent 0.56 0.36 0.90 0.57 0.02 Age 0.002 0.99 1.0 0 0.99 0.74 Gender Male Female 0.41 0.52 0.85 0.67 <0.001 Admit NDI 0.001 0.99 1.01 1.00 0.84 Admit NPR 0.01 0.95 1.08 1.01 0.77

PAGE 108

108 Table 2 9 Adjusted Odds Ratio (aOR) for Prescription Medication during Year of Admission to Physical T herapy aOR p value 95% CI for OR Lower Upper Non narcotic analgesics Group Non Adherent Ref Adherent 1.2 0 0.84 0.21 6.78 LOS (days) 1.00 0.98 0.98 1.02 Visits 0.99 0.93 0.82 1.20 Age 1.03 0.24 0.98 1.08 Gender Male Ref Fe male 7.70 0.06 0.92 6.48 Comorbid LBP 3.12 0.05 0.98 9.90 Admit NPR 1.03 0.78 0.82 1.31 Opioids Group Non Adherent Ref Adherent 1.55 0.43 0.52 4.62 LOS (days) 1.01 0.19 1.00 1.02 Visits 0.95 0.33 0.86 1.05 Age 1.02 0.10 1.00 1.05 Gender Male Ref Female 1.50 0.25 0.76 2.96 Comorbid LBP 1.45 0.25 0.77 2.72 Admit NPR 1.23 0.002 1.08 1.41 Anti inflammatory Medication Group Non Adherent Ref Adherent 0.65 0.39 0.24 1.77 LOS (days) 1.00 0.85 0.99 1.01 Visits 1.02 0 .76 0.92 1.12 Age 0.99 0.52 0.97 1.02 Gender Male Ref Female 1.51 0.22 0.78 2.91 Comorbid LBP 2.24 0.01 1.18 4.24 Admit NPR 1.04 0.51 0.92 1.18

PAGE 109

109 Table 2 9 Continued. aOR p value 95% CI for OR Lower Upper Musculoskeletal Therape utic Agents Group Non Adherent Ref Adherent 1.81 0.27 0.63 5.20 LOS (days) 1.00 0.87 0.99 1.01 Visits 1.00 0.99 0.91 1.09 Age 0.99 0.21 0.96 1.01 Gender Male Ref Female 1.34 0.38 0.70 2.58 Comorbid LBP 1.82 0.05 0.99 3.33 Admit NP R 1.08 0.26 0.95 1.22 Corticosteroids Group Non Adherent Ref Adherent 0.72 0.54 0.25 2.07 LOS (days) 1.00 0.95 0.99 1.01 Visits 0.99 0.84 0.90 1.09 Age 1.00 0.73 0.98 1.03 Gender Male Ref Female 0.80 0.52 0.41 1.58 Comorbid LB P 1.89 0.05 1.0 3.51 Admit NPR 1.04 0.55 0.91 1.19

PAGE 110

110 Table 2 10 Healthcare Utilization the Year after Admission to Physical Therapy Variable B 95%CI for IRR IRR p value Lower Upper Number of Prescriptions the Year after PT Non Adherent Ref Adherent 0.53 1.28 2.26 1.70 <0.001 Age 0.02 1.01 1.02 1.02 <0.001 Gender Male Female 0.09 0.87 1.37 1.09 0.46 Discharge NPRS 0.07 1.03 1.12 1.07 <0.001 Number of Healthcare Visits the Year after PT Non Adherent Ref Ad herent 0.77 0.34 0.42 0.47 <0.001 Age 0.001 0.99 1.01 1.00 0.82 Gender Male Female 0.58 0.48 0.66 0.56 <0.001 Discharge NDI 0.001 0.99 1.01 1.00 0.72 Discharge NPRS 0.05 1.00 1.09 1.05 0.03 Number of Diagnostic Imaging the Year af ter PT Non Adherent Ref Adherent 0.65 0.70 5.19 1.91 0.20 Age 0.02 0.95 1.01 0.98 0.22 Gender Male Female 0.19 0.34 2.03 0.83 0.68 Discharge NDI 0.97 0.97 1.03 1.00 0.99 Discharge NPR S 0.09 0.87 1.36 1.09 0.44

PAGE 111

111 Table 2 11 Costs during the Year of Admission to Physical Therapy Variables e B 95%CI for e B p value Lower Upper Total Costs for episode of care(PT) Group Non Adherent Ref Adherent 0.78 0.69 0.89 <0.001 Age 0.99 0.98 0.99 0.05 Gender Male Female 0.99 0.91 1.08 0.87 Admit NDI 0.996 0.993 0.999 0.04 Admit NPR 1.01 0.99 1.03 0.20 Comorbid LBP 1.13 1.05 1.23 0.002 Number of PT visits 1.15 1.13 1.17 <0.001 LOS (days) 0.99 0.99 1.00 0.05 Cost s for Healthcare Visits the Year of PT G roup Non Adherent Ref Adherent 0.79 0.26 2.42 0.68 Age 0.993 0.97 1.02 0.66 Gender Male Female 0.35 0.15 0.77 0.01 Admit NDI 1.02 0.47 1.91 0.88 Admit NPR 0.89 0.74 1.05 0.18 Number of PT visits 1.09 1.00 1.19 0.0 Comorbid LBP 0 .95 0.47 1.91 0.88 Cost s for Prescriptions the Year of PT Group Non Adherent Ref Adherent 0.74 0.21 2.51 0.62 Age 1.05 1.01 1.07 <0.001 Gender Male Female 0.77 0.32 1.83 0.55 Admission NPRS 1.17 0.98 1.41 0.09 Comorbid LBP 1.78 0.84 3.78 0.13

PAGE 112

112 Table 2 12 Costs during the Year after admission to Physical Therapy Variable e B 95%CI for e B p value Lower Upper Costs for Healthcare Visits the Year after admit to PT Group Non Adherent Ref Adherent 0.24 0.03 2.02 0.19 Age 0.98 0.91 1.03 0.38 Gender Male Female 0.37 0.07 1.86 0.23 Discharge NDI 1.01 0.95 1.07 0.81 Discharge NPRS 1.31 0.88 1.97 0.19 Comorbid LBP 2.12 0.54 8.15 0.28 Cost s for Prescriptions the Year after admit to PT Non Adherent Ref Adherent 2.61 0.68 9.97 0.16 Age 1.05 1.02 1.09 0.002 Gender Male Female 1.10 0.81 0.50 2.43 Discharge NPRS 1.14 0.96 1.36 0.13 Comorbid LBP 3.47 1.57 7.63 0.002 T able 3 1 Comparison of Baseline Characteristics of Clinical Care G roups Variable Acute N=670 Chronic N=1080 Sample N=1750 p value Non Adherent Adherent Non Adherent Adherent Non Adherent Adherent Age 45.32 (15.54) 43.01 (14.56) 52.67 (16.78) 50.82 (16.78) 49.90 (16.07) 47.63 (16.33) <0.001 Gender Female 67.6% 6 0.8% 70.7% 68.5% 69.5% 65.4% 0.24 Admission NDI Score 40.74 (17.28) 37.35 (17.67) 34.55 (15.09) 31.74 (15.77) 36.91 (16.24) 34.02 (16.75) <0.001 Admission NPRS 5.86 (2.15) 5.70 (2.30) 5.29 (2.14) 5.12 (2.13) 5.51 (2.16) 5.35 (2.22) <0.001 Admission SF 6 D 0.37 (0.20) 0.41 (0.20) 0.44 (0.17) 0.47 (0.18) 0.44 (0.18) 0.37 (0.20) <0.001 Reported as mean (standard deviation) or percentage (%)

PAGE 113

113 Table 3 2 Comparison of Clinical Outcomes and Process Variables by Clinical Care Group Acute N=670 Chronic N=1080 Sample N=1750 p value Non Adherent (74) Adherent (590) Non Adherent Adherent Non Adherent Adherent Disability Discharge NDI Score 25.48 (17.75) 20.41 (18.63) 25.67 (16.27) 24.91 (16.94) 25.59 (16.84) 23.08 (17.73) 0.08 Change in NDI Score 15.26 (16.08) 16.95 (16.27) 8.89 (13.11) 6.83 (14.18) 11.32 (14.64) 10.95 (15.82) <0.001 % Achieving Successful Outcome 45.9% 53.1% 34.39% 28.74% 39.10% 39.47% <0.001 Pain Discharge Pain Score 3.29 (2.42) 2.89 (2.76) 3.57 (2.38) 3.32 (2.22) 3.47 (2.39) 3.15 (2.45) 0.02 Change in Pain Score 2.58 (2.49) 2.81 (2.29) 1.72 (2.25) 1.79 (2.21) 2.04 (2.38) 2.21 (2.29) <0.001 % Achieving Pain Management 59.9% 66.6% 48.19% 54.35% 52.05% 59.63% <0.001 SF 6D Discharge SF_6D 0.55 (0.20) 0.60 (0.21) 0.54 (0.19) 0.55 (0.19) 0.54 (0.19) 0.57 (0.20) 0.08 Change SF_6D 0.18 (0.18) 0.19 (0.19) .10 (0.15) 0.08 (0.16) 0.13 (0.17) 0.13 (0.18) <0.001 OTHER Value 2.04(2.64) 3.20 (2.93) 1.22 (2.22) 1.68 (4.15) 1.53 (2.43) 2.29 (3.78) <0.001 Efficiency 2.97(3.71) 5.31 (4.99) 1.79 (2.97) 2.39 (5.99) 2.23 (3.33) 3.58 (5.78) <0.001

PAGE 114

114 Table 3 3 The Adjusted Odds Ratio (aOR) of Achieving a Successful Outcome from PT (50% Improvement in NDI) Variables aOR p value 95% CI for OR Lower Upper Clinical Care Groups Chronic N on Adherent Ref Chronic Adherent 0.65 0.10 0.39 1.09 Acute non Adherent 1.64 <0.001 1.29 2.09 Acute Adherent 1.99 0.01 1.17 3.38 Age 1.00 0.12 0.99 1.00 Gender Male Ref Female 1.00 0.99 0.79 1.27 LOS (days) 0.99 0.02 0.98 0.99 Visi ts 1.03 0.11 0.99 1.07 Admission NDI 0.99 0.00 0.98 0.99

PAGE 115

115 Table 3 4 The Adjusted Odds Ratio (aOR) of Achieving Successful Pain Management (50% Improvement in NPRS) aOR p value 95% CI for OR Lower Upper Clinical Care Groups Chronic Non Adhe rent Ref Chronic Adherent 1.29 0.27 0.82 2.10 Acute non Adherent 1.61 <0.001 1.28 2.02 Acute Adherent 2.36 0.002 1.38 4.04 Age 1.00 0.29 1.00 1.01 Gender Male Ref Female 1.07 0.58 0.85 1.33 LOS 1.00 0.81 1.00 1.01 Visits 1.01 0.56 0.98 1.05 Admission NDI 0.90 <0.001 0.86 0.95

PAGE 116

116 Table 3 5 Model for Prediction of Efficiency of Physical Therapy Intervention B p 95% CI for B Clinical Care Groups Lower Upper Chronic Non Adherent Ref Chronic Adherent 0.19 0.59 0.87 0.5 0 Acute non Adherent 0.92 <0.001 0.55 1.27 Acute Adherent 2.49 <0.001 1.68 3.31 Age 0.0003 0.95 0.01 0.01 Gender Male Ref Female 0.05 0.78 0.29 0.39 LOS 0.01 0.01 0.02 0.002 Admission NDI 0.07 <0.001 0.06 0.08 R 2 =0.18

PAGE 117

117 Table 3 6 M odel for Prediction of Value of Physical Therapy Intervention Variable B p 95% CI for B Clinical Care Groups Lower Upper Chronic Non Adherent Ref Chronic Adherent 0.12 0.63 0.60 0.04 Acute non Adherent 0.64 <0.001 0.04 0.90 Acute Adherent 1.26 <0.001 0.69 1.84 Age 0.001 0.76 0.01 0.01 Gender Male Ref Female 0.08 0.54 0.17 0.32 LOS 0.01 0.02 0.01 0.001 Admission NDI 0.05 0.08 0.08 1.18 R 2 =0.16

PAGE 118

118 Table 3 7 Comparison of Unadjusted He althcare Utilization and Costs by Cli nical Group Chronic Non Adherent Chronic Adherent Acute Non Adherent Acute Adherent p Physical Therapy Visits 5.57 (2.98) 2.60 (0.91) 6.29 (4.75) 3.14 (1.34) <0.00 1 LOS 39.23 (37.41) 8.47 (3.79) 46.86 (56.63) 8.86 (4.94) <0.00 1 Costs 885.84 (51 4.84) 425.57 (194.56) 1018.34 (654.03) 460.86 (167.48) <0.00 1 Healthcare Providers Visits 13.11 (17.55) 6.07 (6.96) 8.32 (11.13) 8.28 (16.09) 0.31 Costs 2200.37 (57.39) 1070.29 (2293.38) 1008.00 (1759.64) 864.57 (1793.88) 0.20 Prescription Medication Number of Prescription s 3.50 (6.15) 4.50 (6.80) 4.32 (6.09) 2.8 (3.34) 0.87 Diagnostic Imaging Number of Diagnostic Images 1.13 (2.13) 0.33 (0.61) 0.65 (1.13) 0.71 (1.88) 0.34 Reported as M ean (Standard deviation)

PAGE 119

119 Table 3 8 Comparison of Physical Th erapy Utilization and Costs by Clinical Care Group Variable IRR 95%CI for IRR P Lower Upper PT visits Clinical Care Groups Chronic Non Adherent Ref Chronic Adherent 0.47 0.34 0.65 <0.001 Acute Non Adherent 1.12 0.15 1.30 0.15 Acute Adheren t 0.55 0.36 0.86 0.01 Age 1.01 1.00 1.01 0.08 Gender Male Female 1.04 0.88 1.22 0.63 Admit NDI 1.00 1.00 1.01 0.08 Admit NPR 1.04 0.99 1.07 0.07 PT LOS Clinical Care Groups Chronic Non Adherent Ref Chronic Adherent 0.21 0.17 0. 26 <0.001 Acute Non Adherent 1.23 1.19 1.33 <0.001 Acute Adherent 0.26 0.20 0.33 <0.001 Age 0.99 0.99 0.99 <0.001 Gender Male Female 1.06 0.99 1.14 0.06 Admit NDI 0.99 0.98 0.99 <0.001 Admit NPR 1.10 1.08 1.11 <0.001 Total Cost s for e pisode of care(PT) Clinical Care Groups Chronic Non Adherent Ref Chronic Adherent 0.79 0.65 0.96 0.02 Acute Non Adherent 0.98 0.86 1.11 0.71 Acute Adherent 0.83 0.63 1.09 0.19 Age Gender Male Female 0.99 0.87 1.11 0.83 Admit NDI 0.99 0.99 1.00 0.78 Admit NPR 1.00 0.97 1.02 0.76 Comorbid LBP 1.16 0.01 1.04 0.13 Visits 1.17 1.15 1.19 <0.001

PAGE 120

120 Table 3 9 Comparison of Health Care Utilization and Costs by Clinical Care Group the Year of Admission to Physical Therapy. Variable IRR 95%CI for IRR p Lower Upper Healthcare Visits the Year of PT Clinical Care Groups Chronic Non Adherent Ref Chronic Adherent 0.52 0.42 0.64 <0.001 Acute Non Adherent 0.54 0.48 0.61 <0.001 Acute Adherent 0.61 0.46 0.80 <0.001 Age 0. 99 0.98 1.00 <0.001 Gender Male Female 0.74 0.67 0.82 <0.001 Admit NDI 1.00 0.99 1.01 0.09 Admit NPR 1.01 0.98 1.04 0.52 Number of PT visits 1.06 1.05 1.07 <0.001 Cost s for Healthcare Visits the Year of PT Clinical Care Groups Chronic Non Adherent Ref Chronic Adherent 0.56 0.20 1.54 0.27 Acute Non Adherent 0.32 0.17 0.60 <0.001 Acute Adherent 0.19 0.05 0.75 0.02 Age Gender Male Female 0.74 0.40 1.34 0.32 Admit NDI 1.01 0.53 1.91 0.97 Admit NPR 0.94 0.81 1.09 0.43 Number of PT visits 1.09 1.01 1.18 0.03 Comorbid LBP 1.01 0.53 1.91 0.97

PAGE 121

121 Table 3 10 Number of Diagnostic Images the Year of Admission to Physical Therapy Number of Diagnostic Imaging the Year of PT Variable IRR 95%CI for IRR p Lower Up per Clinical Care Groups Chronic Non Adherent Ref Chronic Adherent 0.27 0.11 0.67 0.005 Acute Non Adherent 0.54 0.36 0.82 0.004 Acute Adherent 0.54 0.22 1.33 0.54 Age 0.98 0.97 0.99 0.009 Gender Male Female 0.59 0.42 0.84 0. 004

PAGE 122

122 Table 3 11 Number of Prescriptions the Year of Admission to Physical Therapy Variable IRR 95%CI for IRR p value Lower Upper Clinical Care Groups Chronic Non Adherent Ref Chronic Adherent 1.14 0.85 1.54 0.38 Acute Non Adherent 1. 14 0.93 1.41 0.21 Acute Adherent 0.57 0.33 0.98 0.04 Age Gender 0.97 0.99 1.00 0.32 Male Female 0.54 0.45 0.65 <0.001 Admit NPR 1.06 1.02 1.10 0.005

PAGE 123

123 LIST OF REFEREN CES 1. Fejer R, Kyvik KO, Hartvigsen J. The prevalen ce of neck pain in the world population: a systematic critical review of the literature. Eur Spine J. Jun 2006;15(6):834 848. 2. Cassidy JD, Carroll LJ, Cote P. The Saskatchewan health and back pain survey. The prevalence of low back pain and related disab ility in Saskatchewan adults. Spine (Phila Pa 1976). Sep 1 1998;23(17):1860 1866; discussion 1867. 3. Ihlebaek C, Hansson TH, Laerum E, et al. Prevalence of low back pain and sickness absence: a "borderline" study in Norway and Sweden. Scand J Public Healt h. 2006;34(5):555 558. 4. Childs JD, Cleland JA, Elliott JM, et al. Neck pain: Clinical practice guidelines linked to the International Classification of Functioning, Disability, and Health from the Orthopedic Section of the American Physical Therapy Assoc iation. J Orthop Sports Phys Ther. Sep 2008;38(9):A1 A34. 5. Horton R. GBD 2010: understanding disease, injury, and risk. Lancet. Dec 15 2012;380(9859):2053 2054. 6. Decade TFfBaJ. The Burden of Musculoskeletal Diseases in the United States. 2011. 7. Linto n SJ, Hellsing AL, Hallden K. A population based study of spinal pain among 35 45 year old individuals. Prevalence, sick leave, and health care use. Spine (Phila Pa 1976). Jul 1 1998;23(13):1457 1463. 8. Goode AP, Freburger J, Carey T. Prevalence, practice patterns, and evidence for chronic neck pain. Arthritis Care Res (Hoboken). Nov;62(11):1594 1601. 9. Bot SD, van der Waal JM, Terwee CB, et al. Incidence and prevalence of complaints of the neck and upper extremity in general practice. Ann Rheum Dis. Jan 2005;64(1):118 123. 10. Di Fabio RP, Boissonnault W. Physical therapy and health related outcomes for patients with common orthopaedic diagnoses. J Orthop Sports Phys Ther. Mar 1998;27(3):219 230. 11. Jette DU, Jette AM. Physical therapy and health outcome s in patients with spinal impairments. Phys Ther. Sep 1996;76(9):930 941; discussion 942 935. 12. Fritz JM, Cleland JA, Brennan GP. Does adherence to the guideline recommendation for active treatments improve the quality of care for patients with acute low back pain delivered by physical therapists? Med Care. Oct 2007;45(10):973 980.

PAGE 124

124 13. Gellhorn AC, Chan L, Martin B, Friedly J. Management Patterns in Acute Low Back Pain The Role of Physical Therapy. Spine (Phila Pa 1976). Apr 20 2012;37(9):775 782. 14. Ins titute of Medicine Relieving Pain in America: A Blueprint for Transforming Prevention, Care, Education, and Research Washington DC: National Academies Press; 2011. 15. IASP. Classification of chronic pain. Descriptions of chronic pain syndromes and defin itions of pain terms. Prepared by the International Association for the Study of Pain, Subcommittee on Taxonomy. Pain Suppl. 1986;3:S1 226. 16. Walsh NE, Brooks P, Hazes JM, et al. Standards of care for acute and chronic musculoskeletal pain: the Bone and Joint Decade (2000 2010). Arch Phys Med Rehabil. Sep 2008;89(9):1830 1845. 17. Bone and Joint Decade The Burden of Musculoskeletal Dise ases in the United States 2011: http://ww w.boneandjointburden.org/pdfs/BMUS_chpt9_economic.pdf Accessed Accessed April 6, 2011 18. Jacobson, L Lindgren, B. What are the costs of Illness? Socialstyrelsen. National Board of Health and Welfare. Stockhom. 1996. 19. Picavet HS, Hazes JM. Prevalence of self reported musculoskeletal diseases is high. Ann Rheum Dis. Jul 2003;62(7):644 650. 20. Martin BI, Deyo RA, Mirza SK, et al. Expenditures and health status among adults with back and neck problems. JAMA. Feb 13 2008;299(6):656 664. 21. Main CJ, Willi ams AC. Musculoskeletal pain. BMJ. Sep 7 2002;325(7363):534 537. 22. Guzman J, Hurwitz EL, Carroll LJ, et al. A new conceptual model of neck pain: linking onset, course, and care: the Bone and Joint Decade 2000 2010 Task Force on Neck Pain and Its Associat ed Disorders. Spine (Phila Pa 1976). Feb 15 2008;33(4 Suppl):S14 23. 23. Guzman J, Haldeman S, Carroll LJ, et al. Clinical practice implications of the Bone and Joint Decade 2000 2010 Task Force on Neck Pain and Its Associated Disorders: from concepts and findings to recommendations. Spine (Phila Pa 1976). Feb 15 2008;33(4 Suppl):S199 213. 24. Manchikanti L, Singh V, Datta S, Cohen SP, Hirsch JA, American Society of Interventional Pain P. Comprehensive review of epidemiology, scope, and impact of spinal pai n. Pain Physician. Jul Aug 2009;12(4):E35 70.

PAGE 125

125 25. Hoy DG, Protani M, De R, Buchbinder R. The epidemiology of neck pain. Best Pract Res Clin Rheumatol. Dec;24(6):783 792. 26. Hoy DG, Protani M, De R, Buchbinder R. The epidemiology of neck pain. Best Pract R es Clin Rheumatol. Dec 2010;24(6):783 792. 27. Borghouts JA, Koes BW, Vondeling H, Bouter LM. Cost of illness of neck pain in The Netherlands in 1996. Pain. Apr 1999;80(3):629 636. 28. Medical Expenditure Panel Survey. http://www.meps.ahrq.gov/mepsweb Accessed April 6, 2011. 29. Martin BI, Turner JA, Mirza SK, Lee MJ, Comstock BA, Deyo RA. Trends in health care expenditures, utilization, and health status among US adults with spine problems, 1997 2006. Spin e (Phila Pa 1976). Sep 1 2009;34(19):2077 2084. 30. Davis MA, Onega T, Weeks W, Lurie J. Where the United States Spends its Spine Dollars: Expenditures on different ambulatory services for the management of back and neck conditions. Spine (Phila Pa 1976). Mar 16. 31. Cote P, Cassidy JD, Carroll L. The treatment of neck and low back pain: who seeks care? who goes where? Med Care. Sep 2001;39(9):956 967. 32. Ferrari R. Origin of neck pain. Part 3. Clin Biomech (Bristol, Avon). Dec 2001;16(10):930 933. 33. Car roll LJ, Hogg Johnson S, van der Velde G, et al. Course and prognostic factors for neck pain in the general population: results of the Bone and Joint Decade 2000 2010 Task Force on Neck Pain and Its Associated Disorders. Spine (Phila Pa 1976). Feb 15 2008; 33(4 Suppl):S75 82. 34. Stahl M, Mikkelsson M, Kautiainen H, Hakkinen A, Ylinen J, Salminen JJ. Neck pain in adolescence. A 4 year follow up of pain free preadolescents. Pain. Jul 2004;110(1 2):427 431. 35. Croft PR, Lewis M, Papageorgiou AC, et al. Risk f actors for neck pain: a longitudinal study in the general population. Pain. Sep 2001;93(3):317 325. 36. Ehrmann Feldman D, Shrier I, Rossignol M, Abenhaim L. Risk factors for the development of neck and upper limb pain in adolescents. Spine (Phila Pa 1976) Mar 1 2002;27(5):523 528. 37. Cote P, Cassidy JD, Carroll LJ, Kristman V. The annual incidence and course of neck pain in the general population: a population based cohort study. Pain. Dec 2004;112(3):267 273.

PAGE 126

126 38. Lawrence RC, Felson DT, Helmick CG, et a l. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States. Part II. Arthritis Rheum. Jan 2008;58(1):26 35. 39. Cote P, van der Velde G, Cassidy JD, et al. The burden and determinants of neck pain in workers: results of the Bone and Joint Decade 2000 2010 Task Force on Neck Pain and Its Associated Disorders. J Manipulative Physiol Ther. Feb 2009;32(2 Suppl):S70 86. 40. Fejer R, Jordan A, Hartvigsen J. Categorising the severity of neck pain: establishment of cut points fo r use in clinical and epidemiological research. Pain. Dec 15 2005;119(1 3):176 182. 41. Ektor Andersen J, Isacsson SO, Lindgren A, Orbaek P. The experience of pain from the shoulder neck area related to the total body pain, self experienced health and ment al distress. The Malmo Shoulder Neck Study group. Pain. Sep 1999;82(3):289 295. 42. Ferrari R, Russell AS. Regional musculoskeletal conditions: neck pain. Best Pract Res Clin Rheumatol. Feb 2003;17(1):57 70. 43. McBeth J, Macfarlane GJ, Benjamin S, Morris S, Silman AJ. The association between tender points, psychological distress, and adverse childhood experiences: a community based study. Arthritis Rheum. Jul 1999;42(7):1397 1404. 44. Hurwitz EL, Carragee EJ, van der Velde G, et al. Treatment of neck pain: noninvasive interventions: results of the Bone and Joint Decade 2000 2010 Task Force on Neck Pain and Its Associated Disorders. Spine (Phila Pa 1976). Feb 15 2008;33(4 Suppl):S123 152. 45. Carragee EJ, Hurwitz EL, Cheng I, et al. Treatment of neck pain: i njections and surgical interventions: results of the Bone and Joint Decade 2000 2010 Task Force on Neck Pain and Its Associated Disorders. J Manipulative Physiol Ther. Feb 2009;32(2 Suppl):S176 193. 46. Jette AM, Delitto A. Physical therapy treatment choic es for musculoskeletal impairments. Phys Ther. Feb 1997;77(2):145 154. 47. Wang WT, Olson SL, Campbell AH, Hanten WP, Gleeson PB. Effectiveness of physical therapy for patients with neck pain: an individualized approach using a clinical decision making alg orithm. Am J Phys Med Rehabil. Mar 2003;82(3):203 218; quiz 219 221. 48. Fritz JM, Brennan GP. Preliminary examination of a proposed treatment based classification system for patients receiving physical therapy interventions for neck pain. Phys Ther. May 2 007;87(5):513 524.

PAGE 127

127 49. Childs JD, Fritz JM, Piva SR, Whitman JM. Proposal of a classification system for patients with neck pain. J Orthop Sports Phys Ther. Nov 2004;34(11):686 696; discussion 697 700. 50. Borghouts JA, Koes BW, Bouter LM. The clinical cou rse and prognostic factors of non specific neck pain: a systematic review. Pain. Jul 1998;77(1):1 13. 51. Fritz JM, Childs JD, Wainner RS, Flynn TW. Primary care referral of patients with low back pain to physical therapy: impact on future health care util ization and costs. Spine (Phila Pa 1976). Dec 1 2012;37(25):2114 2121. 52. Linton SJ, van Tulder MW. Preventive interventions for back and neck pain problems: what is the evidence? Spine (Phila Pa 1976). Apr 1 2001;26(7):778 787. 53. Delitto A, Erhard RE, Bowling RW. A treatment based classification approach to low back syndrome: identifying and staging patients for conservative treatment. Phys Ther. Jun 1995;75(6):470 485; discussion 485 479. 54. Fritz JM, Cleland JA, Childs JD. Subgrouping patients with l ow back pain: evolution of a classification approach to physical therapy. J Orthop Sports Phys Ther. Jun 2007;37(6):290 302. 55. Fritz JM. Physical therapy in a value based healthcare world. J Orthop Sports Phys Ther. Jan 2012;42(1):1 2. 56. US Physical Th erapy I. Medicare Reimbursement Reduction for Physical Therapy Services. Electronic]. 1/8/2013; http://finance.yahoo.com/news/medicare reimbursem ent reduction physical therapy 211500377.html Accessed 4/8/2013, 2013. 57. Alonso Blanco C, Fernandez de las Penas C, Cleland JA. Preliminary clinical prediction rule for identifying patients with ankylosing spondylitis who are likely to respond to an ex ercise program: a pilot study. Am J Phys Med Rehabil. Jun 2009;88(6):445 454. 58. Childs JD, Cleland JA. Development and application of clinical prediction rules to improve decision making in physical therapist practice. Phys Ther. Jan 2006;86(1):122 131. 59. Cleland JA, Childs JD, Fritz JM, Whitman JM, Eberhart SL. Development of a clinical prediction rule for guiding treatment of a subgroup of patients with neck pain: use of thoracic spine manipulation, exercise, and patient education. Phys Ther. Jan 2007 ;87(1):9 23. 60. Cleland JA, Fritz JM, Childs JD, Kulig K. Comparison of the effectiveness of three manual physical therapy techniques in a subgroup of patients with low back

PAGE 128

128 pain who satisfy a clinical prediction rule: study protocol of a randomized clini cal trial [NCT00257998]. BMC Musculoskelet Disord. 2006;7:11. 61. Cleland JA, Mintken PE, Carpenter K, et al. Examination of a clinical prediction rule to identify patients with neck pain likely to benefit from thoracic spine thrust manipulation and a gene ral cervical range of motion exercise: multi center randomized clinical trial. Phys Ther. Sep 2010;90(9):1239 1250. 62. Puentedura EJ, Cleland JA, Landers MR, Mintken PE, Louw A, Fernandez de Las Penas C. Development of a clinical prediction rule to identi fy patients with neck pain likely to benefit from thrust joint manipulation to the cervical spine. J Orthop Sports Phys Ther. 42(7):577 592. 63. Whitman JM, Cleland J, Mintken P. Clinical prediction rules in physical therapy: coming of age? J Orthop Sports Phys Ther. Mar 2009;39(3):231 233. 64. Gross AR, Hoving JL, Haines TA, et al. A Cochrane review of manipulation and mobilization for mechanical neck disorders. Spine (Phila Pa 1976). Jul 15 2004;29(14):1541 1548. 65. Grimshaw J, Eccles M, Russell I. Develo ping clinically valid practice guidelines. J Eval Clin Pract. Sep 1995;1(1):37 48. 66. Fritz JM, Hunter SJ, Tracy DM, Brennan GP. Utilization and clinical outcomes of outpatient physical therapy for medicare beneficiaries with musculoskeletal conditions. P hys Ther. Mar 2011;91(3):330 345. 67. Pietrobon R, Coeytaux RR, Carey TS, Richardson WJ, DeVellis RF. Standard scales for measurement of functional outcome for cervical pain or dysfunction: a systematic review. Spine (Phila Pa 1976). Mar 1 2002;27(5):515 5 22. 68. Hains F, Waalen J, Mior S. Psychometric properties of the neck disability index. J Manipulative Physiol Ther. Feb 1998;21(2):75 80. 69. Riddle DL, Stratford PW. Use of generic versus region specific functional status measures on patients with cervi cal spine disorders. Phys Ther. Sep 1998;78(9):951 963. 70. Westaway MD, Stratford PW, Binkley JM. The patient specific functional scale: validation of its use in persons with neck dysfunction. J Orthop Sports Phys Ther. May 1998;27(5):331 338. 71. Wheeler AH, Goolkasian P, Baird AC, Darden BV, 2nd. Development of the Neck Pain and Disability Scale. Item analysis, face, and criterion related validity. Spine (Phila Pa 1976). Jul 1 1999;24(13):1290 1294.

PAGE 129

129 72. Cleland JA, Childs JD, Whitman JM. Psychometric pro perties of the Neck Disability Index and Numeric Pain Rating Scale in patients with mechanical neck pain. Arch Phys Med Rehabil. Jan 2008;89(1):69 74. 73. Childs JD, Piva SR, Fritz JM. Responsiveness of the numeric pain rating scale in patients with low ba ck pain. Spine (Phila Pa 1976). Jun 1 2005;30(11):1331 1334. 74. Sheffield TUo. SF 6D. Electronic] Brief Overview of the SF 6D. Available at: http://www.shef.ac.uk/scharr/sections/heds/mv h/sf 6d Accessed 4/8/2013, 2013. 75. Richardson SS, Berven S. The development of a model for translation of the Neck Disability Index to utility scores for cost utility analysis in cervical disorders. Spine J. Jan 2012;12(1):55 62. 76. Statistics BoLa. C onsumer Price Index All Urban Users; 2013:US City Average for Medical Care Services. 77. Bot SDM, van der Waal JM, Terwee CB, et al. Predictors of outcome in neck and shoulder symptoms A cohort study in general practice. Spine (Phila Pa 1976). Aug 15 20 05;30(16):E459 E470. 78. Showalter C, O'Hearn M. Cervical and thoracic mobilization versus manipulation for mechanical neck pain. Letter. J Orthop Sports Phys Ther. Apr;42(4):385 386; author reply 386 392. 79. Puentedura EJ, Cleland JA, Landers MR, Mintken PE, Louw A, Fernandez de Las Penas C. Development of a clinical prediction rule to identify patients with neck pain likely to benefit from thrust joint manipulation to the cervical spine. J Orthop Sports Phys Ther. 2012;42(7):577 592. 80. Blozik E, Laptin skaya D, Herrmann Lingen C, et al. Depression and anxiety as major determinants of neck pain: a cross sectional study in general practice. BMC Musculoskelet Disord. 2009;10:13. 81. McGlynn EA, Asch SM, Adams J, et al. The quality of health care delivered t o adults in the United States. N Engl J Med. Jun 26 2003;348(26):2635 2645. 82. O'Connor GT, Quinton HB, Traven ND, et al. Geographic variation in the treatment of acute myocardial infarction: the Cooperative Cardiovascular Project. JAMA. Feb 17 1999;281(7 ):627 633. 83. Larson E. Status of practice guidelines in the United States: CDC guidelines as an example. Prev Med. May 2003;36(5):519 524. 84. Jull G, Trott P, Potter H, et al. A randomized controlled trial of exercise and manipulative therapy for cervic ogenic headache. Spine (Phila Pa 1976). Sep 1 2002;27(17):1835 1843; discussion 1843.

PAGE 130

130 85. Ylinen J, Takala EP, Nykanen M, et al. Active neck muscle training in the treatment of chronic neck pain in women: a randomized controlled trial. JAMA. May 21 2003;28 9(19):2509 2516. 86. Ylinen J, Kautiainen H, Wiren K, Hakkinen A. Stretching exercises vs manual therapy in treatment of chronic neck pain: a randomized, controlled cross over trial. J Rehabil Med. Mar 2007;39(2):126 132. 87. Flynn T, Fritz J, Whitman J, e t al. A clinical prediction rule for classifying patients with low back pain who demonstrate short term improvement with spinal manipulation. Spine (Phila Pa 1976). Dec 15 2002;27(24):2835 2843. 88. Hicks GE, Fritz JM, Delitto A, McGill SM. Preliminary dev elopment of a clinical prediction rule for determining which patients with low back pain will respond to a stabilization exercise program. Arch Phys Med Rehabil. Sep 2005;86(9):1753 1762. 89. Feuerstein M, Hartzell M, Rogers HL, Marcus SC. Evidence based p ractice for acute low back pain in primary care: patient outcomes and cost of care. Pain. Sep 2006;124(1 2):140 149. 90. Bishop PB, Wing PC. Knowledge transfer in family physicians managing patients with acute low back pain: a prospective randomized contro l trial. Spine J. May Jun 2006;6(3):282 288. 91. Carroll LJ, Hogg Johnson S, Cote P, et al. Course and prognostic factors for neck pain in workers: results of the Bone and Joint Decade 2000 2010 Task Force on Neck Pain and Its Associated Disorders. J Manip ulative Physiol Ther. Feb 2009;32(2 Suppl):S108 116. 92. Luime JJ, Koes BW, Miedem HS, Verhaar JA, Burdorf A. High incidence and recurrence of shoulder and neck pain in nursing home employees was demonstrated during a 2 year follow up. J Clin Epidemiol. Ap r 2005;58(4):407 413. 93. Enthoven P, Skargren E, Oberg B. Clinical course in patients seeking primary care for back or neck pain: a prospective 5 year follow up of outcome and health care consumption with subgroup analysis. Spine (Phila Pa 1976). Nov 1 20 04;29(21):2458 2465. 94. Bertozzi L, Gardenghi I, Turoni F, et al. Effect of Therapeutic Exercise on Pain and Disability in the Management of Chronic Nonspecific Neck Pain: Systematic Review and Meta Analysis of Randomized Trials. Phys Ther. Apr 11 2013. 9 5. Jette AM, Smith K, Haley SM, Davis KD. Physical therapy episodes of care for patients with low back pain. Phys Ther. Feb 1994;74(2):101 110; discussion 110 105.

PAGE 131

131 96. Maetzel A, Li L. The economic burden of low back pain: a review of studies published bet ween 1996 and 2001. Best Pract Res Clin Rheumatol. Jan 2002;16(1):23 30. 97. Asche CV, Kirkness CS, McAdam Marx C, Fritz JM. The societal costs of low back pain: data published between 2001 and 2007. J Pain Palliat Care Pharmacother. 2007;21(4):25 33. 98. Porter ME. What Is Value in Health Care?. New England Journal of Medicine. Dec 23 2010;363(26):2477 2481. 99. Loeser JD, Butler SH, Chapman RC, Turk DC. Bonica's Management of Pain. Vol 3rd Edition: Lippincott Williams & Wilkins; 2001.

PAGE 132

132 BIOGRAPHICAL SKETC H Maggie Horn, PT, DPT, MPH earned her Master of Public Health (MPH) with a concentration in epidemiology from the University of Florida i n 2011. She earned her Doctor of Physical Therapy degree from the University of Florida in 2010 and her Bachelor of Science (BS) in psychology with honors from Oklahoma State University in 2005. Maggie is a member of various professional organizations including the American Physical Therapy Association, the Florida Physical Therapy Association and prior membership in the Florida Public Health As sociation.