1 TRANSLATING COMMUNITY BASED PARTICIPATORY RESEARCH INTO NURSING PRACTICE: CLOSING THE HEALTH DISPARITIES GAP By CATHERINE LEVONIAN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010
2 2010 Catherine Levonian
3 To my daughters: Jennifer and Lisa Levonian
4 ACKNOWLEDGMENTS This dissertation would not be possible without the gentle guidance of my chair, Shawn Kneipp. Through her dedication to her mentorship role in developing nurse researchers she has motivated me dail y Each of my committee members (Barbara Lutz, Ellen Lopez and David Miller) shared their particular research expertise and specific manner of support during this long marathon called doctoral education. I would like to thank the Florida Nurses Association, Florida Public Health Association and the Florida Depart ment of Health for their support of this research. Thank you to all the public health nurses who took their precious time to participate and share their knowledge in either the national or Florida survey. No one gets through a rigorous program without help and at UF the list of my helpers is quite long. It runs from the many faculty members in nursing and public health, the staff of ORS and IT the invincible Cecile Kiley to the awesome CBPR research team Along this journey I met some new friends that I hope will be my colleagues for life. Tina and Ann shared visionary moments, mountains and valleys but we leave UF stronger for them At mile 22 of this marathon of a program, Maryann and I found that we were at the same point in this doctoral education jou rney; it was nice to finish the last miles together. Most of all and in all things I thank my family. My family has been my solid base from which I even hoped to soar to a doctor al education. My brother and sister in law were cheerleaders for me. Somehow my husband Don always knew what I needed during this time and helped me in innumerable ways. Jennifer challenged me and offered her specialized talents for assistance Lisa made me laugh and smile which always lifted my spirit to continue Thank you to al l my friends and family for their unending support.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................................... 4 LIST OF TABLES ................................................................................................................................ 7 LIST OF FIGURES .............................................................................................................................. 8 ABSTRACT .......................................................................................................................................... 9 CHAPTER 1 INTRODUCTION ....................................................................................................................... 11 Problem and Purpose .................................................................................................................. 12 Significance ................................................................................................................................. 13 Definition of Terms ..................................................................................................................... 15 2 LITERATURE REVIEW ........................................................................................................... 18 Partnering to Address Health Disparities .................................................................................. 18 Translating CBPR into Practice ................................................................................................. 19 Call for CBP R ...................................................................................................................... 19 Community-Based Participatory Approach to Practice .................................................... 20 Working with communities ................................................................................................. 21 Review of CBPR studies ..................................................................................................... 22 CBPA: Inno vation in public health nursing practice ......................................................... 26 Trends in Public Health Nursing: A Slow Return to Community Roots ................................. 27 General Public Health Nursing Competencies and Community Based Practices ................... 32 Summary of Relevant Literature ................................................................................................ 35 Si gnificance of Study .................................................................................................................. 36 3 METHODOLGY ......................................................................................................................... 40 Conceptual Framework. .............................................................................................................. 40 Research Design .......................................................................................................................... 45 Study Aims ........................................................................................................................... 45 Samples and Settings ........................................................................................................... 46 Sample .................................................................................................................................. 46 Aim 1 Inclusion criteria ............................................................................................... 47 Aim 1 Exclusion criteria for ........................................................................................ 47 Aims 2 & 3 Inclusion criteria ...................................................................................... 47 Ai ms 2 & 3 Exclusion criteria ..................................................................................... 47 Setting ................................................................................................................................... 48 Procedure for Protection of Human Subjects ..................................................................... 49 Potential Risks ..................................................................................................................... 5 0
6 Measures ...................................................................................................................................... 50 Aim 1: ................................................................................................................................... 50 Content Validity ................................................................................................................... 53 Data Collection for Instrument Testing ..................................................................................... 54 Procedures/Study Protocol for Aims 2 & 3 ............................................................................... 55 Statistical Analyses ..................................................................................................................... 57 Data Screening ..................................................................................................................... 58 Statistical Analysis Assumptions ........................................................................................ 59 4 RESULTS .................................................................................................................................... 63 Part I Aim 1: Instrument Development .................................................................................. 64 Sample .................................................................................................................................. 64 Analysis ................................................................................................................................ 65 Part II Aims 2 & 3 .................................................................................................................... 66 Sample .................................................................................................................................. 66 Sample Demographics ......................................................................................................... 67 Missing D ata ........................................................................................................................ 69 Regression One .................................................................................................................... 75 Regression Two ................................................................................................................... 76 Regression Three ................................................................................................................. 77 5 DISCUSSION AND RECOMMENDATIONS ........................................................................ 90 Overview of Study ...................................................................................................................... 90 Major Findings ............................................................................................................................ 90 Recruitment and Response Rates as Research Barriers in Nursing Studies ............................ 95 Strengths and Limitations ........................................................................................................... 99 Strengths ............................................................................................................................... 99 Limitations ........................................................................................................................... 99 Directions for Future Research ................................................................................................. 100 Implications for Practice ........................................................................................................... 102 Summary .................................................................................................................................... 102 APPENDIX A SURVEY ................................................................................................................................... 105 B MULTIPLE IMP UTATION ..................................................................................................... 122 LIST OF REFERENCES ................................................................................................................. 124 BIOGRAPHICAL SKETCH ........................................................................................................... 138
7 LIST OF TABLES Table page 2 1 Summary of relevant research findings ................................................................................ 37 2 2 A comparison of principles of CBPR (Israel, et al., 2003) translated to CBPA ................ 39 3 1 Diffusion of CBPA: constructs, concepts and survey items ................................................ 62 4 1 Weekly response to survey (Florida Survey) ....................................................................... 78 4 2 Survey race demographic ...................................................................................................... 79 4 3 Factor loadings table for principal axis factoring with promax rotation ............................ 80 4 4 Demographics of Florida sample .......................................................................................... 83 4 5 Measure of extent Aim 2 ....................................................................................................... 84 4 6 Survey subscales (Florida sample) ........................................................................................ 84 4 7 Correlations of CBPA subscales ........................................................................................... 85 4 8 Items used in multiple regression .......................................................................................... 86 4 9 Regression 1: Predictors of self competency in CBPA skills .............................................. 87 4 10 Regression 2: Predictors of health department involvement of community in assessment and planning ........................................................................................................ 88 4 11 Correlations ............................................................................................................................ 89 4 12 Regression 3: Predictors of time s pent in CBPA ................................................................. 89 B1 Multiple Imputation ............................................................................................................. 122
8 LIST OF FIGURES Figure page 3 1 Diffusion of Innovations Framework Related to CBPA (Rogers, 2005). ........................... 61 4 1 Exploratory Factor Analysis: Scree Plot ............................................................................... 79 4 2 Flow chart of participants Florida Study .............................................................................. 82
9 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy TRANSLATING COMMUNITY BASED PARTICIPATORY RESEARCH INTO NURSING PRACTICE: CLOSING THE HEALTH DISPARITIES GAP By Catherine Levonian May 2010 Chair: Shawn Kneipp Major: Nursing Sciences The primary purpose of this study is to examine factors related to adopting a C ommunity Based Participatory Approach (CBPA) to practice among P ublic H ealth N urse s (PHN) in health departments throughout Florida An existing CBPA survey was adapted for use in this study to measure concepts more specific to PHN perceptions of the value of CBPA to reducing health disparities and to assess the amount of time PHNs spend engaging with the community. This adapted survey was sent to PHNs nationally to demonstrat e the new measure is both valid and reliabl e so that it coul d be used in the Florida Public Health Nursing study A web based survey of PHN s working at local health departments in Florida was conducted to examine factors related to adopting a CBPA to practice among PHNs. As the largest portion of the public health workforce it is essential that PHN s are competent for CBPA adoption to be successful. Findings from this study suggest that public health nurses do not feel competent i n skills associated with CBPA. Therefore, in order for public health nursing to feel con fident in their competency in working with communities, they will need continuing education to grasp the value and obtain the necessary skills to work with communities, opportunities to practice these skills and to work with other competent nurses that can serve as preceptors and mentors. The participants reported that pol icy and program decisions were most influenced by funding sources and the Florida State
10 Department of Health. Community residents were reported as having less than 2 % influence in program and policy decisions. Therefore these findings offer direction to future research to examine how finances and local health department organizational and leadership structure influences practice decisions .
11 CHAPTER 1 INTRODUCTION This study evaluated the extent to which what has typically been a research approach (community based participatory research CBPR) is being translated into public health nursing practice (heretofore referred to as a community -based participatory approachCBPA) to reduce health disparities. Health disparities continue to exist, are related to the social determinants of health and a ne w orient ation is needed to understand the complex social and behavioral factors that contribute to health (Minkler & Wallerstein, 2003b; Schultz, Krieger, & Galea, 2002) As the citizenry of the US becomes increasingly diverse, there is a growing demand for culturally appropriate solutions in health care (McKnight, 2000) CBPR principles that emphasize commu nity involvement have been found to augment the outcomes of many health interventions, (Goodman, Yoo, & Jack, 2006; Meade & Calvo, 2001) and the act of participation itself has been shown to enhance health in communities (Wallerstein & Duran, 2006) CBPR has been effective in reducing health disparities, such as in breast cancer screening in rural migrant communities (Meade & Calvo, 2001) healthy food securit y in urban settings (Travers, 1997) and asthma environmental triggers in m ultiple communities (Krieger, Allen, Roberts, Ros s, & Takaro, 2005; Parker et al., 2005) As a research approach, CBPR is highly collaborative, designed to ensure participation by affected communities about the issues being studied (Agency for Healthcare Research and Quality [AHRQ] 2005) CBPR includes a variety of stakeholders (such as directors of local health a gencies or organizations, community leaders and members from the target population) and researchers in all aspects of the research process More specifically, CBPR incorporates: 1) co -learning and reciprocal transfer of expert knowledge, 2) shared decisio n making power and 3) mutual ownership of the processes and products of the research.
12 By applying the principles of CBPR as an approach to practice, CBPA has been called for by public health experts and organizations (Isra el, Schulz, Parker, & Becker, 1998; Savage et al., 2006; Wallerstein & Duran, 2006) Current public health practice, however, continues to overwhelmingly focus on individuals in a more traditional healthcare service model and largely bypasses at risk communities when providing population-focused care and/or programs intended to reduce hea lth disparities. Implicit in recent Institute of Medicine (IOM) recommendations is the recognition that to eliminate health disparities, programs aimed at individual care are inadequate, and that it is essential to bring affected communities together and i nvolve them in all parts of health promotion/disparity reduction programs from inception to implementation, which is consistent with a CBPA to practice (Giachello et al., 2003; I nst itute of Medicine [IOM], 2003; Schwab & Syme, 1997; Lasker & Weiss, 2003). In response to health problems that have not been resolved with current approaches, research and practice has called for a focus on health us ing an ecological approach, integration of research and practice, and more community involvement (Bekemeier, 2008; Butterfield, 2002; Israel, et al., 1998; Minkler & Wallerstein, 2003b) Problem and Purpose The primary purpose of this study is to examine factors related to adopting a CBPA to practice among P ublic H ealth N urse s (PHN) in health departments throughout Florida Findings from this research will yield a better understanding of current PHN beliefs and practices related to using a CBPA in Florida a state with notable health disparities. This work also provides a foundation for future studies that will inform the effective translation of a CBPR approach into PHN practice, which should contribute to meeting the Healthy People 20 20 goal of eliminating health disparities ( Fielding & Kumanyika, 2009) The specific aims of this study are:
13 1 ) To further develop a more comprehensive, valid, and reliable measure of the perceptions and knowledge of CBPA and the degree to which it has been adopted into practice by building upon an existing, recently developed measure of community involvement in public health practice. 2 ) To determine the extent to which PHNs in local health departments in Florida (a) perceive that a CBPA is relevant to reducing health disparities, (b) are educationally prepared to implement a CBPA, (c) perceive that they are competent to implement a CBPA, (d) perceive that their re spective health departments are incorporating a CBPA into practice, and (e) spend time engaging in a CBPA in their practice settings. 3 ) To examine the relationships between : a ) Perceptions of self competency and a) personal factors, (educational preparation, PH N experience, age, race/ethnicity, importance of CBPA skills to reducing health disparities), and b) work factors ( region served job role, percent of time engaged in CBPA) b ) Perceptions of health department engagement in CBPA and personal factors (race, PHN experience, job role, self competency in CBPA, importance of CBPA skills to reducing health disparities) and work factors (funding support for health disparity reduction pr ograms, job role, region served, competency of others) c ) Time spent in CBPA and personal factors (race, PHN experience, job role, self competency in CBPA, importance of CBPA skills to reducing health disparities) and work factors (funding support for health disparity reduction programs, job role, region served, competency of others) This is a descriptiv e/correlational study reporting the current use of CBPA according to PHN self report of the preparedness to use and actual use of a CBPA in practice. Alternati ve approaches such as actual observation w ere not done, thus factors that could be measured using more objective methods were not be applied to this dissertation work, but may be useful in future studies. The sample of RNs/APNs su rveyed was exclusively in Florida and is not generalizable to other states. Finally, it is acknowledged that the response rate affect s the extent to which find ings are generalizable within Florida as well. Significance In the last 100 years, many improvements to population health can be attributed to public health (Anderson & McFarlane, 2000; IOM, 2003) Health disparities, however, continue to be a
14 problem. It has been well documented that disparities in health exist for minorities, low income, and marginalized citi zens of the United States of America in virtually every dimension of health (United States Department of Health and Human Services, 2010) An overarching goal of Healthy People 2000 to decrease health disparities (U .S. Public Health Service., 1991) was elevated to eliminate health disparities by Healthy People 2010 (U .S. Dep artment of Health and Human Services [USDHHS] 2000) Healthy People 2020 goals are now being formulated, the four goals recommen ded are : to eliminate preventable disease, disability, injury and premature death; achieve health equity; eliminate health disparities and improve health for all groups ; create social and physical environments that promote good health for all; and promote health development and healthy behaviors at every stage of life (Fielding & Kumanyika, 2009) To meet these goals requires focusing scientific and practice efforts on developing and utilizing an evidence base to close the health disparities gap. As recomm ended by I nstitute of Medicine (I OM ) Committee on Assuring the Health of the Public in the 21st Century, to meet the n eeds of all Americans, the manner of providing care must include an examination of all determinants of health, principles of equity, and participation by and empowerment of communities of interest (Anderson & McFarlane, 2000) As a scientific approach used predominantly in public health, community based participatory research (CBPR) has been successful in decreasing health disparities (IOM, 2003; Kellogg, 2005; Min kler & Wallerstein, 2003a) Having demonstrated CBPR brings promising results, several leaders in public health have noted the translation of CBPR principles to a community based participatory approach (CBPA) in public health settings should be occur ring nationally; yet studies suggest this approach is not being implemented as widely as expected (Grumbach, Miller, Mertz, & F inocchio, 2004; Head et al., 2004; Margolis, Parker, & Eng, 1999; Turnock et al., 1994) Given
15 the histor y of public health nursing that PHNs comprise the larges t group of public health professionals (Gebbie, 2000) and that PHNs are at the front lines of public health in the United States, no other group is better situated to translate a CBPA into practice. Through empowering communities to assist in improving the health of their members, PHNs can substantially help reshape the environment to tackle health disparities (Anderson & McFarlane, 2000) To be able to engage in this type of nursing practice, however, P HNs need to be competent in implementing a CBPA in their practice settings (Cross et al., 2006) The sentinel IOM report, Who Will Keep the Public Healthy, (Gebbie, Rosenstock, Hernandez, & Institute of Medicine, 2003), identi fied CBPR as a critical new area required in public health professional education with the end goal being to translate a research mo del into practice. This dissertation study examine s the current beliefs and practices of PHNs with respect to the use of C BPA in health departments throughout Florida. Results of this study will direct the development of training and performance tools for PHNs and to facilitate the translation of CBPA into practice in local health department settings, with the end goal of reducing health disparities. Definition of Terms CBPR : Community based participatory research is defined as a collaborative approach to research that equitably involves all partners in the research process and recognizes the unique strengths that each bring s (Minkler & Wallerstein, 2003b p.4 ). T he principles of CBPR which are incorporated into the research process are 1) recognition of the community as a unit of identity; 2) development of existing community strengths and resources; 3) facilitation of collaborative partnerships in all phases of the research process; 4) integration of knowledge and action for benefit of all partners; 5) co learning and empowerment; 6) cycl ical and iterative process ; 7 ) comprehension of health from a positive and ecological view; 8) dissemination
16 of findings and knowledge gained to all partners (Israel, et al., 1998) These principles operate along a continuum and should be at least partial ly reflected in all CBPR projects. CBPA: C ommunity based participatory approach to practice involves participation of at risk populations in identifying problems and designing culturally sensitive interventions to decrease health disparities within their own communities. P ublic health nursing practice embraces a population approach that partner s with the community in endeavors to eliminate health disparities by designing programs to meet the specific needs identified by the community of interest. Public H ealth Nursing : Public health nursing is the practice of promoting and protecting the health of populations using knowledge from nursing, social and public health sciences (Quad Council of Public Health Nursing Organizations, 2003, p.2 ) Public Health Nurse : The first public health nurses in the late 1800s cared for the poor in their homes and advocated for social reform Following the Depression and World War II the growth of government support for public health result ed in many new nursing positions at local health departments and schools. With this support public health nurses move d out of the community, losing their c lose relationships within the community and subsequently their skills related to mobilizing communities for change Today, most PHNs are registered nurses who work in local, state and federal agencies, schools, and other community locations (Ivanov & Blue, 2008) Th e e ducational preparation requirements for public health nurses ( PHN ) differs from state to state ( USDHHS 2005) ; h owever, a baccalaureate education curriculum is the only entry level p rogram that includ es theory and application of population based nursing (A ssociation of C ommunity H ealth N ursing E ducators [ACHNE] 2007) Public health nursing is
17 distinct from other nursing becaus e the focus of practice is on the population rather than an individual. LPHD : Local Public Health Department : The local public health department is the governmental body responsible for the local public health. As a governmental body, both the structure and governance can take many varied forms yet all are guided and their authority set by state and local laws (N ational A ssociation of County and City Health O fficials [NACCHO] 2005) Health Disparity : A population is considered a health disparity population if there is a significant disparity in the overall rate of disease incidence, prevalence, morbidity, mortality or survival rates in the population as compared to the health status of the general population US Public law 106525 (National Cancer Inst itute, 2009, p.1 ). Health Disparity Program: Research or intervention programs focused on inequities in health with the aim of reducing health disparities and improving outcomes among racial and ethnic minorities. In Florida an example is the Closing t he Gap Grant administered by the Florida Department of Health, Office of Minority Health (Office of Minority Heatlh, 2007)
18 CHAPTER 2 LITERATURE REVIEW This chapter presents a r evi ew of relevant literature covering areas specific to aims of this study. Major findings in each of these areas are outlined in Table 2 1 followed by a comprehensive review of these areas. Partnering to Address Health Disparities Health disparities continue to exist between whites and non -whites ethnic minorities, and lower SES groups in the U .S. (A HRQ 2005; Centers for Disease Control and Prevention [CDC] 2009 ; Smedley, Stith, Nelson, & I OM 2002 ), are seen in nearly every area of health throughout the lifespan, and continue to exist when controlling for socio -economic status ( SES) and insurance status (Harrison & Falco, 2005; Smedley, et al., 2002; Sue & Dhindsa, 2006) When compared to whites, racial and ethnic minorities are consistently less healthy, have a shorter life expectancy, higher rates of infant mortality and higher rates of chronic diseases in areas such as c ervical cancer and breast cancer, heart d isease, and diabetes ( Kellogg, 2005; CDC, 200 9 ). Studies have demonstrated there are many factors that contribute to health disparities -a few of which include stress, the physical and social environment, genetics, SES, behavior, access to health care, and the quality of health care received (CDC 2009 ; Hogan, Njoroge, Durant, & Ferre, 2001; Kellogg, 2005; Smedley, et al., 2002) As noted by the IOM, addressing these disparities successfully will require the use of an ecolog ical model that recognizes multiple determinants of health. Such a model should inform and guide public health practice, education, and research (IOM, 2003) Public health researchers have incorporated a CBPR approach to decrease health disparities. This approach, used increasingly as an effective method for conducting community research within the nursing and public health scientific communities, offers promise for the development of successful,
19 evidenced-based nursing interventions [to] address gaps in health and well -being (Savage, et al., 2006, p. 477) To that end, organizations such as the C enters for D isease Control and Prevention (CDC, 2009) have increased funding for programs aimed at partnering with communities to reduce health disparities. In 2000, Florida signed into law the Reducing Racial and Ethnic Health Disparities, Closing the Gap g rant p rogram which made available $5 million in funding for the 20092010 fiscal year in the 2009 funding announcement (Office of Minority Health, 2009) Closing the Gap is focused on fostering partnerships to address priority health needs identified by communities H ealth d epartments receiving this funding are expected to identify projected program results that fulfill one of these three categories: policy development, community organization, or health promotion. Currently, however, the gap in health continues in Florida, s uch as in heart disease, where the hypertension rate is 24% higher for Hispanics and A frican American than for whites, the age adjusted death rate from heart disease is 24% higher for African Americans than for whites, and the stroke death rates are almost double for African Americans when compared to rates for whites Disparities are evident in cancer as well, where the age adjusted mortality rate is 21.6% higher for non -whites than for whites (Office of Minority Health [OMH] 2009). Although a CBPA presents a promising way to address this gap based on sound research, it is unclear to what exte nt a CBPA is being adopted in health departments throughout Florida. This study examines PHN perceptions of and engagement in a CBPA as relevant to decreasing health disparities throughout Florida. Translating CBPR into Practice Call for CBPR Over the past decade, researchers and practitioners have voiced a need for more CBPR to address the multiple determinants of health disparities as identified by members of disadvantaged
20 communities, and this has subsequently lead to increased interest and funding (Israel, et al., 1998; Minkler & Wallerstein, 2003a; Savage, et al., 2006; Wallerstein & Duran, 2006) Community involvement is now required in some funding initiatives (Minkler & Wallerstein, 2003b) and public health programs such as the Healthy Start and the Ryan Whit e Care Act (IOM, 1996) In response to IOMs Future of Public Health Report the W.K. Kellogg Foundation invested $14 million in the CommunityBased Public Health Initiative to link academia with public health agencies and communit ies to form partnerships focused on public health education and practice (Smith & Randolph Back, 2000) The Agency for Health Research and Quality released a Request for Applications (RFA) for CBPR studies examining factors contributing to health disparities. In 2001, in response to that RFA, nine centers termed Excellence Centers To Eliminate Ethnic/Racial Disparities (EXCEED) were funded (AHRQ, 2001) CommunityBased Participatory Approach to Practice A CBPA in practice translates to the community setting the same principles used in a decades worth of public health rese arch utilizing the CBPR approach (see Table 2 2 ). As the focus of this dissertation research, applying a CBPA to practice requires that PHNs are able to enhance the capacity of community members to serve in partnership endeavors, appreciate the role of co mmunity participation, and mobilize community resources to address community defined priorities (Israel et al., 2003) There is a critical distinction, between community-based and community -placed research and practice that is worth emphasizing. In community-placed research, research is being conducted in a community or for a community but not necessarily in partnership with a community (Goodman, et al., 2006) Although the communities hold the necessary components required to assure population health, they are somet imes brought into projects late after planning has begun or they are simply used as informants or subjects of
21 research,(IOM, 2003, p.11 ). In practice, com munity members are often recipients of programs that are designed and implemented with limited involvement of those most affected. In community -based research the community is actively engaged in the research as partners (Israel, et al., 1998) Whereas in a community based practice model community organiz ations, community members and other stakeholders are brought into the process early to first identify community strengths and weaknesses. From this assessment, they work in partnership to set priorities, design interventions, implement interventions and ad vocate for policy change ( Keller, Schaffer, Lia Hoagberg, & Strohschein 2002) A CBPA is likely to have the benefits of empowering the participants, strengthening social engagement, establishing trust, and ensuring accountability (IOM, 2003, p.283). Wo rking with communities A CBPA involves the community stakeholders as partners in health planning, implementation, and evaluation processes. An important and necessary part of this partnership involves building a trustful relationship. In CBPR building trus t is crucial S tudies have found that c ommunities can be distrustful of research and researchers (Ahmed, Beck, Maurana, & Newton, 2004) As one commun ity member responded when asked about participating in research enough research.on the community, what are you going to do for the community (Singer 1993, p.18 ). Effective p rograms cannot be applied to communities but must be developed with and applied in a community context that takes into account social, political, cultural and economic features unique to that community (Potvin, C argo, McComber, Delormier, & Macaulay, 2003) CBPR studies are different from what might be considered more traditional approaches to research, as the community becomes the architect instead of the passive subject of research from which they either ma y not benefit or even suffer harm (A mericna P ublic H ealth A ssociation 2004; Schultz, Israel, Selig, Bayer, & Griffin, 2000) This approach gives the
22 community a voice at the table and a sense of ownership of the project (Israel, et al., 1998) Once at the table community members have questioned how much input they would actually have in the research process (Schultz, et al., 2000) and partnerships have been challenged with differing expecta tions and goals of each member, power issues, and decisions concerning representation (Wallerstein & Duran, 20 06) While there can be struggles when organizing community studies or programs the outcomes of community engagement and partnership building can be positive and rewarding. Minkler, Vasquez, Chang, & Miller, 2008 conducted a case study review of CBPR studies that resulted in policy change. In this review, community participants reported a researcher asked us about the definition of disability at first we talked about personal limits, but no w we talk about the power of voice and another participant r eported CCT (C oncerned Citizens of T illery) is my lifelineWhen you come here, there are no big guys, no little geezers. Everybody is together When there is unity, there is strength. I feel stronger after I leave the meeting (Minkler, Vasquez, Chang, & Miller, 2008, pp. 19 23). Review of CBPR studies It is instructional to review the diversity of published CBPR studies and outcomes to date. Studies using a CBPR approach have addressed health issues such as cancer, heart disease, asthma, homelessness, adolescent pregnancy, immunizations, HIV -AIDS, toxic dumping, access to care, perinatal health, and others (Clougherty, Levy, Hynes, & Spengler, 2006; Hogan, et al., 2001; Minkler, 2005; Paine -Andrews et al., 2000; Van Duyn, Reuben, & Macario, 2006; Wang, 2000; Wilcox & Knapp, 2000; Wing, 1996) Many unique positive health out comes have been achieved as well as other outcomes such as develop ment of successful marketing strategies for community p rograms (Burroughs et al., 2006) ; increased transportation, employment skills and daycare availability (Wilcox & Knapp, 2000) ; higher quality of life and fewer hospitalizations in
23 asthmatic children (Primomo, Johnston, DiBiase, Nodolf, & Noren, 2006) ; and an increase in cancer prevention practices (Linnan et al. 2005) Many of these studies are innovative, such as partnering with trained cosmetologists to deliver cancer prevention messages. In that particular study, after 12 months, 55% of the customers reported making positive changes in their health related to cancer preventi on because of this intervention (Linnan et al., 2005) In another example, the National Cancer Ins titute funded a nationwide program of Special Populations Networks (SPN) projects that sought to reduce health disparities through a community participatory approach (Van Duyn, et al., 2006) The SPN program found that the nations objective for overcoming health disparities will not be readily achieved without the collaboration of the multiple agencies and organizatio ns concerned with cancer prevention and control in minority and underserved communities (Van Duyn et al. 2006, p. 1953). The review of individual studies can illuminate the challenges and benefits of CBPR. A case study analysis conducted on a partnership between nursing researchers and the Healthy Cities Committee found the program is being sustained ten years after completion of the formal study (Minkler, Vasquez, Warner, Steussey, & Facente, 2006) Many positive outcome s and policy changes were reported such as the implementation of smoking ban s on indoor smoking; the construction of a playground; and ongoing work to change land use policy and to build trails. This program is continuing (regular meetings, incorporation a s non profit) yet due to budget and staff restraints detailed monitoring of outcomes and reporting of results are not being completed. This was noted as a limitation to reporting and evaluation of ongoing long term outcomes after the formal study period ends that will impact evaluation of CBPR projects. Another CBPR study was conducted with two Vietnamese American communities in California and Texas for the purpose of increasing the number of women receiving pap smears
24 and increasing community capacity Significant findings include a higher pap receipt in the intervention community than the wait control community, systems change (state policy, new Vietnamese pap clinic initiated); and successful coalition organization and sustainability (Ngu yen et al., 2006) Advice was given by the authors for investigators using a CBPR approach to seek financial support for initial planning phase of the project, they report ed that funding from a C enters for Disease Control and Prevention (CDC) grant allo wed for the one year planning at the beginning of the project. Competing goals from the researchers and communit ies must be dealt with in the project design phase, as seen in this study where the community coalition was adamant that a wait control instead of a no treatment control was used in the design. In four systematic reviews of CBPR studies, varied research designs were utilized to develop a diversity of interventions with modest outcomes reported in both experimental and non experimental designs (Cook, 2008; Merzel & D'Afflitti, 2003; Viswanathan et al., 2004) It was more common for studies to be described as including a non experimental de sign (Cook, 2008; Satterfield et al., 2003) except in the AHRQ review which were predominately quasi experimental (Viswanathan, et al., 2004) The amount of community participation varied in the studies in each review. In these reviews, several authors point to the process of CBPR as crucial to the studys success (Cook, 2008; Viswanathan, et al., 2004) The systematic reviews discussed the importance of the community organizing process (Merzel & D'Af flitti, 2003) initiation of the study from community, and the degree of community collaboration in all aspects of the process (Cook, 2008; Viswanathan, et al., 2004) Even studies that did not show health outcomes should be reviewed for the process benefits (intermediate outcomes) offered by CBPR such as community ownership of the issue and increased community capacity (Minkler, Blackwell, Thompson, &
25 Tamir, 2003) The involvement of community can assist in achieving a more valid and relevant program (Cook, 2008) increased community engagement in intervention activities (Merzel & D'Afflitti, 2003) and sustainability (Wallerstein & Duran, 2006) It is this process outcome that is an important evidence link to the translation of CBPR into a form more suited for public health prac tice. Butterfoss suggests ongoing systematic documentation of community participation should be conducted to be able to demonstrate a link from the participation process to achievement of intermediate and long term goals (Butterfoss, 2006) An action outcome was associated more frequently with programs initiated from the community (Cook, 2008) The action component of CBPR is distinct from other research methods and crucial given that in research the re is often a disconnect between knowledge generation and intervention implementation (Lueng, Yen, & Minkler, 2004) In these systematic reviews cost data was n ot provided, in future CBPR reviews it would be beneficial to also consider costs. This type of data can be use ful in decision making (Polit & Beck, 2008) and needed to garner support for a change in practice. The review authors suggested in the future researchers should seek to achieve a balance between community col laboration and research methodologies (Viswanathan, et al., 2004) .; and consider the type of expected outcomes and magnitude of effect size anticipated in a community study (Merzel & D'Afflitti, 2003) Additionally, funding agencies should expand the criteria for excellence to include the input of community members and a commitment to action (Cook, 2008) Thes e reviews highlight that the CBPR approach is similar to other research approaches in that it seeks to further understand a topic of importance utilizes a variety of research designs and is used in multiple disciplines. It is different than other approac hes in that it involves the community in some or all aspects of the
26 research process, is based on an ecological model, employs RCT s less often involves a longer process up front, and ultimately focuses on action/sustainability to solve community issue. CBPA : I nnovation in public health nursing practice T he s kills that make CBPR a successful research approach can be used by PHNs to successfully apply a CBPA to care for community residents. Studies have shown that current public health nursing practice is focused more on individual level care than working with communities, therefore a change in practice will need to occur to adopt a new public health nursing practice paradigm (Keller et al. 2002; Grumbach et al., 2004) To adopt a CBPA approach, staff must have the necessary educational preparation and work in an environment that supports the use of a CBPA in practice. Increasingly, practitioners are called to use a CBPA in practice. The premise is that by involving the community in all phases of health pro motion and disease prevention programs (assessment, planning, implantation and evaluation), the resultant program will be more ecologically based, specific, and culturally sensitive to the target population, and will thus be more effective in decreasing he alth disparities. Programs cannot be applied to communities but must be developed with and applied in a community context that takes into account social, political, cultural and economic features unique to that community (Potvin, et al., 2003) The IOM report, Future of Public Health (1988), delineated the core functions of public health practice and called for a population based approach to public health. To be able to institute a CBPA to practice, PHNs need knowledge, skills and attitudes that support this developmental, culturally sensitive approach, along with an understanding of the role community context plays in health disparities (Schultz et al., 2005; Travers, 1997) For example, in a CPPR study involving a nutrition program, community residents identified inner city stores of the same chain that had higher prices (and confronted that supermarket chain as a community). This resulted in
27 decreased price inequities between stores (Travers, 1997) Hence, when public health programs socially validate specific needs or contexts in a community that f unctioned as facilitators or barriers to health, only then can interventions truly have a sustained positive effect in reducing health disparities. PHNs, with their unique knowledge of the community and its people, are best suited to recruit community leaders and members, coordinate collaborative efforts, develop interventions, and apply findings to interventions relevant to the community (Savage, et al., 2006) In doing so, PHNs involve at -risk, underserved communi ties in developing focused interventions to decrease health disparities T his study examines the extent to which a CBPA is being adopted into practice to achieve this goal Trends in Public Health Nursing: A Slow Return to Community Roots Nurses comprise the largest group of public health professionals in the U.S (Gebbie, 2000) In 2008, 21% of the local health department workforce was comprised of nurses (NACCHO, 2009). Since PHNs rep resent such a large percentage of the public health professional workforce they are in a position to play a pivotal role in the transition of public health to a more population focus. Public health nursing is defined as a particular form of nursing that focuses on population health (Grumbach, et al., 2004; Parker, Margolis, Eng, & Henriquez Roldan, 2003) The educational preparation for PHNs differs from state to state. However, a baccalaureate education curriculum is the only entry level program that includes theory and application of PHN (Gebbie & Hwang, 2000) As PHNs, nurses have the responsibility to connect with the community they serve (Gebbie, 2005) with the goal of foster ing ownership of health improvement efforts by the entire community (Anderson & McFarlane, 2000) A gap curr ently exists between public health specialty organizations focus on population based, community level practice (including a CBPA), and the actual focus of public health practice today (Grumbach, et al., 2004; Israel, et al., 2003) From the late 18 00s, one of the first
28 public health nurses, Lillian Wald, worked with the community in the tenements of New York City. Walds vision of PHN was one of working for reforms in health, industry, recreation, education and housing (Buhler -Wilkerson, 1993; Holder, 2004) Community involvement continued to be crucial for successful community programs well beyond Walds initial work (Anderson & McFarlane, 2000; Sullivan, 2003) One such nurse -initiated program was the Mom to Tot Center developed in the late 1960s in Detroit by Nancy Milio, in full collaboration with a minority communit y. A tribute to the Centers importance to the community, when riots in Detroit led to setting fire to buildings in protest, the Center was the only building on the block not burned (Milio, 1970) Nevertheless, in response to funding shifts over the years, PHN practice has changed to an individual rather than a community -focus (Iton, 2007) In 2008, 75.3% of Florida registered nu rses working in public/community health report providing direct patient care (FCN, 2009) This change, via local public health departments, was primarily driven by the health care reimbursement structure of the U.S (Anderson & McFarlane, 2000; Grumbach Miller., Mertz Fino cchio 2004; Keller, et al. 1998) Over the past two decades, local health departments have often needed to seek grant monies to fund community projects (Zahner & Gredig, 2005b ). There is, however, a current shift to bring both practice and research more in line with a definition of public health that emphasizes community -based health care (Cross, et al., 2006; Gebbie & Hwang, 2000) This shift in focus includes using effective evidence based methods such as CBPR/CBPA (Gebbie, Rosenstock, Hernandez, & Institute of Medicine, 2003) As stated by the IOM, CBPR and practice should be given a prominent role in public health practice (Gebb ie, et al., 2003, p.89 ). CBPR is so vital that the American Public Health Association (APHA) issued a policy statement in 2004 in
29 support of this research approach (APHA, 2004) After the identification of the three core practices of public health, the Public Health Steering Committee recognized the ten essential service s of public health (Novick, Morrow, & Mays, 2008) Listed among thes e essential services is mobilizing community partnerships for action (CDC, 1994) As stated by the Institute of Medici ne (IOM) CBPR and practice should be given a prominent role in public health practice and is a critical requisite in public health professional education (Gebbie, et al., 2003) The practice of public health nursing is guided by the Public Health Nursing Scope and Standards of Practice (Weierbach, 2007) Within the scope of practice public health nursing is described in terms of collaborating with others and working in partnership with communities by including the per spectives, priorities, and values of the population in interpreting the data, making policy and program decisions, and selecting appropriate strategies (American Nurses Association, 2007, p.8 ). T hough the scope and standards are not an explicit pronouncement of CBPA, the principles and skills associated with CBPA are integrated into each phase of the nursing process. To be able to practice in a CBPA approach, however, a public health workforce competent in providing population -focused care and applying a CBPA to practice is essential (Gebbie, et al., 2003; Grumbach, et al., 2004) Healthy People 2010 calls for an increase in the proportion of public health departments t hat incorporate specific competencies into practice. Competence infe rs that members of the public health workforce possess the skills, knowledge, and attitude s necessary to perform certain functions. Although PHN by definition is population focused care, recent studies have found the focus more often on individuals as opposed to populations or communities (Grumbach, et al., 2004; Head, et al., 2004; Margolis, et al., 1999; Turnock, et al., 1994)
30 Educationally, baccalaureate prepared nurses should be better able to engage in CBPA than those with Associate Degrees. Findings, however, indicate this is not true: PHNs, regardless of educational preparation, report their current practice is at the individual level and that they do not feel prepar ed to practice effectively at the community level (Grumbach, et al., 2004; Jakeway, Cantrell, Cason, & Talley, 2006) A 2004 study of California PHNs found that a minimal amount of nursing time was spent in community based care, mana gers rated individual level interventions such as case management as more important to public health practice and both managers and staff themselves did not feel PHN staff was educationally prepared for community level practice (Grumbach, 2004). Likewise a fter recognizing inadequate skills in population health among PHNs in Georgia an online course focused on population focused health was created for public health nursing. Following the course, 34% of the PHNs reported that would not recommend the course t o other PHNs stating that there was not time in the practice setting to implement population level public health. In addition, l ater f ollow up found that due to budget cuts, Directors could no longer support the course, and PHNs reported practice had not c hanged but remained individual focused (Jakeway et al., 2006). Trends in current practice suggest that PHNs do not feel education ally prepared to use a CBPA (Gebbie & Hwang, 2000; Grumbach, et al., 2004; Zahner & Gredig, 2005 a ) although research findings in this area are limited (Issel, Baldwin, Lyons, & Madamala, 2006; Parker, et al., 2003) While not explicitly examining a CBPA, t wo studies evaluated community based public health nursing and found differing practices. In a 2005 Wisconsin study, a moderate shift to more population focused nursing was reported (Zahner & Gredig, 2005b ); and in Illinois (2006), PHN staff and PHN faculty reported feeling little more than minim ally prepared to perform all PHN competencies inclu ding those related to CBPA (Issel, et al., 2006) As the
31 largest part of the public health workforce, the potential impact PHNs can make to d ecreasing health disparities is significant Discernment of personal and work factors, (such as educational preparation, perceived competency, job role), on the incorporation of a CBPA into practice is necess ary to determine what factors are needed to prom ote the successful adoption of CBPA in the future This study examine d relationships between PHNs perceptions/engagement in a CBPA and personal/work/funding support factors. In practice, a CBPA requires skills such as coalition building, community organi zing, collaboration, advocacy, policy development, and social marketing (Keller, et al., 1998) One study by Parker et al. (2003) examined a CBPA in practice by public heal th staff in North Carolina. This study evaluated community based practice following an educational intervention to build capacity for community practice They found that public health staff perceived only a moderate amount of a CBPA being implemented in pr actice after the intervention (Parker, et al., 2003) Thus, e nhancing CBPA competencies appears to require more than continuing education alone, particularly if a CBPA is not being practiced in the current job setting. To effectively institute a CBPA in practice, factors related to the likelihood of adopting it need to first be identified (Parker, et al., 2003) Role of Heal th Departments: Collaborative Relationships with Served Communities Communication and collaboration is crucial between local public health departments and the communities they serve. In spite of this, in 2003 the IOM reported that communication and collabo ration is often limited, leading to the duplication of effort and an inefficient use of resources (IOM, 2003, p.11 ). Hence, the IOM envisions a public hea lth system that does not rely on government agencies acting alone but functions as a collaborative partnership of public health agencies, academia, health care systems and communities (IOM, 2003) As recommended
32 by the IOM, local public health agencies support community led efforts to inventory resources, assess needs, formulate collaborative responses, and evaluate outcomes for community health improvement and the elimination of health disparities(IOM, 2003, p.203 ). The communication and collaboration currently seen in practice ranges from (1) programs with minimum input from community members (such as public health clinics); (2) collaborative models in which the community joins programs with predetermined practices (such as the Women, Infants and Children or WIC -program); and (3) programs with jointly defined processes and outcomes (such as Racial and Ethnic Approach to Community Health or the REACH program) (Harrison & Falco, 2005,p.88 ). Despite the inherent challenges of organizational change, organizational transition from individual based care to a community based care model is possible. For example, a local public health department in Virginia adopted a new framework for PHN practice that included coalition building and community outreach, which resulted in a shift to the m ajority of practice time being allocated to population based practice (CDC, 1997; Kosidlak, 1999) Administrative support of PHNs in the transition to a CBPA was identified as critical to its success. Jakeway et al. (2006) found that supervisor and agency support was pivotal in whether nurses attended a trai ning course on community focused care and, once trained, whether transfer of knowledge to practice occurred. Local h ealth departments can support or restrain PHNs appreciation and/or use of a CBPA, and directly impacts adoptio n into practice. Given this, t his study examine d PHNs perceptions about the extent to which their respective health departments invest in a CBPA into practice. General Public Health Nursing Competencies and Community Based Practices A competent public health workforce demands skills i n population based practice (Gebbie, et al., 2003; IOM, 2003; Keller, et al., 1998) To transition PHN practice from individual to population based practice the Association of State and Territorial Directors of Nursing (ASTDN)
33 recommends the development of competencies, educational programs based on the competencies and collaboration with schools of public health and nursing to develop population health educational programs (Association of State and Territorial Directors of Nursing, 2000) Competency based performance requires both the identification in measurable indictors of the specific skills required to practice and the evalu ation of these skills (Gebbie, 2004) Several different PHN competencies were developed such as the Quad Council Co mpetencies and the Competencies for PHN Practice instrument (PHNCI) which are utilized for practice and as curriculum frameworks (Swider et al., 2006) The knowledge, attitudes and skills associated with community based participatory practice are weaved through several of the competency tools but none are specific to CBPA ; rather, the competencies reference other domains of practice. Two current public health nursing competencies were examined to assess for skills specific to community based participatory practice. The Quad Council Competencies were developed by the Quad Council of Public Health Nursing Organization. The Quad Council of Public Health Nursing Organizations is a collaboration of four national nursing groups, including the American Nurses Association, American Public Health Association, t he Association of Community Health Nursing Educators and the Association of State and Territorial Directors of Nursing. The Quad Council published the Core Public Health Nursing Competencies in 2003 as standards for the skills, knowledge and attitudes public health nurses need to have to practice competently as public health nurses (Quad Council, 2004) These competencies were based on the Council on Linkages between Academia and Public Health Practice s set of core competencies for public health professionals (Kulbok & Reed, 2006) The competencies are based on eight domains of knowledge that are translated into measurable behaviors at three levels of com petence: awaren ess, knowledge and proficiency.
34 F our domains contain skills specific to CBPA (analytic assessment, communication, community dimensions of practice and leadership and systems thinking skills) as re flected in the PHN Intervention Wheel a mode l used by PHNs across the nation that defines more clearly the practice of public health nursing as a population based practice and better describe the work of PHNs at the community and systems level of practice(Keller, et al., 1998, p.207). Studie s suggest the translation of the Quad Council competencies into practice, however, has been remarkably slow. In 2006, Oppenwal examined the current use of these competencies in practice (n=334). One third of PHNs were not familiar with them, clearly indicating a need for further work (Oppewal, Lamanna, & Lee Glenn, 2006) Cross, Block and Josten et al (2006) described the Quad Council Competencies as 1) difficult to measure; 2) organized inconsistently with the nursing process and 3) insufficiently specific and sensitive to measure public health nursing. In response the PHNCI competencies were developed b ased on the Minnesota Public Health Intervention Model and the Public Health Intervention Wheel This is competency tool the PHNCI was first developed to evaluate the Minnesota Department of Health HRSA funded grant Public Health Nursing Practice for the 21st century (Cros s, et al., 2006) B eyond evaluating the continuing education program the PHNCI intended use was as a competency instr ument for PHNs and students. Fitting with the principles of CBPA, the philosophy statement forming this tool is that collaboration with com mu nity and other professionals is a requirement for practice. The tool was noted to be comprehensive yet lengthy (195 measurable items) which could be a barrier for practical use. Within the PHNCI are many activities matching skills required in CBPA such a s collaborating with the community in assessment, planning and evaluation and coalition building.
35 Both the Quad Council and PHNCI were developed to use as tools to assess individual PHN competency. To assess all public health staff at a health department, another competency measure is the Community Based Public Health Initiative (CBPHI). Following a continuing education program titled Community Based Public Health Initiative (CBPHI), Margolis, Parker and Eng (1999) measured competency in community based pub lic health among health department staff. The CBPHI tool was used to assess agency performance in CBPA as a collective from the perspectives of the health department staff. The CBPHI tool developed by Parker et al (Parker, et al., 2003) which was basically psychometrically sound but lacked items specific to the pro posed research was tailored and pilot tested for use in this study Thus, a major shift in PHN practice now includes competency expectations around implementing a CBPA into practice; however, the extent to which these competencies are being met is largely unknown. Although, varied public health competency tools have been developed none are specific for CBPA. For states with diverse populations and wide disparities in health such as Florida it is reasonable to think translating a CBPR into a CBPA for pr actice would be a public health priority; however, it remains unclear the extent to which this is happening. To address this knowledge gap, this study examine d the extent to which PHNs in local health departments in Florida are meeting expected competencie s with respect to implementing a CBPA in practice, and will identify areas in need of development to effectively meet them. Summary of Relevant Literature To adopt a CBPA into practice requires an understanding of PHNs current strengths and weaknesses for engagement in a CBPA (Ge bbie, et al., 2003; Parker, et al., 2003) Community involvement in public health programs is the responsibility of health departments (IOM, 2003) and is essential for culturally sensitive, effective collaborative programs. PHNs need the support of adm inistrators and agencies to make a change in practice a reality (Jakeway, et al., 2006)
36 There is limited evidence of the scope of current PHN practice and further study is ne eded (Parker, et al., 2003) Comprehension of current beliefs and practices of PHNs in FL in the use of a CBPA to d ecrease health disparities begin s the process of ultimately fostering adoption of a CBPA into practice. Significance of Study Applying a CBPA to practice involves the engagement of whole communities in the development of specific, culturally sensitive interventions to decrease health disparities. Despite efforts to prepare PHNs to use a CBPA and to promote community based practice, actual implementation of a CBPA is sparsely reported in the literature This research study examine d the extent to which a gap persists in PHN/health department implementation of an effective approach (CBPA) for addressing these disparities in a state with wide disparities in health.
37 Table 2 1 Summary of r elevant r esearch f indings Research Area Major Findings 1. Partnering to Address Health Disparities Health disparities exist even when controlling for SES and insurance coverage (Harrison & Falco, 2005; Sme dley, et al., 2002; Sue & Dhindsa, 2006; AHRQ, 2005). -Minority populations are increasing in the U.S (Britto, Pandzik, Meeks, & Kotagal, 2006; Centers for Disease Control and Prevention, 2009 ). -Physiologic, social, cultural, genetic and environmental factors affect health disparities (CDC, 2009 ; Kellogg, 2005; MacQueen et al., 2001) -Assessing and addressing health disparities are the mission of public health (Kellogg, 2005; Parker, et al., 2003; U SDHHS 2000) Eliminating health disparities requires working in partnerships with affected co mmunities (CDC 2009 ; Israel, et al., 2003) 2. Translating CBPR into Practice CBPA is grounded in the same principles of CBPR (Israel, et al., 2003) CBPR research has had positive outcomes in reducing disparities (Clougherty, et al., 2006; Gany, Shah, & Changrani, 2006; Giachello, et al., 2003; Paine -Andrews, et al., 2000; Primom o, et al., 2006; Wilcox & Knapp, 2000) There is an increased call for CBPR to address health disparities (Gany, et al., 2006; Goodman, et al., 2006; Israel, et al., 1998; MacQueen, et al., 2001; Minkler & Wallerstein, 2003a; Parker, et al., 2003; Savage, et al., 2006; Wallerstein & Duran, 2006; Zahner & Gredig, 2005b ). There is increased funding for health disparity program from agencies such as the CDC (CDC, 200 9 ). Education in CBPR is needed for public health researchers/practitioners to transfer t his knowledge to practice using a CBPA (Gebbie, et al., 2003; Parker, et al., 2003)
38 Table 2 1. Continued Research Area Major Findings 3. Trends in Public Health Nursing: A Slow Return to Community Roots Nurses are the largest occupational group within the public health workforce (Gebbie, 2000) Current emphasis is on transi tioning the focus of PHNs to a population focused care (Gebbie & Hwang, 2000; Grumbach, et al., 2004; Israel, et al., 2003; Zahner & Gredig, 2005b ). There is insufficient data on the current scope of population based PHN practice that includes community participation (Grumbach, et al., 2004) -Measuring competencies of PHNs in the use of CBPA is needed (Grumbach, et al., 2004; Israel, et al., 2003) 4. Role of Health Dept. in Collaboration with Communities Vital collaborative relationships between health departments/community less than optimal (Gebbie, et al., 2003; IOM, 2003) Investment of health departments and administrators in CBPA is pivotal in PHNs decision to adopt CBPA (Jakeway, et al., 2006) 5. Public Health Competencies & Community Based Practice Recently developed public health nursing competencies and the Intervention Wheel include the components of CBPA (collaboration with community partners, coalition building, community assessment, advocating for needs in the community) (Keller, Strohschein, Lia Hoagberg, & Schaffer, 2004)
39 Table 2 2 A comp arison of principles of CBPR (Is rael, et al., 2003) translated to CBPA CBPR CBPA Collaborates equitable partnership with community in all phases, recognizes community as a unit of identity, integrates and achieves a balance between research and action for the mutual benefits for all. Local Health Department (LHD) recognizes the community as a unit of identity and partners with the community in all phases of program. Action is for the mutual benefit of community and LHD needs. Builds on strengths of the community LHD partners to assess community strengths and help increase community capacity Disseminates findings and knowledge gained to all partners The community is involved in evaluation and given opportunity to give feedback Emphasis on local relevance of pub lic health problems; ecological perspective encompassing multiple determinants of health/disease, empowering process that attends to social inequities. LHD is invested and staff is competent in skills necessary to use an ecological approach to health and in empowering communities Involves system development through cyclical process LHD is involved in program development in a cyclic iterative process. Long term commitments to reduce health disparities LHD is engaged in community partnerships/coalitions to reduce health disparities
40 CHAPTER 3 METHODOLGY Beginning with an explanation of the conceptual framework utilized for this study, this chapter continues with a detailed description of the study design, sample, and analysis plan. Conceptual Framework. The conceptual underpinnings for developing the survey items to measure a CBPA to practice in Aim 1 are derived from the principles of CBPR (Israel, et al., 2003) applied to practice, the PHN competencies (Quad Council, 2004) and essential health department responsibilities (Hofrichter, 2007) T he conceptual framework that guided this study Rogers Diffusion of Innovation theory is described here An innovation is described by Rogers as a technology, idea, or practice that is seen as new by an individual or organization (Rogers, 2003) This theory has been used extensively to describe the factors related to the adoption of an innovation For the purposes of this study the innovation is defined as a CBPA to public health nursing practice. A diffusion of innovat ion framework has been used widely to examine and explain many varied innovations such as the adoption of clinical practice guidelines (Hader et al., 2007) evidence based practice (Shirey, 2006) health and human service programs (Racine, 2006) genetics, (Horner, Abel, Taylor, & Sands, 2004; Jenkins & Calzone, 2007) tobacco control, (Stud lar, 1999) clinical information technology, (Hilz, 2000; Huffstutler, Wyatt, & Wright, 2002; Jaana, Ward, Pare, & Sicotte, 2006; Lee, 2004) social programming, (Dearing, 2004) and the use of PHN core competencies (Oppewal, et al., 2006) Rogers (2003) defines diffusion theory as the process by which an innovation is communicated through certain channels over time among the members of a social system
41 (Rogers, 2003, p.11 ). The adoption decision process is described in five stages: knowledge, persuasion, decision, implement atio n and confirmation During the process an individual or decision making unit passes from first awareness of an innovation (knowledge stage), to forming a favorable or unfavorable attitude about the innovation (persuasion stage ), to taking actions which lea ds to a decision to adopt or reject the innovation (decision stage) to using the innovation (implementation stage), and lastly to confirming the decision and seeking reinforcement (confirmation stage). Rogers describes certain attributes of the innovation that can affect adoption. During the adoption process, five attributes of the innovation have been identified as influential to adoption, including: relative advantage (better or more advantageous over current programs ), compatibility (consistent with va lues, beliefs, past experiences, and needs), complexity (perceived as difficult to understand and use), trialability (ability to try out on a limited basis before adoption), and, observability (results of innovat ion are apparent to others) which can sway adoption. As an example, one study examined factors influencing the adoption of new surgical endoscopic procedures This endoscopy study examined whether competition, available budget for new equipment, surgical technique and skill training, infor mation and support at conferences or media, extra benefit, characteristics of the technology, patient demand, planning and availability of equipment, reimbursement, and equipment servicing had a stimulating or restraining influence on adoption of the innov ative surgical technique Two factors, budget and planning were viewed by participants as limiting factors. The most important stimulating factor found that influenc ed adoption was the perception of additional benefits from the new p rocedure. Surgeons repo rted that the additional benefit from the new technique such as clinical effectiveness, morbidity and cost effectiveness had the greatest influence in the final decision
42 (Dirksen, Ament, & Go, 1996) Conside ration of influencing factors and being able to describe how the innovation offers an a relative advantage over current practice is essential to consider when marketing a new innovation (Cain & Mittman, 2002) In addition to innovation attributes, characteristics of the social system can also influence the knowledge and persuas ion stage in the decision process. The soc ial system can affect the innovations diffusion by the influence of social norms, social structure /networks the role of o pinion leaders and the consequences of the innovation. As an example, Jerome -DEmilia and Begun (2005) found that although there was research evidence supporting the use of breast conserving surgery rather than mastectomy the uptake in practice within the United States varied widely. To understand this variation, a study was designed focusing on the degree of influence hospital type had on the adoption of breast conserving surgery. The study found that the type of hospital (social context) had a significant influence in the adoption of the use of breast conserving surgery over mastectomy with ac ademic teaching hospitals being more likely to adopt than non academic teaching hospitals and community hospitals being the lowest adopters. The researchers hypothesized that this difference is related to early adopter characteristics (Jerome -D'Emilia & Begun, 2005) This finding aligns with Rogers description of early adopters as more likely to have higher education, higher status, a favorable attitude toward science, more connected social network, more outward oriented from the social system and tend to work in larger size organizations (Rogers, 2003) This understanding of social system characteristics highlights that awareness of the spec ific characteristics of the social system in which the new innovation is being diffused is important for considering potential adoption. E ven once a decision is made to adopt the innovation, several barriers can impede implementation. As an example of bar riers, one study by Britto, Pandzik, Meeks, and Kotagul
43 (2006) examined the diffusion of an influenza immunization program. The investigators identified that the lack of physician support and inadequate staff time to implement the intervention were obstacl es to full adoption of the program (Britto, et al., 2006) Following implementation, c onfirmation is when a reinforcement of an innovation decision is sought. Validating the impact of the innovation is necessary for further implementation of the innovation and if viewed negatively decision to adopt could be reversed (Rogers, 2003) Thus from the literature there are known select factors that have affected adoption of innovations budget, planning, perceived additional benefits, social context, administrative support and adequate time. These are represented in the in Rogers theory of Diffusion of Innovation as factor influencing adoption of new innovation. An abbreviated version of the framework by Rogers (Rogers, 2003) guide d this research (Figure 3 1 ) to examine factors related to adopting a CBPA into practice. Included in the figure are the variables used in measuring the relevant constructs. Note the far right construct represented in the model (impact of innovation) represents the outcome that would be expected following adoption/implementation of a CBPA to practice; however, this outcome variable was not a focal area of inquiry for this study (thus the grey text), although it will be an outcome of interest in future stu dies Given the current gap between recommendations and actual practice related to CPBA, it is imperative to examine the degree of diffusion of CBPA into the complex organizational structure of local health departments. According to Rogers, for a new pra ct ice to be adopted, the knowledge and persuasion steps have a linear effect on decision making. The adoption or rejection of a new practice (an innovation) is influenced directly by an individuals knowledge and persuasions (i.e., perceptions, social system and willingness/ability to "adopt or engage in
44 an innovation) (Rogers, 2003) In the knowledge stage, PHNs become aware of CBPA and can acquire the s kills req uired to be c ompetent to implement into practice There are two ways that this stage can start, one way is that a person recognizes a need and seeks a solution (innovation) or a person becomes aware of an innovation and develops a need to have this innovation. Using th e adoption of CBPA as an example, a local health department might seek a new more effective way to practice and consider CBPA or become aware of CBPA through a conference or social network and consider changing their practice to use more participatory appr oach. The persuasion stage inv olves the forming of an opinion about CBPA and whether it is advantageous to practice. Persuasion is directly influenced by characteristics of the innovation itself, such as the perceived relevance to practice, and/or as a pra ctice requirement for funding, as is the case for CDC REACH and Floridas Department of Health Closing the Gap funds granted to local health departments. Overall, diffusion theory will guide inquiry into the dissemination of this practice innovation by examining knowledge/skills (education, competency), persuasion (perception of importance of CBPA) and the adoption/implementation (community based participation in assessment/planning) of a CBPA into public health practice. The ability of PHNs to adopt/engag e in an innovation (CBPA) is evaluated in terms of personal factors (education, competency, age, and experience) work factors (area worked, job role, region served ) and funding support. By examining the characteristics of those PHNs who adopted a CBPA, wa ys to encourage adoption and to shorten time for adoption/implementation will be revealed (Landrum, 1998) Once CBPA is implemented into PHN practice, the specific impact of this practice change in decreasing health disparities can be measured in future studies.
45 A community-based participatory focus in public health concentrates on improving outcomes by partnering with the community in assessment, planning, implementation and evaluation activities. Adoption of a CBPA can result in increased community empowerment and engagement and more effective community specific programs because the community voice is included in all step s of practice. Therefore, an understanding of factors that influence the adoption of a new practice such as CBPA is crucial. Knowing that all new innovations are not well adopted into practice, utilizing the diffusion of innovation framework assisted in un derstanding possible reasons why an innovation might be adopted or not and the driving and restraining forces influencing adoption (Dearing, 2004) Research Design A descripti ve, correlational design was used to meet all aims. As necessary preliminary work for the disser tation ( a ims 2 and 3), Aim 1 was met prior to examining a im 2 and 3, whereby the investigator adapted an existing measure of CBPA (Parker, et al. 2003) to measure concepts more specific to PHN perceptions of the value of CBPA to reduce health disparities the amount of practice ti me spent engaging in CBPA, and demonstrated sufficient validity and reliability of the new measure to use it to meet a ims 2 and 3. Study Aims 1 To further develop a more comprehensive, valid and reliable measure of the perceptions and knowledge of CBPA and the degree to which it has been adopted into practice by building upon an existing, recently developed measure of c ommunity involvement in public health practice. 2 To determine the extent to which PHNs in local health departments in Florida (a) perceive that a CBPA is relevant to reducing health disparities, (b) are educationally prepared to implement a CBPA, (c) perceive that they are competent to implement a CBPA, (d) perceive that their respective health departments are incorporating a CBPA into practice, and (e) spend time engaging in a CBPA in their practice settings. 3 To examine the relationships between :
46 a Perc eptions of self competency and a) personal factors, (educational preparation, PHN experience, age, race/ethnicity, importance of CBPA skills to reducing health disparities), and b) work factors (geographic location, job role, percent of time engaged in CBP A) b Perceptions of health department engagement in CBPA and personal factors (race, PHN experience, job role, self competency in CBPA, importance of CBPA skills to reducing health disparities) and work factors (funding support for health disparity reductio n programs, job role, region served, competency of others) c Time spent in CBPA and personal factors (race, PHN experience, job role, self competency in CBPA, importance of CBPA skills to reducing health disparities) and work factors (funding support for hea lth disparity reduction programs, job role, region served, competency of others) For a im 1, a national sample of currently practicing PHN s completed a web based survey (see sample description, Table 4 2 and full details of instrument development in the me asure s section). For a im 2, a sample of currently practicing PHNs working at local health departments in Florida completed a web based survey. This web based survey approach was used to collect data with SNAP software, which is capable of collecting and st oring data, as well as importing it into SPSS 14 (SPSS Inc., Chicago IL) for analysis. This web based method using SNAP was selected for subject ease of completion, avoidance of data entry error, the ability to reach a representative sample of subjects geo graphically dispersed (Morris, Fenton, & Mercer, 2004) Sample s and Setting s Sample Two samples were required to meet study aims. For a im 1, a convenience sample of PHNs across the U .S. responded to a web based survey within a three week period. The survey was sent via email to nurses through an American Public Health Association Public Health Nurse Section listserv and also through the State Directors of Public Health Nursing in North Ca rolina and Alaska to their respective state PHN distribution list.
47 Aim 1 Inclusion criteria Include: (a) participants in a staff or supervisory position, practicing as PHNs in a local / county health department in U S (b) licensed in U S as a Register ed Nurse (RN) or Advanced Practice Nurse (APN), (c) willingness to complete the questionnaire, and (d) English speaking. Aim 1 Exclusion criteria for Include: Nurses with a l icensed practical nurse (LPN) degree only are excluded To meet a ims 2 & 3, sub jects i nclude d all 1,795 PHNs (RNs or ARNPs) working in Florida at local health departments (L. Hill p ersonal c ommunication September 9, 2009) Of the 67 health departments in FL, 33 are rural and 34 are urban. Aims 2 & 3 Inclusion criteria Include d : (a) participants in a staff or supervisory position, practicing as PHNs in a local / county health department in FL, (b) licensed in FL as a Registered Nurse (RN) or Advanced Practice Nurse (APN), (c) willingness to complete the questionnaire, and 4) Eng lish speaking. Aims 2 & 3 Exclusion criteria The only exclusion criteria was licensure as a licensed practical nurse. A power analyses using G Power 3.0.3 was used to calculate sample size for Aims 2 & 3. For a multiple regression full model (8 predict ors), with an effect size of 0.15 (small effect size of a 0.25 adjusted R2), and an alpha of .05, a sample size of 160 achieve d a power of .95. For a im s 2 & 3 a sample size of 547 survey responses was achieved. Varied r esponse rates from 6% 60% with an av erage of 3 9.6 % have been reported for web or internet based surveys (Cook, Heath, & Thompson, 2000; Morris, et al., 2004) To increase the respo nse rate, th is study use d initial presurvey contact, reminder contacts and personal contacts which have been associated with larger response rates in web based and internet survey s (Cook, et al., 2000)
48 Recruitment of subjects for this study start ed with the researcher making personal contact with the Director s of Nursing at local health departments. This was done by partici pating in a Department of Health Public Health Nurse Leaders conference call in August 2009. A dditional survey information was sent to the Directors of Nursing through the conference call minutes and an email which announced the survey and contained specific information describing how health department nurses could participate. Hulley 2001 suggests it is important to contact the organization in which the survey will be conducted presurvey and seek their endorseme nt During th e initial contact the goal was to introduce the survey, emphasize the significance, and develop interest in participation. Given the close working relationships of PHN faculty at the U niversity of F lorida College of Nursing with the State of Florida O ffice of Public Health Nursing, ass istance was given to send the survey email to all PHN s on the PHN listserv. Further an announcement about the survey was placed in the Florida Public Health Nurses Association Newsletter. Email notification of upcoming survey was sent to minority nurse or ganizations for their input and to encourage participation from their members. Setting This study was conducted in two parts and in two settings. Part I (aim 1) was conducted via a web based survey to PHNs across the U.S. The survey was sent via email to nurses through (a) an American Public Health Association Public Health Nurse Section listserv, (b) the State Directors of Public Health Nursing in North Carolina and Alaska to their state PHN distribution list. Part II (aims 2 and 3) of the study was conducted exclusively in the state of Florida. The survey was sent via email to PHNs through a Florida Department of Health, Office of Public Health Nursing distribution list.
49 Procedure for Protection of Human Subjects Two IRB applications were submitted and a pproved by the Institutional Review Board at the University of Florida to complete th e research, with approval sought for Parts I and II individually. For Part I an additional IRB approval was secured from the North Carolina Department of Health. For Part II, an additional IRB approval was sought and granted from the Florida Department of Health. Both Parts I & II of the study are consistent with the NIH definition of clinical research outcomes research and health services research as defined in Part II, p.8 of the PHS 4161 U.S. DHHS NIH Individual Fellowship Application Instructions, and are consistent with scenario D of the Decision Table for Human Subjects Research, Protection and the Inclusion of Women, Minorities, and Children in Part I, p.25 of the PHS 416 1 U.S. DHHS NIH Individual Fellowship Application Instructions. This study involve d minimal risk and me t the criteria for expedited review, waived documented (signed) consent, and a HIPAA waiver per the University of Florida IRB and Federal Regulations (45 CFR 46.110) and Expedited Category #7: Research on individual or group characteristics or behavior (including, but not limited to, research on perception, cognition, motivation, identity, language, communication, cultural beliefs or practices, a nd social behavior) or research employing survey, interview, oral history, focus group, program evaluation, human factors evaluation, or quality assurance methodologies ( University of Florida Institutional Review Board 01 Introductory Questionnaire p.28) For Aims 1 3 of the research, no personal health information was collected, thus meeting the criteria for obtaining a HIPAA waiver of documented (signed) consent. No personal identifiers were directly obtained or maintained via linking documents to study ID numbers
50 No special populations were solicited to participate in this research (e.g., pregnant women, prisoners, etc.). There is, however, the potential a PHN who participates could be pregnant or under the age of 21 (considered a child by NIH criteria ). There are no collaborating sites involved in conducting this research other than for the purposes of participant recruitment, to be conducted solely by the applicant / study PI; thus, no Federal Wide Assurance (FWA) is being sought. The survey was web based and was completed by PHNs at their individual health departments The web based survey utilize d a web -based survey collection software program (SNAP). The data was automatically download ed into a secure, encrypted web -based data collection system. SNA P security allows only the authorized user to open the response data; if an unauthorized person attempts to open the response data are automatically deleted Potential Risks No serious physical, psychological, social, or legal risks were incurred by subjec ts as a result of participating in the web based survey. There is always the possibility that responding to some of the questions posed in the survey particularly those around how their immediate supervisor or health department administrative staff are performing with respect to implementing a CBPA into practice may make some participants uncomfortable. The introductory letter to the survey conveyed how the participant information is being kept confidential, would be reported anonymously, and allow ed respondents to choose not to answer specific questions if they do not wish to do so. Measures Aim 1: To further develop a more comprehensive, valid and reliable measure of the perceptions and knowledge of CBPA and the degree to which it has been adopted into practice by building
51 upon an existing, recently developed measure of community involvement in public health practice termed the communitybased public health initiative survey (CBPHI) This aim has been met through preliminary work. During the docto ral training period, a course in survey design was ta ken to tailor the CBPHI tool developed by Parker et al (Parker, et al., 2003) which had demonstrated initial validity and reliability but lacked items specific to the proposed research. The CBPHI tool is a measure of community -based participatory practice. The i nvestigator contacted one of the CBPHI tool developers and received permission to use this tool for this study (E Parker personal communication March 13, 2007). This instrument was originally developed to measure the capacity of local health department s to engage in a community -based participatory approach to public health practice as part of the W.K. Kellogg Foundation Community -Based Public Health (CBPH) Initiative (Parker, et al., 2003) The original paper and pencil CBPHI tool was administered to staff at four local health departments in North Carolina with the purpose of measuring competencies in community-based participatory public health practice. The sample inclusion criterion was staff that provides public health services to the community. Exclusion criteria were specific staff roles such as clerical, se curity or home health personnel. Of eligible staff (429), a little more than half (282) completed the survey and 102 subjects with complete data were used in the analyses. The CBPHI tool contained 42 items that focused on individual performance and the hea lth departments performance, 27 were specific to community -based participatory practice. Data analysis for the CBPHI consisted of exploratory factor analysis (EFA), Cronbachs alpha for the subscale s and descriptive statistics. The results of EFA revealed four factors loading above 0 .4 on separate factors (skills of other staff, own skills, community participation, networking). The internal
52 consistency was measured using Cronbachs alpha and ranged from .63 to .87 for the four subscales (Parker, et al., 2003) In Parkers study, assessment, a proposed dimension d id not emerge as such in the factor analysis. In this revised survey, the following additional questions were added to capture assessmen t. Has your health department presented the findings from the community assessment activities to community members? Is community assessment data used when making program decisions at your health department? Is the target population asked for their opinions or perceptions concerning the health status of their community? Are the community leaders (clergy, government leaders, school leaders) asked for their opinions or perceptions concerning the health status of their community? Are community organizations/agencies asked for their opinions or perceptions concerning the health status of their community: Is the target population asked for their opinions or perceptions concerning the health status of their community in program planning/implementation? Are the community leaders (clergy, government leaders, school leaders) asked for their opinions or perceptions concerning the healt h status of their community in program planning/implementation? Are community organizations/agencies asked for their opinions or perceptions concerning the health status of their community in program planning/implementation? The revised web based sur vey (see Appendix A) contains 73 items total. Thirty items solicit demographic characteristics (rural/urban areas served, region of Florida served, job role, management status, race, ethnicity, job area, work status, age, yrs of PHN experience, receiving dispa rity funding, frequency of community assessment, feedback procedure, agency influence on health department engagement and time spent in individual or CBPA practice ). There are two narrative items asking (a) how the participants would define CBPA to pract ice and (b) anything
53 else they would like to address. There are 41 items that focus on conceptual dimensions consistent with Rogers Diffusion of Innovation Theory as it relates to adopting a CBPA into practice which were the focus of factor analysis and p sychometric testing completed with a im 1. These include: (a) dimensions of community involvement in the core public health areas of assessment and planning, (HD community involvement) (b ) competency of self to use CBPA in practice (self competency CBPA) (c ) competency of others to apply CBPA in practice (other competency CBPA) and (d ) the importance of these CBPA skills in reducing health disparities (importance CBPA). All items included in the CBPA scale (41) for factor analysis were on a 6 point Likert s cale Within each subscale the same 6 -point Likert scale was used which was worded to best fit that set of items. Nineteen questions on assessment, planning and implementation (question 21 -question 39) used a 6 point Likert scale with the choice of respons e numbered 16 with 1=never, 2=a little of the time, 3=some of the time, 4=a good bit of the time, 5=most of the time, and 6=all the time. This subscale was reverse coded for analysis to be similar to the other subscales. Questions on competency of self, others and importance of competency to reducing health disparities (questions 44a -f ,45a -f and 47 a -f) used a 6 point Likert scale with the choice of response numbered from 1 to 6 with 1=excellent, 2=very good, 3=good, 4=fa ir, 5=poor, 6=very poor. To score this survey, those items (41) measuring conceptual dimensions of CBPA can be summed as subscales or total score. The subscales are able to measure relevant constructs in the theoretical model, and allow ed the research to m ove forward in meeting a ims 2 and 3. Content Validity A panel comprised of internal community health experts (from University of Florida College of Nursing and the College of Public Health & Health Professions) and external national
54 public health experts reviewed the survey for content validity and to assess if questions were appropriate, flowed logically to meet study aims, and were reflective of CBPA constructs. Four external national experts reviewed the survey: Edith Parker MPH, DrPH, School of Public Health, University of Michigan; Doris Glick PhD, RN, University of Virginia School of Nursing; Pamela A. Kulbok, DNSc, APRN, BC; University of Virginia School of Nursing; and Joy Reed EdD, RN, North Carolina Department of Health and Human Services. Five i nternal University of Florida nursing, public health, and survey development experts reviewed the survey. The reviewers provided narrative reviews of the tool and offered suggestions for revision. Following the expert review, revisions to the measure were made; revisions and generation of items were reflective of the panel review, conceptual framework and review of literature for example: i nvestment of health departments and administrators in CBPA is pivotal to adoption of CBPA, (Jakeway, et al., 2006) and recently developed PHN competencies and a widely used population based practice model in PHN (the Intervention Wheel) include d components of CBPA (collaboration with communi ty partners, coalition building, community assessment, advocating for needs in the community, community organizing) (Keller, et al., 2004) Data Co llection for Instrument T esting In part I of the study an initial paper and pencil survey was conducted. The survey in paper and pencil form was distributed at the 2007 APHA conference in Washington DC to PHNs attending the conference; however an adequate sample was not achieved. Following scientific review recommendations and in consultation with a psychometric expert the response scale was changed to aid in analysis and scoring. The survey was changed to a web based format utilizing SNAP Software, and was distributed via email to PHNs working in health departments through
55 key contacts with State Directors of Public H ealth Nursing and the APHA Public Health Nursing Listserv from July 14, 2008 to August 8, 2008. Following part I the newly adapted survey was utilized in part II of the study to meet a im s 2 and 3 Aim 2 : To determine the extent to which PHNs in local heal th departments in Florida (a) perceive a CBPA is relevant to reducing health disparities, (b) the adequacy of their educational preparation to implement a CBPA to practice, (c) perceive their individual competence to implement a CBPA in practice, (d) perce ive that their respective health departments are incorporating a CBPA into practice, and (e) spend time engaging in a CBPA in their practice setting. Aim 3: To examine the relationships between : Perceptions of self competency and a) personal factors, (edu cational preparation, PHN experience, age, race/ethnicity, importance of CBPA skills to reducing health disparities), and b) work factors (geographic location, job role, percent of time engaged in CBPA) Perceptions of health department engagement in CBPA and personal factors (race, PHN experience, job role, self competency in CBPA, importance of CBPA skills to reducing health disparities) and work factors (funding support for health disparity reduction programs, job role, region served, competency of othe rs) Time spent in CBPA and personal factors (race, PHN experience, job role, self competency in CBPA, importance of CBPA skills to reducing health disparities) and work factors (funding support for health disparity reduction programs, job role, region serv ed, competency of others) T he survey measure adapted in a im 1 originally from Parker et al (Parker, et al., 2003) was used to measure CBPA practice among PHNs in Florida. Procedures/Study Protocol for Aims 2 & 3 Institutional Review Board approval was obtained separately for Part I ( a im1) and Part II (a ims 2 and 3). For Part I ( a im 1) IRB approval was obtained fro m the University of Florida IRB and from the North Carolina Department of Health IRB For Part II (Aims 2 and 3), IRB approval was obtained from the University of Florida IRB and the Florida Department of Health IRB prior to beginning data collection for Aims 2 and 3.
56 Aims 2 & 3 : The present sur vey conducted to meet study aims 2 & 3 utilize d SNAP software for development and delivery to participants. SNAP is available to students and faculty at the U niversity of F lorida, College of Nursing an d has been used by the researcher s sponsor. Prior to sending the survey to PHNs, extensive marketing of the research study and recruitment for increasing participation was conducted. Marketing strategies include d : Sending a colorful p oster that described the survey via email to Directors of Nursing at each of Florida local health departments to be forwarded to PHN staff (September 1, 2009) Discuss ing the upcoming survey with Florida public health leadership during a PHN Leaders conference call (August 20, 2009). T he Nursing Services Director, Carol Wright Tanner endorsed the survey and encouraged participation to Directors of Nursing from local health departments during call Placed s tudy announcement in the Sept ember issue of FL PHN newsletter The survey was distributed via an email from the Executive Community Health Nursing Director (Florida Office of Public Health Nursing) to all Florida PHNs working via a Department of Health (DOH) listserv on September 2, 2009. The email contained information about the survey, a web link to the survey and an attached participant letter explaining the survey and PI contact information. To ensure inclusion criteria are met, only RNs/APNs on the PHN listserv which includes nurse s from local health departments and Childrens Medical Services were sent the survey in English. Children Medical Services (CMS) nurses did not meet criteria for inclusion but were part of the DOH listserv. The introduction letter clearly included the crit eria asking only those nurses working at local health departments take the survey. The survey was sent to PHNs personal email addresses, with completed surveys submitted directly to the SNAP database. SNAP allows the researcher to check completed questionn aires as they are sent, and monitor response rates. To further facilitate high response rates Dillman (2007), recommends multiple reminders to complete the survey. E ncouraging reminders were planned to be sent at 1, 2 and 4 weeks, based on findings by Morr is et al. (2004) indicating that email survey reminders
57 should be sent at shorter intervals (than the usual 1, 3 and 7 weeks for postal surveys) to be more effective E mail reminders were sent as follows: to PHNs at 1, 2, and 6 weeks and to Local Health De partments Directors of Nursing at 4 weeks. During the last week of the survey the researcher placed phone calls to the Directors of Nursing at large health departments in Florida to insure their nurses had received the survey email and had the opportunity to participate. Several of the Directors asked the researcher to send the survey link again, which was then resent to the Director. One Director explained that nurses worked in clinics and only got to a computer one time a week, so completing the survey w as difficult while another Director said that her nurses liked having the opportunity to participate and survey content had reminded them of how they used to practice but no longer practice. Results w ere computed following the analysis plan detailed in the section below. The data will be reported with no individual identifiers. Statistical Analyses To me e t aim 1: ( To further develop a more comprehensive, valid and reliable measure of the perceptions and knowledge of CBPA and the degree to which it has been adopted into practice by building upon an existing, recently developed measure of community involvement in public health practice ) exploratory factor analysis was conducted. Exploratory factor analysis is a multivariate technique used for data reduction to determine the underlying structure and to summarize underlying variables this technique does not put any a priori constraints on the number of factors to be extracted This method of factor analysis is used when support for on the actual factor structure is unknown (Hair, Anderson, Tatham, & Black, 1998) For the exploratory factor analysis principal axis factoring with promax rotation was the chosen method to estimate the common variance among the items. Promax is an oblique rotation method that assum es the factors are correlated (Pett, Lackey, & Sullivan, 2003)
58 Aims 2 & 3 : To meet aim 2: To determine the extent to which PHNs in local health departments in Florida (a) perceive that a CBPA is relevant to reducing health disparities, (b) are educationally prepared to implement a CBPA, (c) perceive that they are competent to implement a CBPA, (d) perceive that their respective health departments are incorporating a CBPA into practice, and (e) spend time engaging in a CBPA in their practice settings, descriptive statistical analy sis of central tendency (mean, standard deviation) was conducted for interval level data and frequencies for categorical and dichotomous variables. To meet aim 3: Three separate multiple regressions were planned to meet aim 3. The three outcome variables of interest were not highly correlated (r = .23 .29, p=.01) so separate multiple regressions were conducted. (a) A multiple regression was conducted to examine the relationship between: p erceptions of self competency and personal factors, (educational preparation, PHN experience, age, race/ethnicity importance of CBPA skills ), work factors ( region served, job role, percent of time spent in CBPA); (b) A multiple regression was conducted to examine the relationship between: p erceptions of health department en gagement in CBPA and work factors (funding support for health disparity reduction programs region served, job role, competency of others) and personal factors ( PH N experience, race/ethnicity self competency in CBPA importance of CBPA skills) (c) A mul tiple regression was conducted to examine the relationship between percent of time spent in CBPA and personal factors ( PHN experience, race, importance of CBPA skills, self competency CBPA) and work factors (funding support, region, job role, competency of others). Data Screening Data cleaning was conducted to code variables and determine the amount of missing data in the dataset. Analysis of missing data was conducted to determine the pattern of missing data. Multiple imputation using S tata was conducted prior to data use in analysis. Descriptive statistic s
59 were used to describe the frequency of race/ethnicity, educational preparation, job role, region served and funding sources. Statistical Analysis Assumptions Prior to use in analysis th e data was examined for violations of assumptions. For the regression models, diagnostics were conducted to ensure assumptions are met. These include examining whether standard residuals or Dfbetas exceeds the cut off point (SR>3, DfBetas>1), correlations and collinearity diagnostics among predictor variables t o assess if multicollinearity was present, tests of independence, homoscadesticity, normal distribution and linearity of each IV/DV relationship. An alpha of .05 was used as the level of significance for all analyses. Regression one: A histogram of self competency CBPA was examined and this variable visually appeared normally distributed with a slight positive skewness distribution The skewness for this variable was 0.58 and kurtosis 0.75, which is w ithin the + 1 acceptable range (Meyers, Gamst, & Guarino, 2006) In larg e samples it is preferred to inspect the histogram for the shape of the distribution as formal inference tests can easily reject even a slight departure from normality in large samples (Tabacachnick & Fidell, 2007) Examining the normal probability plot, most of the data points fall on the diagonal line indicating a normal distribution. Examining the distribution, there are 96 % of the cases with abs olute values less than 1.96, 2.7% absolute values between 1.96 and 2.58, a nd 0.7% with values between 2.59 and 3.29. There are three outliers with absolute values greater than 3.29. These cases scores were examined, were reas onable /legitimate response values and will be used in the analysis Regression two: A histogram of the subscale, HD community involvement, was examined for normality. The histogram appears to be normally distributed. The skewness 0.059 and
60 kurtosis 0.162 indicates a normally distribution that is a bit flattened. The normal probability plot shows the points fall on the line fairly well. No outliers were noted on the boxplot. Regression three: A histogram of the item, time spent in CBPA was examined for normality. The skewness 0 .346 and kurtosis 0 .578 indicates a negatively skewed flattened distribution that is within the acceptable + 1 range (Meyers, et al., 2006) Examination of bivariate scatter plots showed linearity assumptions were met for all dependent variables. Homoscadascity was met, the scatter plot was examined visually and show ed a similar dispersion across residuals and dense near best fit line indicating homoscadescity. The data was examined for outliers and influential cases, regression two had no cases with a DfBeta >1.0 and no cases with ZRE > 3.0. Regression one and three h ad one case with ZRE of 3.00 and no other cases were > 3.0. Multivariate outliers were examined by computing Mahalanobis distance for each case on the three dependent variables and evaluated with chi -square distribution, no multivariate outliers were iden tified. Multicollinearity was explored for each regression by examining independent to independent variable correlations (none were greater than .9), Tolerance (all above .01), VIF ( all less than 10) (Meyers, et al., 2006) Bivariate correlations between the subscales were examined. The subscale competency of self had a large positive correlation with competency of other PHNs in CBPA skills (r=.58, p<.001) and a moderate correlation with subscale importance of CBPA skills (r=.41, p<.01). The subscale HD community involvement was moderately correlated with competency of other (r= .33, p<.01) and self competency (r= .29, p<.01). The subscale importance CBPA had a small correlation with HD community involvement ( r= .14, p<.01) (see Table 4 7).
61 Figure 3 1. Diffusion of Innovations framework r elated to CBPA (Rogers, 2003 ).
62 Table 3 1. Diffusion of CBPA: c onstructs, c oncepts and s urvey i tems Construct Concept Survey Items Sample Questions Knowledge Individual PHN personal characteristics and perceptions of personal characteristics Personal factors (experience, age, race/ethnicity, educational preparation) Competency Work factors (job role, major area of work and work setting) Personal factors : Examples items surveyed as categorical questions: Education (diploma, associate, bachelor or graduate); PHN years of exp erience (numerical); age (numerical): race (American Indian or Alaska Native, Asian, African American or Black, Native Hawaiian or Other Pacific Islander or White) and ethnicity (Hispanic or Latino yes or no). *Competency : Example items will ask: In general, how would you rate your competency in the following areas Working with community groups? Community organizing? Program planning? Advocating for the community? Community assessment? Influencing public health policy? (6point Likert scale) Health Department characteristics Investment in CBPA *Example items ask: How often have you used findings from the assessment data in your work? How often are the community residents asked for their opinions? How often do the programs you work with use communit y feedback in making decisions? Persuasion Innovation characteristics Perceived relevance of CBPA to reduce health disparities/Funding Example items ask: How important is the use of a CBPA to reducing health disparities? Decision to adopt Amount of time PHNs spend implementing CBPA These items ask: What percentage of your time is spent in CBPA?
63 CHAPTER 4 RESULTS This chapter describes the results and analysis of the web based surveys conducted to answer aim 1 (national sample) and aims 2 and 3 (Florida sample). All data was analyzed using SPSS 14 (SPSS, Chicago IL) for Windows statistical software with the alpha set apriori at 0.05. To meet Aim one, the data from Part I (national survey) was analyzed to further develop a more comprehensive, valid and reliable measure of the perceptions and knowledge of CBPA and the degree to which it has been adopted into practice by building upon an existing, recently developed measure of community involvement in public health practice (Parker, et al., 2003) Following the determination of a valid and reliable t ool, the Florida study was conducted to determin e the extent to which PHNs in local health departments in Florida (a) perceive that a CBPA is relevant to reducing health disparities, (b) are educationally prepared to implement a CBPA, (c) perceive that the y are competent to implement a CBPA, (d) perceive that their respective health departments are incorporating a CBPA into practice, and (e) spend time engaging in a CBPA in their practice settings. Additionally, regression analysis focused on examining the relationship between: Perceptions of self competency in CBPA skills and (a) personal factors, (educational preparation, PHN experience, age, race perceived importance of CBPA skills to reducing health disparities ), work factors ( region served, job role, competency of other PHNs at health department, percent of work time spent in CBPA ) Perceptions of health department involvement of community in assessment and planning and a) personal fa ctors, ( PHN experience, race perceived importance of CBPA skills t o reducing health disparities and perceived self competency in CBPA skills ), work factors (region served, job role, competency of other PHNs at health department, funding support for health disparity reduction programs ) Time spent in CBPA and a) personal fa ctors, ( PHN experience, race, perceived importance of CBPA skills to reducing health disparities and perceived self competency in CBPA skills ), work factors ( region served, job role, competency of other PHNs at health department, funding support for he alth disparity reduction programs )
64 Part I Aim 1: Instrument Development Sample In Part I, the survey was sent via email to nurses through a n American Public Health Association Public Health Nurse Section listserv t he State Directors of Public Health Nursing in North Carolina and Alaska to their respective state PHN distribution list and personal contacts in public health. This study achieved a sample size of 846 survey responses within a three week period (July Au gust 2008) based on a national sampling frame. It was unknown how many nurses were reached by the listserv email distribution lists and personal contact and so it was not p ossible to calculate a response rate. The final sample included 545 complete cases which was of sufficient size for conducting factor analysis and psychometric testing of a 41 item scale based on traditional 5 and 10subjects per item standards (Meyers, et al., 2006) The Kaiser -Meyer Olkin measure of sampling adequacy was considered excellent (.938). The following characteristics describe the sample in the national su rvey. The survey participants were predominately white (90%) and nonHispanic (97%) (see Table 4 2). In the 2004 National Sample Survey of Registered Nurses, nurses working in public health were primarily white ( 94% ) (US Department of Health and Human Ser vices, 2004) In this national sample, t he average age reported was 49 yrs with half under 50 yrs old. This sample was comprised of 40.3% staff nurses and 59.7% nurses employed as supervisor, director or other. Over half (62%) described the area they work as rural. The number of years of public health experience ranged from less than one year to 47 yrs with an average of 14.6 (9.6). The highest level of nursing education attained was a baccalaureate degree for 49% of the sample while 7% attained a diploma and 28% an associate degree. Sixteen percent had attained a graduate degree,
65 with the largest portion earning a masters (15.6%) and a smaller portion a doctorate degree (0.5%). Analysis A total of 846 PHNs responded, with 545 cases having complete data, see Table 4 2 for sample demographics. Only the cases with complete data (n=545) were used in the statistical analysis. The Kaiser -Meyer Olkin measure of sampling adequacy which examines intercorrleations between items was .94, above 0 .8 is can be interpreted as meritorious (Hair, et al., 1998) Construct validity was assessed by performing exploratory factor analysis (EFA) using Principal Axis Factoring to analyze common variance. To maximize interpretability of the factors, o blique rotation using Promax was chosen as the applied rotation method given the expected correlation amo ng factors (Thompson, 2004) O blique (Promax) rotation was conducted with kappa set at 4 to maximi ze separation of item loadings on different factors. To determine the number of factors to extract the several criteria were examined (scree plot, latent root, percentage of variance explained) as well as the interpretability of the factors as best represe ntation of the data (Hair, et al., 1998) The Scree plot illustrated a four factor structure (see Figure 4 1). The first six factors had eigenvalues greater than 1 which is used as criteria for importance of factor (Hair, et al., 1998) The first factor contributed the largest amount of the variance (3 7 %) and the next three facto rs explained diminishing amounts of variance. After examining the analysis four factors were extracted as the best representation of the data and explaining 6 2 % of the variance. The pattern matrix presented the unique contribution of each item to the facto r (factor loadings), the factor solution had 60% of loadings above 0.70. The four factors include: Perceived importance of a CBPA to Practice to Reduce Health Disparities ; Perceived Self Competency in use of CBPA to Practice ;
66 Perceived Competency of PHN Co -workers in use of CBPA to Practice ; Health departments i nvolvement of community in Assessment/Planning Processes This factor solution is depicted in the theoretical model (Figure 3 1). Eigenvalues were 1 6.6 ; 6.0 ; 3. 5 ; and 2.0 respectively. See Table 4 3 for a detailed EFA table. Reliability analysis was conducted. The total scale Cronbachs alpha was .86, for each of the subscales, Cronbachs alpha s ranged from .88 -.96. Item discrimination evaluation was conducted with no substantial change in alpha noted. The Corrected Item Total Correlations for each subscale and items ranged from .58 .85 Table 4 6 shows the scale mean, standard deviation, and Cronbachs alpha s for subscales of the CBPA tool. The results of this preliminary CBPA survey psy chometric testing show the tool is a valid and reliable tool to use in the dissertation study. Part II Aims 2 & 3 Sample Using reminders is recommended in the literature (Dillman, 2007) in this present study f indings indicate that sending reminders improved the overall response rat e. Each time a reminder was sent, a spike in responses was noted for 1 3 days after ( see Table 4 1 ). Initial response during the first week was 33 %, this is similar to the reported expectation of receiving 30% of responses in the first week reported by Yoon & Horne, 2004 but less than the 50% reported by Shannon and Bradshaw, 2002 More than half of the total responses (56%) were received at two weeks. The survey was kept active for 8 weeks total resulting in 547 survey responses. The emails with survey link were sent to a potential sample of 1795 PHNs working local health departments, a small number of emails (10) were returned marked undeliverable to sender, decreasing the available sample to 1785. The response rate was calculated as follows: Completed surveys: a ll items answered = 252 Partials surveys: with missing data = 295 Refusal: replied to email will not participate = 0
67 Eligible no response = 1238 252+295 /(252+295)+1238= 0.306 100=30.6 This calculates to a response rate of 30.6. A large portion (82%) of the total responses was achieved by five weeks. Continuing the survey for three more weeks resulted in 76 more responses ending with a sample of 547. This response rate is within the reported expected survey response rates the mean response rate in a meta analysis of 68 studies was 39.6% (C. Cook, et al., 2000) To check the representativeness of the sample achieved, the responders were compared to the demographics of those reported for public heal th nursing, information on nonresponders is not available to compare responders and nonresponders (Kramer, Schmalenber, Brewer, Verran, & Keller -Unger, 2009) Sample Demographics The following characteristics describe the Florida sample and are compared to other Florida nursing demographics (see Table 4 4). It is important to note that it can be difficult to compare demographics with other surveys of public health nursing due to lack of a clear definition of which employment categories s hould be included within public and or community health nursing when conducting nursing surveys (A CHNE, 2010; FCN, 2009) Consequently demographics can be compared only broadly. The sample achieved in this Florida study was largely (97%) female reflecting the re lative smaller proportion of the nursing workforce that is male, however this sample is not reflective of Florida workforce which is 90% female (FCN, 2009) The part icipants ranged from age 23 75 yrs old with a mean age of 54 yrs old. The majority of participants (65%) were greater than 50 yrs old. This PHN sample is older than the reported 48 yrs old average for all Florida registered nurses (FCN, 2009) This sample had many years of public health nursing experience ranging from 1 42 yrs, with 66% reporting greater than ten years experience The aging of the public
68 health nursing workforce coupled with their many years of experience which will be lost with retirement is concerning and demands a plan for recruitment and succession planning as stated in the recent Index of Nursing Wo rkforce and Education Report (FCN, 2009). This sample was reflective of the race/ethnicity breakdown of Florida PHN as reported by the Florida Department of Health (DOH) with the exception of having slightly less reporting Hispanic (3.4% vs. DOH 5.2%) and African American (15% vs. DOH 20.6%) racial/ethnicity category ( Grant, DOH, 2007). The highest level of nursing education attained was a baccalaureate degree for 39% of the sample while 9.6% attained a diploma and 32.8% an associate degree. Almost 19% had attained a graduate degree, with the largest portion earning a masters (18 %) and a smaller portion a doctorate degree (0.7%). This is similar to the educational attainment reported for nurses working in the public/community health setting by FCN (FCN, 2009) The nurses role at the health department was split between less than half reporting a staff position (42.9%) whil e slightly more than half reported a management position or other role (56.9%). To identify who is included in other role and comprehend the makeup of the management/director/other group, the respondents who answered other where then asked to describe that role. The role other than manager/director included in the 56.9% was described as nurse practitioner, coordinator, consultant, program specialist, school nurse, assistant director of nursing, health department administrator and program manager. Nearly al l PHNs in this sample are employed i n full time positions (92.3%). All regions of Florida were represented in the sample with the breakdown as follows: West (12.7%), North 28.4%, Central ( 30.1%) and South (28.7%). PHNs were asked if they worked primarily one program, 61% responded yes, however when asked to describe which program, several chose more than one program. This might be related to working most of the time in one program but assisting in others. Question w ording in
69 future studies should ask specifically which is your primary program. The most frequent program worked was the immunization program (40%), followed closely by family planning, STD and womens health. Previous studies reported that one of the mos t frequent services offered at U.S. local health departments is immunizations (NACCHO, 2009). Over half (57.4 %) reported that the health department where they worked received funding for health disparity projects. The participants reported that policy and program decisions (75%) were most influenced by funding sources and the Florida State Department of Health. Community residents were reported as having less than 2% influence in program and policy decisions. Conducting a community assessment is part of t he ten essential public health services yet in a recent study of local health departments only 63% of local health departments had conducted a community health assessment in the last three years (NACCHO, 2009,). In the CBPA study of Florida health departme nts, 63% reported conducting a community assessment in the last four years. Missing D ata Prior to analysis data screening was conducted to identify any implausible data and to deal with missing data for this dataset There were 547 responses to the survey of those 250 were not missing responses any items on the survey ( 46%). Frequencies were run to establish how many missing data points there were for each item in the total survey (demographics, survey subscales and naturally skipped items) Specific to t he s urvey scale items (41 items) these items were reviewed to identify missing values which ranged from none missing to all 41 missing The mean number of missing survey items was 9.2 from a 41 item scale. In t he Parker et al. 2003 CBPHI
70 study 36% of the returned surveys contained missing data. In this present CBPA study, 54% were missing at least one item on the 41 item scale. Once the number of missing items on the survey items was identified; a dummy coding procedure was conducted to label those respondents missing none as zero and those missing any survey subscale data as one. In addition, respondents missing no data w ere compared to those missing 10, 15 and 20 and 50% (20 questions) of data across all survey subscale items. To test the differen ces between cases missing and not missing data, Independent T tests were conducted on continuous variables to test mean differences and Chi Square tests on categorical and dichotomous variables. The distribution of missing data was identified as missing at random based on the pattern of missing in which those missing data do not deviate in a systematic way from those without missing data. Missing data is described in one of three possible patterns, missing completely at random (MCAR), missing at random (MAR ) or missing not at random (MNAR). A MNAR pattern is indicated w hen the pattern of missing is not related to other variables being measured but is dependent on the unobserved missing value itself (Fox Wasylyshyn & El -Masri, 2005) Dat a is described as MCAR when the distribution of missing data is unpredictable whereas in MAR the missing data can be predicted from other available data in the dataset. The pattern of missingness can be examined by comparing mean differences between group s missing and not missing data, if the missingness is predicted by items other than the dependent variable it can be termed MAR (Tabacachnick & Fidell, 2007) There is not a clear number that represents excessive missingness and it has been suggested that it is the pattern of missing that is more significant than the amount (Fox Wasylyshyn & El -Masri, 2005; Tabacachnick & Fidell, 2007) Cases missing 0 20% of responses were not statistically different on most items than cases missing greater than 20 on most items ; how ever, these were related to
71 management role. While the literature has not offered clear guidance of an acceptable extent of missing data for MI, and although MI h as been suggested to be sensible and accurate when missing over 50% of data, for this study a more moderate, or prudent approach was taken based on methodology findings in the MI literature Respondents that were missing >20 items (50%) were eliminated from the analysis. Respondents (440) missing 020 items were examined for missing demographic var iables (race, ethnicity, gender, nursing education) that are characteristics important for developing MI models 40 8 cases were not missing those demographic variables and these were the cases kept (see Figure 4 2 for flow chart of participants) Multiple imputation (MI) was conducted to impute item level values using the S tata SE Version 11.0 imputation program This program uses a simulation based repeated sub -sampling procedure to replace missing values with p redicted values. When data are missing, the use of MI preserves data producing more unbiased estimations, builds error into the standard errors resulting in slight increases in standard errors, and reduces bias more so than most other typical approaches for handling missing data in cluding complete case analysis. By minimizing bias, yet retaining a larger sample size, MI is able to increase statistical power without inflating the likelihood of a Type I error. Different than single imputation that adds a random normal residual once; multiple imputation adds more randomness in each imputation. For most data, i t is recommended that five imputations ( M ) are adequate to estimate missing data and for this sample five imputations were conducted (Schafer & Yucel, 2002; Tabacachnick & Fidell, 2007) Stata SE 11.0 MI uses a simulation procedure to generate multiple imputations that includes between imputation variability. The MI procedure can be broken down into three steps: (1) imputations are generated, (2) the chosen analysis is conducted on each imputed dataset (in this
72 case m=five times) and (3) the results from the analyse s are combined into a single analysis result In order to impute a variable that contained missing data, a model containing other predictor variables correlated to that variable as well as variables related to the missing data mechanism w as built to predict the missing value (Donders, van der Heijden, Stijnen, & Moons, 2006) The MI program requires all respondents for whom you are imputing X value all have observed values on the predictors in the imputation model. Due to this requirement, the variable funding for health disparity projects was unable to impute data for 15 cases that were missing data on the predictors used in the MI mod el To include the 15 cases in this model would have required giving up or excluding at least 1 variable that was one of the most relevant predictors of missingness. It was decided to retain the more robust predictor, and lose the relatively insignifican t number of cases (n=15) in order to do so for this particular analysis. Multiple regression was conducted separately for each of the five imputed datasets using Stata SE 11.0, which then combines the results from the five MR analyses into a single resul t Items were imputed for items embedded in each of the following subscales : competency of others, self competency, HD community involvement, importance of CBPA skills In addition, the variables Funding for health disparity programs, PHN experience and age required imputation Prior to MI complete cases were available as follows t he Subscal e HD community involvement (n=274); self competency CBPA (n=377); competency of others (n=353); importance of CBPA (n= 374) ; age (n=391 ) and years of PHN experience ( n=384). After MI was completed the dataset had 408 complete responses for each of these items. Funding for health disparity programs was missing
73 43 responses, n=365. After MI was completed the dataset had 393 complete responses for this item. Analysis T he extent to which PHNs in local health departments in Florida (a) perceive that a CBPA is relevant to reducing health disparities, (b) are educationally prepared to implement a CBPA, (c) perceive that they are competent to implement a CBPA, (d) perceive that their respective health departments are incorporating a CBPA into practice, and (e) spend time engaging in a CBPA in their practice settings was examined to measure factors thought to be associated with adopting CBPA into practice PHNs were asked how i mportant overall is CBPA to decreasing health disparities, the mean of 1.82 (0.86) translates to between vital to very important to decreasing health disparities (s ee Table 4 5 ). Although participants perceived CBPA as very important, they reported not bei ng competent in CBPA skills the average competency sum fell almost midpoint of the possible range of 7 42, M= 19.97 (6.4) More than half (57.6%) of the participants reported their highest level of nursing education at a bachelor degree or higher therefo re a large portion of the sample are educationally prepared to engage in CBPA. Even though a large portion should have the skills to participate in CBPA, respondents reported spending about 25 50% on their time on average engaging in CBPA W hile they also reported spending a great deal (76 99%) to moderate (51 75%) amount of time in care focused on individuals, This is congruent with the FCN report of PHNs spending 75% of their time in individual work as opposed to population based service (FCN, 2009) This also supports the findings of a study of California PHNs that reported PHNs were more likely to perform individual le vel interventions than community or systems level interventions (Grumbach et. al. 2004).
74 A series of 19 questions were asked about t he health department involvement of community in assessment and planning activities this subscale overall reported that the HD spent some to a good bit of the time involving community in assessment and planning activities, Median = 63, M 63.06 (19.90) (possible range 19 114) The most frequent community involvement task reported was in answer to Has your health departments c ommunity assessment tried to identify the strengths and weaknesses of the community it serves falling between a good bit of the time and most of the time. Slightly more than half reported that community members were asked their opinion about their communi ty most of the time. This is similar to a 2008 NACCHO report that 50% of local health departments reported engaging community members to discuss unmet healthcare services. The lowest performed task was to the question Were you as an employee asked for your perceptions concerning the health status of the communities your serve falling between a little of the time and some of the time. Thus there is room for improvement in involving not only th e community but the public health staff in assessment and planning activities. With Florida s high proportion of minority and poverty rates, intractable health disparities continue and individual level interventions have shown to be ineffective necessitati ng consideration of new ways to improve the health of our communities. In part I of the study, through exploratory factor analysis, four subscales were identified in exploratory factor analysis which were importance of nursing competency in CBPA skills to decreasing health disparities (Importance skills -eight items ), rating of self competency in CBPA skills (Self competency CBPA -seven items ), rating of other nurses in health department in CBPA skills (Other competency CBPA -seven items ) and Health department of involving community in assessment and planning (HD community involvement -nineteen items ).These subscales were
75 summed and the mean and standard deviation calculated (see Table 4 6 ). The subscales were moderately correlated with each other ranging from r =.14 .58, p<.01 (see Table 4 7). To determine the relationship between what work and personal factors influence PHN self competency, HD involvement with community and time spent in CBPA, t hree multiple regressions were conducted In this exploratory study without a theoretical reason to enter variables in a particular order all variables were entered at one time using a standard multiple regression approach. The three regression s sought to meet aim 3 of this study which examined the relationship between: P erceptions of self competency in CBPA skills and (a) personal factors, (educational preparation, PHN experience, age, race perceived importance of CBPA skills to reducing health disparities ), work factors ( region served, job role, competency of other PH Ns at health department, percent of work time spent in CBPA ) Perceptions of health department involvement of community in assessment and planning and a) personal fa ctors, ( PHN experience, race perceived importance of CBPA skills to reducing health disp arities and perceived self competency in CBPA skills ), work factors (region served, job role, competency of other PHNs at health department, funding support for health disparity reduction programs ) Time spent in CBPA and a) personal fa ctors, ( PHN experience, race, perceived importance of CBPA skills to reducing health disparities and perceived self competency in CBPA skills ), work factors ( region served, job role, competency of other PHNs at health department, funding support for health disparity reduction programs ) Regression One The first regression entered the following predictors measured at the interval level ( time spent in CBPA, importance of CBPA subscale, age and PHN experience, ) and dichotomous level ( job role, race, region of Florida se rved and education). The dependent variable self competency in CBPA was measured at the interval level (see Table 4 8) This sample n=408 has no missing data after MI A moderate amount of variance (46%) in self competency was accounted for by this set of independent variables (Adjusted R2=.46) Table 4 9 displays the results of the regression analysis
76 for self competency regressed on time spent in CBPA, importance of CBPA, age, PHN experience, job role, region and education. A significant relationship was found between self competency, F (12, 375.7) =25.5, p<.001 and time spent in CBPA (b=0.82, p<.001), the importance of CBPA (b=0.21, p<.001), the competency of other PHNs at health department in CBPA (b=.48, p<.001) and job role (b= 1.3, p<.05). The greater time spent engaging in CBPA, the more important CBPA was perceived to decreasing health disparities the more competent your PHN coworkers were and if you were in a staff position was related to higher self competency scores. Being in a managem ent/director/other job role was related to lower levels of self competency. This finding is further discussed in C hapter 5. Regression Two The second regression entered the following predictors measured at the interval level (importance of skills subscale, other competency CBPA, competency of self, PHN experience) and at a dichotomous level (job role, race region of Florida and funding for health disparity programs). The dependent variable HD community involvement was measured at the interval level. The var iable funding for health disparity project is missing 15 cases that were not able to be imputed, the sample for this regression after MI is n=3 93. These regression results are summarized in Table 4 10. A significant relationship was found between HD community involvement F( 11,371.5 )= 6.89, p<.001 A d justed R2 = .1 5 and three of the eight predictors (competency of others, job role and yrs of PHN experience) Respondents who perceived that coworkers were competent reported higher scores in HD community involvement (b=0.87, p<.001) Being in a manager/director/other role was significantly associated with lower reports of HD community involvement than those in a staff position In total a small amount (1 5 %) of
77 the variance in HD community involvement was explained by these variables. Respondents that reported more PHN experience reported lower score in HD community involvement. Regression Three The third regression entered these predictors measured at the interval level (importance of skills subscale, other competency in CBPA, self competency CBPA PHN experience) and at a dichotomous level (f unding for health disparities job role race, and region ). The dependent variable was t he percentage of time spent engaged in CBPA in practice The variable funding for health disparity project is missing 15 cases after multiple imputation the sample for this regression after MI is n= 393. These regression results are summarized in Table 4 1 2. A significant relationship was found between time spent in CBPA F ( 11,343. 1 ) = 3. 62, p <.001, and one of the predictors. A higher degree of s elf competency in CBPA was associated with a greater amount of time in CBPA. This relationship is consistent with the findings in regression one between time spent in CBPA and self competency Funding for health disparities programs was not associated with the amount of time spent in CBPA challengin g the thought that increased calls from funding bodies for inclusion of the community in programs will increase the time spent in CBPA. This model accounted for a small amount ( 8 .4%) of the variance in time spent in CBPA (Adjusted R2 =.0 8 4 ).
78 Table 4 1 Weekly response to s urvey (Florida Survey) Week Action Day Number Responses Weekly Responses Cumulative number R esponses % of total Responses 1 Survey sent 1 94 180 180 33 2 58 3 6 2 7 26 2 PHN Reminder Day 8 8 48 127 307 56 9 57 10 13 15 14 7 3 PHN Reminder Day 15 15 4 74 381 70 16 46 17 11 20 7 21 6 4 DON Reminder Day 27 22 26 13 53 434 79 27 25 28 15 5 29 8 17 451 82 30 35 9 6 PHN Reminder Day 42 38 41 11 20 471 86 42 9 7 43 39 61 532 97 45 10 46 2 47 49 10 8 50 59 15 15 547 100
79 Table 4 2 Survey race d emographic Figure 4 1 Exploratory f actor a nalysis: s cree p lot % Race/ethnicity White Black Asian Am Indian/Alaska Native Nat Hawaiian /PI Other Hispanic Source Part I Nat l sample achieved Part I CBPA study 90 6.0 1.3 2.4 0 0 3.0 Survey Part I US RN Working in public health 94 4.3 1.4 0.4 0.25 0 2.0 USDHHS HRSA NSSRN 2004 Part II FL Study Part II 81.4 15 1.7 0.5 0 1.5 3.4 Survey part II FL PHN 76.6 20.6 2.2 0.4 0.2 1.0 5.2 Grant, M. FL DOH 2007 FL RN 72.0 11.6 6.2 0.2 0 1.5 7.8 FCN 2009
80 Table 4 3 Factor l oadings t able for p rincipal a xis f actoring with p romax r otation Scale H D involvement in Community Importance of CBPA Competency of other PHNs at HD in CBPA Competency of self in CBPA Cronbachs alpha 0.96 0.94 0.92 0.88 Number of items 19 8 7 7 Health Department involvement in community assessment/planning Employee input 0.55 Community members input 0.8 Identify the strengths and weaknesses 0.82 Presented the findings to all employees 0.8 Presented the findings to community members 0.85 Comm. assessment data used in program decisions 0.87 Target population asked for their opinions 0.88 Community leaders asked opinions about community 0.89 Community organizations asked about community 0.89 Programs address problems identified by the community when public health statistics point to different problems 0.74 Used findings community assessment data in your work 0.83 Community assessment data used in health dept program decisions 0.83 Target population opinions in program planning 0.85 Community leaders asked for opinions in program planning 0.86 Community organizations opinions in program planning 0.86 Consult community members prior to program implementation 0.67 Network about health dept activities with other local agencies 0.7
81 Table 4 3 Continued. Scale H D involvement in Community Importan ce of CBPA Competency of other PHNs at HD in CBPA Competency of self in CBPA Health dept jointly plan program with other agencies 0.67 Health dept share resources with other agencies 0.53 Competency of self in:CBPA Community Assessment 0.63 Working with Community groups 0.75 Community organizing 0.78 Program planning 0.85 Advocating for the Community 0.8 Influencing public health policy 0.65 Communicating with minority populations 0.45 Competency of other PHNs at your health department in CBPA Community Assessment 0.76 Working with Community groups 0.77 Community organizing 0.91 Program planning 0.89 Advocating for the Community 0.67 Influencing public health policy 0.81 Communicating with minority populations 0.51 Rate the importance of PHN competency in CBPA: Community Assessment 0.83 Working with Community groups 0.89 Community organizing 0.89 Program planning 0.89 Advocating for the Community 0.88 Influencing public health policy 0.84 Communicating with minority populations 0.78 How important is CBPA to reducing health disparities 0.53
82 Figure 4 2 Flow chart of participants Florida s tudy Florida PHNs 1632 RNs & 163 ARNPs eligible 547 responses 540 cases 250 without missing data Missing 0 20 on survey =440 408 not missing on specific demographics 408 used in analysis 7 responses worked CMS ineligible 10 emails returned to sender
Table 4 4 Demographics of Florida s ample Characteristic N M (SD) Age 408 54 (8.3) 20 50 127 (31.1%) 51 65 265 (65%) 66 > 16 (3.9%) Gender 408 Male 11 (3%) Female 397 (97%) Number of yrs PHN experience 15.7(9.2) 0 10 139 (34.1%) 11 20 158 (38.7%) 21> 111 (27.2%) Highest level nursing education 408 Diploma 39 (9.6%) Associates 134 (32.8%) Bachelors 159 (39%) Masters 73 (17.9%) Doctorate 3 (0.7%) Job role 408 Staff 175 (42.9%) Manager 113 (27.7%) Director 31 (7.6) Other 89 (21.8%) Region served 408 West 52 (12.7%) North 116 (28.4%) Central 123 (30.1%) South 117 (28.7%) Time spent in CBPA 408 100% 8 (2%) 76 99% 44 (10.8%) 51 75% 76 (18.6%) 25 50% 112 (27.5%) 1 24% 121 (29.7%) 0% 47 (11.5%)
Table 4 5. Measure of e xtent A im 2 Item N No. items Possible score Actual score M (SD) CBPA is relevant to decreasing health disparities 390 1* 1 6 1 6 1.82 (0.86) Educationally prepared at bachelor degree or higher 408 1 57.6% (253) Self competency in CBPA 408 7*** 7 42 7 42 19.97 (6.37) HD is incorporating CBPA into practice 408 19 19 114 19 114 63.06 (19.90) Spend time engaging in CBPA 382 1** 1 6 1 6 4.06 (1.25) *6 -point Likert scale 1=vital 6=not important, ** 6 -point Likert 1=all 6=none, ***6 point Likert scale 1=excellent 6=very poor Table 4 6. Survey s ubscales (Florida sample) Subscales No. Items Possible score range Actual score range Median M (SD) Cronbachs alpha Importance skills 8* 8 48 8 41 14 16.11 (7.42) 0.9 5 Self competency CBPA 7** 7 42 7 42 19 19.97 (6.37) 0.8 9 Other competency CBPA 7** 7 42 7 42 21 21.08 (6.39) 0.9 5 HD community involvement 19** 19 114 19 114 63 63.06 (19.91) 0.9 6 *6 -point Likert scale 1=vital 6=not important, ** 6 -point Likert 1=all 6=none, ***6 point Likert scale 1=excellent 6=very poor
Table 4 7. Correlations o f C BPA subscales Competency HD performance Self skills Other skills Importance skills HD 1 29 33 14 Self 29 1 58 41 Other 33 58 1 28 Import .14 41 28 1 *P<.01
Table 4 8 Items us ed in m ultiple r egression Number of items Scale Possible score Time spent in CBPA 1 6 point L ikert scale Importance of skills subscale is a sum of an 8 6 point L ikert scale 8 48 Self competency in CBPA 7 6 point L ikert scale 7 42 HD community involvement was measured as a sum on a 19 item scale (range 19 114) 19 6 point L ikert scale 19 114 Other competency CBPA is sum 7 6 point Likert scale 7 42 Education 1 Categorical r ecoded to dichotomous 1 baccalaureate or higher 0=less than baccalaureate PHN experience 1 Interval 0 Job role 1 Dichotomous 1=management or other 0 staff Race 1 Categorical recoded to dichotomous 1=white 0=minority Region 3 Categorical recoded to dichotomous dummy coded Funding for health disparity projects 1 Dichotomous 0=no 1=yes
Table 4 9 Regression 1 : Predictors of s elf c ompetency in CBPA s kills Variable p value SE 95% CI Constant 3.40 2.12 [0.78,7.58] Time spent CBPA 0.8 2 <.001 0.21 [0.40,1.23] Sum Importance skills 0.2 1 <.001 0.03 [0.14,0.28] Bachelor degree or less 0. 12 0.52 [ 1.14,.0.90] PHN experience 0.0 5 0.04 [ 0.13,0.02] Job role (staff/other) 1. 34 .01 0.51 [ 2.35,0.33] Race 2 0.33 0.71 [ 1.73,1.07] 3 0.39 1.30 [ 2.93,2.17] Age 0. 03 0.035 [ 0.06,0.10] Region of Florida served 2 0.16 0.81 [ 1.76,1.44] 3 0.27 0.81 [ 1.85,1.44] 4 0.22 0.84 [ 1.86,1.42] Sum Other competency CBPA 0. 48 <.001 0.04 [.41,.56] Adjusted R 2 .466 F (12,375.7) 25.5 <.001 Dependent variable Self competency in CBPA
Table 4 10. Regression 2 : Predictors o f h ealth d epartment involvement of c ommunity in a ssessment and p lanning Variable p value Standard Error 95% CI Constant 48.82 5. 32 [ 38.31 59.32 ] Funding for disparity 3.07 2.0 4 [ 7.09 0.95 ] Sum Importance Skills 0 .1 0 0.1 4 [ 0.1 9 ,0. 38 ] PHN experience 0. 2 6 0.02 9 0.12 [ 0.50, 0.0 3 ] Job Role 5. 38 0.00 8 2.00 [ 9. 33 1. 43 ] Race 2 2. 68 2. 68 [ 2. 5 8,7. 95 ] 3 2. 96 4.9 4 [ 6. 76 ,12. 69 ] Region of Florida served 2 1. 3 2 3.2 1 [ 7. 6 3 ,4.9 9 ] 3 3. 40 3.1 7 [ 9. 64 ,2. 83 ] 4 2. 39 3.2 2 [ 8.7 2 ,3. 93 ] Sum Other competency CBPA 0.87 <.001 0.1 9 [0. 50 ,1. 24 ] Sum Self Competency CBPA 0 2 5 0.2 0 [ 0.1 4 ,0.6 4 ] Adjusted R 2 0. 151 F (1 1, 3 7 1.5 ) 6.89 <.001 Dependent variable HD community involvement
Table 4 1 1 Correlations Item Correlation Time CBPA Funding Staff/other Yrs PHN Self Competency Importance CBPA Competency others Time in CBPA 1 .06 .10 .001 .29** .13** .19** Funding 1 .07 .02 .06 .03 .01 Staff/Other 1 .25** .15** .10* .02 Yrs PHN 1 .03 .02 .10 ** Self Competency 1 .41** .58** Importance CBPA skills 1 .27** Competency other CBPA 1 p<.05 **p<. .01 Table 4 1 2 Regression 3 : P redictors of time spent in CBPA Variable p value SE 95% CI Constant 3. 127 0.338 [ 2.46,3.79 ] Funding for disparity 0.158 0. 147 [ 0.45,0.14 ] Sum Importance of CBPA 0. 0012 0.009 [ 0.02.0.02] PHN experience 0.0017 0.007 [ 0.01 3 ,0.01 6 ] Job Role 0.04 59 0.137 [ 0.2 2 ,0.3 2 ] Race 0.2 20 0.175 [ 0.563,0.124 ] 0.35 5 0.360 [ 1.0 74,0.363 ] Region of Florida served 0.0 05 0.2 09 [ 0.4 16 ,0. 406 ] 0. 203 0.206 [ 0.6 07,0.201 ] 0. 164 0.21 5 [ 0.589,0.260 ] Sum Competency of other PHNs in CBPA 0.003 7 0.01 2 [ 0.0 20 ,0.02 8 ] Sum Competency of self 0.0 526 <.001 0.01 3 [0.02 7 ,0.07 9 ] Adjusted R 2 0.0 84 F (11, 3 43.1 ) 3. 62 <.001 Dependent variable time spent in CBPA
CHAPTER 5 DISCUSSION AND RECOMMENDATIONS This chapter discusses the study findings and presents how these findings relate to previous research. The theoretical framework, Diffusion of Innovation is discusse d as it relates to these findings. Future areas of research, implications to practice are offered and limitations are acknowledged. Overview of Study The primary purpose of this study was to examine factors related to adopting a CBPA to practice among P ubli c H ealth N urse s (PHN) in health departments throughout Florida Diff usion of Innovation theory was used as a guide to examine the extent health departments in Florida are adapting a CBPA approach to practice. By applying the principles of CBPR as an approach to practice, CBPA has been called for by public health experts and organizations (Israel, et al., 1998; Savage, et al., 2006; Wallerstein & Duran, 2006) however c urrent public health practice continues to overwhelmingly focus on individuals To understand more fully what factors influence adoption of CBPA, this study examined personal and work factors that could impact health departments implementing a CBPA approach to practice. Major Findings The data suggests that time spent engaging in C BPA, the competency of other PHN coworkers and how important PHNs perceived that competency in CBPA skills are to reducing health disparities were related to their level of perceived self competency in CBPA skills It is not surprising that the more time a nurse spends working in CBPA the greater level of competency reported or the less time spent practicing a skill results in less competency Studies have shown that a supportive environment which allows time to practice and develop skills builds competenc e (L indberg, 2006). Likewise nurses who are novice (in CBPA) can be
mentored by working with other nurses competent in CBPA. In fact, Kramer & Schmalenberg (2004) noted that the ability to work with other competent nurses fostered a positive work environm ent and ultimately improved care in their study of Magnet hospitals. Further whether the respondent was in a staff or management position was related to perceived self competency in CBPA with those in management positions reporting they were less c ompete nt in CBPA skills. Th e management group included managers, directors and others. A portion 6% (n=24) of this group identified as others were Nurse Practitioner (NP) which could imply that both their education and current job duties are clinic all y focused on care at the individual level only, and signify why the management group reported less competency related to CBPA. The regression was rerun excluding those reporting that they were a N urse Practitioner; the results were the same as those in a management role felt less competent in the skills associated with CBPA. This finding is concerning given that leadership makes the policy and practice decisions and sets the direction of practice yet they are reporting less self competen cy than staff nurses in skills such as working with communities, community organizing, program planning, and influencing public health policy. This warrants further investigatio n of public health nursing leadership at local health departments to determine (1) how does this lea dership competency level impact current initiatives by public health to incorporate more community based practice and (2) what continuing education is required to augment these skills in public health leadership Weiss, Anderson and Lasker (2002) reporte d that successful community partnerships were related to effective leadership. Although competent leadership is necessary ingredient for community partnerships, the r esults of this present study suggest that leaders do not feel adequately prepared in skill s associated with working with communities To address this
leadership training programs such as the Northeast Public Health Leadership Institute (NEPHLI) have been developed. Previous studies have documented improved skills such as working with communiti es to address public health issues following attendance at NEPHLI continuing education programs (Saleh, Williams, Balougan, 2004). As noted additional to job role, the amount of time spent engaging in CBPA and the competency of PHN coworkers were predictor s of self competency This is understandable in that the more time spent working with the community and working with other nurses who are competent that a respondent would feel more confident in CBPA skills by gaining valuable application experiences and mentorship This finding supports that of Grumbach et al (2004) who found PHNs who spent more time in community and systems level practice felt more confident in their educational preparation to work with individuals, communities and systems. And lastly, t he more that a respondent thought that CBPA skills were important to decreasing health disparities the higher the level of self competency reported in CBPA. If you value something, if you think it is important then you are more apt to engage in the activity (Rogers, 2003) It is interesting that educational preparation and PHN experience was not related to self competency. If the skills needed for CBPA are acquired in b ac ca laureate or higher nursing education or in on the -job training as a public health nurse it was expected that the relationships between education level/PHN experience and self competency to be significant. Public health nursing is only taught in baccalaurea te or higher nursing program s hence it was expected that nurses prepared at that level would feel more competent in CBPA. As the largest portion of the public health workforce it is essential that PHN are competent for CBPA adoption to be successful. Ther efore, in order for public health nursing to feel confident in their competency in
working with communities, they will need continuing education to grasp the value and obtain the necessary skills to work with communities opportunities to practice these sk ills and to work with other competent nurses that can serve as preceptors and mentors Nineteen questions comprised a subscale that questioned the HD involve ment of community in program planning and community assessment. A small amount (15%) of the varian ce in HD involvement was explained by three of the predictors. Coworkers competency in CBPA predicted higher levels of CBPA, while respondents who reported more PHN experience reported lower score in HD community involvement. This supports findings by Zahner & Gredig (2005a ) that nurses with long term public health experience grapple more with the shift in practice paradigm and identified the need for continuing education for PHNs in program planning, building coalitions, and community assessment. Being in a manager/director/other role was significantly associated with lower reports of HD community involvement than those in a staff position. In total a small amount (1 5 %) of the variance in HD community involvement was explained by these variables. As stated earlier, CBPR is being used increasingly as an effective method for conducting community research and organizations such as the CDC (CDC, 200 9 ) have increased funding for programs aimed at partnering with communities to reduce health disparities. In this study over half (57.4 %) reported that the health department where they worked received funding for health disparity projects. The funding was reported coming from Florida Closing the Gap, CDC, Migrant health, HIV/AIDS and other state and federal programs. The amount of funding was not asked so what percentage of the budget is funding for hea lth disparity programs is not know n The participants reported that policy and program decisions (75%) were most influenced by funding sources and the Florida State Department of Health. Community residents were reported
as having less than 2% influence in program and policy decisions. This is direct ly opposit e to the objective from public health standards and experts to involve community organizations and community members in assessment and planning activities. Therefore these findings offer guidance to f uture research to examine how finance and organizational structure influences the decisions and directions of public health practice. Health departments frequenc ies of conducting community health assessment w ere reported as between 1 2 yrs (29%), 34 yrs (33%) 5 8yrs (26%) 9 11 yrs (5%), 12 or greater (2%), and never (5%). It is vital to conduct an assessment to enable the health department to evaluate the strengths and weaknesses in the community with the community input. The findings from community ass essment should be the basis for prioritizing and planning strategies for public health practice ( Keller et al. 2002) It is promising to see that 62% of the health departments reported conducting a community assessment every 1 4 yrs. This is similar to the findings in a 2008 NACCHO surve y of local health departments that found that 63% of local health departments had conducted a community assessment in the last three years (NACCHO, 2009). It is concerning to find that according to respondents, that 7% of health departments conduct a community assessment every 9 yrs or greater with 5% never conducting a community assessment. This may reflect that although a community assessment was done, PHNs were not included in the planning or feedback so they were unaware of when a survey is conducted at their health department. Applying a CBPA to practice requires that public health workers have the skills to develop the capacity of community members to serve in partnership endeavors, value the role of community participation, and mobilize community resources to address community defined priorities (Israel, et al., 2003) Over half the sample was prepared at the baccalaureate or higher
level yet overall Florida PHNs perceived that they are only moderately competent to implement a CBPA requiring skills in community as sessment, working with community groups, community organizing, program planning, advocating for the community, influencing public policy, communicating with minority populations. This result is compatible with earlier findings, that PHNs, regardless of edu cational preparation, report their current practice is at the individual level and that they do not feel prepared to practice effectively at the community level (Grumbach, et al., 2004; Jakeway, et al., 2006) The basic skills associated with a CBPA to practice are only taught at the baccalaureate level or higher in nursing education. Although m ore than half of the participants reported their highest level of nursing education was at a baccalaureate level (57% ) or higher there were still 43 % reporting an Associa te degree or less (see Table 4 4). In Florida, educational preparation that would include CBPA skills has been attained by a little over half of the PHNs. While 65% of the management group was prepared at the baccalaureate level or higher they felt less co mpetent in CBPA skills than staff PHNs of whom less than half held a baccalaureate degree (47% ). This finding shows that leadership with more experience in public health nursing and higher education are still not prepared in CBPA. Clearly more than educational preparation is influencing perceived competency in CBPA skills. Future training and workforce development efforts in CBPA must include all public health nursing staff. Recruitment and Response Rates as Research Barriers in Nursing Studies Aim 1 of this study sought to further develop a more comprehensive, valid and reliable measure of the perceptions and knowledge of CBPA and the degree to which it has been adopted into practice by building upon an existing, recently developed measure of community involvement in public health practice. Initially, a paper and pencil mode of survey delivery was used which failed to achieve an adequate sample this method was changed to an online web
based survey. As a web based survey, a national sample adequate for the planned statistical analysis was achieved (n=547). The initial survey in paper and pencil could have been mailed to a distribution list of PHNs as an alternative to a web based approach. Mail would involve additional cost to the investigator, addit ional burden of data entry with possible data entry error but has been reported to have higher response rates than web based delivery (Kramer, et al., 2009; Shannon & Bradshaw, 2002). Although the sample was adequate for the planned exploratory factor analysis, it was not a random selection from a national sample of public health nurses but a convenience sample obtained through the American Public Health Association public health nursing section listserv and contacts with Alaska and North Carolina S tate Directors of Nursing therefore the results are not genera lizable beyond this sample All Florida public health nurses working at local health departments n=1785 were invited via email to participate in th e Florida survey. To encourage PHNs to complete the survey several techniques identified in the literature were used such a s pre -notification, frequent friendly reminders which included a statement that others had responded, and offer s to share survey results (Dillman, 2007; Edwards et al., 2009) The Florida Department of Health Office of Pub lic Health Nursing was instrumental in accessing the PHNs sample. The Office of Public Health Nursing used their PHN distribution list to send the original participant letter with link to the survey and several reminders via email. The Office of Public Health not only was helpful in sending emails to the nurses but also added their endorsement in the emails and encouraged nurses to respond. Several nurses sent questions about the survey to DOH who was quick to forward the question t o the Principal Investigator to aid in a quick reply. All PHNs at Florida local health departments were sent the CBPA survey and 30. 6 % responded. This response rate is slightly lower than a meta analysis of 68 survey studies
reported mean response rate of 39.6% (Cook, et al., 2000) Reviewing studies of nursing in the last three years, t he response rates for paper and electronic surveys has varied from 12 % (Doorenbo s et al., 2008) to 58% (Adams & Barron, 2009) for electronic surveys. Recent surveys in Florida have reported difficulty in achieving high response rates. A recent multistate mail survey of nurses the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS), which included Florida nurses reported a 3 9 % response rate (Kutney -Lee et al., 2009) The Florida Center for Nursing (FCN) surveyed the Florida nursing workforce in 2004 and 2007 wi th response rates less than 9 % in both surveys. Following these surveys, FCN partnered with the Florida Board of Nursing (FBON) to include the 2008 survey as part of the online license renewal. This timing was fortuitous because this partnership came when FBON changed their method of licensure renewal to one that was easier to complete online increasing substantially the nu mber of nurses that renewed online O nly after FCN includ ed the 2008 survey as part of online renewal were t hey able to reach a 92% response rate. A 30.6% response rate in this study is similar yet slightly lower than recent surveys of nurses. The follow ing factors impacting response rate in surveys were consider ed Timing, salience, survey method and lack of incentive s are possible reasons that a higher response rate was not achieved. The CBPA survey was sent via email on September 2, 20 09 and was activ e for 8 weeks. During this time period nurses at local health departments were involved in vaccination projects during a H1N1 flu pandemic that was consuming much of their time and could impact their ability to participate in the survey (Trossman, 2009) Although this present survey had the endorsement of the Office of Public Health Nursing prior to sending it out perhaps in the future timing should be considered more closely prior to sending out the survey. Another factor that could have impacted response rate is the degree that PHNs were interested in the s urvey topic.
Salience of the topic impacts response rate with topics of more interest garnering higher response rates (Dillman, 2007) Although this topic is abo ut public health nursing some PHNs responses included comments about not knowing all the answers since they are an average PHN and this survey should exclude direct care providers since my involvement with the subject matter is limited these comments r eveal at some health departments PHNs are not included in program assessment and planning endeavors Additionally, t his was a web based survey that used work email addresses to contact respondents. A kl, Maroun, Klocke, Montori, and Schunemann (2005) found that use of an institutional email account resulted in a lower response rate than a commercial email in medical residents and non respondents reported that the institutional account was not checked regularly. In a telephone follow up to a large county health department nursing director, the researcher was informed that nurses mostly work in clinics and only get to check their email weekly This would limit their time to participate in this survey. Finally, i ncentives are a well documented way to increas e response rates (Dillman, 2007) Non monetary monetary or lottery incentives were considered but funding for this dissertation was limited and a lottery for in centive was not supported by IRB. Whereas better timing, greater topic salience, more accessible survey method, and incentive could have increased response rate the response rate achieved is comparable to other recently published rates. While in future st udies concurrent events such as flu pandemic might be avoided this is a potential occurrence in research. Perhaps greater pre survey contact with PHNs to discuss the upcoming survey along with further description in the survey letter would create more interest in the topic to PHNs. Given the comparable response rates in the HCAHPS survey, the difficulty to obtain PHNs personal email accounts, it is not recommended to change the survey method. Above all to increase response rates, future research should incl ude incentives as funding allows.
Strengths and Limitations Strengths This study has several strengths to identify. The purpose of this study, to better understand factors influencing local health departments adoption of CBPA, is both important and timely in respect to the call for a focus on health using an ecological approach, integration of research and practice more community involvement and the widening gap in health disparities. Current practice in public health nursing was assessed from the perspect ive of both staff and leadership which allowed comparison of these viewpoints. Prior to use in the Florida study, the adapted tool was tested in preliminary work using a national sample. This national study achieved a large sample in part I for the analysi s planned to use exploratory factor analysis to understand the underlying dimensions of this scale. The results of this preliminary national survey psychometric testing show the tool is a valid and reliable tool to use in the Florida study I n part II this study achieved a good response rate for a web based survey in this population. All PHNs in Florida had the opportunity to participate in the survey. This study provides an initial step forward in advancing our understanding about local health departments use of a CBPA to practice in the State of Florida More specifically, this study identifies current competency and points to the inclusion of leadership and staff in future CBPA training efforts. Limitations This study relied on the self report of PHNs and their perceptions of current practice environment. The weakness of the self report survey is that actual observation was not done but rel ied on perceptions of PHNs The PHNs responding might not have been fully aware of the health departments inclusion of community in assessment and planning activities. Future studies could include observations to quantify the type of community involvement in public health nurse pr actice. This study sample part II was specific to Florida PHNs and is not generalizable to other
states The national sample in part I was not a random sample but a convenience sample of available PHNs to participate in the survey. In this study it was not possible to compare the data at multiple levels. In future data could be collected at the health department, region and perhaps state level. This would allow a comparison of the data at each level to identify differences at each level. Additionally collec ting the size of the population served by each health department would allow for analysis of differences dependent on health department size. Directions for Future Research Findings from this study begin to create a better understanding of current PHN beli efs and practices related to u sing a CBPA in Florida This work provides a foundation for future studies that will inform the effective translation of a CBPR approach into PHN practice, which aims to meet the Healthy People 202 0 goal of eliminating health disparities. There are varied directions that future research can use to approach understanding the adoption of a CBPA to practice at local health departments Results of the present study suggest that educational preparation at the baccalaureate level alo ne was not a predictor of self competency in CBPA This finding supports that of Parker et al. 2003 in which only a modest gain in competency was seen after training. Parker suggests that effective translation of CBPA to practice demands the examination of organizational factors related to the likelihood of adopting it (Parker, et al., 2003) First, o ne approach is to identify a health department that has successfully transitioned from a public health model that does work for the community to a model that does work with the community. Investigate what processes i nfluenced their adoption of CBPA Rogers describes certain attributes of the innovation that can affect adoption such as relative advantage, compatibility, complexity, trialability, and observability Future studies should explore w hat best practice s exist within health departments, and what role these attributes played in their adoption of CBPA. Findings could be discussed with other health departments to better understand their
local context as it relates to a CBPA in practice In the end local health dep artments can strategize to foster implementation at their organization such as promoting the relative advantages of using CBPA by promoting the outcomes achieved when working with the community In addition, i n this study the respondents reported that 44% of the program and policy decisions are influenced by funding bodies and 38% by s tate DOH Along with proximal and distal outcomes achieved, cost data should be inclu ded in a study of health departments that h ave adopted CBPA Although cost data can be useful in decision making (Polit & Beck, 2008) systematic reviews of CBPR studies were examined and cost data was missing (Cook, 2008; Merzel & D'Afflitti, 2003; Viswanathan, et al., 2004) Supplying decision makers with information on outcomes, best practices and costs from health departments who successfully transitioned to a community based participatory practice model could aid in persu asion to adopt CBPA. Organizations such as health departments can learn from each other, Lasker and Weiss (2003) formed a work group to examine community projects from several Turning Pont initiatives, they found that effective community partnerships can be achieved but also commun ities can benefit by learning specific strategies that strengthen collaborations to be successful in community lead problem solving. Another direction that future research must take is to seek to understand further who and what factors are the drivers of public health at the health department level This present study examined work and personal factors related to the adoption of CBPA. When the findings from practice are compared to the ideal put forth by standards and experts, findings s uggest that improvements are needed. Thi s finding supports that of Smith, Minyard, Parker, Van Valkenburg and Schoemaker (2007) that found the public health in Georgia was not aligned with the e ssential s ervices of public health practice given that direc t medical services was the dominant
service. T hey found that the primary drivers of public health in Georgia are not community need but finance related. Adoption of CBPA is influenced by the system wide drivers of public health in Florida. Further e xamina tion of the drivers of public health practice from a systems level is needed Implications for Practice Study findings indicate that PHNs value community based participatory public health practice and think it is important to reducing health disparities ye t do not feel they are adequately prepared in the skills associated with CBPA. This study examined the current beliefs and practices of PHNs with respect to the use of CBPA in health departments throughout Florida The factors associated with whether healt h departments adop t a CBPA are complex, but for it to happen will require workforce competency (staff and management), and leadership, financial and organizati onal support. As suggested by Parket et al (2003 ) this tool could be useful measures of both workforce capacity and level of health departments investment in community based pr actice Local health departments can use this survey or the subscales to assist in monitoring the amount of community involvement and the competency of staff in skills assoc iated with community based participatory practice. From these findings health departments can explore the incongruence between the call for population based practice and the current practice today. Although local health department could use this tool or se parate subscales in development of training and performance tools for health department to facilitate the t ranslation of CBPA into practice what is also needed is a greater understanding how funding sources and financial polic ies in public health impact C BPA adoption at local health department level. Summary The majority of this sample was white, female, over 50 yrs old, PHNs who have a baccalaureate or higher level of education. Most PHNs reported feeling moderately competent at
CBPA skills yet thought CB PA skills were very important to vital decreasing health disparities. The amount of current community level involvement in CBPA activities was reported limited c onsistent with observed trend in public health over the past two decades. PHN time was describ ed spending a greater amount of their time in care of individuals than working with communities. Multiple regression analysis was used to examine the relationship between work and personal factors and self competency in CBPA, Health department involvement of community and time spent engaging in CBPA. Self competency was associated with time spent in CBPA, competency of PHN coworkers and perceived importance of CBPA. In addition PHNs in job role s at the administrative or supervisory levels perceived themselv es as less competent at practicing using a CBPA than their staff counterparts. The health department involvement of community was associated with competency of PHN coworkers, amount of PHN experience and job role in that those who perceived there coworkers are competent in CBPA report higher levels of HD community involvement. Managers/directors report lower levels of community involvement while having experience in public health nursing was associated with lower levels of HD community involvement. Although over the last one hundred years Americans are healthier and living longer with many of the improvements credited to public health, with these gains Americans still fall behind other industrialized nations in he alth indicators such as age expectancy and infant mortality rates. Even more concerning is a persistent and widening gap in health between racial ethnic and socioeconomic groups. In the next hundred years future health improvements will require a broader ecological view of health which encompasses the social determinants of health (Turnock, 2004) To eliminate hea lth disparities and meet the challenges of attaining and maintain ing healthy communities the manner of public health practice will need to change from
the traditional public health model to a new community based participatory model This present study found that Florida local health departments are somewhat engaged with the community in assessment and planning efforts, and moderately competent in CBPA skills. To fully engage community organizations and community members in collaborative equal partnerships i mprovements are needed in the extent of current CBPA practice and the competency level of PHNs. A collaborative participatory partnership model must bring the communit y together to focus on the social determinants that impact current health issues and work together to develop effective multilevel strategies with the end goal of improving health and eliminating health disparities.
APPENDIX A SURVEY
APPENDIX B MULTIPLE IMPUTATION Table B 1 Multiple Imputation MI Data Mean Std. Dev. Actual Data Mean Std. Dev. age_yr 53.72 8.52 age_yr 53.67 8.55 yrs_phn 15.71 9.10 yrs_phn 15.64 9.11 Rass_emp_percep 4.11 1.61 Rass_emp_percep 4.11 1.61 Rass_comm 3.20 1.49 Rass_comm 3.19 1.49 Rass_strweak 2.73 1.36 Rass_strweak 2.72 1.36 Rass_find_emp 3.31 1.62 Rass_find_emp 3.28 1.63 Rass_find_comm 3.10 1.51 Rass_find_comm 3.11 1.52 Rass_data_prog 3.07 1.42 Rass_data_prog 3.08 1.42 Rass_targ 3.31 1.53 Rass_targ 3.33 1.52 Rass_leader 3.26 1.46 Rass_leader 3.23 1.46 Rass_probidcom 3.31 1.39 Rass_probidcom 3.32 1.37 Rplan_asswk 3.75 1.51 Rplan_asswk 3.74 1.52 Rplan_assdec 3.34 1.36 Rplan_assdec 3.35 1.35 Rplan_targ 3.69 1.43 Rplan_targ 3.69 1.43 Rplan_leader 3.54 1.45 Rplan_leader 3.54 1.43 Rplan_comorg 3.57 1.38 Rplan_comorg 3.54 1.38 Rplan_conscom 3.43 1.38 Rplan_conscom 3.47 1.37 Rplan_netwk 2.91 1.36 Rplan_netwk 2.89 1.35 Rplan_joint 2.96 1.28 Rplan_joint 2.95 1.27 Rplan_share 3.30 1.43 Rplan_share 3.28 1.42 Rass_comorg 3.16 1.41 Rass_comorg 3.13 1.42 compsf_ass 3.04 1.12 compsf_ass 3.04 1.12 comsf_comgrp 2.47 1.08 comsf_comgrp 2.47 1.09 comsf_organ 3.16 1.16 comsf_organ 3.16 1.16 comsf_plan 2.84 1.17 comsf_plan 2.83 1.17 comsf_advoc 2.50 1.08 comsf_advoc 2.50 1.09 comsf_policy 3.41 1.22 comsf_policy 3.40 1.22 comsf_minor 2.54 1.07 comsf_minor 2.54 1.07 compoth_ass 3.17 1.06 compoth_ass 3.18 1.06 compoth_comgrp 2.85 1.02 compoth_comgrp 2.85 1.02 compoth_organ 3.09 1.10 compoth_organ 3.10 1.09 compoth_plan 3.09 1.15 compoth_plan 3.10 1.16 compoth_advoc 2.79 1.09 compoth_advoc 2.80 1.09 compoth_policy 3.35 1.21 compoth_policy 3.36 1.20 compoth_minor 2.70 1.07 compoth_minor 2.69 1.06 import_ass 2.12 1.20 import_ass 2.10 1.16 import_comgrp 1.94 1.06 import_comgrp 1.95 1.07 import_organ 2.23 1.17 import_organ 2.22 1.16 import_plan 2.10 1.10 import_plan 2.09 1.09 import_advoc 1.92 1.14 import_advoc 1.86 1.04 import_policy 2.09 1.16 import_policy 2.06 1.11 import_minor 1.80 1.02 import_minor 1.80 1.01 importcbpa_disp 1.90 1.02 importcbpa_disp 1.82 0.86
Table B 1. Continued MI Data Mean Std. Dev. Actual Data Mean Std. Dev. time_cbpa 4.06 1.25 time_cbpa 4.06 1.25 time_cbpa2cat 0.31 0.46 time_cbpa2cat 0.32 0.47 mi_sumHDcommass_plan 63.06 19.91 sumHDcommass_plan 62.38 22.13 mi_sumcompsefl 19.97 6.36 sumcompsefl 20.01 6.54 mi_sumcompoth 21.08 6.38 S umcompoth 21.14 6.67 mi_sumimportCBPA 15.87 7.34 sumimportCBPA 15.79 7.43 mi2_sumimportCBPA 16.11 7.42
L IST OF REFERENCES Adams, S., & Barron, S. (2009). Use of evidence based practice in school nursing: prevalence, associated variables and perceived needs. Worldviews on Evidence Based Nursing, 6(1), 1626. Agency for Healthcare Research and Quality. (2001). Excellence Center s to Eliminate Ethnic/Racial Disparities (EXCEED) Retrieved February 14, 2010, from Agency for Healthcare Research and Quality http://www.ahrq.gov/research/exceed.pdf Agency for Healt hcare Research and Quality. (2005). National Healthcare Disparities Report Retrieved January 17, 2007, from http://www.ahrq.gov/qual/nhdr05/nhdr05.pdf Ahmed, S., Beck, B., Maurana, C., & Newton, G (2004). Overcoming barriers to effective community -based participatory research in US medical schools. Education for Health, 17(2), 141151. Akl, E., Maroun, N., Klocke, R., Montori, V. & Schunemann, H. (2005). Electronic mail was not better than postal mail for surveying residents and faculty. Journal of Clinical Epidemiology 58, 425429. American Nurses Asso ciation. (20 07). Public Health Nursing: Scope and Standards Silver Spring MD: Nursesbooks.org. American Public Health Association. (2004). Poli cy Statement : Support for Community -Based Participatory Research in Public Health. Retreived February 14, 2010, from American Public Health Association. http://www.apha.org/advocacy/policy/policysearch/default.htm?id=1298 Anderson, E. T., & McFarlane, J. M. (2000). Community as partne: T heory and practice in nursing (3rd ed.). Philadelphia: Lippincott Williams & Wilkins. A ssociation of C ommunity H ealth N ursing E ducators (2007). Graduate Education for Advanced Practice Public Health Nursing: At the Crossroads. Retreived February 15, 2010, from http://www.achne.org/files/public/GraduateEducationDocument.pdf Association of Community Health Nursing Educators. (2010). Resea rch Priorities for Public Health Nursing. Public Health Nursing, 27(1), 94 100. Association of State and Territorial Directors of Nu rsing. (2000). Public Health Nursing; A partner for healthy populations Washington DC: American Nurses Publishing. Bekem eier, B. (2008). Upstream nursing practice and research. Applied Nursing Research, 21, 5052.
Britto, M., Pandzik, G., Meeks, C., & Kotagal, U. (2006). Combining evidence and Diffusion of Innovation theory to enhance influenza immunization. Jounal of Quality and Patient Safety, 32(8), 426432. Buhler Wilkerson, K. (1993). Bringing care to the people: Lillian Wald's legacy to public health nursing. American Journal of Public Health, 83(12), 17781786. Burroughs, E., Peck, L. E., Sharpe, P. A., Granner, M. L., Bryant, C. A., & Fields, R. (2006). Using focus groups in the consumer research phase of a social marketing program to promote moderate -intensity physical activity and walking trail use in Sumter C ounty, South Carolina. Preventing of Chronic Disease, 3(1), A08. Butterfield, P. (2002). Upstream reflections on environmental health: An abbreviated history and framework for action. Advanced Nursi ng Science, 25 (1), 32 49. Butterfoss, F. (2006). Proce ss evaluation for community participation. Annual Review Public Health, 27, 323340. Cain, M., & Mittman, R. (2002). Diffusion of Innovation in Health Care ..Retreived February 14, 2010, from California Healthcare Foundation, http://www.chcf.org/documents/healthit/DiffusionofInnovation.pdf Centers for Disease Control and Prevention. (1994). Th e Essential Public Health Services Retrieved July 10, 2008, from Centers for Disease Control and Prevention. Websire: http://www.cdc.gov/od/ocphp/nphpsp/essentialphservices.htm Centers for Disease Cont rol and Prevention.(1997). Princi ples of community engagement. Centers for Disease Control and Prevention. Retreived February 14, 2010, from http://www.cdc.gov/phppo/pce/ Centers for Disease Control and Prevention. (2009). Eliminating Racial & Ethnic Health Disparities, Retreived February 15, 2010 from http://www.cdc.gov/omhd/About/disparities.htm Clougherty, J. E., Levy, J. I., Hynes, H. P., & Spengler, J. D. (2006). A longitudinal analysis of the efficacy of environmental interventions on asthma related quality of life and symptoms among children in urban public housing. Journal of Asthma, 43(5), 335343. Cook, C. Heath, F., & Thompson, R. (2000). A meta analysis of response rates in web or internet based surveys. Educational and Psychological Measurement, 60, 821. Cook, J., O Dickinson, H., & Eccles, M. (2009). Response rates in postal surveys of healthcare pr ofessionals between 1996 and 2005: An observational study. BMC Health Services Research, 9 (160), 1 8.
Cook, W. K. (2008). Integrating research and action: a systematic review of community-based participatory research to address health disparities in environmental and occupational health. Journal of Epidemiology Community Health, 62 668676. Cross, S., B lock, D., Josten, L., Reckinger, D., Olson Keller L., Strohschein, S., et al. (2006). Development of the public health nursing competency instrument. Public Health Nursing, 23(2), 108114. Dearing, J. (2004). Improving the state of health programming by using diffusion theory. Journal of Health Communication, 9, 21 36. Dillman, D. (2007). Mail and Internet Surveys (2nd ed.). Hoboken: John Wiley and Sons. Dirksen, C., Ament, A., & Go, P. (1996). Diffusion of six surgical endoscopic procedures in the Netherlands. Stimulating and restraining factors. Health Policy, 37, 91 104. Donders, A. R., van der Heijden, G., Stijnen, T., & Moons, K. (2006). Review: A gentle introuction to impu tation of missing values. Journal of Clinical Epidemiology, 59, 10871091. Doorenbos, A., Berger, A., Brohard Holbert, C., Eaton, L., Kozachik, S., LoBiondo-Wood, G., Mallory, G, Varricho, C. (2008). 2008 ONS Research Priorities Survey. Oncolcogy Nursin g Society Report, 35(6), 100106. Edwards, P., Clarke, R., DiGuisppi, C., Wentz, R., Kwan, I., Cooper, R., ..Pratrap, S. (2009). Methods to increase response to postal and elctronic questionnaries (reveiw). Cochrane Database of Systematic Reveiws (3). E mmons, K. (2000). Health behaviors in a social context. In L. Berkman & I. Kawachi (Eds.), Social Epidemiology (pp. 242 266). New York: Oxford University Press. Fielding, J. & Kumanyika, S. (2009). Recommendations for the concepts and form of Healthy Peop le 2020. American Journal of Preventive Medicine 37 (3), 255257. Florida Center for Nursing. (2009). An alysis of Workforce Survey Data from the RN Cohorts Renewing in 2008. Orlando: F lorida Center for N ursing Retrieved February 14, 2010, from http://www.flcenterfornursing.org/files/2008_Workforce_Survey.pdf Fox Wasylyshyn, S., & El -Masri, M. (2005). Handling missing data in self report measures. Research in Nursing and Health, 28, 488495. Gany, F. M., Shah, S. M., & Changrani, J. (2006). Ne w York City's immigrant minorities: reducing cancer health disparities. Cancer, 107(8 Suppl), 20712081. Gebbie, K. (2004). Competencyto Curriculum Toolkit: Developing curricula for public health workforce. retreived from Columbia Universit y School of N ursing Center for Health
Policy.Retreived February 14, 2010, from http://www.sphtc.org/Compt_to_Curric_Toolkit.pdf Gebbie, K. M. (2000). The public health work force : enumeration 2000. Health Resources and Service Agency. Retreived February 14, 2010, from ftp://ftp.hrsa.gov/bhpr/nationalcenter/phworkforce2000.pdf Gebbie, K. M. (2005). Public health: community and a shared future. Managed Care, 14(9 Suppl), 2. Gebbie, K. M., & Hwang, I. (2000). Preparing currently employed public health nurses for chan ges in the health system. American Journal of Public Health, 90(5), 716721. Gebbie, K. M., Rosenstock, L., Hernandez, L. M., & Institute of Medicine. (2003). Who will keep the public h ealthy Washington DC: The National Academies Press. Giachello, A. L., Arrom, J. O., Davis, M., Sayad, J. V., Ramirez, D., Nan di, C., et al. Chicago Southeast Diabetes Community Action Coalition (2003). Reducing diabetes health disparities through community -based participatory action research: the Chicago Southeast Diabet es Community Action Coalition. Public Health Report, 118(4), 309323. Goodman, R. M., Yoo, S., & Jack, L., Jr. (2006). Applying comprehensive community-based approaches in diabetes prevention: rationale, principles, and models. Journal of Public Health Management and Practice, 12(6), 545555. Grant, M. Florida Department of Health Human Resource Mangement, Personal Communication, 2007. Grumbach, K., Miller, J., Mertz, E., & Finocchio, L. (2004). How much public health in public health nursing pract ice. Public Health Nursing, 21(3), 266276. Hader, J., White, R., Lewis, S., Foreman, J., McDonald, P., & Thompson, L. (2007). Doctors' views of clinical practice gudelines: a qualitative exploration using innovation theory. Journal of Evaluation in Clinical Practice, 13 601606. Hair, J., Anderson, R., Tatham, R., & Black, W. (1998). Multivariate Data Analysis (5th ed.). Upper Saddle River: Prentice Hall. Harrison, E., & Falco, S. M. (2005). Health disparity and the nurse advocate: reaching out to al leviate suffering. Advances in Nursing Science, 28(3), 252 264. Hart, A., Brennan, C., Sym, D., & Larson, E. (2009). The impact of personalized prenotification on response rate to an electronic survey. Western Journal of Nursing Research, 31(1), 17 23.
Head, B. J., Aquilino, M. L., Johnson, M., Reed, D., Maas, M., & Moorhead, S. (2004). Content validity and nursing sensitivity of community -level outcomes from the Nursing Outcomes Classification. Journal of Nursing Scholarship, 36(3), 251259. Hill, C ., Fahrney, K., Wheeless, S., & Carson, C. (2006). Survey response inducements for registered nurses. Western Journal of Nursing Research, 28(3), 322334. Hilz, L. (2000). The informatics nurse specialist as change agent: Application of innovationdiffus ion theory. Computers in Nursing, 18(6), 272281. Hofrichter, R. (2007). Transforming elements of public health practice; Preliminary considerations. In R. Hofrichter (Ed.), Tackling Health Inequities Through Public Health Practice Lansing: NACCHO. Hogan, V. K., Njoroge, T., Durant, T. M., & Ferre, C. D. (2001). Eliminating disparities in perinatal outcomes lessons learned. Journal of Maternal Child Health, 5(2), 135140. Holder, V. L. (2004). From handmaiden to right hand-the infancy of nursing. AORN Journal, 79(2), 374(312). Horner, S., Abel, E., Taylor, K., & Sands, D. (2004). Using theory to guide the diffusion of genetics content in nursing curricula. Nursing Outlook, 52, 80 84. Health Resources and Services Administration.(2005). Public H ealth Workforce Study Health Resources and Services Administration. Retrieved February 14, 2010 from http://bhpr.hrsa.gov/healthworkforce/reports/publichealth/default.htm Huffstutler, S., Wyatt, T., & Wright, C. (2002). The use of handheld technology in nursing education. Nurse Educator, 27(6), 271275. Hulley, S. B. (2001). Designing clinical research : an epidemiologic approach (2nd ed.). Philadelphia: Lippincott Williams & Wilkins. Institute of M edicine. (1988). The future of Public Health. Washington, D.C.: National Academie s Press. Institute of Medicine. (1996). Healthy Communities: New Partnerships for the Future Washington DC: National Academy Press.Retreived February 14, 2010, from http:// www.nap.edu/openbook.php?isbn=030905625X Institute of Medicine. (2003). The Future of Public's Health in the 21st Century Washington, D.C: National Academies Press. Israel, B. A., Schulz, A. J., Parker, E., Becker, A. B., Allen, A., & Guzman, J. R. (200 3). Critical issues in developing and following community based participatory research principles. In
M. Minkler & N. B. Wallerstein (Eds.), Community -Based Participatory Research for Heatlh San Francisco: Jossey Bass. Israel, B. A., Schulz, A. J., Parke r, E. A., & Becker, A. B. (1998). Review of community-based research: assessing partnership approaches to improve public health. Annual Review of Public Health, 19, 173202. Issel, L. M., Baldwin, K. A., Lyons, R. L., & Madamala, K. (2006). Self -reported competency of public health nurses and faculty in Illinois. Public Health Nursing, 23(2), 168177. Iton, A. (2007). Tackling the root cause of health disparities through community capacity building. In R. Hofrichter (Ed.), Tackling Health Inequities Thr ough Public Health Practice Lansing : NACCHO. Ivanov, L., & Blue, C. (2008). Public Health Nursing: Leadership, Policy and Practice Clifton Park: Delmar Cengage Learning. Jaana, M., Ward, M., Pare, G., & Sicotte, C. (2006). Antecedents of clinical information technology sophistication in hospitals. Health Care Management Review, 31(4), 289299. Jakeway, C. C., Cantrell, E. E., Cason, J. B., & Talley, B. S. (2006). Developing population health competencies among public health nurses in Georgia. Public Health Nursing, 23(2), 161167. Jenkins, J., & Calzone, K. (2007). Establishing the essential nursing competencies for genetics and genomics. J ournal of Nursing Scholarship, 39(1), 10 16. Jerome D'Emilia, B., & Begun, J. (2005). Diffusion of breast conserving surgery in medical communities. Social Science Medicine, 60 143151. Keller, L., Schaffer, M. A., Lia Hoagberg, B., & Strohschein, S. (2002). Assessment, Program Planning, and Evaluation in PopulationBased Public Health Practice. Journal of Public Health Management & Practice, 8(5), 30 43. Keller, L, Strohschein, S., Lia Hoagberg, B., & Schaffer, M. (1998). Populationbased public he alth nursing interventions: a model from practice. Public Health Nursing, 15(3), 207215. Keller, L Strohschein, S., Lia Hoagberg, B., & Schaffer, M. A. (2004). Populationbased public health interventions: practice -based and evidence -supported. Part I Public Health Nursing, 21(5), 453468. Kellogg, W. K. (2005). Racial and Ethnic Health Disparities : Schools of Public Health Respond as Engaged Institutions Retrieved February 16, 2010, from http://www.asph.org/UserFiles/WKKF75thFinalProceedings.pdf
Kosidlak, J. G. (1999). The Development and Implementation of a Population-Based Intervention Model for Public Health Nursing Practice. Public Health Nursing, 16(5), 311320. Kramer, M., Schmalenber, C., Brewer, B., Verran, J., & Keller -Unger, J. (2009). Accurate assessment of clinical nurses' work environments: Response rate needed. Research in Nursing and Health, 32, 229 240. doi: 10.1002/nur.20315 Krieger, J., Allen, C., Roberts, J., Ross, L., & Takaro, T. (2005). What's with the wheezing. In B. A. Israel, E. Eng, A. Schultz & E. Parker (Eds.), Methods in Community -based Participatory Research for Heatlh. San Francisco: Jossey -Bass. Kulbok, P. A., & Reed, J. (2006). PHN competencies. Public Health Nursing, 23(2), 97 98. Kutney -Lee, A., McHugh, M., Sloane, D., Cimiotti, J., Neff, D. & Aiken, L. (2009). Nursing: A key to patient satisfaction. Healtlh Affairs 28 (4)669677. Landrum, B. (1998). Marketing innovations to nurses, part 1: How people adopt innovations. Journal of W ound O stomy and Contience Nursing 25 (4),194199. Lasker, R. & Weiss, E. (2003) Broadening participation in community problem solving: a multidisciplinary model to support collaborative practice and research. Journal of Urban Health: Bulletin of the N ew York Academy of Medicine 80 (1),14 60. Lee, T. T. (2004). Nurses' adoption of technology: Application of Rogers' innovationdiffusion model. Applied Nursing Research, 17(4), 231238. Lindgerg, E. (2006). Competence in critical care. Dimensions of Critical Care Nursing 25(2), 7781. Linnan, L. A., Ferguson, Y. O., Wasilewski, Y., Lee, A. M., Yang, J., Solomon, F., et al. (2005). Using community -based participatory research methods to reach women with health messages: results from the North Carolin a Beauty and Health Pilot Project. Health Promotion Practice, 6(2), 164173. Lueng, M., Yen, I., & Minkler, M. (2004). Community-based participatory approach for increasing epidemiology's relevance in the 21st century. International Journal of Epidemiolo gy, 33, 499506. MacQueen, K. M., McLellan, E., Metzger, D. S., Kegeles, S., Strauss, R. P., Scotti, R., et al. (2001). What is community. An evidence based definition for participatory public health. American Journal of Public Health, 91(12), 19291938. Margolis, L. H., Parker, E. A., & Eng, E. (1999). Who speaks for public health agencies: assessing the core functions in local health departments. Journal of Public Health Management Practice, 5(3), 47 53.
McKnight, J. (2000). Rationale for a community approach to health improvement. In T. Bruce & S. McKane (Eds.), A Partnership Model (pp. 13 18). Washington DC: American Public Health Association. Meade, C., & Calvo, A. (2001). Developing communityacademic partnerships to enhance breast health among rural and hispanic migrant seasonal farmworker women. Oncology Nursing Forum, 28(10), 15771584. Merzel, C., & D'Afflitti, J. (2003). Reconsidering communitybased health promotion: Promise, performance, and p otential. American Journal of Public Health, 93(4), 557-574. Meyers, L. W., Gamst, G., & Guarino, A. J. (2006). Applied Multivariate Research: Design and Interpretation Thousand Oaks: Sage Publications. Milio, N. (1970). 9226 Kercheval; the storefront that did not burn. Ann Arbor,: University of Michigan Press. Miller, S., Alpert, P., & Cross, C. (2008). Overweight and obesity in nurses, advanced practice nurses, and nurse educators. Journal of the American Academy of Nurse Practitioners, 20, 259265. Minkler, M. (2005). Community-based research partnerships: challenges and opportunities. Journal of Urban Health, 82(2 Suppl 2), ii3 12. Minkler, M., Blackwell, A., Thompson, M., & Tamir, H. (2003). Communitybased participatory research: Implications for public health funding. American Journal of Public Health, 93(8), 12101213. Minkler, M., Vasquez, V., Chang, C., & Miller, J. (2008). Promoting Heatlhy Public Policy through Community -Based Participatory Research: Ten Case Studies, University of California, Berkeley, School of Public Health and Policy Link W.K. Kellogg Minkler, M., Vasquez, V., Warner, J., Steusse y, H., & Facente, S. (2006). Sowing the seeds for sustainable change: a community -based participatory research partnership for health promotion in Indiana, USA and its aftermath. Health Promotion International, 21(4), 293300. Minkler, M., & Wallerstein, N. (2003a). Community based participatory research for health. San Francisco, CA: Jossey Bass. Minkler, M., & Wallerstein, N. (2003b). Introduction of community-based participatory research. In M. Minkler & N. Walllerstine (Eds.), Community -Based Partici patory Research for Health San Francisco: Jossey -Bass.
Morris, D., Fenton, M. V., & Mercer, Z. B. (2004). Identification of national trends in nursing education through the use of an online survey. Nursing Outlook, 52(5), 248254. National Association of County and City Health Officials. (2005). Operational definition of a functional local health department National Association of County and City Health Officials (Ed.) Retrieved February 14, 2010, from http://www.naccho.org/topics/infrastructure/accreditation/upload/OperationalDefinitionB rochure 2.pdf National Association of County and City Health Officials. (2009). 2008 National Profile of Local Health Departments. Retrieved February 26, 2010 from http://www.naccho.org/topics/infrastructure/profile/resources/2008reports/upload/NACC HO_2008_ProfileReport _post to -website 2.pdf National Cancer Institute. (2009). Health Disparities Defined. Retrieved February 10, 2010 from http://crchd.cancer.gov/disparities/defined.html Nguyen, T., McPhee, S., Bui Tong, N., Luong, T., Ha Iasconis, R., Nguyen, T., et al. (2006). Community-based participatory research increases cervical cancer screening among Vietnamese -Americans. Journal of Health Care for the Poor and Underserved, 17, 31 54. Novick, L., Morrow, C., & Mays, G. (2008). A f ramework for p ublic h ealth a dministration an d Practice. In Public Health Administration: Principles for Population-Based Management Sudbury: Jones and Bartlett. Office of M inority Health, Florida Department of Health. Funding Announcement (2009). Florida Department of Health Retrieved February 16, 2010, from http://www.doh.state.fl.us/Minority/CTG/Funding.htm Office of M inority Health, Florida Department of Health. Reducing Health Disparities among Minorities. Grant Prog ram. (2009 ). Florida Department of Health Retrieved February 1 6 2010, from http://www.doh.state.fl.us/Minority/CTG/index.htm Office of Minority Hea lt h, Florida Department of Health. (2007). Closing the Gap Approved Funding FY 06/07 Retrieved Februra ry 12, 2010, from http://www.doh.state.fl.us/equopp/ctg/indexCTG.html Oppewal, S., Lamanna, B. F., & Lee Glenn, L. (2006). Comparison of the dissemination and implementation of standardized public health nursing competencies in academic and practice settings. Public Health Nursing, 23(2), 99 107. Paine -Andrews, A., Fisher, J., Harris, K., Lewis, R., Williams, E., Vincent, M., et al. (2000). Some experiential lessons in supporting and evaluating community based initiatives for preventing adolescent pregnancy. Health Promotion Practice, 1 (1), 66 76.
Parker, E., Margolis, L. H., Eng, E., & Henriquez Roldan, C. (2003). Assessing the capacity of health departments to engage in community -based participatory public health. American Journal of Public Health, 93(3), 472476. Parker, E., Robins, T., Israel, B. A., Brakefield Caldwell, W., Edgren, K., & Wilkens, D. (2005). Developing and implementing guidelines for dissemination: The experience of the community action against asthma project. In B. A. Israel, E. Eng, A. Schultz & E. Parker (Eds.), Methods in Community -Based Participartory Research for Health San Francisco: Jossey Bass. Pett, M., Lackey, N., & Sullivan, J. (2003). Making Sense of Factor Analysis: The use of Factor analysis for instrument development in healthcare research Thousand Oaks: Sage Publications. Polit, D., & Beck, C. (2008). Nursing Research: Generating and Assessing Evidence for Nursing Practice (8 ed.). Philadelphia: Lippencott, Williams and Wilkens. Potvin, L., Cargo, M., McComber, A., Delormier, T ., & Macaulay, A. (2003). Implementing participatory intervention and research in communities: lessons from the Kahnawake Schools Diabetes Prevention Project in Canada. Social Science and Medicine, 56,12951305. Price, J. (2000). Show me the money. Medic al Directors' responses to a mail survey. The American Journal of Managed Care, 6(11), 12571260. Primomo, J., Johnston, S., DiBiase, F., Nodolf, J., & Noren, L. (2006). Evaluation of a community -based outreach worker program for children with asthma. Pu blic Health Nursing, 23(3), 234241. Quad Council of Public Health Nursing Organizations. (2003). Scope and Standards of Public Health Nursing Practice Washington DC: American Nurses Publishing. Quad Council of Public Health Nursing Organizations (2004). Public health nursing competencies. Public Health Nursing, 21(5), 443452. Racine, D. (2006). Reliable Effectivenss: a theory on sustaining and replicating worthwhile innovations. Administrative Policy Mental Health and Mental Health Services Researc h, 33, 356387. .. Rogers, E. (2003). Diffusion of Innovations (5th ed.). New York: Free Press. Satterfield, D., Volansky, M., Caspersen, C., Engelgau, M., Bowman, B., Gregg, E., et al. (2003). Community-based lifesyle interventions to prevent type 2 dia betes. Diabetes Care, 26(9), 26432652.
Saleh,S., Williams, D., & Balougan, M. (2004). Evaluating the effectiveness of public health leadership training: The NEPHLI experience. American Journal of Public Health 94(7),12451249. Savage, C. L., Xu, Y., Lee, R., Rose, B. L., Kappesser, M., & Anthony, J. S. (2006). A case study in the use of communitybased participatory research in public health nursing. Public Health Nursing, 23(5), 472478. Schafer, J., & Yucel, R. (2002). Compu tational strategies for multivariate linear mixed effects models with missing values. Journal of Computational and Graphical Statistics, 11(2), 437457. Schultz, A., Israel, B., Selig, S., Bayer, I., & Griffin, C. B. (2000). The research p erspective: d ev elopment and i mplementation of p rinciples for c ommunity-ba sed r esearch in p ublic h ealth. In T. Bruce & S. McKane (Eds.), A Partnership Model (pp. 5369). Washington DC: American Public Health Association. Schultz, A., Krieger, J., & Galea, S. (2002). Addr essing social determinants of health: community -based participatory approaches to research and practice. Health Education and Behavior, 29(3), 287 295. Schultz, A., Zenk, S., Odoms Yong, A., HollisNeely, T., Nwankwo, R., Lockett, M., Kannan, S. (2005) Healthy eating and exercising to reduce diabtes: exploring the potential of social determinants of health frameworks withn the context of communitybased participatory diabetes prevention. American Journal of Public Health, 95(4), 645651. Schwab, M., & Syme, S. L. (1997). On paradigms, community participation, and the future of public health. American Journal of Public Health, 87(12), 20492051; Shannon, D. & Bradshaw, C. (2002). A comparison of response rate, reponse rate time, and costs of mail and elctronic surveys. The Journal of Experimental Education, 70(2),179192. Shirey, M. (2006). Evidence based practice: How nurse leaders can facilitate innovation. Nursing Administration Quarterly, 30 (3), 252265. Singer M. (1993). Knowledge for use: A nthropology and communitycentered substance abuse research. Social Science Medicine, 37 (1), 14 25. Smedley, B., Stith, A., Nelson, A., & Institute of Medicine, (2002). Unequal treatment : confronting racial and ethnic disparities in health care Washington, D.C.: National Academy Press. Smith, G., & Randolph -Back, K. (2000). A Foundation Perspective on the CommunityBased Public Health Initiative. In T. Bruce & S. McKane (Eds.), A Partnership Model (pp. 109116). Washington DC: American Public He alth Association.
Smith, T. Minyard, K., Parker, C., Van Valkenburg, R., & Shoemaker, J. (2007). From theory to practice: What drives the core business of public health. Journal Public Health Management Practice 13(2),169172. Studlar, D. (1999). Diffusi on of tobacco control in North America. The Annals of the American Academy of Political and Social Science, 566, 68 79. Sue, S., & Dhindsa, M. K. (2006). Ethnic and racial health disparities research: issues and problems. Health Education and Behavior, 33(4), 459 469. Sullivan, M., Chao, S., Allen, C., Kone, A., Pierre -Louis, M. & Krieger, J. (2003). Community Researcher Partnerships; Perspectives form the field. In M. W. Minkler, N (Ed.), Community Based Participatory Research for Health (pp. 113130). San Francisco: Jossey Bass. Swider, S., Levin, P., Ailey, S., Breakwell, S., Cowell, J., McNaugton, D., et al. (2006). Matching a graduate curriculum in public/community health nursing to practice competencies: The Rush University Experien ce. Public Health Nursing, 23(2), 190195. Tabacachnick, B. G., & Fidell, L. S. (2007). Using Multivariate Statistics Boston: Pearson/Allyn and Bacon. Thompson, B. (2004). Exploratory and Confirmatory Factor Analysis: Understanding concepts and applications Washington, DC: American Psychological Association. Travers, K. (1997). Reducing inequities through participatory reserch and community empowerment. Health Education and Behavior, 24(3), 344356. Trossman, S. (2009). Fleeting or pandemic: RNs dr aw on experience, planning to repond to current influenza outbreak, The American Nurse, p.1. Turnock, B. (2004). Public Health: What is it and how does it work. (3rd ed.). Sudbury: Jones and Barlett Publishers. Turnock, B. J., Handler, A., Hall, W., Pot sic, S., Nalluri, R., & Vaughn, E. H. (1994). Local health department effectiveness in addressing the core functions of public health. Public Health Report, 109(5), 653658. U.S. Department of Health and Human Services. (2000). Healthy people 2010 : U nde rstanding and improving health (2nd ed.). Retreived February 14, 2010 from http://www.healthypeople.gov/Document/tableofcontents.htm#volume1 U.S. Department of Health and Human Services. (2004). National Sample Survey of Registered Nurses. Gallup Poll. Retreived February 14, 2010, from http://bhpr.hrsa.gov/healthworkforce/rnsurvey04/preface.htm
U.S. Department of Health and Human Services. (2005). Public Health Workforce Survey Health Resources and Services Administration. Retreived Febr uary 17, 2010, from ftp://ftp.hrsa.gov/bhpr/nationalcenter/publichealth2005.pdf U.S. Department of Health and Human Servies. (2010). National Partnership for Action. Retrieved March 1, 2010 from http://minorityhealth.hhs.gov/npa/templates/brow se.aspx?lvl=1&lvlid=13 U.S. Public Health Service. (1991). Healthy people 2000 : National health promotion and disease prevention objectives : F ull report, with commentary Washington, D.C University of Florida Institutional Review Board (2010). Retreived February 17, 2010 from http://irb.ufl.edu/irb01/forms.htm Van Duyn, M. A., Reuben, S. H., & Macario, E. (2006). Special populations networks: Themes and lessons learned. Cancer, 107(Suppl.8), 19451954. Viswanathan, M., Ammerman, A., Eng, E., Gartlehner, G., Lohr, K., Griffith, D., et al. (2004). Community Based Participatory Reseach: Assessing the evidence (Publication No. 04E0222). Retrieved February 14, 2010, from http://www.ahrq.gov/clinic/tp/cbprtp.htm Wallerstein, N. B., & Duran, B. (2006). Using communitybased participatory research to address health disparities. Health Promotion Practice, 7(3), 312323. Wang, C., Cash, J., & Powers, L. (2000). Who knows the streets as well as the homeless. Promoting personal and community action through photovoice. Health Promotion Practice, 1 (1), 81 89. Weierbach, F. (2007). Public Heatlh Nursing Competencies 19531966: Effective and Efficient. Puib lic Health Nursing, 24(5), 480487. Weiss, E., Anderson, R., & Lasker, R. (2002). Making the most of collaboration: Exploring the relationship between partnership synergy and partnership functioning. Health Education & Behavior 29(6), 683698. Wilcox, R., & Knapp, A. (2000). Building communities that create health. Public Health Report, 115(2 3), 139143. Wing, S., Grant, G., Green, M,. & Stewart, C. (1996). Community based collaboration for environmental justice: South -East Halifax environmental reaw akening. Environment and Urbanization, 8(2), 129140. Yoon, S., & Horne, C. (2004). Accruing the sample in survey research. Southern Online Journal of Nursing Research, 2(5), 1 17.
Zahner, S., & Gredig, Q. (2005a ). Improving public health nursing educa tion: recommendations of local public health nurses. Public Health Nursing, 22(5), 445450. Zahner, S., & Gredig, Q. (2005b ). Public health nursing practice change and recommendations for improvement. Public Health Nursing, 22(5), 422428.
BIOGRAPHICAL SKETCH Catherine Levonian received her initial nursing degree a Bachelor of Science in Nursing from Marycrest College in Davenport, Iowa in 1987. Upon completion of her degree, she worked in acute care and public health nursing. She received her Master of Science in Nursing (MSN) in 1990 from the University of Virginia, with a concentration in Management of Health Systems. After completion of her MSN, she worked as a nurse educator at acute care adult and pediatric medical centers. In addi tion to her doctoral studies she was awarded a Master in Public Health at the University of Florida. During her first and fourth year of doctoral studies, she worked as a Graduate Teaching Assistant, teaching public health nursing to undergraduate student s. She had valuable research experience under her mentor Dr Shawn Kneipp, as a Graduate Research Assistant working three years on a RO1 NIH study. Catherine will graduate in May 2010 with her PhD in n ursing s cience s and a Master of Public Health. She pl ans to continue her research in health disparities as a researcher and educator.