USF Libraries
USF Digital Collections

Factors affecting the utilization and quality of long-term care

MISSING IMAGE

Material Information

Title:
Factors affecting the utilization and quality of long-term care
Physical Description:
Book
Language:
English
Creator:
Temple, April
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Nursing home placement
Dementia
Nursing assistants
Turnover
Benefits
Dissertations, Academic -- Aging Studies -- Doctoral -- USF   ( lcsh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: The aging of the nation in the coming decades will contribute to an increased demand for long-term care. Given this trend, it is becoming increasingly important to understand utilization of services along the continuum of care and to determine factors that influence the provision of quality care. These insights are needed to reduce national expenditures on long-term care and to maximize independence and well-being among older adults. The purpose of the present dissertation was to explore factors that influence the utilization and quality of long-term care by conducting three separate studies. The first study examined risk of nursing home (NH) placement in a frail sample of older adults receiving publicly-funded long-term care in two settings: home and community-based services (HCBS) or assisted living (AL). Specifically, it explored whether these settings of care modify the relationship between dementia and NH placement.The second study examined NH organizational characteristics and job characteristics of staff in relation to turnover of nursing assistants (NAs) in a nationally representative sample of NHs. Using the same sample, the final study examined the presence of employment-based benefits of NAs and explored the relationship between NH organizational characteristics and the availability of staff benefits. Findings from this dissertation can contribute to a greater understanding of the use of long-term care services and the retention of staff to provide quality care. In the first study, AL was associated with longer time to NH placement for individuals with dementia when compared to HCBS. In the second study, the provision of competitive wages and benefits, involvement of NAs in resident care planning, and high levels of nurse staffing were associated with lower NA turnover, which may be important for quality of NH care.In the final study, for-profit ownership and high Medicaid occupancy were negatively associated with NA benefits whereas facility size, chain membership, occupancy level, union involvement, and education of the NH administrator were positively associated with benefits offered to these staff. Collectively, the findings from this dissertation may have important implications for policy makers, providers, and consumers of long-term care.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2009.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by April Temple.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 83 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002317244
oclc - 659862355
usfldc doi - E14-SFE0002990
usfldc handle - e14.2990
System ID:
SFS0027307:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Factors Affecting the Utilization and Quality of Long-Term Care by April Temple A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy School of Aging Studies College of Behavioral and Community Sciences University of South Florida Co-Major Professor: Ross Andel, Ph.D. Co-Major Professor: Debra Dobbs, Ph.D. Elizabeth Bass, Ph.D. Cathy L. McEvoy, Ph.D. Jennifer R. Salmon, Ph.D. Date of Approval: June 25, 2009 Keywords: nursing home placement, dementia, nursing assistants, turnover, benefits Copyright 2009, April Temple

PAGE 2

Acknowledgements I would like to gratefully acknowledge all who have contributed to the successful completion of this dissertation. This work would not have been possible without the support of many individuals. I extend my sincerest gratitude to my major professors, Dr. Debra Dobbs and Dr. Ross Andel, for their generous commitment to my doctoral work and my professional development. I appreciate their guidance, patience, and encouragement over the past four years. Their mentorship has offered an i deal example of how I wish to mentor my future students. I also wish to thank the following members of my doctoral committee: Dr. Cathy McEvoy, Dr. Jennifer Salmon, and Dr. Elizabeth Bass. I greatly appreciate their multidisciplinary insights and helpful comments on the conceptualization of this research. I owe many thanks to Dr. Glenn Mitchell and Su Wang for their technical assistance with the state administrative data sets. I am very grateful to the faculty and staff at the School of Aging Studies for their support and assistance in my educational endeavors. Lastly, I owe a special note of gratitude to my family. I am grateful to my parents for their inspiration and support as I have pursued my academic goals. I extend a heartfelt thank you to my husband for his unwavering love and encouragement throughout graduate school. And to my late grandparents, I will always be thankful for their role in instilling in me a compassion for older adults.

PAGE 3

i Table of Contents List of Tables ................................................................................................................ .... iii List of Figures ............................................................................................................... .... iv Abstract ...................................................................................................................... ....... v Chapter One: Introduction ................................................................................................. 1 Utilization of Long-Term Care................................................................................ 1 Nursing Home Staffing and Quality of Care .......................................................... 2 Summary ............................................................................................................... 4 Chapter Two: Study I ........................................................................................................ 5 Abstract ................................................................................................................. 6 Introduction ............................................................................................................ 7 Methods ................................................................................................................. 9 Sample ...................................................................................................... 9 Measures ................................................................................................. 11 Dependent Variable ..................................................................... 11 Independent Variables ................................................................. 11 Analyses .................................................................................................. 12 Results ................................................................................................................ 13 Sample Characteristics ............................................................................ 13 Risk of NH Placement .............................................................................. 14 Modifying Effect ....................................................................................... 15 Discussion ........................................................................................................... 15 Limitations................................................................................................ 17 Conclusion ............................................................................................... 18 Chapter Three: Study II ................................................................................................... 22 Abstract ............................................................................................................... 23 Introduction .......................................................................................................... 24 Organizational Characteristics Associated with NA Turnover ................. 25 Job Characteristics Associated with NA Turnover ................................... 26 Theoretical Framework ............................................................................ 26 Methods ............................................................................................................... 28 Sample .................................................................................................... 28 Measures ................................................................................................. 29 Turnover ....................................................................................... 29 Organizational Characteristics ..................................................... 29 Job Characteristics ...................................................................... 30 Analyses .................................................................................................. 31

PAGE 4

ii Results ................................................................................................................ 32 Discussion ........................................................................................................... 34 Limitations................................................................................................ 38 Conclusion ............................................................................................... 39 Chapter Four: Study III .................................................................................................... 43 Abstract ............................................................................................................... 44 Introduction .......................................................................................................... 45 The Importance of Benefits ...................................................................... 46 Theoretical Framework for the Availability of Benefits in Nursing Homes .................................................................................. 46 Nursing Home Characteristics and the Availability of Benefits ................ 47 The Present Study ................................................................................... 49 Methods ............................................................................................................... 49 Sample .................................................................................................... 49 Measures ................................................................................................. 50 Benefits ........................................................................................ 50 Organizational Characteristics ..................................................... 52 Analyses .................................................................................................. 53 Results ................................................................................................................ 54 Discussion ........................................................................................................... 55 Limitations................................................................................................ 58 Conclusion ............................................................................................... 59 Chapter Five: Concluding Remarks ................................................................................ 64 Discussion of Findings ........................................................................................ 64 Limitations ........................................................................................................... 66 Future Directions ................................................................................................. 68 References .................................................................................................................... .. 70 Appendices .................................................................................................................... 77 Appendix A: Action Letter .................................................................................... 78 Appendix B: Curriculum Vitae ............................................................................. 79 About the Author ................................................................................................. End Page

PAGE 5

iii List of Tables Table 2.1 Predisposing, Enabling, and Need Characteristics of the Sample .................. 19 Table 2.2 Risk of NH Placement ..................................................................................... 20 Table 2.3 Risk of NH Placement by Setting of Care ....................................................... 21 Table 3.1 Nursing Home Organizational Characteristics, Job Characteristics, and Turnover ............................................................................................................ 40 Table 3.2 Organizational and Job Characteristics Associated with Low and High NA Turnover Compared to Moderate NA Turnover .......................................... 41 Table 4.1 Weighting of NA Benefits ................................................................................ 60 Table 4.2 Organizational Characteristics of the Sample ................................................. 61 Table 4.3 Organizational Characteristics Associated with NA Benefits .......................... 62

PAGE 6

iv List of Figures Figure 3.1 Framework of Analysis for Nursing Assistant (NA) Turnover ......................... 42 Figure 4.1 Percentage of Nursing Homes with Benefits Offered to Nursing Assistants .............................................................................................................. .... 63

PAGE 7

v Factors Affecting the Utilization and Quality of Long-Term Care April Temple ABSTRACT The aging of the nation in the coming decades will contribute to an increased demand for long-term care. Given this trend, it is becoming increasingly important to understand utilization of services along the continuum of care and to determine factors that influence the provision of quality care. These insights are needed to reduce national expenditures on long-term care and to maximize independence and well-being among older adults. The purpose of the present dissertation was to explore factors that influence the utilization and quality of long-term care by conducting three separate studies. The first study examined risk of nursing home (NH) plac ement in a frail sample of older adults receiving publicly-funded long-term care in two settings: home and community-based services (HCBS) or assisted living (AL). Spec ifically, it explored whether these settings of care modify the relationship between dementia and NH placement. The second study examined NH organizational characteristics and j ob characteristics of staff in relation to turnover of nursing assistants (NAs) in a nationally representative sample of NHs. Using the same sample, the final study examined the presence of employment-based benefits of NAs and explored the relationship between NH organizational characteristics and the availability of staff benefits. Findings from this dissertation can contribute to a greater understanding of the use of long-term care services and the retention of staff to provide quality care. In the

PAGE 8

vi first study, AL was associated with longer time to NH placement for individuals with dementia when compared to HCBS. In the second study, the provision of competitive wages and benefits, involvement of NAs in resident care planning, and high levels of nurse staffing were associated with lower NA turnover, which may be important for quality of NH care. In the final study, for-profit ownership and high Medicaid occupancy were negatively associated with NA benefits whereas facility size, chain membership, occupancy level, union involvement, and education of the NH administrator were positively associated with benefits offered to these staff. Collectively, the findings from this dissertation may have important implications for policy makers, providers, and consumers of long-term care.

PAGE 9

1 Chapter One: Introduction The aging of the population in the ensuing decades will likely contribute to an increased demand for long-term care. Approximately 9 million older adults in the United States currently need long-term care, and this number is projected to rise to 12 million by the year 2020 (U.S. Department of Health and Human Services, 2007). Although the majority of long-term care is provided by informal caregivers, greater numbers of older adults will need formal care provided by home and community-based services (HCBS) or provided in an assisted living (AL) or nursing home (NH) setting. Given these trends, understanding outcomes of care in long-term care settings is an increasingly important issue. The purpose of this dissertation is to examine factors affecting the utilization and quality of long-term care. Utilization of Long-Term Care Long-term care provides assistance to persons with physical or cognitive impairments that result in functional limitations in activities of daily living. The continuum of long-term care includes HCBS, AL, and NH care. HCBS offers an array of services including case management, home health care, and personal care assistance with activities of daily living to approximately 2. 8 million clients in the United States (Ng, Harrington, & OMalley, 2008). AL refers to a variety of residential care settings that provide personal care and 24-hour assistance with activities of daily living within an environment that focuses on maximizing digni ty, privacy, and autonomy (National Center for Assisted Living, 2006). There are approximately 68,000 AL communities in the United States that provide care to nearly one million older adults (National Center for

PAGE 10

2 Assisted Living, 2006). NHs operate under a medical model of care and provide acute rehabilitation and 24-hour skilled nursing care to the most functionally impaired. There are approximately 16,500 NHs nationwide that provide long-term care to approximately 1.3 million residents (Harrington, Carrillo, & Blank, 2008). In an effort to provide care in the least restrictive setting and eliminate the institutional bias inherent in public funding of long-term care, HCBS and AL are increasingly serving frail older adults as less institutional and less expensive alternatives to traditional NH care. As a result, the number of older adults with dementia or cognitive impairment in these settings is substantial. Approximately one-quarter of HCBS clients and more than one-half of AL residents have dementia or other cognitive impairment (Alzheimers Association, 2007; Zimmerman et al., 2007). In comparison to HCBS, AL may be better able to meet the needs of residents with dementia since a core philosophy of AL is to adapt its service provision to allow residents to age in place (Chapin & DobbsKepper, 2001). However, the differential effect of HCBS and AL on risk of NH placement for individuals with dementia remains unknown. The purpose of the first dissertation study is to explore the modifying effect of setting of care in HCBS or AL on the relationship between dementia and NH placement. The findings may contribute greater understanding of the role of HCBS and AL along the continuum of care. Nursing Home Staffing and Quality of Care For frail older adults who need medical or skilled nursing care, NHs are often the most appropriate setting of long-term care. In this setting, quality of care has been a persistent concern. In 2007, nearly one in five NHs received a serious deficiency during state inspection for causing actual harm to residents or placing residents in immediate jeopardy (Harrington et al., 2008). In addition, failure to provide quality care is the third

PAGE 11

3 most frequently cited deficiency in NHs across the country (Harrington et al., 2008). Therefore, improving the quality of NH care is a priority for consumers, providers, and policy makers. Central to the delivery of quality NH care are registered nurses (RNs), licensed practical nurses (LPNs), and nursing assistants (NAs). A large body of research has demonstrated a direct relationship between nurse staffing levels and the quality of care provided to residents (e.g., Bleismer, Smayling, Kane, & Shannon, 1998; Harrington, Zimmerman, Karon, Robinson, & Beutel, 2000; Schnelle et al., 2004). Although research and policy have primarily focused on achieving adequate staffing levels to ensure optimal resident care (Centers for Medicare and Medicaid Services, 2001; Harrington et al., 2000), the presence of high nursing staff turnover has also been associated with poor quality of care outcomes. For example, studies have documented that high turnover of nursing staff is associated with physical restraint use, catheter use, contractures, pressure ulcers, psychoactive drug use, hospitalizations, infection rates, and quality of care deficiencies (Castle & Engberg, 2005; 2006; Spector & Takada, 1991; Zimmerman, Gruber-Baldini, Hebel, Sloane, & Magaziner, 2002). Given the deleterious consequences to quality of care associated with high turnover, there is a need to determine factors that influence turnover of NH staff. Since NAs provide the majority of direct care to nursing home residents and generally have the highest turnover rates in the industry (Cohen-Mansfield, 1997), turnover of NAs may be particularly important. The purpose of the second dissertation study is to examine NH organizational characteristics and job characteristics of staff in relation to turnover of NAs. The findings may provide insight into possible retention strategies, which may help reduce turnover and subsequently improve the delivery of resident care.

PAGE 12

4 With respect to maintaining a qualified workforce of NAs, high turnover and labor shortages represent major challenges faced by the NH industry. Moreover, the demand for these direct care workers is expected to increase with the aging of the nation (U. S. Bureau of Labor Statistics, 2000). One factor affecting recruitment and retention of NAs is the limited availability of employee benef its. For example, greater availability of benefits has been associated with fewer intentions to leave the job among NAs (Stearns & DArcy, 2008). However, little is known about the range of benefits offered to NAs and the variation of benefits across the NH industry. The aims of the third dissertation study are to describe the employment-based benef its offered to NAs and to explore the relationship between NH organizational characteristics and the availability of NA staff benefits. Findings from this study may be informative for strategies to enhance the provision of benefits to attract and retain NA staff. Summary In summary, the present doctoral dissertation is organized as three individual studies that address two primary goals of long-term care: to allow older adults to age in place in the least restrictive setting and to provide high quality of care. Since NHs are the most expensive and least preferable option, the first study examines the viability of HCBS and AL to prevent or delay NH placement among frail older adults, including those with dementia. In an effort to improve the quality of care for older adults residing in NHs, the second and third dissertation studies explore correlates of turnover and benefit provision of NAs. Collectively, the findings may have important implications for understanding the continuum of long-term care and maintaining a direct care workforce to provide quality care. The three dissertation studies are presented in turn in the following chapters.

PAGE 13

5 Chapter Two: Study I Setting of Care Modifies Risk of Nursing Home Placement for Older Adults with Dementia April Temple, NHA, Ross Andel, PhD & Debra Dobbs, PhD

PAGE 14

6 Abstract Purpose: The purpose of this study was to examine risk of nursing home (NH) placement among older adults receiving hom e and community-based services (HCBS) or assisted living (AL) and to explore w hether these settings of care modify the relationship between dementia and risk of NH placement. Methods : The sample consisted of dually eligible Medicare and Medicaid beneficiaries age 65 and over who received HCBS (n=1,630) or resided in AL (n=836) in Florida between July 1999 and June 2000. Cox proportional hazard regression models were used to estimate risk of NH placement over a five year study period and to test the interaction of setting of care by dementia status. Results : In all, 15% of HCBS participants were placed in a NH compared to 26% of AL participants. As indicated by a significant interaction term in the regression model, setting of care modified the relationship between dementia and NH placement (HR=0.45, CI =0.31-0.66). In post-hoc analyses stratified by setting of care, dementia was associated with a 50% increased risk of NH placement from HCBS ( HR=1.50, CI =1.122.02) but was not associated with placement from AL ( HR =0.86, CI =0.63-1.16). Implications : The findings suggest that AL may prevent or delay nursing home placement for individuals with dementia when compared to HCBS.

PAGE 15

7 Introduction The number of older adults needing long-term care is projected to rise from 9 million to 12 million by the year 2020 (U.S. Department of Health and Human Services, 2007). This growth will contribute to an incr eased demand for long-term care services across the continuum of care including home and community-based services (HCBS), assisted living (AL), and nursing home (NH) care. NH care represents the most expensive option for long-term care services and consumes a majority of public funding under the Medicaid program (Centers for Medicare and Medicaid Services, 2007a). As a result, states have increased the availab ility of Medicaid-funded waivers for HCBS and AL to provide long-term care in less expensive and less institutional settings to individuals who would otherwise require institutional care. The number of cognitivelyimpaired older adults in these alternative settings is substantial, with one-quarter of HCBS recipients and more than one-half of AL residents having dementia or other cognitive impairment (Alzheimers Association, 2007; Zimmerman et al., 2007). Thus, the purpose of this study was to examine the effect of these settings of care on subsequent NH placement, particularly for persons with dementia. A large body of research has examined the role of dementia in regards to NH placement from the community. Dement ia has been associated with an increased likelihood of NH utilization compared to HCBS and AL in a sample of publicly-funded clients in Florida (Borrayo, Salmon, Polivka, & Dunlop, 2002). Dementia has also been found to be a risk factor for NH placement among HCBS clients (e.g., Andel, Hyer, & Slack, 2007; Scott, Edwards, Davis, Cornman, & Macera, 1997), yet the findings remain equivocal from AL. In a national survey of AL facilities, dementia was associated with an increased likelihood of discharge to a NH (Phillips, Munoz, Sherman, Rose, Spector, &

PAGE 16

8 Hawes, 2003). A few studies, however, reported no association between dementia or cognitive impairment and risk of NH placement from AL (e.g., Dobbs, Hayes, Chapin, & Oslund, 2006; Kopetz et al., 2000; Rosenberg et al., 2006). Collectively, these findings suggest that the community setting of care for persons with dementia may have an important influence on future NH utilization. In comparison to HCBS, AL settings may be better able to meet the needs of residents with dementia since a core component of the AL philosophy is to allow residents to age in place (Chapin & Dobbs-Kepper, 2001). To achieve this goal, AL facilities adjust their service provision and level of care criteria to meet residents changing needs in avoidance of discharge to a higher level of care in a NH. Research also suggests that AL may be an appropriate (Sloane, Zimmerman, Gruber-Baldini, Hebel, Magaziner, & Konrad, 2005) and less costly (Leon & Moyer, 1999) setting of care compared to a NH for persons with mild to moderate dementia. Thus, AL may be superior to HCBS for older adults with dementia to age in place; however, the differential effect of HCBS and AL on risk of NH placement for persons with dementia remains unknown. The objective of this study was to explore the modifying effect of setting of care on the relationship between dementia and NH placement in a sample of frail older adults receiving publicly-funded long-term care in HCBS or AL. Potential risk factors for NH utilization were classified according to the behavioral model of health services use (Andersen, 1995; Andersen & Newman, 1973). This model, commonly used in health services research, suggests that peoples use of health services is a function of three factors: predisposition to use of services, presence of enabling resources, and health characteristics representing need for care. Setting of care in HCBS or AL represents an

PAGE 17

9 enabling resource while dementia represents an important need for care characteristic. We hypothesized that setting of care would modify the association between dementia and risk of NH placement, so that the effect of dementia would be greater among HCBS participants compared to AL residents while controlling for other important factors. Methods Sample The sample for this study consisted of older adults age 65 and over who were enrolled in a Medicaid waiver program for HCBS or AL in the state of Florida between July 1999 and June 2000. HCBS participants were derived from the Florida Medicaid Aged and Disabled Adult waiver program. This program provides services such as personal care, case management, home health care, homemaker assistance, and home delivered meals. AL participants were derived from the Florida Medicaid Assisted Living for the Elderly waiver program, which provides services within the scope of AL care. Both of these programs provide long-term care to low income, community dwelling older adults who are at risk for NH placement. Participants had to be dually eligible for Medicare and Medicaid and have complete data for all study variables. A total of 2,894 individuals met these criteria. To assure that the long-term care assignment was not temporary or transitional, participants were al so required to remain continuously enrolled in HCBS or AL for the first three months in the study. This criterion led to the exclusion of 428 participants. Those excluded were less likely to be White or have an available caregiver and were more likely to be in HCBS and have a greater number of chronic diseases (p < .05). The analytic sample consisted of 2,466 participants, 1,630 who were receiving HCBS and 836 who were residing in AL. These individuals were followed until June 2004 or until death or placement in a NH.

PAGE 18

10 Data used in this study were derived from three state administrative data sources: Agency for Health Care Administration (AHCA) Medicaid claims data, Department of Health death records, and Department of Elder Affairs (DOEA) assessments in the Comprehensive Assessment Review and Evaluation Services (CARES) system or the Client Information Registration and Tracking System (CIRTS). These data sets reside in a data warehouse established by the State Data Center at the Florida Policy Exchange Center on Aging. Permission to use the data was obtained from AHCA and DOEA, and the study procedures were approved by the Institutional Review Board at the University of South Florida. Florida Medicaid claims data were used to determine demographic information, dual eligibility status, enrollment in the AL and HCBS waiver programs, and NH use. The Department of Health death certificate files were matched with Medicaid claims data to determine date of death. The Medicaid claims data and death records were subsequently matched with assessment data in CARES and CIRTS, from which sociodemographic characteristics, functional limitations, and health and cognitive status were obtained. CARES and CIRTS assessments are administered to community dwelling elders who apply for the Medicaid waiver programs used in this study and are designed to help the DOEA evaluate need and elig ibility for long-term care services. These assessments are administered by trained nurses or social workers and reviewed by physicians and program managers to ev aluate service eligibility and ensure data accuracy. An informed proxy, a family member, or a caregiver is used to obtain information when the client cannot participate in the assessment due to severe physical or cognitive impairment. Participants may have multiple assessments prior to and during enrollment in the programs, and the time between assessments and actual enrollment

PAGE 19

11 tends to vary widely due to administrative delays. Therefore, a twelve month window was used to identify the assessment closest to the first month in HCBS or AL. On average, the assessment took place within 6 months ( SD = 3.9 months). Measures Dependent Variable. The dependent variable was time to NH placement in months. NH placement was defined as a NH stay of 30 or more days with Medicaid as the primary payer to reflect long term, custodial care. Independent Variables. Predisposing characteristics of the participants included age, gender, and race. Age was measured as a continuous variable, gender was dichotomized into female (1) and male (0), and race was dichotomized into White (1) versus non-White (0). Enabling resources included marital status dichotomized into married (1) and other (0), and the availability of a family caregiver was coded as present (1) or absent (0). In addition, setting of care was included as an enabling resource and dichotomized to reflect participants residing in AL (1) versus the reference group receiving HCBS (0) at baseline. To measure need for care characteristics, a diagnosis of the following health conditions at baseline were included as dichotomous variables: arthritis, cancer, dementia, diabetes, emphysema/chronic obstruc tive pulmonary disease (COPD), heart disease, incontinence, and stroke. To create a parsimonious regression model, arthritis, cancer, diabetes, emphysema/COPD, heart disease, and stroke were summed to reflect a chronic disease score ranging from 0 to 6. Dementia and incontinence were not included in the chronic disease score but examined separately since both have been found previously to be important predictors of NH placement (Andel et al., 2007; Friedman, Steinwachs, Rathouz, Burton, & Mukamel, 2005). In addition, difficulty with

PAGE 20

12 activities of daily living (ADLs) and instrument al activities of daily living (IADLs) were used to assess functional status. ADLs included bathing, dressing, eating, toileting, transferring, and walking. IADLs included heavy and light housework, preparing meals, taking medications, managing finances, shopping, talking on the telephone, and transportation. The scale for both ADL and IADL variables was coded as no difficulties (0), needing some assistance (1), or cannot perform (2), and the scores reflect a sum of scores ranging from 0 to 12 for ADLs and 0 to 16 for IADLs. Analyses Descriptive statistics were calculated for all study variables. Chi-square and ttests were used to test for frequency and mean score differences, respectively, between HCBS and AL participants. A Cox proportional hazards regression model in the SAS procedure PHREG (SAS Institute, 2003) was used to estimate risk of NH placement. Cox proportional hazards regression is one approach to survival analysis that is used to study the occurrence and timing of events in longitudinal data (Allison, 2007). Analyzing longitudinal data with this method allows for determining the predictive risk of NH placement independently associated with each study variable while accounting for the proportion of time spent in the study. Participants were right censored if placement in a NH was not observed during the study period. Of the sample, 22% (n = 534) were censored for the number of months for which data were available during the follow-up period and 59% (n = 1,465) were censored due to death. The results are presented as the adjusted hazard ratio or relative risk of NH placement associated with each predictor controlling for the covariates. A hazard ratio greater than 1.00 indicates an increased risk of NH placement and below 1.00 indicates a decreased risk of NH placement. A 95%

PAGE 21

13 confidence interval was used with a significance level at a two-tailed p =0.05. To test the hypothesis that setting of care modifies the association between dementia and risk of NH placement, an interaction term between setting of care and dementia was included in a second Cox proportional hazards regression model. Operationally, a modifying effect (or moder ation) occurs when the modifying variable affects the direction or strength of the relationship between an independent variable and a dependent variable and the interaction between the modifying and independent variable is significant (Baron & Kenny, 1986). Since the interaction was statistically significant, post-hoc analyses were conducted in which the sample was stratified by the modifying variable, setting of care, to interpret the interaction effect. We used the likelihood ratio chi-square test to assess change in model fit with the inclusion of the interaction term. Specifically, the difference in -2 log likelihood values between the two models corresponds approximately to a chi-square test for goodness of fit. A statistically significant chi-square test indicates a change in model fit in which a smaller -2 log likelihood is a better fit. Results Sample Characteristics Baseline predisposing, enabling, and need for care characteristics for all participants and by setting of care are presented in Table 2.1. The average age of the sample was 81.56 years ( SD = 8.56), the majority were female (77%) and white (67%), and few were married (17%) or had an available caregiver (31%). In regards to need for care, nearly all participants had arthritis (99%), 64% had heart disease, and 55% were incontinent. Less than half of the sample had dementia (43%), diabetes (30%), stroke (30%), emphysema/COPD (24%), or cancer (16%). On average, participants had at

PAGE 22

14 least two chronic diseases as represented by a mean chronic disease score of 2.62 ( SD = 1.09). The mean ADL score was 5.57 ( SD = 2.85), and the sample had substantial limitations in IADLs as represented by a mean IADL score of 12.49 ( SD = 3.22). Of the entire sample, 19% (n = 467) were placed in a NH during the duration of the study with an average time to placement of 16.92 months ( SD = 12.40 months). The baseline characteristics of the sample varied greatly by setting of care as indicated by the t-test and chi-square statistics presented in Table 2.1. HCBS participants were younger, less likely to be female or White, and more likely to be married or have an available caregiver. HCBS participants were also less likely to have dementia compared to AL participants. In addition, HCBS participants were less likely than AL participants to have diabetes, incontinence, and difficulties with IADLs but more likely to have emphysema/COPD or stroke. In relation to NH placement, HCBS participants were less likely to be placed (15% versus 26%) but had a shorter average time to placement (15.14 months, SD = 12.57 months) compared to AL participants (18.38 months, SD = 12.08 months). Risk of NH Placement Results from the Cox proportional hazard models adjusted for all study variables and the inclusion of the interaction term are presented in Table 2.2. In the first model, each year of age increased the risk of NH placement by 3%, being White increased the risk by 24% compared to non-White, and having an available caregiver reduced the risk by 24%. Although chronic diseases and IADLs approached significance, the only need for care characteristic that increased the risk of NH placement was incontinence. The second model, which included the addition of the interaction term, provided a better fit to the data as indicated by a statistically significant change in -2 log likelihood (degrees of

PAGE 23

15 freedom = 1, 2 = 16.31, p < .01). In this model, setting of care, dementia, and the interaction term emerged as significant predictors but the main effects are not interpreted with the inclusion of the interaction (Baron & Kenny, 1986). Modifying Effect The sample was stratified by the modifying variable, setting of care, to interpret the significant interaction effect between setting of care and dementia. As presented in Table 2.3, dementia was a significant risk factor for NH placement from HCBS but not AL. Other differences in risk between the two settings were that chronic diseases, incontinence, and ADLs increased the risk of placement from AL but not HCBS. In contrast, having a caregiver reduced the risk among HCBS participants but had no effect on risk among AL participants. Age and IADL limitations increased the risk from both settings of care. Discussion In this study, we examined risk of NH placement among dually eligible older adults receiving a Medicaid waiver for HCBS or AL while controlling for factors important for the utilization of health services (Andersen, 1995; Andersen & Newman, 1973). We also explored whether these settings modi fy the relationship between dementia and risk of NH placement. One of the strengths of this study was the ability to examine risk of NH placement among recipients of two community settings of long-term care, and we found that dementia emerged as a significant risk factor for NH placement from HCBS but not from AL. Specifically, a diagnosis of dementia increased the risk of subsequent NH placement by 50% among HCBS recipients only. This finding supports our hypothesis and suggests that differences in care provided in HCBS or AL affect future NH utilization among individuals with dementia.

PAGE 24

16 The results of this study are consistent with prior research in which dementia is a particularly strong determinant of NH placement among HCBS recipients (Andel et al., 2007; Scott et al., 1997). For HCBS recipients with dementia to remain in the community, it is likely that they receive care from a family caregiver. Even though availability of a caregiver significantly reduced the risk of placement among HCBS recipients, the burden and stress associated with caregiving for a person with dementia often precipitates the need for NH care (Gaugler, Leach, Clay, & Newcomer, 2004; Yaffe et al., 2002). HCBS may not provide enough assistance to these caregivers to prevent institutionalization. Therefore, HCBS clients with dementia may need additional case management and more comprehensive services. A challenge for policy makers would be to balance the costs of increased care at home with the potential cost-savings of delayed NH placement. Similar to recent studies (e.g. Dobbs et al., 2006; Rosenberg et al., 2006), dementia was not associated with NH placement from AL in this study. The AL industry has played an increasingly important role in caring for individuals with dementia, and our study provides additional evidence that these residents may be able to age in place. In relation to health outcomes of care, research has demonstrated that AL can provide adequate care to residents with dementia that do not have significant medical needs (Sloane et al., 2005). If the preference among consumers is to avoid NH placement, those with dementia may be better served in AL rather than at home with HCBS. From a policy perspective, national expenditures on long-term care could also be reduced by an estimated 14% if individuals with dementia received care in AL instead of a NH (Leon & Moyer, 1999). More research is needed, however, to examine the long-term cost effectiveness of providing care in AL versus HCBS to publicly-funded clients

PAGE 25

17 with dementia. Another important finding of our study was that only a relatively small percentage of the sample (19%) was placed in a NH during the duration of the study. Other research indicates that dually eligible older adults are much more likely to be institutionalized (Clark & Hulbert, 1998). Although our finding may partially reflect the high death rate in our sample (59%), it may also suggest that these beneficiaries with significant long-term care needs may be able to avoid NH placement by utilizing HCBS or AL. Similarly, another study found that individuals at risk of NH placement were largely able to remain in the community and avoid NH placement wi th the provision of publicly-funded HCBS or AL (Chapin, Baca, Macmillan, Rachlin, & Zimmerman, 2009). Avoiding placement in a NH is not only consistent with the preferences of older adults but also has significant cost-saving implications for the Medicare and Medicaid programs. Moreover, the results provide support for current state and federal initiatives that aim to eliminate the institutional bias and expand access to community alternatives to rebalance the delivery of publicly-funded long-term care (e.g., Kane, Kane, Priester, & Homyak, 2008). This study identified other important risk factors for NH placement. Consistent with similar studies of the dually-eligible population (e.g., Andel et al., 2007; Friedman et al., 2005), increasing age, White race, and incontinence increased the risk of institutionalization. The only need for care characteristic that significantly increased the risk of placement was incontinence, which may reflect the overall frailty of our sample. This finding may also reflect policies to discharge residents with incontinence in many AL facilities (Chapin & Dobbs-Kepper, 2001). Limitations Several study limitations should be mentioned. First, the sample consisted of

PAGE 26

18 relatively frail, dually eligible older adults in one state receiving a Medicaid waiver for HCBS or AL. Although this sample is unique, the results cannot be generalized to other publicly-funded or private pay long-term care clients. Second, setting of care and health assessments were derived at baseline. We were not able to capture whether HCBS recipients transitioned into AL, although movement between Medicaid waiver programs in Florida has been estimated at less than 10% (Mitchell, Salmon, Polivka, & SoberonFerrer, 2006). There is a need for future longitudinal studies to use multiple assessments to account for changes in utilization of services and health status over time. Lastly, we were not able to assess the severity of dementia or cognitive impairment. Although we accounted for overall disability with measures of limitations in ADLs and IADLs, the inclusion of dementia severity would be more desirable. Conclusion It is of significance to policy makers and consumers alike to understand the viability of less costly and less institutional settings of care, namely HCBS and AL, to prevent or delay NH placement. Given these settings, the results of this study suggest that individuals with dementia in HCBS remain at risk of NH placement but those in AL may be able to age in place longer. The findi ngs have important implications for understanding the role of HCBS and AL along the continuum of care and improving wellbeing and independence among older adults with dementia.

PAGE 27

19 All HCBS AL p-valueaN 2,4661,630 836 Predisposing Characteristics Age 81.56 (8.56)80.47 (8.67)83.69 (7.57)<.01 Female 77 76 80 0.03 White 67 61 78 <.01 Enabling Resources Married 17 21 9 <.01 Available caregiver 31 40 14 <.01 Need for Care Arthritis 99 99 100 0.10 Cancer 16 17 14 0.09 Dementia 43 33 63 <.01 Diabetes 30 20 34 <.01 Emphysema/COPD 24 26 19 <.01 Heart disease 64 66 62 0.07 Incontinence 55 51 62 <.01 Stroke 30 35 22 <.01 Chronic diseasesb 2.62 (1.09) 2.76 (1.11) 2.36 (0.96)<.01 ADLs 5.57 (2.85) 5.64 (3.10) 5.44 (2.28)0.10 IADLs 12.49 (3.22)11.93 (3.44)13.58 (2.40)<.01 Outcome Nursing home placement 19 15 26 <.01 Notes M = mean; SD = standard deviation; HCBS = home and community-based services; AL = assisted living; COPD = chronic obstruc tive pulmonary disease; ADLs = activities of daily living; IADLs = instrum ental activities of daily living.aChi-square tests were used to examine differences between frequencies; t-test statistics were used to assess mean score differences. % or M ( SD ) Table 2.1 Predisposing, Enabling, and Need Characteristics of the Sample b Composite of arthritis, cancer, diabetes, heart disease, emphysema/COPD, and stroke.

PAGE 28

HRCIp-valueHRCIp-value Predisposing Characteristics Age 1.03(1.02-1.05)<.011.03(1.02-1.05)<.01 Gender (Ref = male) 1.01(0.78-1.31)0.901.02(0.78-1.33)0.89 Race (Ref = non-white) 1.24(1.00-1.53)0.051.23(0.99-1.52)0.06 Enabling Resources Marital status (Ref = not married) 0.99(0.75-1.31)0.950.99(0.76-1.31)0.97 Available caregiver (Ref = no caregiver)0.76(0.60-0.96)0.020.70(0.55-0.88)<.01 Setting of care (Ref = HCBS) 1.18(0.94-1.48)0.171.71(1.29-2.27)<.01 Need for Care Chronic diseasesa1.09(0.99-1.19)0.071.08(0.99-1.18)0.09 Dementia 1.14(0.92-1.42)0.251.70(1.27-2.25)<.01 Incontinence 1.23(1.00-1.51)0.041.27(1.03-1.55)0.02 ADLs 1.03(0.99-1.08)0.121.03(0.99-1.07)0.17 IADLs 1.04(0.99-1.08)0.091.04(0.99-1.08)0.11 Interaction Setting of care x dementia 0.45(0.31-0.66)<.01 -2 log likelihood 5831.84 5815.53 community-based services; ADLs = activities of daily living; IADLs = instrumental activities of daily living. Table 2.2 Risk of NH Placement Model 1 Model 2 Notes N = 2,466. HR = hazard ratio; CI = confidence interval; Ref = reference group; HCBS = home and aComposite of arthritis, cancer, diabetes, heart disease, emphysema/COPD, and stroke. 20

PAGE 29

HRCIp-valueHRCIp-value Predisposing Characteristics Age 1.04(1.02-1.05)<.011.03(1.01-1.05)<.01 Gender (Ref = male) 0.85(0.61-1.17)0.311.54(0.97-2.45)0.07 Race (Ref = non-white) 1.22(0.94-1.58)0.151.34(0.94-1.93)0.11 Enabling Resources Marital status (Ref = not married) 0.92(0.66-1.29)0.631.25(0.77-2.04)0.37 Available caregiver (Ref = no caregiver)0.56(0.42-0.74)<.010.91(0.60-1.37)0.65 Need for Care Chronic diseasesa1.05(0.92-1.17)0.511.17(1.02-1.35)0.03 Dementia 1.50(1.12-2.02)0.010.86(0.63-1.16)0.32 Incontinence 1.20(0.90-1.60)0.211.51(1.12-2.03)0.01 ADLs 0.98(0.92-1.03)0.361.12(1.05-1.19)<.01 IADLs 1.11(1.05-1.17)<.010.94(0.88-1.00)0.04aComposite of arthritis, cancer, diabetes, heart disease, emphysema/COPD, and stroke. Table 2.3 Risk of NH Placement by Setting of Care AL HCBS Notes N = 2,466. HCBS = home and community-based services; AL = assisted living; HR = hazard ratio; CI = confidence interval; Ref = reference group; ADLs = activities of daily living; IADLs = instrumental activities of daily living. 21

PAGE 30

22 Chapter Three: Study II Exploring Correlates of Turnover among Nursing Assistants in the National Nursing Home Survey April Temple, NHA, Debra Dobbs, PhD & Ross Andel, PhD

PAGE 31

23 Abstract Purpose: This study examined nursing home organiza tional characteristics and specific job characteristics of staff in relation to turnover of NAs. Methods : Cross-sectional data on 944 nationally representative nursing homes were derived from the 2004 National Nursing Home Survey. Using a 3-month turnover rate, 25% of facilities with the lowest turnover rates were classified as low turnover, 25% of facilities with the highest turnover were classified as high turnover, and the remaining 50% of facilities were classified as moderate turnover. Multinomial logistic regression was used to examine organizational and job characteristics associated with low and high turnover compared to moderate turnover. Results : One organizational characteristic, total staffing levels at or above 4.0 hours per patient day, was associated with greater odds of low NA turnover and reduced odds of high NA turnover. Job characteristics including higher wages and union membership were associated with greater odds of low NA turnover and higher wages, fully paid health insurance, employee assistance benefits, and involvement in resident care planning were associated with reduced odds of high NA turnover. Implications : The results of this study suggest that job characteristics of NA staff may be particularly important for turnover. Specifically, the provision of competitive wages and benefits (particularly health insurance) and involvement of NAs in resident care planning could potentially reduce NA turnover, as could maintaining high levels of nurse staffing.

PAGE 32

24 Introduction Nursing assistants (NAs) are central to the delivery of quality nursing home (NH) care. Although research and policy have primarily focused on achieving adequate staffing levels to ensure optimal resident care (e.g., Centers for Medicare and Medicaid Services, 2001; Harrington et al., 2000), the presence of high NA turnover may also be deleterious to the provision of resident care (Castle & Engberg, 2005; 2006). Staff turnover of NAs represents a major challenge faced by the NH industry, measuring 71% annually (American Health Care Association, 2003). Furthermore, these staff members generally have the highest turnover rates and provide the majority of direct care to NH residents (Cohen-Mansfield, 1997). Previous research has demonstrated t hat NH organizational characteristics (e.g., size, ownership, staffing) and job characteristics of staff (e.g., wages, involvement in care planning) may explain some of the variation in NA staff turnover rates (BanaszakHoll & Hines, 1996; Brannon, Zinn, Mor, & Davis, 2002; Castle & Engberg, 2006; Kash, Castle, & Phillips, 2007). Although the relationship between organizational characteristics and turnover has been well studied, research including job characteristics of NA staff is limited. Moreover, commonly us ed data sets in turnover research including the Minimum Data Set (MDS), Online Survey, Certification and Reporting data (OSCAR), and state cost reports lack information on job characteristics of NA staff. Thus, the aims of this study were to confir m findings from prior studies on organizational characteristics associated with NA turnover and to explore specific job characteristics associated with turnover of NAs using a large, nationally representative sample of NHs.

PAGE 33

25 Organizational Characteristics Associated with NA Turnover NH organizational characteristics including size and ownership have been found to be associated with NA turnover in prior research. The size of a facility is likely to influence available resources and management practices, and research suggests that turnover is lower in smaller facilities (Castle & Engberg, 2006). For-profit facilities, which are often concerned with profit margins and may not invest as much in their employees, are consistently associated with higher NA turnover (Banaszak-Holl & Hines, 1996; Brannon et al., 2002; Castle & Engberg, 2006). Chain membership has also been associated with higher NA turnover (Castle, 2005), possibly because facilities belonging to chains have greater corporate oversight and stringency that may contribute to staff dissatisfaction. Financial and staffing resources of a NH also represent organizational characteristics that have been found to be associated with turnover of nursing staff. Medicaid provides lower reimbursement fo r NH care than Medicare and private pay, and revenue from these sources is likely to affect the financial resources available to a facility to invest in nurse staffing. For example, higher proportions of Medicaid residents have been associated with greater nursing staff turnover (Harrington & Swan, 2003) whereas higher proportions of private pay residents have been associated with lower NA turnover (Castle, 2005). Facilities with higher occupancy rates, which may be indicative of greater revenue and better performance, have been found to have lower nursing staff turnover (Harrington & Swan, 2003). Higher nurse staffing levels, which likely reduce workload, have also been associated with lower NA turnover (Castle & Engberg, 2006; Donoghue & Castle, 2007). Finally, unstable leadership, as measured by director of nursing and administrator turnover, has previously been associated with an increased likelihood of

PAGE 34

26 high and decreased likelihood of low NA turnover (Castle, 2005). Job Characteristics Associated with NA Turnover The relationship between job characteristics of NA staff and turnover is less clear. Higher wages have been associated with lower NA turnover (Kash et al., 2007), and limited availability of employee benefits in cluding health insurance and pensions has been cited as a factor affecting retention of NAs (Institute of Medicine, 2008; Stone & Weiner, 2001). Moreover, union membership has been associated with lower NA turnover (Brannon et al., 2002), likely because of its effect on wages and benefits. Banaszak-Holl and Hines (1996) found that NA turnover was significantly reduced by involvement in resident care planning, although a similar study by Brannon and colleagues (2002) did not support this finding. In these same studies, increased training, permanent resident assignments, and involvem ent in resident assessments were not associated with NA turnover (Banaszak-Holl & Hines, 1996; Brannon et al., 2002). This study examined nursing home organizational characteristics and specific job characteristics of staff in relation to tur nover of NAs. It builds on prior research by utilizing a large, nationally representative sample of NHs, which allows for broad generalizability of findings. In addition, it incl udes several previously identified correlates of NA turnover within one study, allowing for an assessment of their relative importance. It further extends prior research by exploring more specific job characteristics of NA staff including several employment-based benefits. Theoretical Framework Numerous theoretical models of turnover exist in the literature that attribute turnover to structural, environmental, and individual factors (e.g., Price, 2001; Abelson & Baysinger, 1984). One causal model developed specifically through research in NHs

PAGE 35

27 focuses on a structural approach to explain the variation in nursing staff turnover rates (Halbur, 1983). This model consists of three factors which are assumed to influence turnover: the opportunity structure, organizational structure, and control structure. The opportunity structure broadly refers to labor market conditions (i.e., unemployment rate and availability of alternative employment), the organizational structure includes organizational characteristics (i.e., ownership and size), and the control structure refers to job characteristics (i.e., pay, benefits, promotion opportunities, and involvement in decision-making). All three factors were associated with nursing staff turnover, and the explained variance of turnover significantly improved, particularly for NAs, with the addition of the control structure to a model including the opportunity and organizational structures (Halbur, 1983). We used a modified version of the structural model proposed by Halbur (1983) that focused specifically on the organizational and control structures. Within our framework of analysis (Figure 3.1), the organizational structure was measured by NH organizational characteristics (i.e., size, profit status, chain membership, financial and staffing resources) and the control structure was measured by job characteristics of NA staff (i.e., pay, benefits, unionization, training, and work-related tasks). We expected that both organizational and job characteristics would be associated with NA turnover but that job characteristics would be more important. We thus hypothesized that organizational characteristics would be associated with NA turnover when job characteristics were not examined. We furt her hypothesized that job characteristics would explain the relationship between organiza tional characteristics and NA turnover.

PAGE 36

28 Methods Sample Data were derived from the 2004 National Nursing Home Survey (NNHS), a publicly available dataset provided by the Na tional Center for Health Statistics that surveys a nationally representative sample of NHs in the U.S. The 2004 NNHS was administered between August and December 2004 and includes components on staffing and facility characteristics at the organizational level. The staffing questionnaire is a selfadministered questionnaire completed by the NH administrator. It contains data on education, staffing levels, staff turnover and tenure, staffing practices, and wages and benefits of various categories of staff within the facility. The facility questionnaire is completed by an in-person, computer-assisted interview with the administrator. This component contains organizational characterist ics and services provided by the facility. Both of these components are included toget her in one publicly available data set. For this survey, a sampling frame of 16,628 NHs was stratified by bed size and metropolitan area and then sorted by certification status, hospital relationship, ownership, geographic region, state, county, and zip code. NHs included in the study were selected using systematic sampling based on probabilities proportional to bed size. A total of 1,500 NHs were selected from the sampling frame. Of these, 283 refused to participate and 43 were ineligible for the study. The final sample yielded 1,174 participating NHs included in the NNHS for a response rate of 81%. Participating facilities had at least 3 beds and were certified by Medicare or Medicaid or had a State license to operate. The analytic sample for this study consisted of 944 of the 1,174 nursing facilities (80%) from the NNHS. Due to missing data on study variables, 230 facilities were

PAGE 37

29 deleted. The 230 omitted facilities did not differ significantly ( p > .05) on bed size, profit status, or occupancy rate, although those omitted were more likely to belong to a NH chain ( 2 = 4.45, p = .04). Measures Turnover For NA turnover, the NNHS ascertains the number of certified NAs who left during the last 3-month period for full and part-time staff. To calculate an average 3-month turnover rate for each facility, these full and part-time figures were combined together to arrive at the number of NA full-time equivalents (FTEs) that left employment in the last 3 months and then divided by the number of established positions as given by NA FTEs. This definition of turnover includes both voluntary and involuntary separations from the organization and both full-time and part-time staff on all work shifts. With the exception of a 3-month time period used in the NNHS, this definition is consistent with previous work assessing the measurement of NH staff turnover (Castle, 2006). The 3-month NA turnover rate in this sample ranged from 0% to 148%, with a mean of 19% ( SD = 20%). Organizational Characteristics A total of nine organizational characteristics that may influence turnover of NA staff were included in this study. Several of these were ordinal level variables that needed to be recoded to attain adequate data distribution across levels of measurement. Facility bed size was measured as 3-49 beds, 50-99, 100-199, and >200 beds and was dichotomized for this study into <100 beds (0) and 100 beds (1). Profit status was measured as government or private not-for-profit (0) versus for-profit (1), and chain membersh ip was measured as belonging to a NH chain (1) versus not (0). Medicaid census, or the proportion of Medicaid residents in the facility, was measured as <19%, 20-39%, 40-59%, 60-79%, and 80%. For this analysis,

PAGE 38

30 it was dummy coded into three variables to represent low (0-59%), moderate (60-79%), and high ( 80%) levels of Medicaid occupancy. Medicare census was measured as <9%, 10-19%, and 20% and was recoded into <9% (0) versus 10% (1). Furthermore, total facility occupancy was measured as <70%, 70-79%, 80-89%, 90-94%, and 95%. For this analysis, it was dummy coded into three variables to represent low (<70%), moderate (80-94%), and high ( 95%) facility occupancy levels. The level of total nurse staffing (i.e., RN, LPN, and NA) hours per patient day (HPPD) was also included. This variable was measured as <1.0, 1.0-1.99, 2.0-2.99, 3-3.99, and 4.0 HPPD and was dichotomized to reflect <4.0 HPPD (0) versus 4.0 HPPD (1) of total nursing care based on an optimal level of 4.1 HPPD recommended by the Centers for Medicare and Medicaid Services (2001. Lastly, administrative and nursing stability were included as measured by the NH administrator and director of nursing tenure, respectively, in months at the surveyed facility. Job Characteristics Several job characteristics of NAs at the facility level were available for inclusion in this study. The average NA hourly wage was included as a continuous measure. The following benefits were selected from a list of twelve benefits offered by the facility based on significant correlations ( p < .05) with NA turnover: fully paid health insurance for the employee or fam ily, partially paid health insurance for the employee or family, retirement, paid sick leave, paid personal days, and employee assistance benefits. Partially paid health insurance refers to health benefits in which NA staff were responsible for a portion of the premiums. Few facilities offered fully paid health insurance, so employee only and family coverage were combined. Involvement of NAs in a labor union at the facility was included as a dichotomous variable. Participation of NAs in resident care planning was dummy coded into three variables to represent

PAGE 39

31 seldom/never, some of the time, or always/most of the time involved in resident care planning. The presence of permanent resident assignments, in which NAs are assigned to the care of the same residents, was included as a dichotomous variable. Since the survey did not include specific training of NAs, a dichotomous variable was included to indicate whether a facility had at least one of the following programs which utilized specialty trained staff as a proxy for staff training: hospice, palliative care, pain management, behavioral problem managemen t, skin wounds, continence management, dementia, and restorative care. Lastly, the number of overtime shifts worked by NAs at the facility in the past week was included as a continuous measure. Analyses Descriptive statistics were calculated for all study variables, and bivariate correlations among variables were examined prior to the multivariate analysis to examine the possibility of multicollinearity. The variables showed no substantial indication of collinearity. The highest correlations were found between fully paid health insurance and partially paid health insurance for the employee ( r = -.41) and for tenure of the NH administrator and director of nursing ( r = .37). Several previous studies have demonstrated that turnover is nonlinear and that factors associated with low and high turnover are different (e.g., Brannon et al., 2002; Castle & Engberg, 2006). Therefore, turnover was modeled as a discrete variable with low, moderate, and high categories, where moderate turnover was the reference group. We were particularly interested in identifying organizational and job characteristics that distinguish facilities with the lowest and highest NA turnover. Thus, 25% of facilities with the lowest turnover rates were classified as low turnover (3-month turnover rates less than 6%), 25% of facilities with the highest turnover were classified as high turnover (3-

PAGE 40

32 month turnover rates greater than 25%), and the remaining 50% of facilities were classified as moderate turnover to represent the norm in the sample. Multinomial logistic regression in the SAS procedure CATMOD (SAS Institute, 2003) was used to determine organizational and job characteristics associated with low and high NA turnover relative to facilities wi th moderate NA turnover. Multinomial logistic regression is an extension of binary logistic regression and is appropriate when there are more than two categories of a dependent variable that are mutually exclusive and exhaustive. To test the hypotheses, organizational characteristics were entered into the model first followed by job characteristics to examine the associations of organizational characteristics with NA turnover prior to and subsequent to entering job characteristics. Odds ratios and confidence intervals were calculated for each independent variable for low and high turnover compared to the reference group, moderate turnover. Odds ratios above 1.00 indicate an increase in odds and below 1.00 indicate a decrease in odds of being in the low or high turnover group compared to the moderate turnover group. Results Descriptive statistics for the facility sample are presented in Table 3.1. Nearly half of the facilities surveyed had more than 100 beds and approximately half belonged to a NH chain. A majority (59%) was for-profit and three-fourths had occupancy rates exceeding 80%. These facilities served predominantly Medicaid residents, with the majority having greater than 60% occupancy of Medicaid recipients. Forty-five percent had greater than 10% occupancy of residents reimbursed under Medicare. In terms of clinical resources, only 20% of facilities offered at least 4.0 HPPD of total nursing care. Tenure of the NH administrator and director of nursing in the facility was comparatively long for the industry, with slightly more than 5 years for the NH administrator and

PAGE 41

33 exceeding 3 years for the director of nursing. In relation to job characteristics, NAs were paid an average hourly wage of $8.70 per hour and were largely offered partially paid health insurance for the employee (69%) or for family coverage (58%), retirement (60%), paid sick days (79%), and paid personal days (64%). Relatively few facilities offered fully paid health insurance for the employee or family (13%) and access to employee assi stance programs (30%). Union involvement of NAs was found in only 15% of facilities. The majority of facilities offered programs employing specialty trained staff (80%), suggesting widespread availability of staff training in the facilities. Furthermore, most facilities assigned residents permanently to NAs (68%) and involved NAs in resident care planning some or most of the time (65%). Facilities averaged 10 overtime shifts per week for NA staff. Table 3.2 presents odds ratios and standard errors from the multinomial logistic regression models. To test the hypotheses, the first model examined associations with organizational characteristics only. Larger facility size, moderate occupancy, nursing care above 4.0 HPPD, and longer tenure of the director of nursing reduced the likelihood of high turnover whereas for-profit status increased the likelihood of high turnover compared to moderate turnover. A high Medicaid census and nursing care above 4.0 HPPD increased the likelihood of low tur nover compared to moderate turnover. The main analysis, which included the addition of job characteristics, is presented in the second model in Table 3.2. The second model accounted for more variance in turnover as represented by an increase in pseudo r-squared values from 0.15 to 0.22. Only one organizational characteristic, nursing HPPD, was statistically significant while several job characteristics of staff were significantly associated with NA turnover. Specifically, greater than 4.0 HPPD of nursing care was associated with over a

PAGE 42

34 four-fold increase in the odds of low turnover and a 43% decrease in the odds of high turnover compared to moderate turnover. Othe r organizational characteristics including Medicare census and director of nursing tenure approached statistical significance ( p < .10). NA wage was significantly associated with both low and high turnover. Specifically, a one dollar increase in wages increased the likelihood of low turnover by 21% and decreased the likelihood of high turnover by 20% in comparison to moderate turnover. In addition, NA involvement in a labor union increased the likelihood of being in the low turnover group compared to the moderate turnover group by 66%. Two employmentbased benefits, fully paid health insurance and availability of employee assistance programs, were significantly associated with a reduced odds of high turnover compared to moderate turnover by 50% and 38%, res pectively. Finally, involvement in care planning some or most of the time compared to seldom or never was significantly associated with a lower likelihood of high tu rnover compared to moderate turnover by 44% and 38%, respectively. To assess the robustness of these findings, a sensitivity analysis was conducted (not shown) using cut-off points for low and high turnover based on prior studies (e.g., Brannon et al., 2002; Castle, 2005; Castle & Engberg, 2006). No substantive or directional changes in the results were found, supporting the validity of the findings. Discussion Recruitment and retention of NAs has been an ongoing challenge with implications for the quality of resident care. In this study, we examined NH organizational characteristics and specific job characteristics of staff in relation to turnover of NAs. Only one organizational characteristic (albeit an important one for practice and policy), nurse staffing levels, was associated with NA turnover in the full model. In support of our

PAGE 43

35 hypotheses, several job characteristics including unionization, wages, fully paid health insurance, employee assistance benefits, and involvement in resident care planning were associated with NA turnover. Consistent with our framework of analysis (Figure 3.1), these findings suggest that organizational characteristics are associated with NA turnover but their influence is mostly ex plained by job characteristics of staff. The findings also suggest that job characteristics appear to be particularly important for turnover of NAs. A practical aspect of our findings is that the organizational and job characteristics associated with NA turnover are all potentially modifiable, and the resulting implications are discussed below in detail. The results of this study indicate that high levels of nurse staffing increased the likelihood of low turnover and decreased the likelihood of high turnover of NAs. Others have also found that greater staffing levels are associated with lower NA turnover (e.g., Castle, 2005; Castle & Engberg, 2006; Donoghue & Castle, 2007), although it is possible that a reciprocal relationship exists as high turnover has also been shown to have a negative effect on nurse staffing levels (Harrington & Swan, 2003). A possible explanation for the finding in this study is that greater staffing levels reduce workload and allow NAs to spend more time providing quality care and interacting with residents, which have previously been shown to be important to NA staff (Bowers, Esmond, & Jacobson, 2000; Chou, Boldy, & Lee, 2002). Thus, staffing at higher levels may potentially increase NA job satisfaction and subsequently reduce turnover. An important policy implication is that mandated staffing at or above the 4.1 HPPD recommended by the Centers for Medicare and Medicaid Services (2001) may not only improve resident care but may also reduce turnover. Although increasing staffing levels may initially be costly, it may save expenses associated with NA turnover which is estimated at $4.1

PAGE 44

36 billion annually (Seavey, 2004). Similar to the findings of Kash and colleagues (2007), this study provides evidence that hourly pay is an important fact or contributing to NA turnover. Moreover, our finding that union involvement was associated with lower turnover further underscores the importance of pay, as unions tend to improve pay. Increasing wages may be a promising solution to the turnover crisis, yet it is also a challenging endeavor for many NHs that rely on fixed payments fr om Medicaid and Medicare. One possibility is for NH providers to reduce unnecessary spending in other cost centers, such as advertising expenses (Kash & Boyer, 2008), to increase spending on NA wages. Other initiatives that have successfully been used to increase wages of direct care workers include state utilization of wage pass-throughs to increase reimbursement from public funds, collective bargaining from unions or consumer groups, and improvements in minimum wage laws (Seavey & Salter, 2006). Policies such as these should continue to be implemented and dedicated to increasing NA wages. In addition, availability of fully paid health insurance reduced the likelihood of high turnover. These findings provide evidence for the importance of health benefits on staff retention in the NH industry. The finding that only fully paid health insurance was significant suggests that even partially paid health insurance may not be as important to NAs, likely because of difficulty paying their portion of premiums. Ongoing strategies to increase the availability and affordability of health insurance coverage for direct care workers are needed and include subsidizi ng employer-sponsored health plans, expanding employer-based coverage, and increa sing eligibility for publicly-funded plans (Seavey & Salter, 2006). On a much broader level, health care reform by policy makers is needed to improve the accessibility and affordability of health insurance, particularly

PAGE 45

37 for low wage workers such as NAs. These findings are particularly relevant considering the increasing expenditures on health care, and national health care policy decisions will likely influence actions within the NH industry to offer affordable health insurance to its employees. Findings from this study also highlight the importance of employee assistance programs, an employment-based benefit. Specific ally, the availability of these programs reduced the likelihood of high turnover co mpared to moderate turnover. NAs face several personal and job-related stressors that may affect their ability to stay in their jobs. For example, Foner (1994) found that NAs face a multitude of personal problems including difficulties with finances, housing, child care, transportation, and domestic relationships. In the workplace, stressors may include demanding work schedules (Geiger-Brown, Muntaner, Lipscomb, & Trinkoff, 2004), heavy workloads (Chou et al., 2002), and the physically and emotionally cha llenging tasks of resident care (Myers, Silverstein, & Nelson, 2002). It may be that these employee assistance programs help NA staff more effectively manage their personal and work-related stress, which may increase job satisfaction and reduce turnover. Providers should consider the addition of this employment-based benefit to potentially reduce NA turnover. Involvement in resident care planning significantly reduced the likelihood of high NA turnover, which is consistent with the findings of Banaszak-Holl and Hines (1996). Similarly, culture change movements such as the Wellspring model which involve direct care workers in decision-making have been found to successfully reduce staff turnover (Stone et al., 2002). Involvement in decis ion-making provides NAs with greater responsibility and empowerment, which may subsequently increase job satisfaction and reduce turnover. Involving NAs in resident care planning on a regular basis may be an

PAGE 46

38 important and cost-effective intervention to implement in NHs to reduce turnover. Limitations There are a few notable limitations to this study. First, the study is crosssectional, so only associations between or ganizational and job characteristics and NA turnover can be identified. For example, we cannot infer from the cross-sectional findings that higher wages reduce turnover since low turnover may also contribute to higher average wages. The measure of turnover also included both voluntary and involuntary separations from the facility, which likely have different correlates (Castle, 2006). Future research would benefit from utilizing longitudinal data to examine predictors of voluntary or involuntary turnover of NA staff. There are additional limitations with the NNHS public use data. First, many of the variables were measured at the ordinal level when a continuous measurement may have been preferable. There may also be other potentially important organizational or job characteristics related to NA turnover that were not available in the data, such as acuity of residents (Castle, 2005) or quality of care indicators (Castle & Engberg, 2006). Moreover, we could not control for local market characteristics due to lack of information about facility location. Others have found that market factors such as unemployment rate, per capita income, and number of local nursing home beds may affect turnover rates (Banaszack-Holl & Hines, 1996; Castle & Engberg, 2006), although it is unlikely that these controls would have changed the fi ndings. Lastly, we were not able to include individual-level factors such as job satisfaction or intent to leave, which may also explain turnover (Price, 2001). Future research efforts are needed to examine facility, market, and individual level factors associated with turnover within the same model.

PAGE 47

39 Conclusion More direct care workers will be needed in the coming decades to provide care for the increasing older adult population. Addressing the high turnover of these paraprofessionals may be important to ensure quality of resident care. The results of this study show that NH organizational and job characteristics of staff are important factors associated with turnover of NAs. Specifically, nurse staffing levels, wages, health insurance, employee assistance benefits, unionization, and involvement in resident care planning were significantly associated with NA turnover. Attention to these organizational and job characteristics may help reduce tur nover and subsequently improve the delivery of resident care.

PAGE 48

Table 3.1 Nursing Home OrganizationalCharacte ristics, Job Characteristics, and Turnover Variable Operational Definition % or M (SD) Organizational Characteristics Facility size 100 beds 48 Chain membership Member of a nursing home chain 51 For-profit ownership For-profit ownership of facility 59 Occupancy Proportion of beds utilized by residents < 80% 24 80-94% 42 95% 34 Medicaid census Proportion of residents paid by Medicaid < 60% 37 60-79% 42 80% 21 Medicare census 10% of residents paid by Medicare 45 Nurse HPPD 4.0 HPPD of total nursing staff (RN+LPN+NA) 20 NHA tenure Nursing home administrator months at current facility64.3 (74.6) DON tenure Director of nursing months at current facility 41.0 (47.8) Job Characteristics NA wage Average NA wage per hour 8.7 (1.5) Full insurance Fully paid health insurance for employee or family 13 Partial employee insurancePartially paid health insurance for employee 69 Partial family insurancePartially paid health insurance for employee's family 58 Retirement Retirement or pension program 60 Sick days Paid sick days 79 Personal days Paid personal days 64 Employee assistanceEmployee assistance program 30 NA union Presence of NA union involvement 15 Staff training Facility offers programs with specialty trained staff 80 Resident assignmentNA assigned to same residents 68 Care planning NA frequency of involvement in resident care planning Seldom/never 35 Some of the time 28 Always/most of the time 37 Overtime shifts NA overtime shifts per week at facility 10.1 (19.8) Turnover Low turnover 3 month NA turnover rate < 6% 25 Moderate turnover 3 month NA turnover rate 6% and 25% 50 High turnover 3 month NA turnover > 25% 25 Notes N = 944 facilities. M = mean; SD = standard deviation; HPPD = hours per patient day; RN = registered nurse; LPN = licensed practical nurse; NA = nursing assistant; NHA = nursing home 40 administrator; DON = director of nursing.

PAGE 49

OR(SE)OR(SE)OR(SE)OR(SE) Organizational Characteristics Facility size (Ref < 100 beds) 0.82(.17)0.70(.17)*0.73(.20)0.74(.19) Chain membership (Ref = no) 0.89(.19)0.84(.19)0.96(.19)0.87(.20) For-profit ownership (Ref = nonprofit)0.81(.19)1.96(.20)**0.85(.22)1.41(.23) Occupancy (Ref < 80%) 80-94% 0.89(.22)0.67(.21)*0.91(.23)0.75(.22) 95% 1.29(.23)0.72(.22)1.26(.24)0.90(.24) Medicaid census (Ref < 60% ) 60-79% 0.94(.20)1.08(.19)0.90(.20)1.00(.20) 80% 1.64(.25)*1.11(.25)1.42(.26)0.95(.27) Medicare census (Ref < 10%) 1.35(.19)1.26(.19)1.38(.19)^1.29(.20) Nurse HPPD (Ref < 4.0) 4.35(.19)**0.59(.27)*4.60(.20)**0.57(.28)* NHA tenure 1.00(.01)1.00(.01)1.00(.01)1.00(.01) DON tenure 1.00(.01)0.99(.01)*1.00(.01)0.99(.01)^ Job Characteristics NA wage 1.21(.06)**0.80(.07)** Full insurance (Ref = no) 1.05(.28)0.50(.35)* Partial employee insurance (Ref = no) 0.86(.22)0.95(.21) Partial family insurance (Ref = no) 0.81(.20)0.82(.20) Retirement (Ref = no) 0.88(.20)0.98(.19) 1.12(.24)0.68(.21)^ Personal days (Ref = no) 1.12(.19)1.15(.18) Employee assistance (Ref = no) 0.85(.20)0.62(.23)* NA union (Ref = no) 1.66(.25)*1.04(.29) Staff training (Ref = no) 0.73(.22)1.01(.22) Resident assignment (Ref = no) 1.26(.20)1.32(.19) Care planning (Ref = Seldom/never) Some of the time 0.83(.23)0.56(.22)** Always/most of the time 1.14(.20)0.62(.20)* Overtime shifts 1.00(.01)1.00(.01) Pseudo R41 2 Notes N = 944 facilities. NA = nursing assistant; OR = odds ratio; SE = standard error; Ref = reference; HPPD = hours per patient day; NHA = nursing home administrator; DON = director of nursing; 0.15 0.22 Sick days (Ref = no) Table 3.2 Organizational and Job Characteristics Associated with Low and High NA Turnover Compared to Moderate NA Turnover LowHigh High Low Model 1 Model 2 ^p<.10; *p<.05; **p<.01.

PAGE 50

42 Figure 3.1 Framework of Analysis for Nursing Assistant (NA) Turnover

PAGE 51

43 Chapter Four: Study III Nursing Home Characteristics and the Availability of Benefits for Nursing Assistants: Results from the National Nursing Home Survey April Temple, NHA, Debra Dobbs, PhD & Ross Andel, PhD

PAGE 52

44 Abstract Purpose: The purpose of this study was to descr ibe the presence of employment-based benefits of nursing assistants (NAs) and to explore nursing home organizational characteristics associated with benefits offered to these staff. Methods : Cross-sectional data on 944 nationally representative nursing homes were derived from the 2004 National Nursing Home Survey. Benefits were measured as a composite variable of five NA benefits weighted by cost. Multivariate ordinary least squares regression was used to examine the relationship between organizational characteristics and NA benefits. Results : For-profit ownership and Medicaid occupancy were negatively associated with NA benefits. Facility size, chain membership, occupancy level, union involvement, and education of the nursing home administrator were positively associated with NA benefits. Implications : Organizational characteristics may explain some of the variation in the availability of NA benefits across nursing homes and may be informative for policy and practice to improve NA staff recruitment and retention. In particular, facilities with forprofit ownership and high proportions of Medicaid residents should consider costeffective ways to enhance the benefit package offered to NAs.

PAGE 53

45 Introduction Nursing assistants (NAs) deliver the majority of direct care that is central to a residents quality of care and quality of life in a nursing home (NH). With respect to maintaining a qualified direct care workforce, high vacancy and turnover rates represent major challenges faced by the NH industry. NAs hold approximately 1.1 million jobs in NHs (Smith & Baughman, 2007), yet vacancy rates indicate this number is insufficient to meet the current demand (American Health Care Association, 2003). Furthermore, historically high turnover rates of NAs, measuring 71% annually across the nations NHs, exacerbate the unmet need (American Health Care Association, 2003). As the demand for these paraprofessional staff is projected to increase with the aging of the baby boomers (U. S. Bureau of Labor Statistics, 2000; U. S. Department of Health and Human Services, 2003), attention to recruitment and retention of NAs is of importance. A significant factor affecting the recruitment and retention of NAs that has received less attention in the literature is the limited availability of employee benefits (U. S. Department of Health and Human Services, 2003; Stone & Weiner, 2001). For example, less than half of NAs receive health insurance through their employer (Potter, Churilla, & Smith, 2006; Smith & Baughman, 2007), and three-fourths have no pension benefits (U.S. General Accounting Office, 2001). A recent report by the Institute of Medicine (2008) highlighted the importance of benefits for the recruitment and retention of direct care workers including NAs and recommended increasing the availability of benefits. However, little is known about the range of benefits offered to NAs and the variation of benefits across the NH industry. The primary purpose of this study was to examine employment-based benefits offered to NAs and to explore the relationship between NH organizational characteristics and the availability of NA staff benefits.

PAGE 54

46 The Importance of Benefits Employment-based benefits are an important component of employee compensation. In addition to those that ar e legally required, benefits including health insurance, paid time off, retirement programs, and other fringe benefits constitute approximately 30% of employee compensation (U.S. Bureau of Labor Statistics, 2007a). The provision of benefits has been associated with improvements in job satisfaction (Price, 2001), job performance (Lambert, 2000), and recruitment and retention of employees (Sethi, 2001; Spetz & Adams, 2006). In the NH setting, recent studies have found that benefits were associated with fewer intentions to leave (Stearns & DArcy, 2008) and lower turnover among NAs (Temple, Dobbs, & Andel, 2009). Furthermore, research resulting from the Better Jobs Better Care initiative found that benefits were positively associated with job commitment (Bishop et al., 2008) and job satisfaction (Ejaz, Noelker, Menne, & Bagakas, 2008) of direct care workers. Theoretical Framework for the Availability of Benefits in Nursing Homes For decades now, critical perspectives on social policy and aging have been used to explain the quality of care delivery for older adults, including care provided in NHs (e.g., Estes, 1979; Estes, Harrington, & Pellow, 2001; Estes, Wallace, Linkins & Binney, 2001). The political economy of aging perspective (Estes, 1999a; 1999b) emphasizes the influence of political, ec onomic, and sociocultural factors on the distribution of resources for the care of older adults. Within this framework, a marketdriven healthcare system reinforced by public policies has facilitated the creation of the aging enterprise (Estes, 1979; Estes, 1999b; Estes, Harrington, et al., 2001). The aging enterprise refers to burgeoning industries that have had a considerable financial interest in serving the needs of older adults (Estes, 1979; Estes, 1999b). Thus

PAGE 55

47 healthcare, including NH care, has become a profitable commodity through cost containment measures, growth in the for-profit sector, and corporate consolidation (Estes, 1999b; Estes, Harrington, et al., 2001). These market-driven changes in the industry have influenced the organizational context and operations of many NHs with pot entially negative consequences for staffing. Consistent with the political economy of aging perspective, NH structural characteristics including size, profit status, chain membership, and occupancy by payer source may provide insight into the variation of NH investments in benefits of NA staff. Given the relationship between benefits and staff retention found previously (Bishop et al., 2008; Stearns & DArcy, 2008; Temple et al., 2009), understanding correlates of the extent of benefit provision across facilities may have important implications for strategies to improve staff retention. Nursing Home Characteristics and the Availability of Benefits Aging policies in the U.S. have reinforced market-driven systems over the years, which has created the aging enterprise and resulted in a NH industry that is mostly profit motivated (Estes, Harrington, et al., 2001). The NH industry experienced a significant increase in for-profit and chain membership from the 1970s to 1990s (Estes, Harrington, et al., 2001). For-profit facilities are primarily concerned with profit maximization and as a result may not invest as much in employee compensation as nonprofits. Using 54 NHs from Wisconsin, Haley-Lock and Kruzich (2008) found that public and nonprofit NHs were associated with better wages and health insurance contributions for NAs. Similarly, Kash and colleagues (2007) found that expense ratios related to nursing staff benefits was significantly higher in nonprofit versus for-profit NHs. In contrast to these findings, Hunter (2000) did not find an association between profit status and an index of NA

PAGE 56

48 wages and benefits in a sample of 149 NHs in Massachusetts. In addition to profit status, NHs belonging to a multi-facility chain overseen by the same corporate organization may benefit from greater economies of scale and sources of capital. This may explain the finding that those facilities belonging to a corporate chain are more likely to provide better NA wages and benefits (Haley-Lock & Kruzich, 2008). With the goal of profit maximization present in the aging enterprise (Estes, 1979; Estes, 1999b), sources of revenue may also affect the capacity of NHs to invest in benefits for NA staff. Facilities with larger numbers of beds and higher occupancy rates may be indicative of greater revenue and better performance, and research has shown that these characteristics are associated wi th better NA wages and benefits (Haley-Lock & Kruzich, 2008). Occupancy rates by payer source may also influence profits and expenditures on staffing. Medicaid is the lowest third-party payer of NH care in contrast to Medicare and private pay, and research has shown that greater proportions of private pay residents have been associated with higher job quality of NAs in terms of wages and benefits (Hunter, 2000). Other important organizational influences on the availability of staff benefits may exist. Because of the focus on profit maximization apparent in the aging enterprise (Estes, 1979), collective bargaining of employee unions may have an important influence on minimum wages and benefits offered to staff. For example, union presence was associated with higher job quality of NAs in one study (Hunter, 2000). Additionally, NH administrators with better training and education may be better able to balance the competing demands of profit maximization and investment in human capital. One study found that NH administrators that completed a standardized training program provided better wages and benefits to NAs (Hunter, 2000) while another study found that NH

PAGE 57

49 administrators with Masters degrees improved job quality of NAs only in nonprofit facilities (Haley-Lock & Kruzich, 2008). The Present Study Although some research suggests that organizational characteristics may play a role in the availability of benefits offered to NA staff across NHs, previous studies (e.g., Hunter, 2000; Haley-Lock & Kruzich, 2008) have been limited by small samples from a single state and the inclusion of only a few benefits in the outcome measure. Using the political economy of aging as a framework, the present study explored the presence of a full range of NA benefits and investigated the relationship between NH organizational characteristics and benefits offered to NA staff. It utilized a large, nationally representative sample of 1,174 NHs, allowing for generalizability of the findings. In addition, it included several benefits in the outcome measure that are likely considered the most important to employees. Consistent with the framework of the political economy of aging perspective (Estes, 1979; 1999b; Estes, Harrington, et al., 2001), we expected that the NH organizational context would be related to the provision of benefits. Specifically, we hypothesized that organizational characteristics including profit status, chain membership, occupancy by payer source, and union membership would be associated with the availability of benefits offered to NAs. Methods Sample Data were derived from the 2004 National Nursing Home Survey (NNHS), a publicly available dataset provided by the Na tional Center for Health Statistics that surveys a nationally representative sample of NHs in the U.S. The 2004 NNHS was administered between August and December 2004 and includes components on staffing

PAGE 58

50 and facility characteristics at the organizational level. The staffing questionnaire is a selfadministered questionnaire completed by the NH administrator. It contains data on education, staffing levels, staff turnover and tenure, staffing practices, and wages and benefits of various categories of staff within the facility. The facility questionnaire is completed by an in-person, computer-assisted interview with the administrator. This component contains organizational characterist ics and services provided by the facility. Both of these components are included together in one publicly available dataset. For this survey, a sampling frame of 16,628 NHs was stratified by bed size and metropolitan area and then sorted by certification status, hospital relationship, ownership, geographic region, state, county, and zip code. NHs included in the study were selected using systematic sampling based on probabilities proportional to bed size. A total of 1,500 NHs were selected from the sampling frame. Of these, 283 refused to participate and 43 were ineligible for the study. The final sample yielded 1,174 participating NHs included in the NNHS for a response rate of 81%. Participating facilities had at least 3 beds and were certified by Medicare or Medicaid or had a State license to operate. The analytic sample for this study consisted of 944 of the 1,174 NHs (80%) from the NNHS. Due to missing data on study variables, 230 facilities were deleted. The 230 omitted facilities did not differ significantly ( p > .05) on bed size, profit status, or occupancy rate, although those omitted were more likely to belong to a NH chain ( 2 = 4.45, p = .04). Measures Benefits For the main analysis, benefits were measured as a composite variable based on the following five weighted benefits: health insurance, retirement, paid

PAGE 59

51 vacation, paid sick days, and paid personal days. These benefits, which are among the most important in employee compensation (Christensen, 2002), were selected based on availability of weightings. The weightings were derived from the hourly cost in dollars of these benefits to employers in the privat e industry from the Employer Costs for Employee Compensation (ECEC; U.S. Bureau of Labor Statistics, 2007a). Costs for private industry were selected because NHs in the sample are largely based in the private industry. Weighting these benefits in terms of their respective costs to employers allowed for assessing the relative value of each benefit and capturing greater variation in the outcome rather than summing the number of available benefits. For health insurance, facilities received a corresponding weight based on the most comprehensive level of health coverage offered to NAs: partially paid health insurance for the employee, fully paid health insurance for the employee, partially paid health insurance for the employees family, and fully paid health insurance for the employees family. The ECEC includes only an average hourly cost to employers for all health plans combined ($1.85; U.S. Bureau of Labor Statistics, 2007a). However, monthly employer costs of the four levels of health insurance coverage are included in the National Compensation Survey (NCS; U. S. Bureau of Labor Statistics, 2007b), from which the ECEC is derived. Using this information, the estimated hourly costs of the four plans in this study were averaged to equal the hourly cost of all health insurance plans presented in the ECEC and were calculated in the same relative proportions as the monthly costs in the NCS. For example, the least expensive plan, partial employee coverage, is approximately 33% of the monthly cost of the most expensive plan, full family coverage. Therefore, the calculated hour ly cost of partial employee coverage is also 33% of the calculated hourly cost of full family coverage.

PAGE 60

52 Facilities received the corresponding weight in Table 4.1 if they offered the benefit to NA employees. The total benefits score for each facility was calculated as the sum of the weights of available benefits with a range from 0.06 to 4.96 and a mean of 3.27 (SD = 1.11). There were additional benefits surveyed in the NNHS which could not be included in the benefits score due to lack of cost data to generate weights. However, these benefits were still included in this study to explore the full range of NA staff benefits. These benefits included child daycare, transportation allowance, employee assistance, and career promotion. Overall, the most important benefits from an employees perspective were included in the benefits score (Christensen, 2002). Organizational Characteristics. A total of eight organizational characteristics of NHs were included in this study, several of which needed to be recoded to attain adequate data distribution across levels of measurement. Facility bed size was measured as 3-49 beds, 50-99, 100-199, and >200 beds and was dichotomized for this study into <100 beds (0) and 100 beds (1). Profit status was measured as government or private not-for-profit (0) versus for-profit (1), and chain membership was measured as belonging to a NH chain (1) versus not (0). Medicaid census, or the proportion of Medicaid residents in the facility, was measured as <19%, 20-39%, 40-59%, 60-79%, and 80%. For this analysis, it was dummy coded into three variables to represent low (0-59%), moderate (60-79%), and high ( 80%) levels of Medicaid occupancy. Medicare census was measured as <9%, 10-19%, and 20% and was recoded into <9% (0) versus 10% (1). Total facility occupancy was measured as <70%, 70-79%, 80-89%, 9094%, and 95%. For this analysis, it was dummy coded into three variables to represent low (<70%), moderate (80-94%), and high ( 95%) facility occupancy levels. The

PAGE 61

53 education of the NH administrator was dichotomized as having a graduate degree (1) versus an undergraduate degree or lower level of education (0). Lastly, involvement of NAs in a labor union was included as a dichotomous variable. Analyses Descriptive statistics were calculated for all study variables, and bivariate correlations among variables were examined prior to the multivariate analysis to examine the possibility of multicollinearity. The variables showed no substantial indication of collinearity. The highest correlations were found between chain membership and for-profit ownership ( r = .38) and between Medicaid and Medicare census (r = -.34). All of the organizational characteristics with the exception of chain membership were significantly correlated with the benefits score. Ordinary Least Squares regression in the SAS program procedure REG (SAS Institute, 2003) was used to examine associations between organizational characteristics and the benefits score. Ordinary Least Squares regression examines the relationship between each independent variable and the dependent variable while controlling for all the other independent variables in the model. To test the hypothesis, all organizational characteristics were entered into the model, and two-tailed tests of significance ( p < .05) were conducted to determine the statistical significance of each independent variable. The results are presented as unstandardized and standardized regression coefficients for each independent variable. The unstandardized regression coefficients represent the amount of change in the benefits score for a one-unit change in each respective independent variable whereas the standardized regression coefficients allow for the comparison of the relative influence of each independent variable on the outcome.

PAGE 62

54 Results Descriptive statistics for the facility sample are presented in Table 4.2. Nearly half of the facilities surveyed had more than 100 beds and approximately half belonged to a NH chain. A majority (59%) was for-profit and three-fourths had occupancy rates exceeding 80%. These facilities served predominately Medicaid residents, with the majority having greater than 60% occupancy of Medicaid recipients. Forty-five percent had greater than 10% occupancy of residents reimbursed under Medicare. Approximately 29% of NH administrators had a graduate degree. NA union involvement was present in only 15% of facilities. In relation to the availability of NA benefits (Figure 4.1), the majority of facilities (55%) offered partially paid health insurance for family coverage as the most comprehensive health plan followed by partially paid health insurance for the individual employee (20%). Few facilities offered fully paid health insurance for the employee (10%) or family (3%), and approximately 12% offered no form of health insurance. Nearly all facilities offered paid vacation days (96%), and the majority offered retirement or pension plans (60%), paid sick days (79%), paid personal days (64%), and career promotion opportunities (71%). Relative ly few NHs offered access to employee assistance programs (30%), child daycare assistance (5%), or transportation allowance (5%). Table 4.3 presents the unstandardized and standardized regression coefficients from the multivariate regression mo del examining the association between organizational characteristics and NA benefits. Facility size and chain membership were positively associated with the NA benefits score, whereas for-profit ownership was negatively associated with the benefits score. NHs with higher occupancy levels were

PAGE 63

55 associated with a higher benefits score whereas NHs with higher Medicaid occupancy levels were associated with a lower benefits score for NA staff. There was no association with Medicare occupancy. The presence of NA union involvement and graduate-level education of the nursing home administrator were both associated with a higher NA benefits score. With respect to the standardized regression coefficients, forprofit ownership and higher occupancy levels had the highest levels of association with the NA benefits score. The model accounted for 16% of the variation in the benefits score (Adjusted R2 = .16). Discussion Recruitment and retention of NAs has been an ongoing challenge with direct implications for maintaining a qualified workforce to provide long-term care to the increasing population of older adults. Since employee benefits may be important for NA job satisfaction (Ejaz et al., 2008) and retention (Bishop et al., 2008; Stearns & DArcy, 2008), we examined the availability of NA staff benefits and explored the relationship between NH organizational characteristics and NA benefits in a large, nationally representative sample of NHs. In support of our hypotheses, we found that several organizational characteristics including size, profit status, chain membership, occupancy levels, Medicaid occupancy, NA unionization, and education of the NH administrator were associated with the level of benefits offered to NA staff. These findings are consistent with theoretically-based expectations within the political economy framework (e.g., Estes, 1999a) that systemic differences of NHs in the aging enterprise may influence the availability of benefits for NA staff. The results of this study indicate that for-profit facilities offered fewer benefits to NA staff, and that the magnitude of this association was among the largest in the

PAGE 64

56 analytic model. Previously, Haley-Lock and Kruzich (2008) found that public and nonprofit NHs provide better compensation to NAs in terms of wages and health insurance contributions. In context of t he political economy framework (Estes, 1979; Estes, 1999a; Estes, Harrington, et al., 2001), the provision of fringe benefits may be counter to profit maximization present in the aging enterprise. Nonprofits may invest more in staff compensation due to ideological differences in mission as well as from favorable tax exemptions (Estes, Alford, & Egan, 2001). It may be advantageous for forprofit facilities to model nonprofit facilities in their provision of benefits to recruit and retain NAs. Although this may initially be detrimental to profits, providers may save direct and indirect expenses associated with NA turnover at an estimated $3,500 per worker (Seavey, 2004). Similar to the findings of Haley-Lock and Kruzich (2008), this study provides additional evidence that chain membership may be positively associated with NA benefits. With the transformation of health services to a profit-seeking aging enterprise (Estes, Alford, et al., 2001), there has been a decrease in the number of free standing facilities and an increase in multi-facility chains. Multi-facility chains are more likely to take advantage of greater economies of scale and sources of capital, which may enable these facilities to offer more generous benefits. For example, NHs belonging to a chain may be able to negotiate more competitive health insurance plans for their employees. Therefore, free standing facilities may benefit economically from pooling together with other affiliated organizations to increase the availability of staff benefits. NHs generally target clients and services toward those that generate the highest profits (Estes, Alford, et al., 2001). In this study, we found that larger bed size and higher occupancy levels were positively associated with the level of NA benefits. In contrast,

PAGE 65

57 facilities with higher levels of Medicaid occupancy, the lowest third-party payer, offered fewer NA benefits. Others have also found these relationships between size and occupancy measures and the wages and benefits of NAs (Haley-Lock & Kruzich, 2008; Hunter, 2000). It may be that larger facilities and highly occupied facilities bring in greater sources of revenue, which in turn could be used to invest in staff benefits. Contrastingly, higher proportions of Medicaid residents may be indicative of facilities that bring in lower revenues, thus not able to offer as many fringe benefits to staff. In light of these findings, it may be important for policy makers to recognize that low Medicaid reimbursement rates may contribute to fewer benefits for NAs. In a previous study, union membership was associated with lower turnover among NAs (Brannon, Zinn, Mor, & Davis, 2002), which is likely due to the fact that unions tend to improve pay and benefits. Results from this study provide evidence that facilities in which NAs are members of a union offer higher levels of benefits to staff. Similarly, Hunter (2000) found that unionization was associated with better job quality of NAs in terms of wages and benefits. In relation to the commodification of care delivery in the aging enterprise (Estes, Harrington, et al., 2001), collective bargaining from unions may be one of few alternatives to increase spending on NA staff benefits. Thus, investment in employee benefits for NAs may be one strategy for providers to avoid unionization in their facilities. Lastly, education of the NH administrator was positively associated with NA benefits in the fully adjusted model. Specifically, facilities with administrators earning a graduate level of education were associated with a higher level of benefits. This association was also found in a study by Haley-Lock & Kruzich (2008), although the relationship existed only among administrators in nonprofit facilities. In a similar study,

PAGE 66

58 Hunter (2000) found a positive association between job quality of NAs and NHs with administrators who completed a state-approved training program. Administrators who are highly educated may possess additional knowledge in business or human resource management that allows them to recognize the importance of employee compensation in addition to earning profits. More research is needed in this area to understand the contributions of education of the administrator on staffing decisions. Limitations There are a few notable limitations to this study. First, the study is crosssectional, so only associations between organizational characteristics and NA benefits can be identified. For example, we cannot infer from the cross-sectional findings that higher occupancy levels increase benefits offered to staff. Future research would benefit from utilizing longitudinal data to examine predictors of NA benefits. In addition, there may be other important organizational characteristics related to NA benefits that were not available in the data, such as facility expense ratios or profit margins. Moreover, we could not control for local market conditions due to limitations of the data. Future research efforts are needed to examine additional organizational and market characteristics associated with benefits provision within the same model. Lastly, we could not include all of the available benefits in the benefits score due to lack of data to generate weightings. Nonetheless, we feel we included the most important NA benefits in the outcome measure based on previous research (Christensen, 2002). Despite these limitations, this study also has important strengths. To the best of our knowledge, this is the first study to explore the prevalence of a range of benefits offered to NA staff, which could be useful for future comparisons. In addition, this study utilized a large, nationally representative sample of NHs from the NNHS and included

PAGE 67

59 the most important employee benefits in the outcome measure. Finally, the weighting of the benefits score by cost to employers represents a novel way to capture the relative worth of various benefits offered to NA staff. A direction for future research would be to determine the value of various benefits from the perspective of NAs. NAs may place little importance on select benefits or have low expectations of employers to offer certain benefits, which may also influence the types of benefits that are widely available to these staff. Conclusion Inadequate staffing levels and high turnover of NAs has direct implications for the quality of resident care (Castle, 2008; Castle & Engberg, 2005; Harrington, Zimmerman, Karon, Robinson, & Beutel, 2000). Increasing the availability of employment-based benefits is receiving greater attention as one possible solution to address the ongoing shortage and high turnover of NAs (Institute of Medicine, 2008). The results of this study suggest that NH organizational characteristics may explain some of the variation in the availability of benefits offered to NA staff. Attention to these systemic differences may be informative for strategies for providers and policy makers to enhance the provision of benefits to recruit and retain NA staff with the intention of improving quality of care.

PAGE 68

60 Table 4.1 Weighting of NA Benefits Benefit Cost/MoaCost/Hrb Health insurance Partial employeec 265.74 0.94 Full employeec 382.19 1.34 Partial familyc 642.02 2.26 Full familyc 814.44 2.86 Retirement 0.92 Paid vacation 0.90 Paid sick days 0.22 Paid personal days 0.06aNational Compensation Survey. bEmployer Costs for Employee Compensation. cCost per month used to calculate cost per hour where the average is $1.85.

PAGE 69

61 nursing assistant. Table 4.2 Organizatinal Characteristics of the Sample Variable Operational Definition % Facility size 100 beds 48 For-profit ownershipFor-profit ownership of facility 59 Chain membershipMember of a multi-facility chain 51 Occupancy Percentage of beds utilized by residents 79% 24 80-94% 42 95% 34 Medicaid censusPercentage of residents paid by Medicaid 59% 37 60-79% 42 80% 21 Medicare census 10% of residents paid by Medicare 45 NHA education NHA earned a Master's degree or higher29 NA union Presence of NA union involvement 15 Notes N = 944 facilities. NHA = nursing home administrator; NA =

PAGE 70

62 *p<.05; **p<.01; ***p<.001. b SE Facility size (Ref < 100 beds) .20.07.09** For-profit ownership (Ref = nonprofit) -.50.08-.22*** Chain membership (Ref = no) .25.07.11*** Occupancy (Ref 79%) 80-94% .33.09.15*** 95% .54.09.23*** Medicaid census (Ref 59% ) 60-79% -.19.08-.08* 80% -.31.10-.11** Medicare census (Ref < 10%) .09.08.04 NHA education (Ref < graduate degree) .21.07.09** NA union (Ref = no) .47.10.15*** Notes N = 944 facilities. NA = nursing assistant; b = unstandardized regression coefficient; SE = standard error; = standardized regression coefficient; Ref = reference; NHA = nursing home administrator. Adjusted R2 = 0.16 Table 4.3 Organizational Characteristics Associated with NA Benefits

PAGE 71

63 Figure 4.1 Percentage of Nursing Homes with Benefits Offered to Nursing Assistants 0 10 20 30 40 50 60 70 80 90 100Part Emp Health Part Fam H ealth F u l l E m p Healt h Full Fam Health Retirement S ic k P ers onal Vacation A s s ist an ce Day car e Transportation P rom oti onNA BenefitsPercent of NHs Notes N = 944 facilities. NH = nursing home; NA = nursing assistant.

PAGE 72

64 Chapter Five: Concluding Remarks With the projected increased demand for long-term care, the utilization of less costly alternatives to NH care may be important to contain public and private expenditures on long-term care. The quality of care provided to increasing numbers of older adults in these long-term care settings is also of significance. In an attempt to contribute to the state of knowledge on the utilization and quality of long-term care, the purpose of this dissertation was to conduct three studies to understand factors affecting utilization along the continuum of care and the retention of direct care workers to improve quality care. In this section, the main findings of the three studies will be summarized and discussed. In addition, limitations of this dissertation and directions for future research will be addressed. Discussion of Findings The first study offered novel information on the viability of HCBS compared to AL as an alternative to NH care in a frail sample of dually eligible older adults. The results suggest that setting of care modifies the association between dementia and risk of NH placement. Specifically, dementia emerged as a significant risk factor for NH placement from HCBS but not from AL. This suggests that individuals with dementia who reside in AL may be able to age in place and avoid NH placement whereas those who receive HCBS remain at a relatively high risk for NH placement. These results are consistent with the aging in place philosophy of AL and the evolving role the AL industry has assumed in providing care to individuals with dementia.

PAGE 73

65 Along the continuum, NHs are the most expensive and least desirable setting of long-term care. In response, states and the federal government have made concerted efforts to increase access to communitybased long-term care. For example, Money Follows the Person federal demonstration grants aim to eliminate the bias towards NH care and rebalance the delivery of long-term care in 30 states (Centers for Medicare and Medicaid Services, 2007b). Findings of the first dissertation offer greater understanding of service utilization along the continuum and provide support for state and federal initiatives to increase access to community-based settings of long-term care, even for older adults with dementia. The second study highlighted the importance of job characteristics of staff in relation to turnover of nursing assistants (NAs) using a nationally representative sample of NHs. Specifically, the provision of competitive wages and benefits, involvement in resident care planning, unionization, and high levels of nurse staffing could potentially reduce NA turnover. The inclusion of specific job characteristics of staff was a unique contribution to the literature, and the findings offer practice and policy implications with respect to increasing retention of NAs to improve the delivery of resident care. The final study was unique in that it examined the prevalence of employmentbased benefits of NAs and the relationship between NH organizational characteristics and benefits offered to these staff in a nationally representative sample of NHs. The results suggest that NH organizational characteristics are important for the availability of NA benefits. Facility size, chain membership, occupancy level, union involvement, and education of the administrator were positively associated with NA benefits whereas forprofit ownership and Medicaid occupancy were negatively associated with the provision of NA benefits. These findings are consistent with the political economy of aging

PAGE 74

66 framework (e.g., Estes, 1999a) and suggest that systemic differences of NHs in the aging enterprise may influence an organizations investment in benefits for NA staff. For those older adults who rely on NH care, quality of care in this setting has been a persistent concern. Several states have attempted to improve quality of care by increasing staffing levels (Mueller et al., 2006); however, little attention has been given to addressing the high turnover of nursing staff. The second and third studies of this dissertation highlight several strategies that NHs may employ to retain NAs and to improve the benefit package offered to these staff. Strategies that effectively reduce turnover of NA staff may subsequently contribute to improvements in quality of NH care. At the same time, many of these approaches may also require greater reimbursement to NHs from public payers, including Medicare and Medicaid. Collectively, the three studies of this dissertation addressed two primary goals of long-term care: to facilitate aging in place in the least restrictive setting and to provide high quality of care to consumers. Taken together, reductions in expenditures achieved by rebalancing the delivery of long-term care may provide for increased spending on staff retention to improve the quality of NH care. In practice, however, these two goals may be at odds. For example, mandated staffing increases in Florida were financed by reducing state expenditures on HCBS (Florida Policy Exchange Center on Aging, 2001). Within a limited budget, policy makers will have to grapple with balancing the costs of providing care to greater numbers of older adults in the community versus investing in retention of qualified staff to improve quality of NH care. Limitations Although each of the studies has unique contributions, their limitations should also be acknowledged. In the first study, the sample consisted of a relatively frail sample

PAGE 75

67 of dually eligible older adults in one state. Although this sample is unique, the generalizability of the findings is limited. Future studies should include a more representative sample of older adults in HCBS or AL which include both publicly and privately funded clients. Secondly, the measures were collected at baseline. There is a possibility that health status and setting of care changed over the follow-up period. Consequently, there is a need for future longitudinal studies to use multiple assessments to account for these potential changes in the analyses. Lastly, the severity of dementia or cognitive impairment was not available. Although we accounted for functional limitations, including these measures would strengthen the findings. The second study was a cross-sectional examination of organizational and job characteristics in relation to NA turnover. This type of analysis limits the ability to conclude whether these characteristics influence turnover. Future research in this area would benefit from using longitudinal data to examine the directionality of the relationship between organizational or job characteristics and turnover of NAs. In addition, several of the measures in the publicly available data set used in this research were provided at the ordinal level when a continuous measure may have been preferable. Lastly, other previously identified factors relating to nursi ng staff turnover such as resident, market and individual characteristics could not be examined. The final study was limited by its cross-sectional analysis of the relationship between NH organizational characteristics and NA benefits. Longitudinal research is needed to infer how NH organizational characteristics affect the availability of NA benefits. In addition, there may be other important factors related to NA benefits that were not available in the National Nursing Home Survey, such as facility expense ratios, profit margins, or local market conditions. Future studies should look more closely at

PAGE 76

68 these factors which may be associated with the provision of benefits. Lastly, we could not include all of the surveyed benefits in the outcome measure due to lack of data to generate weightings. As information on the employer costs of these benefits becomes available, future research would be strengthened by the inclusion of all available benefits in the outcome. Future Directions This dissertation research was undertaken to examine two salient issues in longterm care: utilization and quality. Although several important research questions were addressed, other key questions beyond the scope of this dissertation remain. For example, what are the potential costs and ben efits of serving frail older adults with significant long-term care needs in the community? If high turnover of direct care staff was addressed, would quality of care subsequent ly improve? It is imperative that these questions are answered to achieve positive change in the long-term care system. The first direction for future research is to determine the potential costs and benefits of providing long-term care to frail older adults in HCBS and AL compared to a NH. More studies are needed to assess clinical outcomes of care in HCBS and AL for individuals with significant physical or cognitive impairments to ensure placement in these settings is appropriate. Moreover, it is important to determine the overall costeffectiveness of care provided in HCBS or AL as an alternative to NH care. Expanded access to publicly funded HCBS and AL has the potential to reduce Medicaid expenditures on long-term care, but a greater understanding of these savings balanced with the potential cost shifting to Medicare for increased hospitalizations and acute rehabilitation is needed. This dissertation was not able to examine clinical outcomes or

PAGE 77

69 costs of care, and this information may be useful to policy makers to support increased funding of HCBS and AL. One final direction for future research is to further explore the role of amenable factors related to turnover of NAs and the delivery of quality NH care. Quality of care was indirectly assessed in this dissertati on by using turnover as a proxy measure; therefore, future research is needed to test directly the relationship between nursing staff turnover and quality of care. Many of the limitations of this dissertation research could be addressed in future studies by applying for use of the full data in the National Nursing Home Survey. This would allow researchers to access the raw data and to merge the data with other data sets that provide resident information (e.g., Online Survey, Certification and Reporting data), market characteristics (e.g., the Area Resource File), individual characteristics of NAs (e.g., the National Nursing Assistant Survey), and measures of quality (e.g., Nursing Home Compare). This type of research study would provide greater insight on the relative importance of various correlates of NA turnover and on the relationship between NA turnover and quality of care. Consumers of long-term care desire expanded access to HCBS and AL and improvements in the quality of NH care. Although the current dissertation has its limitations, the findings do offer greater insight on the utilization of services along the continuum of care and the retention of direct care workers to improve the quality of care. Most importantly, the findings may contribute to improvements in the independence and well-being of older adults.

PAGE 78

70 References Abelson, M. A., & Baysiner, B. D. (1984). Optimal and dysfunctional turnover: Toward an organizational level model. Academy of Management Review, 9 331-341. Allison, P. D. (2007). Survival analysis using SAS: A practical guide Cary, NC: SAS Institute. Alzheimers Association. (2007). Alzheimers disease facts and figures 2007 Chicago, IL: Author. American Health Care Association. (2003). Results of the 2002 AHCA survey of nursing staff vacancy and turnover in nursing homes Washington, DC: Author. Andel, R., Hyer, K., & Slack, A. (2007). Risk factors for nursing home placement in older adults with and without dementia. Journal of Aging and Health, 19, 213228. Andersen, R. (1995). Revisiting the behavioral model and access to medical care: Does it matter? Journal of Health and Social Behavior, 36, 1-10. Andersen, R. M., & Newman, J. F. (1973). Societal and individual determinants of medical care utilization in the United States. Milbank Quarterly, 511, 95-124. Banaszak-Holl, J., & Hines, M. (1996). Factors associated with nursing home staff turnover. The Gerontologist, 36 512-517. Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182. Bishop, C. E., Weinberg, D. B., Leutz, W., Dossa, A., Pfefferle, S. G., & Zincavage, R. M. (2008). Nursing assistants job commitment: Effect of nursing home organizational factors and impact on resident well-being. The Gerontologist, 48 36-45. Bleismer, M. M., Smayling, M., Kane, R. L., & Shannon, I. (1998). The relationship between nursing staffing levels and nursing home outcomes. Journal of Aging and Health, 10 351-371. Borrayo, E. A., Salmon, J. R., Polivka, L., & Dunlop, B. D. (2002). Utilization across the continuum of long-term care services. The Gerontologist, 42 603-612.

PAGE 79

71 Bowers, B. J., Esmond, S., & Jacobson, N. (2000). The relationship between staffing and quality in long-term care facilities: Exploring the views of nurse aides. Journal of Nursing Care Quality, 14 55-64. Brannon, D., Zinn, J. S., Mor, V., & Davis, J. (2002). An exploration of job, organizational, and environmental factor s associated with high and low nursing assistant turnover. The Gerontologist, 42 159-168. Castle, N. G. (2005). Turnover begets turnover. The Gerontologist, 45 186-195. Castle, N. G. (2006). Measuring staff turnover in nursing homes. The Gerontologist, 46, 210-219. Castle, N. G. (2008). Nursing home caregiver staffing levels and quality of care: A literature review. Journal of Applied Gerontology, 27 375-405. Castle, N. G., & Engberg, J. (2005). Staff turnover and quality of care in nursing homes. Medical Care Research and Review, 43 616-626. Castle, N. G., & Engberg, J. (2006). Organizational characteristics associated with staff turnover in nursing homes. The Gerontologist, 46 62-73. Centers for Medicare and Medicaid Services. (2001). Report to Congress: Appropriateness of minimum nurse staffing ratios in nursing homes phase II final report Baltimore, MD: Author. Centers for Medicare and Medicaid Services. (2007a). National health expenditures by type of service and source of funds, CY 1960-2005 Baltimore, MD: Author. Centers for Medicare and Medicaid Services. (2007b). Money follows the person rebalancing demonstration fact sheet. Baltimore, MD: Author. Chapin, R., Baca, B., Macmillan, K., Rachlin, R., & Zimmerman, M. (2009). Residential outcomes for nursing facility applicants who have been diverted: Where are they 5 years later? The Gerontologist, 49, 46-56 Chapin, R., & Dobbs-Kepper, D. (2001). Aging in place in assisted living: Philosophy versus policy. The Gerontologist, 41, 43-50. Chou, S. C., Boldy, D. P., & Lee, A. H. (2002). Staff satisfaction and its components in residential aged care. International Journal for Quality in Health Care, 14 207217. Christensen, R. (2002). Value of benefits constant in a changing world: Findings from the 2001 EBRI/MGA value of benefits survey. Employee Benefit Research Institute Notes 23, 1-3.

PAGE 80

72 Clark, W. D., & Hulbert, M. M. (1998). Research issues: Dually eligible Medicare and Medicaid beneficiaries, challenges and opportunities. Health Care Financing Review, 20, 1-10. Cohen-Mansfield, J. (1997). Turnover among nursing home staff: A review. Nursing Management, 28, 59-64. Dobbs, D., Hayes, J., Chapin, R., & Oslund, P. (2006). The relationship between psychiatric disorders and the ability to age in place in assisted living. American Journal of Geriatric Psychiatry, 14, 613-620. Donoghue, C., & Castle, N. G. (2007). Organizational and environmental effects on voluntary and involuntary turnover. Health Care Management Review, 32 360369. Ejaz, F. K., Noelker, L. S., Menne, H. L., & Bagakas, J. G. (2008). The impact of stress and support on direct care workers job satisfaction. The Gerontologist, 48, 6070. Estes, C. L. (1979). The aging enterprise San Francisco, CA: Jossey-Bass. Estes, C. L. (1999a). Clinical gerontology and the new political economy of aging. In M. Minkler & C. L. Estes (Eds.), Critical gerontology: Perspectives from political and moral economy (pp. 17-35). Amityville, NY: Baywood Publishing Company. Estes, C. L. (1999b). The aging enterprise revisted. In M. Minkler & C. L. Estes (Eds.), Critical gerontology: Perspectives from political and moral economy (pp. 136146). Amityville, NY: Baywood Publishing Company. Estes, C. L., Alford, R. R., & Egan, A. H. (2001). The transformation of the nonprofit sector: Systemic crisis and the political economy of aging services. In C. L. Estes (Ed.), Social policy & aging (pp. 61-117). Thousand Oaks, CA: Sage Publications. Estes, C. L., Harrington, C., & Pellow, D. N. (2001). The medical industrial complex and the aging enterprise. In C. L. Estes (Ed.), Social policy & aging (pp. 165-185). Thousand Oaks, CA: Sage Publications. Estes, C. L., Wallace, S. P., Linkins, K. W., & Binney, E. A. (2001). The medicalization and commodification of aging and the privatization and rationalization of old age policy. In C. L. Estes (Ed.), Social policy & aging (pp. 45-60). Thousand Oaks, CA: Sage Publications. Florida Policy Exchange Center on Aging. (2001). Informational report of the task force on the availability and affordability of long-term care for the Florida legislature in response to House Bill 1993 Tampa, FL: Author.

PAGE 81

73 Foner, N. (1994). Nursing home aides: Saints or monsters? The Gerontologist, 34 245250. Friedman, S. M., Steinwachs, D. M., Rathouz, P. J., Burton, L. C., & Mukamel, D. B. (2005). Characteristics predicting nursing home admission in the Program of AllInclusive Care for Elderly People. The Gerontologist, 45, 157-166. Gaugler, J. E., Leach, C. R., Clay, T., & Newcomer, R. C. (2004). Predictors of nursing home placement in African Americans with dementia. Journal of the American Geriatric Society, 52, 445-452. Geiger-Brown, J., Muntaner, C., Lipscomb, J., & Trinkoff, A. (2004). Demanding work schedules and mental health in nursing assistants working in nursing homes. Work & Stress, 18 292-304. Halbur, B. T. (1983). Nursing personnel in nursing homes: A structural approach to turnover. Work and Occupations, 10 381-411. Haley-Lock, A. & Kruzich, J. (2008). Serving workers in the human services: The roles of organizational ownership, chain affiliation, and professional leadership in frontline job benefits. Nonprofit and Voluntary Sector Quarterly 37, 443-467. Harrington, C., Carrillo, H., & Blank, B. W. (2008). Nursing facilities, staffing, residents and facility deficiencies, 2001 through 2007. San Francisco, CA: University of California, Department of Social and Behavior Sciences. Harrington, C., Kovner, C., Mezey, M., Kayser-Jones, J., Burger, S., Mohler, M., et al. (2000). Experts recommend minimum nurse staffing standards for nursing facilities in the United States. The Gerontologist, 40 5-16. Harrington, C., & Swan, J. H. (2003). Nursing home staffing, turnover, and case mix. Medical Care Research and Review, 60 366-392. Harrington, C., Zimmerman, D., Karon, S. L., Robinson, J., & Beutel, P. (2000). Nursing home staffing and its relationship to deficiencies. Journal of Gerontology: Social Sciences, 55 S278-S287. Hunter, L. W. (2000). What determines job quality in nursing homes? Industrial and Labor Relations Review 53, 463-481. Institute of Medicine. (2008). Retooling for an aging America: Building the health care workforce. Washington, DC: Author. Kane, R. A., Kane, R. L., Priester, R., & Homyak, P. (2008). Research on state management practices for the rebalancing of state long-term care systems: Final report Baltimore, MD: Centers for Medicare and Medicaid Services.

PAGE 82

74 Kash, B. A., & Boyer, G. J. (2008). Advertising expenditures in the nursing home sector: Evaluating the need for and purpose of advertising. Journal of Healthcare Management, 53, 242-256. Kash, B. A., Castle, N. G., & Phillips, C. D. (2007). Nursing home spending, staffing, and turnover. Health Care Management Review, 32 253-262. Kopetz, S., Steele, C. D., Brandt, J., Baker, A., Kronberg, M., Galik, E., et al. (2000). Characteristics and outcomes of dementia residents in an assisted living facility. International Journal of Geriatric Psychiatry, 15 586-593 Lambert, S. J. (2000). Added benefits: The link between work-life benefits and organizational citizenship behavior. Academy of Management Journal, 43 801815. Leon, J., & Moyer, D. (1999). Potential cost savings in residential care for Alzheimers disease patients. The Gerontologist, 39, 440-449. Mitchell G., Salmon, J. R., Polivka, L., & Soberon-Ferrer, H. (2006). The relative benefits and cost of Medicaid home and community-based services in Florida. The Gerontologist, 46, 483-494. Mueller, C., Arling, G., Kane, R., Bershadsky, J., Holland, D., & Joy, A. (2006). Nursing home staffing standards: Their relationship to nurse staffing levels. The Gerontologist, 46, 74-80. Myers, D., Silverstein, B., & Nelson, N. A. (2002). Predictors of shoulder and back injuries in nursing home workers: A prospective study. American Journal of Industrial Medicine, 41 466-476. National Center for Assisted Living. (2006). Guiding principals for assisted living Washington, DC: Author. Retrieved May 21, 2007 from http://www.ncal.org/about/principles.cfm. Ng, T., Harrington, C., & OMalley, M. (2008). Medicaid home and community-based service programs: Data update Washington, DC: Kaiser Family Foundation. Phillips, C. D., Munoz, Y., Sherman, M., Rose, M., Spector, W., & Hawes, C. (2003). Effects of facility characteristics on departures from assisted living: Results from a national study. The Gerontologist, 43 690-696. Potter, S. J., Churilla, A., & Smith, K. (2006). An examination of full-time employment in the direct care workforce. Journal of Applied Gerontology, 25 356-374. Price, J. (2001). Reflections on the determinants of voluntary turnover. International Journal of Manpower, 22 600-624.

PAGE 83

75 Rosenberg, P. B., Mielke, M. M., Samus, Q. M., Rosenblatt, A., Baker, A., Brandt, J., et al. (2006). Transition to nursing home from assisted living is not associated with dementia or dementia-related problem behaviors. Journal of American Medical Directors Association, 7, 73-78. SAS Institute. (2003). SAS version 9.1.3 [Computer software]. Cary, NC: Author. Schnelle, J.F., Simmons, S.F., Harrington, C., Cadogan, M., Garcia, E., & Bates-Jensen, B. (2004). Relationship of nursing home staffing to quality of care. Health Services Research, 39 225-250. Scott, W. K., Edwards, K. B., Davis, D. R., Cornman, C. B., & Macera, C. A. (1997). Risk of institutionalization among community long-term care clients with dementia. The Gerontologist, 37, 46-51. Seavey, D. (2004). The cost of frontline turnover in long-term care Washington, DC: Institute for the Future of Aging Services and the American Association of Homes and Services for the Aging. Retrieved May 19, 2008 from http://www.bjbc.org/content/docts/tocostreport.pdf. Seavey, D., & Salter, V. (2006). Paying for quality care: State and local strategies for improving wages and benefits for personal care assistants Washington, DC: AARP Public Policy Institute. Sethi, R. C. (2001). Value of benefits constant in a changing world: Findings from the 2001 EBRI/MGA value of benefits survey Washington, DC: Employee Benefit Research Institute. Sloane, P. D., Zimmerman, S., Gruber-Baldini, A. L., Hebel, J. R., Magaziner, J., & Konrad, T. R. (2005). Health and functional outcomes and health care utilization of persons with dementia in residential care and assisted living facilities: Comparison with nursing homes. The Gerontologist, 45 124-132. Smith, K., & Baughman, R. (2007). Caring for Americas aging population: A profile of the direct-care workforce. Monthly Labor Review 130, 20-26. Spector, W., & Takada, H.A. (1991). Characteristics of nursing homes that affect resident outcomes. Journal of Aging and Health, 3, 427-454. Spetz, J., & Adams, S. (2006). How can employment-based benefits help the nurse shortage? Health Affairs, 25 212-218. Stearns, S. C., & DArcy, L. P. (2008). Staying the course: Facility and profession retention among nursing assistants in nursing homes. Journals of Gerontology: Social Sciences, 63B S113-S121.

PAGE 84

76 Stone, R. I., Reinhard, S. C., Bowers, B., Zimmerman, D., Phillips, C. D., Hawes, C. et al. (2002). Evaluation of the Wellspring model for improving nursing home quality New York, NY: The Commonwealth Fund. Stone, R. I., & Wiener, J. M. (2001). Who will care for us? Addressing the long-term care workforce crisis Washington, DC: The Urban Institute and the American Association of Homes and Services for the Aging. Temple, A., Dobbs, D., & Andel, R. (2009). Exploring correlates of turnover among nursing assistants in the National Nursing Home Survey. Health Care Management Review 34, 182-190. U. S. Bureau of Labor Statistics. (2000). The employment situation: May 2000. Washington, DC: Department of Labor. U. S. Bureau of Labor Statistics. (2007a). Employer costs for employee compensation summary Washington, DC: Department of Labor. Retrieved July 14, 2008 from http://www.bls.gov/news.release/pdf/ecec.pdf. U. S. Bureau of Labor Statistics (2007b). National compensation survey: Employee benefits in private industry in the United States, March 2007 Washington, DC: Department of Labor. Retrieved July 14, 2008 from http://www.bls.gov/ncs/ebs/sp/ ebsm0006.pdf. U. S. Department of Health and Human Services. (2003). The future supply of longterm care workers in relation to the aging baby boom generation. Report to Congress Washington, DC: Author. Retrieved July 14, 2008 from http://aspe.hhs.gov/daltcp/reports/ltcwork.htm. U.S. Department of Health and Human Services. (2007). Long-term care Washington, DC: Author. U. S. General Accounting Office. (2001). Recruitment and retention of nurses and nurse aides is a growing concern (GAO-01-750T). Washington, DC: Author. Yaffe, K., Fox, P., Newcomer, R., Sands, L., Lindquist, K., Dane, K., et al. (2002). Patient and caregiver characteristics and nursing home placement in patients with dementia. Journal of the American Medical Association, 287, 2091-2097. Zimmerman, S., Gruber-Baldini, A. L., Hebel, J. R., Sloane, P. D., & Magaziner, J. (2002). Nursing home facility risk factors and hospitalization: Importance of registered nurse turnover, administration, and social factors. Journal of the American Geriatrics Society, 50 1987-1995. Zimmerman, S., Sloane, P. D., Williams, C. S., Dobbs, D., Ellajosyula, R., Braaten, A., et al. (2007). Residential care/assisted living staff successfully detect undiagnosed dementia using the Minimum Data Set Cognition Scale (MDSCOGS). Journal of the American Geriatrics Society, 55, 1349-1355.

PAGE 85

77 Appendices

PAGE 86

78 Appendix A: Action Letter 10-21-2008 RE: 08-70R1, entitled "Exploring Correlates of Turnover Among Nursing Assistants in the National Nursing Home Survey" Dear April Temple, I am pleased to inform you that your manuscript is now accepted for publication in Health Care Management Review. All manuscript materials will be forwarded to the production staff for placement in an upcoming issue, Volume 34, number 1. The next communications you will receive will be from the production staff regarding the article proofs. Thank you for submitting your interesting and important work to HCMR. Sincerely, L. Michele Issel, PhD, RN Editor-in-Chief Health Care Management Review

PAGE 87

79 Appendix B: Curriculum Vitae April Temple University of South Florida School of Aging Studies 4202 E. Fowler Avenue MHC 1352 Tampa, Florida 33620 Phone: (813) 974-3237 Fax: (813) 974-5788 Email: aslack@cas.usf.edu EDUCATION 2004-2009 Ph.D. in Aging Studies University of South Florida, Tampa, Florida Co-Major Professors: Dr. Ross Andel and Dr. Debra Dobbs Dissertation: Factors Affecting the Utilization and Quality of Long-Term Care 2001-2004 B.S. in Gerontology, Minor in Business Administration University of South Florida, Tampa, Florida Summa Cum Laude RESEARCH INTERESTS Medicare and Medicaid health policy; Utilizati on of long-term care services; Quality of care in long-term care; Nurse staffing and workforce development in nursing homes. RESEARCH EXPERIENCE Summer 2006 Graduate Research Assistant Florida Policy Exchange Center on Aging, USF Tampa, FL Project Title: The Effect of Sociodemographic, Health, and Functional Characteristics on Utilization of Long-Term Care Options Among Older Medicare and Medicaid Beneficiaries PI: Ross Andel, PhD Explored the utility of alternatives to nursing home care among dually eligible beneficiaries 2005-2006 Graduate Research Assistant School of Aging Studies, USF Tampa, FL Project Title: Provider Response to Nurse Staffing in Florida Nursing Homes PI: Kathryn Hyer, PhD Conducted interviews with key stakeholders to review Floridas efforts to increase nurse staffing in nursing homes.

PAGE 88

80 2005-2006 Graduate Research Assistant Florida Policy Exchange Center on Aging, USF Tampa, FL Project Title: Floridas Dually Eligible Beneficiaries: Health and Functional Risk Factors for Nursing Home Placement PI: Ross Andel, PhD Identified risk factors for nursing home placement among dually eligible beneficiaries 2004-2005 Principal Investigator (Health and Cost Outcomes Component) Florida Policy Exchange Center on Aging, USF Tampa, FL Project Title: Affordable Assisted Living Evaluation PI: Jennifer R. Salmon, PhD Evaluated the health and cost outcomes of Medicaid waiver and Medicaid state plan funding of assisted living in Florida. TEACHING COMPETENCIES Introduction to Gerontology; Sociocultural Aspects of Aging; Long-Term Care Administration; Health Policy; Research Methods TEACHING EXPERIENCE 2006-2009 Instructor School of Aging Studies, USF Tampa, FL Courses: Introduction to Gerontology (GEY 2000) Sociocultural Aspects of Aging (GEY 3625) 2005-2009 Graduate Teaching Assistant School of Aging Studies, USF Tampa, FL Courses: Psychology of Aging (GEY 4612) Long-Term Care Administration I (GEY 4327) Long-Term Care Administration II (GEY 4329) 2005-2007 Guest Lecturer School of Aging Studies, USF Tampa, FL Courses: Psychology of Aging (GEY 4612) PROFESSIONAL LICENSURES 2005-present Nursing Home Administrator, Florida PROFESSIONAL EXPERIENCE Summer 2007 Intern AARP Office of Policy Integration Washington, DC Assisted with policy research on state health care reform and rebalancing Medicaid long-term care

PAGE 89

81 Summer 2004 Administrator-in-Training Central Park Healthcare and Rehab Center Brandon, FL Learned state and federal nursing home regulations and completed comprehensive rotations in all departments for Nursing Home Administrator licensure. JOURNAL PUBLICATIONS Hyer, K., Temple, A. & Johnson, C. (in press). Floridas efforts to improve quality of nursing home care through nurse staffing mandates, regulation, and Medicaid reimbursement. Journal of Aging & Social Policy Temple, A. Dobbs, D., & Andel, R. (2009). Exploring correlates of turnover among nursing assistants in the National Nursing Home Survey. Health Care Management Review 34, 182-190. Andel, R., Hyer, K., & Slack, A. (2007). Risk factors for nursing home placement in older adults with and without dementia. Journal of Aging and Health 19, 213228. MANUSCRIPTS IN PROGRESS Temple, A. Andel, R., & Dobbs, D. (submitted). Setting of care modifies risk of nursing home placement for older adults with dementia. TECHNICAL REPORTS Andel, R., Slack, A. & Polivka, L. (2006). The effect of sociodemographic, health, and functional characteristics on utilization of long-term care options among older Medicare and Medicaid beneficiaries Tampa, FL: Florida Policy Exchange Center on Aging. Slack, A. Salmon, J., Mitchell, G., Hinton, S., & Soberon-Ferrer, H. (2005). Differential benefits of Medicaid state plan or waiver funding of assisted living Tampa, FL: Florida Policy Exchange Center on Aging. Slack, A. Salmon, J., Mitchell, G., Hinton, S., & Soberon-Ferrer, H. (2005). Medicaid funding of affordable assisted living in Florida Tampa, FL: Florida Policy Exchange Center on Aging. CONFERENCE PRESENTATIONS Temple, A. Andel, R., & Dobbs, D. (2009, March). Turnover of nursing assistants: Results from the National Nursing Home Survey. Poster presented at the annual meeting of the American Society on Aging, Las Vegas, NV.

PAGE 90

82 Temple, A. Andel, R., & Dobbs, D. (2008, November). Turnover of nursing assistants: Results from the National Nursing Home Survey. Paper presented at the annual meeting of the Gerontological Society of America, National Harbor, MD. Slack, A. Andel, R., & Dobbs, D. (2007, November). Assisted living moderates the association between dementia and risk of nursing home placement. Poster presented at the annual meeting of the Gerontological Society of America, San Francisco, CA. Hyer, K., Johnson, C., Popa, M., & Slack, A. (2006, November). Financial incentives to increase staffing in Florida nursing homes: Who were the winners and losers? Paper presented at the annual meeting of the Gerontological Society of America, Dallas, TX. Slack, A. Salmon, J.R., & Mitchell, G. (2006, November). Health and cost outcomes of affordable assisted living in Florida. Poster presented at the annual meeting of the Gerontological Society of America, Dallas, TX. Salmon, J.R., Mills, W.L., & Slack, A. (2006, August). Florida affordable assisted living evaluation. Presented at the annual Florida Conference on Aging, Tampa, FL. Slack, A. Andel, R., & Hyer, K. (2006, April). Differential risk factors for nursing home placement in older adults with and without dementia. Presented at the annual meeting of the Southern Gerontological Society, Lexington, KY. Slack, A. Andel, R., & Hyer, K. (2006, April). Differential risk factors for nursing home placement in older adults with and without dementia. Presented at the annual Southeastern Regional Student Mentoring Conference in Gerontology and Geriatrics, Lexington, KY. Slack, A. Salmon, J.R., & Mitchell, G. (2006, April). Health and cost outcomes of Medicaid funded assisted living. Presented at the University of South Florida Fifth Annual Graduate Research Symposium, Tampa, FL. AWARDS AND HONORS 2009 Provosts Award for Outstanding Teaching by a Graduate Teaching Assistant, University of South Florida 2008 Presidential Management Fellow Nominee, University of South Florida 2007 Admission to Doctoral Candidacy with Distinction, University of South Florida 2007 PEO Scholar Award Nominee

PAGE 91

83 2006 Presentation Award Recipient Graduate Research Symposium, University of South Florida 2006 Second Place Presentation Award Southeastern Regional Student Mentoring Conference in Gerontology and Geriatrics 2005 University of South Florida Graduate Fellowship 2004 University of South Florida King ONeal Scholar Award for 4.0 GPA 2004 University of South Florida Honors College Graduate PROFESSIONAL MEMBERSHIPS 2006-present American College of Health Care Administrators 2005-present Florida Council on Aging 2004-present Gerontological Society of America 2004-present Student Association for Aging Studies, University of South Florida 2002-2005 Sigma Phi Omega, National Gerontology Honor Society

PAGE 92

About the Author April Temple received her Bachelors of Science degree in Gerontology from the University of South Florida in August 2004 and her state licensure as a Nursing Home Administrator in January 2005. She entered the Ph.D. in Aging Studies program in August 2004 with an interest in quality of nursing home care and long-term care policy. While in the Ph.D. program, Ms. Temple was employed as a Graduate Research Assistant in the School of Aging Studies in which she assisted with health services research. In addition, she was employed as a Graduate Teaching Assistant in the School of Aging Studies where she taught undergraduate courses including Introduction to Gerontology and Sociocultural Aspects of Aging. Ms. Temple has co-authored three peer-reviewed publications and presented her research at multiple national conferences. Upon graduation with her doctorate, she will begin as Assistant Professor in Health Services Administration at James Madison University.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200397Ka 4500
controlfield tag 001 002317244
005 20100831154313.0
007 cr cnu|||uuuuu
008 100831s2009 flua ob 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002990
035
(OCoLC)659862355
040
FHM
c FHM
049
FHMM
090
HQ1061 (Online)
1 100
Temple, April.
0 245
Factors affecting the utilization and quality of long-term care
h [electronic resource] /
by April Temple.
260
[Tampa, Fla.] :
b University of South Florida,
2009.
500
Title from PDF of title page.
Document formatted into pages; contains 83 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2009.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
520
ABSTRACT: The aging of the nation in the coming decades will contribute to an increased demand for long-term care. Given this trend, it is becoming increasingly important to understand utilization of services along the continuum of care and to determine factors that influence the provision of quality care. These insights are needed to reduce national expenditures on long-term care and to maximize independence and well-being among older adults. The purpose of the present dissertation was to explore factors that influence the utilization and quality of long-term care by conducting three separate studies. The first study examined risk of nursing home (NH) placement in a frail sample of older adults receiving publicly-funded long-term care in two settings: home and community-based services (HCBS) or assisted living (AL). Specifically, it explored whether these settings of care modify the relationship between dementia and NH placement.The second study examined NH organizational characteristics and job characteristics of staff in relation to turnover of nursing assistants (NAs) in a nationally representative sample of NHs. Using the same sample, the final study examined the presence of employment-based benefits of NAs and explored the relationship between NH organizational characteristics and the availability of staff benefits. Findings from this dissertation can contribute to a greater understanding of the use of long-term care services and the retention of staff to provide quality care. In the first study, AL was associated with longer time to NH placement for individuals with dementia when compared to HCBS. In the second study, the provision of competitive wages and benefits, involvement of NAs in resident care planning, and high levels of nurse staffing were associated with lower NA turnover, which may be important for quality of NH care.In the final study, for-profit ownership and high Medicaid occupancy were negatively associated with NA benefits whereas facility size, chain membership, occupancy level, union involvement, and education of the NH administrator were positively associated with benefits offered to these staff. Collectively, the findings from this dissertation may have important implications for policy makers, providers, and consumers of long-term care.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Co-Advisor: Ross Andel, Ph.D.
Co-Advisor: Debra Dobbs, Ph.D.
653
Nursing home placement
Dementia
Nursing assistants
Turnover
Benefits
690
Dissertations, Academic
z USF
x Aging Studies
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2990