Profile of Home Care Aides, Nursing Home Aides, and Hospital Aides: Historical Changes and Data Recommendations

Transcription

1 The Gerontologist Vol. 42, No. 2, Copyright 2002 by The Gerontological Society of America Profile of Home Care Aides, Nursing Home Aides, and Hospital Aides: Historical Changes and Data Recommendations Yoshiko Yamada, MSW 1 Purpose: To examine demographic characteristics and work conditions of home care aides, nursing home aides, and hospital aides in the late 1980s and late 1990s. Design and Methods: This study replicated a previous study which examined the Current Population Survey (CPS) March supplement from 1987 to The present study examined CPS data from 1997 to 1999 both to obtain a profile of aides in the late 1990s and to compare 1990 s work conditions and demographic characteristics with those of aides in the late 1980s. Results: Whereas demographic characteristics and work conditions of hospital and nursing home aides show little change over the decade, home care aides today are generally younger, more educated, more likely to hold full-time positions, and more likely to have health benefits than their counterparts 10 years ago. Yet work conditions are still poorer for home care aides than for the other types of aides. Implications: Work conditions need further improvement, particularly for home care aides. More accurate data to describe paraprofessional care workers are also necessary to address the serious worker shortage problems. Key Words: Hospital aides, Nursing home aides, Home care aides, Paraprofessionals This research was substantially supported by the Center for Health Workforce Studies at University at Albany. I wish to thank Dr. Paul Wing, Mr. Edward Salsberg, Dr. Susan Sherman, and Dr. Philip McCallion for their encouragement and helpful comments. Address correspondence to Yoshiko Yamada, MSW, Center for Health Workforce Studies, University at Albany, One University Place, Rensselaer, NY yoshiko_yamada@yahoo.com 1 Center for Health Workforce Studies, University at Albany, Rensselaer, NY. The United States has been facing a serious shortage of direct care workers in long-term care settings in the last few decades. Several factors, including the aging population and changes in health care systems, have resulted in a dramatic increase in the demand for direct care workers. The combination of persistent worker shortages, increasing demands, and recordlow unemployment rates has exacerbated the shortages of direct care workers in the late 1990s. A recent nationwide survey found that for 42 states recruitment and retention of paraprofessional aide workers was a major workforce issue in 1999 (North Carolina Division of Facility Services, 1999). Another study, composed of 86 telephone interviews of a variety of long-term care providers in New York state, found that the direct care worker shortage has hit all settings, including 92% of nursing homes, 82% of home health agencies, and 70% of adult care facilities (New York Association of Homes and Services for the Aging, 2000). Previous studies suggest that recruitment and retention of direct care workers are closely associated with a number of factors, including workers demographic characteristics, work conditions, organizational culture, and macroeconomic conditions (Banaszak-Holl & Hines, 1996; Friedman, Daub, Crisci, & Keyser, 1999; Gilbert, 1991; Halbur, 1983; Hill Simonton Bell, 1998; Kiyak, Namazi, & Kahana, 1997; Waxman, Carner, & Berkenstock, 1984). However, little is known about these workers, such as basic demographic characteristics and work conditions. Most existing studies profiling the characteristics of paraprofessional care workers are limited to particular agencies or local areas with very limited generalizability (Crown, Ahlburg, & MacAdam, 1995). Since those variables appear to affect recruitment and retention of the paraprofessional care workforce, accurate information gathering and reporting on key characteristics of these workers are essential to identify strategies to address this growing labor shortage problem. The only studies with nationally representative samples of direct care workers were undertaken by Leon and Franco (1998) and Crown and colleagues (1995). Leon and Franco developed a profile of agencybased and self-employed home health care workers at a national level. They conducted telephone interviews with 623 home care workers identified by the respondents of the Medicare Current Beneficiary Survey (MCBS). Despite the seemingly sufficient sample size, Leon and Franco (1998) pointed out that the response rate was lower than preferred (1,368 MCBS clients Vol. 42, No. 2,

2 were originally contacted) due in part to concurrent government investigations in the industry and the high turnover rate among home care workers. Hence, their estimates should be interpreted with caution. Another study, conducted by Crown and colleagues (1995), compared demographic characteristics and work conditions for three types of aides: hospital aides, nursing home aides, and home care aides. This study used data from the Current Population Survey (CPS) March supplement. The CPS is a monthly survey of about 50,000 households, conducted by the U.S. Census Bureau for the Bureau of Labor Statistics (BLS; BLS, 1996). The sample is representative of national civilian noninstitutional populations aged 15 and older. The March supplement is the primary source of detailed information on income, work experience, geographic mobility, educational attainment, and poverty status (BLS, 2001). Despite difficulties in identifying different types of aides, particularly home care aides, in the CPS data, Crown and colleagues study provides comparisons of different types of aides. Given the growing labor force problem, it seems opportune to examine more recent data and facilitate comparisons with earlier studies. For these reasons this study will replicate the approach taken by Crown and colleagues (1995). This study will also examine the continued usefulness of the CPS to describe direct care workers. Methods Crown and colleagues (1995) merged 1987, 1988, and 1989 data from the CPS March supplement. The present study used a similar approach utilizing 1997, 1998, and 1999 data to examine the demographic characteristics and work conditions of aides. I compared findings from the data set with those from the data set. To make the variables as consistent as possible with Crown and colleagues study, I used the same definitions to identify the three types of aides from the CPS data by using the cross tabulation of occupational (OCC) and industry (IND) codes. Table 1 summarizes the definition of each aide group in the present study. The sample size for each type of aides is presented in Table 2. For the 1990s data, I used chi-squared for categorical variables (e.g., marital status, educational attainment), and analysis of variance for quantitative variables (e.g., income, wages). Because the 1980s data were available only in the aggregated form, I conducted no statistical tests to compare the significance of differences in data for the two different time periods. Results Demographic Characteristics Table 2 provides demographic characteristics of nursing home aides, home care aides, and hospital aides. The age distributions of each occupation group show that home care aides tended to be older than both nursing home and hospital aides in both the late 1980s and 1990s. However, the mean age of home care aides declined in the 10 year period. As for gender distribution, although the proportion of males increased somewhat in the nursing home and home care industries during the 1990s, the direct care workforce is still predominantly female. Hospital aides who work as orderlies may be a possible explanation for the higher proportion of males in hospital settings (Crown et al., 1995). Racial distributions did not change dramatically in the last decade, and the 1990s data showed no significant differences in racial characteristics among the occupations. However, the data showed that home care aides were significantly less likely to be native-born U.S. citizens than were nursing home and hospital aides in the late 1990s. The late 1990s data also showed that nursing home aides were more likely (52%) to have children younger than age 18 than were home care aides (40%) or hospital aides (46%). There were consistent differences in educational attainment among the occupation groups in the two periods studied. Despite overall improvement in educational attainment in the last decade, home care aides still have less education than other aides. Yet it should also be noted that almost one third of nursing home aides and home care aides now have at least some college education, which seems contrary to the public perception of aides. Table 1. CPS Industry Occupation Matrix Occupation Code 446 Occupation Code 447 Occupation Code 465 Code Title Health Aides Nursing Aides Welfare Service Aides Industry Code 761 Private home Home care aide Home care aide Industry Code 831 Hospital Hospital aide Hospital aide Hospital aide Industry Code 832 Nursing facility Nursing home aide Nursing home aide Nursing home aide Industry Code 840 Health service nec. Home care aide Industry Code 861 Job training Home care aide Industry Code 871 Social service nec. Home care aide Industry Code 880 Religious organization Home care aide Note: nec not elsewhere classified. 200 The Gerontologist

3 Table 2. Demographic Characteristics of Nursing Home Aides, Home Care Aides, and Hospital Aides, and Nursing Home Aides Home Care Aides Hospital Aides Characteristic N 1, N 1, N N N N 761 Age, %* Mean age* Gender, %* Man Woman Race, %*** White Black Other Marital status, %* Married Widowed/divorced/separated Never married Education, %* High school High school graduate Some college years of college With children under age 18, %* Yes No Citizenship, %** Native-born U.S. N/A 85.4 N/A 74.5 N/A 85.3 Native-born U.S. outlying area N/A 1.6 N/A 2.8 N/A 0.9 Native-born abroad U.S. parent N/A 0.8 N/A 0.9 N/A 1.3 Foreign born (naturalized) N/A 4.0 N/A 5.6 N/A 6.0 Not a U.S. citizen N/A 8.2 N/A 16.2 N/A 6.4 Source: Crown and colleagues (1995) for data; Author s tabulation of Current Population Survey (March Supplement; Bureau of Labor Statistics, 2001). *p 0.01 for both and **p 0.01 for ; data not available. ***p 0.01 for only. Work Experience Table 3 reports labor force participation for the three types of aides. Although the proportions of fulltime, full-year aides did not change for hospital and nursing home aides, home care aides were more likely to work full time and full year in the 1990s (46%) than in the late 1980s (29%). Even in the late 1990s, however, home care aides were still less likely to work full time and full year than nursing home aides (55%) and hospital aides (63%). Home care aides were also somewhat more likely to work part time and part year (16%) than nursing home aides (11%) and hospital aides (8%). Whereas more than half of part-time hospital aides (54%) and nursing home aides (50%) said they wanted part-time jobs, only 42% of parttime home care aides reported a preference for part-time jobs. Similarly, part-time home care aides were more likely (18%) than hospital aides (12%) and nursing home aides (14%) to report that they could only find part-time jobs. Although the differences in reasons for being part-time workers were not statistically significant (p.286), one should not ignore the fact that a significant minority of home care aides would have preferred a full-time job but worked part time involuntarily. The number of weeks worked per year and the weekly work hours for each occupation did not change dramatically. In the late 1990s, home care aides spent fewer weeks (41) at work than nursing home aides (43) and hospital aides (46), reflecting the higher proportion of part-year workers. Consistent with the 1980s data, data from the late 1990s also showed no statistically significant difference in the average work hours per week. Data from the 1990s also showed that home care aides were somewhat more likely to have earnings from other work (23%) than were nursing home aides (20%) and hospital aides Vol. 42, No. 2,

4 Table 3. Work Experience of Nursing Home Aides, Home Care Aides, and Hospital Aides Nursing Home Aides Home Care Aides Hospital Aides Characteristic Labor force participation, %* FY FT FY PT PY FT PY PT Nonworker N/A 4.5 N/A 7.4 N/A 3.7 Average number of weeks worked per year* Hours usually worked per week Notes: FY full-year; FT full-time; PY part-time; PT part-time. Source: Crown and colleagues (1995) for data; Author s tabulation of Current Population Survey (March Supplement; Bureau of Labor Statistics, 2001). *p 0.01 for both and (16%). Hence, as Crown and colleagues (1995) suggested, many home care aides seemed to hold more than one job and work full-time hours but without the benefits of full-time status. One puzzling finding is that despite the increase in home care aides who work full time and full year, the average number of weeks worked per year and hours usually worked per week declined slightly in the decade studied. Although there is no definite explanation for this finding, one possibility is that home care aides are more likely now than 10 years ago to have one full-time full-year job instead of two part-time jobs. Hence, the increasing proportion of full-time fullyear home care aides did not necessarily result in an increasing number of weeks and hours worked. Another possible explanation is the decline of work weeks and hours among part-time workers in the last decade, which may have brought down the mean number of weeks and hours worked. Future research is needed to test these hypotheses. Wages and Income Table 4 shows wages, earnings, and family income for different types of aides. Wages and income in the 1980s are adjusted to 1998 dollars based on the Consumer Price Index (all items) to make for easier comparison. The table shows that, except for mean hourly wages, wages and income of home care aides increased slightly in the last 10 years, whereas those of Table 4. Wages and Income (in U.S. dollars) of Nursing Home Aides, Home Care Aides, and Hospital Aides Nursing Home Aides Home Care Aides Hospital Aides Wages and Income Hourly wage Mean* Median Mean weekly wage* N/A N/A N/A Personal total income Mean** N/A 13, N/A 12, N/A 16, Median N/A 12, N/A 9, N/A 15, Family total income Mean* 35, , , , , , Median 27, , , , , , Poverty ratio, %* 1.00 N/A 16.1 N/A 22.2 N/A N/A 13.4 N/A 15.7 N/A N/A 16.0 N/A 9.3 N/A N/A 14.2 N/A 8.8 N/A N/A 11.1 N/A 12.5 N/A N/A 29.3 N/A 31.5 N/A 45.5 Notes: No statistical tests were conducted for median wages and income. Wages and income in 1980s are adjusted to 1998 dollars based on Consumer Price Index (all items). Source: Crown and colleagues (1995) for data; Author s tabulation of Current Population Survey (March Supplement; Bureau of Labor Statistics, 2001). *p 0.01 for both and **p 0.01 for ; data not available. 202 The Gerontologist

5 nursing home aides and hospital aides declined in the same time period. The decline was particularly steep among hospital aides. Nonetheless, hospital aides continue to have the highest wages, annual earnings, and family incomes, followed by nursing home aides and home care aides. As Crown and colleagues (1995) pointed out, wage information in the CPS is available only for a minority of the sample, which could be problematic. (Only a 25% subsample of workers reported their wages in the 1980s, compared with a 20% subsample in the 1990s). Nonetheless, the reported CPS wage data are similar to data from the Occupational Employment Statistics, a larger survey with wage data conducted by the BLS. The 1990s data also presented a poverty ratio for each occupation. The poverty ratio is a ratio of income to the poverty level. For example, a poverty ratio of 1.00 is equal to the poverty level, and a ratio of 2.00 is equal to 200% of the poverty level. Compared to average national poverty rates of 12 to 13% in 1998 and 1999 (U.S. Census Bureau, 2000), nursing home aides and home care aides have much higher poverty rates of 16% and 22%, respectively. An additional 29% of nursing home aides and 25% of home care aides live near the poverty level (i.e., poverty ratios between 1.00 and 1.99). In contrast, only 9% of hospital aides live below the poverty level, and almost one half of them have a poverty ratio that is 3.00 and above. Fringe Benefits Differences in availability of fringe benefits among the occupations as well as over time are shown in Table 5. Whereas the proportions of health insurance provided by current employers or unions did not change in the decade studied for nursing home aides (approximately 40%) and hospital aides (approximately 60%), a growing proportion of home care aides now receive health insurance through their employment (14% in the 1980s and 26% in the 1990s). This growth may reflect the increasing availability of full-time positions for home care aides. Yet compared with the other two occupations, home care aides are still substantially less likely to receive health insurance from their employers or unions, and are more likely to receive health insurance through government programs such as Medicaid and Medicare. Table 5 also shows that the nursing homes and hospitals that provide health insurance to their employees are less likely now than they were 10 years ago to pay all of the premium, while home care employers are more likely now than a decade ago to do so. Nonetheless, it should also be noted that about 1 in every 11 nursing home and home care employers who provide health insurance do not pay any premium. Considering the low income of nursing home and home care aides, it is very likely that such benefits are out of reach for these aides without financial support from employers. As for pension coverage, nursing homes and hospitals showed a slight increase in pension provision by employers in the last 10 years, while the proportion stayed the same for home care providers. Hospitals were substantially more likely (72% in the 1990s) to provide pensions for aides than nursing homes (44%) and home care providers (24%). Discussion Data Summary and Policy Implications Whereas nursing home aides and hospital aides showed little change in their demographic character- Vol. 42, No. 2, 2002 Table 5. Fringe Benefits of Nursing Home Aides, Home Care Aides, and Hospital Aides Nursing Home Aides (%) Home Care Aides (%) Hospital Aides (%) Benefits Health insurance Medicare* Medicaid* VA/Military/CHAMPUS Current employer/ union* None*** 28.5 N/A 38.9 N/A 14.5 N/A Health insurance premium**** Employer pays all Employer pays part Employer pays none Pension Plan Provided by employer* Participate in plan if provided** Notes: VA Veterans Administration; CHAMPUS Civilian Health and Medical Program Uniformed Service. Source: Crown and colleagues (1995) for data; Author s tabulation of Current Population Survey (March Supplement; Bureau of Labor Statistics, 2001). *p 0.01 for both and **p 0.01 only for ; data not available or insignificant. ***p 0.01 only for ; data not available. ****p 0.05 for both and

6 istics and work conditions over the last 10 years, the picture looks quite different for home care aides. Compared to the late 1980s, home care aides in the late 1990s were younger, more educated, and more likely to have children. Also, home care providers were more likely to offer full-time/full-year jobs and health insurance for aides in the late 1990s than in the late 1980s. Nevertheless, compared to nursing home aides and hospital aides, home care aides are still less likely to hold full-time jobs and receive benefits. Home care aides are also more likely than other aides to be poor and to hold more than one job. Moreover, despite the real wage decline among hospital and nursing home aides and the slight wage increase among home care aides, data showed that hospital aides still tend to have the best paying and benefited jobs and home care aides have the least. Given the growing and continuing labor shortages for nursing home aide and home care aide positions, these data confirm the need to improve work conditions for long-term care aide workers. Wage and benefits are not the only factors contributing to recruitment and retention of aide workers. As mentioned before, studies have found a number of nonmonetary factors that are closely associated with turnover rates of care aides. These studies do not negate the potential to improve recruitment and retention of care aides through individual employers efforts to address these issues. However, research also shows that a significant proportion of care aides who have left or who are considering leaving their jobs reported money and benefits as reasons for job resignation (Gilbert, 1991; Hill Simonton Bell, 1998). There is also evidence that programs that provided additional benefits, such as a career ladder with enhanced status and wages, more affordable benefits, and guaranteed hours, improved retention of care aides (Feldman, 1993; Gunsch, 1993; Mlynarek & Mondoux, 1996; Moffatte, Stefanini, & Hardke- Peck, 2000). Addressing wages and benefits appears to be critical. However, employers argue that they have very limited control over wages and benefits because the government is the major payer of nursing home and home health care costs. In 1998, 53% of home health care expenditures and 60% of nursing home expenditures were paid through public funds, mostly from Medicare and Medicaid (Health Care Financing Administration, 2000). Hence, wages and benefits for care aides are largely influenced by public health care reimbursement policies. Several states have passed legislation to address the labor shortage in the longterm care workforce. As of 1999, 16 states had approved or implemented some form of a wage pass through (WPT), which mandates that certain amounts or percentages of the increased reimbursement for Medicaid and/or state long-term care programs be given to care paraprofessionals as wages and/or benefits. By 2000, at least two additional states had started to implement WPT (North Carolina Division of Facility Services, 1999, 2000). However, WPT is such a new initiative in most of the states that its impact on recruitment and retention of care workers cannot yet be judged (North Carolina Division of Facility Services, 2000). The data also suggest that a significant minority of home care aides is not U.S. citizens. Some observers have suggested that federal policy makers reconsider current restrictions on immigration to increase the supply of direct care workers (New York Association of Homes and Services for the Aging, 2000). This approach may be particularly effective in addressing the shortage of home care aides, considering the higher proportion of noncitizen workers than for the other two types of aides. Although fewer limitations on immigration may solve the labor problems in the short term, this strategy may not address the fundamental problems identified in this study for direct care workers: low pay and poor benefits. Because immigrants are more likely to accept lower-paying jobs with poor benefits, there is even a danger that the growth of noncitizen care aides will lead to even lower wages and poorer benefits. Limitations of the Present Study and CPS Data Set and Recommendations In this study, paraprofessional care workers were identified in the CPS March supplement based on occupation and industry codes. Although CPS is one of the few national data sets that provides valuable nationally representative data on various occupations, there are several data limitations. Crown and colleagues (1995) pointed out that because the CPS industry and occupation codes were not sufficiently well defined, some aides may have been excluded and people who are not aides may have been included. Although Crown and colleagues (1995) believed that the misclassifications were small in number, closer examination in the present study found potential problems in both occupation and industry codes in CPS related to the study of direct care workers. First, CPS data set does not distinguish between occupation codes for home health aides, who provide assistance with both activities of daily living and instrumental activities of daily living (IADL), and personal care aides, who mainly provide IADL assistance. According to the BLS s Occupational Employment Statistics (n.d.), personal care aides make much less money per hour (average $7.40) than home health aides (average $8.08). Hence, the CPS data may be missing this potentially significant distinction. Another potential problem in CPS occupation codes is the inclusion of orderlies in the same occupation group as nursing aides. The present study found a higher proportion of males in hospital settings, which is perhaps a result of the inclusion of orderlies as care workers. However, orderlies tend to have a different nature of work and different demographic characteristics. By mixing dissimilar groups of workers, the survey data may not provide precise pictures of each occupation. As for the industry codes, the CPS data set 204 The Gerontologist

7 combines several long-term care settings with other industries, which makes it harder to identify paraprofessional care workers in relevant settings. For example, identification of paraprofessional care workers in relatively new residential settings for the elderly, such as assisted living facilities, is extremely difficult in the current CPS industry coding system. The number of assisted living facilities has increased dramatically in the last decade, but the CPS data set does not provide a clear definition of what industry category the aides in those facilities should belong to (e.g., residential care, nursing and personal care facilities). As assisted living facilities generally rely much less on government funding (e.g., Medicare, Medicaid) than other health and long-term care providers, the employers may have more freedom to improve the work conditions of staff, including care aides. Another problem with the CPS industry codes is that home health care agencies, which are the major employers of home care aides, are combined with several other industries in CPS industry codes (IND 840). Despite the fact that most of the direct care workers in the IND 840 belong to home care agencies (BLS, n.d.), the present study does not include health aides (OCC 446) and nursing aides (OCC 447) in the IND 840, which may have resulted in underrepresentation of agency-based home care aides in the present study. Another limitation, more of the present study than the CPS data itself, is the lack of adequate definitions of home care aides. For the purpose of data consistency in the two time periods, the present study used the same occupation and industry codes as Crown and colleagues (1995) to define the three groups of aides. As Table 1 shows, although the CPS has separate industry codes for private-duty (IND 761) and agency-based (e.g., IND 840, 861, 871, 880) home care aides, the present study made no distinctions. Leon and Franco (1998) found that agency-based and self-employed home care workers showed differences in some demographic characteristics and working conditions. Hence, distinctions between agency-based and selfemployed home care workers may be critical to understanding home care workers in different settings. In short, the present study has two limitations in defining home care aides: not including workers who are employed by home care agencies, and not distinguishing between agency-based and self-employed workers. In order to make the best use of the existing data set, it is recommended that future analyses of home care aides using CPS include aide workers in IND 840 (health services, not elsewhere classified) and separate IND 761 (private home) from other home care industries. Research on direct care workers will be greatly facilitated if care workers are properly distinguished from other occupations and industries. Therefore, accurate occupation and industry classification is critical. The workforce crisis in health and long-term care should encourage efforts to make more accurate classifications. Conclusion This study examined the characteristics and work conditions of different types of care aides in the late 1990s and compared this data to similar data from the late 1980s. Some work-related improvements were noted but too many care aides are still at or below poverty level with limited benefits. This is particularly true for home care aides and, to a lesser degree, for nursing home aides. Existing national data sets, although useful, were found to be at risk of missing key groups of workers; this should be considered in planned revisions of occupation and industry codes. As worker shortages continue and demand for care services increases, a number of efforts have been made at organizational, local, state, and national levels to improve recruitment and retention of care staff. Yet without accurate data on current conditions, such efforts may be targeted at the wrong concerns and issues and may fail to bring the desired results. Continued attention to systematic data collection about care workers at various levels (e.g., demographic characteristics, work conditions, and issues workers face) seems essential to secure the quantity and quality of care. References Banaszak-Holl, J., & Hines, M. A. (1996). Factors associated with nursing home staff turnover. The Gerontologist, 36, Bureau of Labor Statistics. (n.d.). Occupational employment statistics survey. Retrieved March 15, 2001, from Bureau of Labor Statistics. (1996). CPS overview. Retrieved March 15, 2001, from http// Bureau of Labor Statistics. (2001). CPS Annual demographic survey (March supplement). Retrieved March 15, 2001, from gov/cps/ads/adsdes.htm Crown, W. H., Ahlburg, D. A., & MacAdam, M. (1995). The demographic and employment characteristics of home care aides: A comparison with nursing home aides, hospital aides, and other workers. The Gerontologist, 35, Feldman, P. H. (1993). Work life improvements for home care workers: Impact and feasibility. The Gerontologist, 33, Friedman, S. M., Daub, C., Crisci, K., & Keyser, R. (1999). A comparison of job satisfaction among nursing assistants in nursing homes and the Program of All-inclusive Care for the Elderly (PACE). The Gerontologist, 39, Gilbert, N. J. (1991). Home care worker resignations: A study of the major contributing factors. Home Health Care Services Quarterly, 12(1), Gunsch, D. (1993). Benefits program helps retain frontline workers. Personnel Journal, 72(2), Halbur, B. T. (1983). Nursing personnel in nursing homes. A structural approach to turnover. Work and Occupations, 10(4), Health Care Financing Administration. (2000, January). National health expenditure projections: Retrieved March 15, 2001, from Hill Simonton Bell. (1998). Certified nursing assistant recruitment and retention project, phase I: Survey results (Prepared for Iowa Caregivers Association). Des Moines, IA: Author. Kiyak, H. A., Namazi, K. H., & Kahana, E. F. (1997). Job commitment and turnover among women working in facilities serving older persons. Research on Aging, 19(2), Leon, J., & Franco, S. J. (1998). Home and community-based workforce. Bethesda, MD: Project HOPE. Mlynarek, L. S., & Mondoux, L. C. (1996). Energizing staff to improve care. Nursing Homes, 45(8), Moffatte, M., Stefanini, K., & Hardke-Peck, R. (2000). How we developed a nursing assistant career ladder. Nursing Homes, 49(3), New York Association of Homes and Services for the Aging. (2000). The staffing crisis in New York s continuing care system: A comprehensive analysis and recommendations. Albany, NY: Author. North Carolina Division of Facility Services. (1999). Comparing state efforts to address the recruitment and retention of nurse aide and other Vol. 42, No. 2,

8 paraprofessional aide workers. Retrieved March 15, 2001, from facility-services.state.nc.us/recruit.pdf North Carolina Division of Facility Services. (2000). Results of a follow-up survey to states on wage supplements for Medicaid and other public funding to address aide recruitment and retention in long-term care settings. Retrieved March 15, 2001, from state.nc.us/survy.pdf U.S. Census Bureau. (2000). Poverty Retrieved March 15, 2001, from Waxman, H. M., Carner, E. A., & Berkenstock, G. (1984). Job turnover and job satisfaction among nursing home aides. The Gerontologist, 24, Received July 17, 2001 Accepted November 20, 2001 Decision Editor: Laurence G. Branch, PhD 206 The Gerontologist