hmgt 320 week 4

Who Hires Social Workers? Structural and Contextual Determinants of Social Service

Staffing in Nursing Homes Amy Restorick Roberts and John R. Bowblis

Although nurse staffing has been extensively studied within nursing homes (NHs), social services has received less attention. The study describes how social service departments are organized in NHs and examines the structural characteristics of NHs and other macro- focused contextual factors that explain differences in social service staffing patterns using longitudinal national data (Certification and Survey Provider Enhanced Reports, 2009– 2012). NHs have three patterns of staffing for social services, using qualified social workers (QSWs); paraprofessional social service staff; and interprofessional teams, consisting of both QSWs and paraprofessionals. Although most NHs employ a QSW (89 percent), nearly half provide social services through interprofessional teams, and 11 percent rely exclusively on paraprofessionals. Along with state and federal regulations that depend on facility size, other contextual and structural factors within NHs also influence staffing. NHs most likely to hire QSWs are large facilities in urban areas within a health care complex, owned by nonprofit organizations, with more payer mixes associated with more profitable reimbursement. QSWs are least likely to be hired in small facilities in rural areas. The influence of policy in supporting the professionalization of social service staff and the need for QSWs with exper- tise in gerontology, especially in rural NHs, are discussed.

KEY WORDS: hiring decisions; interprofessional teams; nursing homes; social services

In the United States, approximately 1.4 millionindividuals live in one of the nation’s 15,700nursing homes (NHs) on any given day (Harris- Kojetin, Sengupta, Park-Lee, & Valverde, 2013). Given the complex medical and psychosocial needs of NH residents, staff working in social services are integral to helping residents adjust to living in an institutional care environment, adapt to new levels of functional impairment, and cope with mental health issues such as depression and dementia. Social service staff have many responsibilities (Bern-Klug & Kramer, 2013; Vourlekis, Zlotnik, & Simons, 2005). They complete psychosocial assessments and care planning, provide emotional support and psycho- social interventions to enhance coping skills for residents and families, and promote individualized decision making to facilitate maximum resident choice and preference. In addition, social service staff assist with transitions in long-term care, such as admissions, referrals to other resources, and dis- charge planning. The psychosocial care provided supports the maintenance or enhancement of physical and mental health to promote the highest

possible quality of life (National Association of Social Workers [NASW], 2003b).

Greater qualifications of social service and mental health staff in NHs have been associated with higher quality, such as fewer psychosocial deficiencies re- lated to psychosocial care (Simons, 2006; Zhang, Gammonley, Seung Chung, & Frahm, 2009). Still, the educational background and preparedness for this role varies a great deal (Bern-Klug et al., 2009). Training in social work is unique, as it involves an accredited curriculum with an extensive field place- ment to develop competencies for practice (Council on Social Work Education [CSWE], 2015), practic- ing in a manner consistent with a code of ethics (NASW, 2015), and ongoing continuing education (NASW, 2003a). However, there are important in- consistencies in definitions of qualifications among NH social workers between professional organiza- tions, licensing boards, and federal guidelines.

Federal guidelines broadly define a qualified social worker (QSW) as an employee with “(i) A bachelor’s degree in social work or a bachelor’s degree in a human services field including but not

doi: 10.1093/hsw/hlw058 © 2016 National Association of Social Workers 15

limited to sociology, special education, rehabili- tation counseling, and psychology; and (ii) One year of supervised social work experience in a health care setting working directly with indivi- duals” (Centers for Medicare and Medicaid Ser- vices [CMS], 2015, pp. 122–123). In contrast, NASW (2003b) specifies that NH social workers should possess at least a bachelor’s degree from an accredited school of social work and have two years of postgraduate experience, in addition to state licensure or credentialing by NASW. Con- sequently, professional social workers differ in several important ways from other social service staff who do not have a bachelor’s degree, a license, or the professional obligation to complete continuing education (Bern-Klug et al., 2009; Cassie, 2015).

Most NH residents have at least one psychosocial service need, and a national review found gaps in addressing these needs (Levinson, 2013; Rehnquist, 2003). Although adequate staffing in NHs has re- ceived much attention, much of the focus has been on nursing rather than social services (Dellefield, Castle, McGilton, & Spilsbury, 2015; Harrington, Zimmerman, Karon, Robinson, & Beutel, 2000). In NHs, social service staff have a vital role in pro- viding good psychosocial care and making sure adequate care plans are developed and followed (American Geriatrics Society and American Associ- ation for Geriatric Psychiatry, 2003; Zhang et al., 2009), but little is known about how social service departments are organized to provide psychosocial care and the determinants of social service staffing patterns in NHs.

Examining the structure of social service staffing in NHs is the first step to understanding how psy- chosocial services for NH residents may be im- proved (Simons, Bern-Klug, & An, 2012). NHs may employ QSWs, paraprofessional staff, or an interprofessional social service team comprised of both types of staff working together to deliver medically relevant social services and provide psy- chosocial care. This study investigates the structure of social service departments across the continental United States and examines the regulatory, struc- tural, and contextual characteristics that account for differences in social service staffing.

Whereas others have studied the prevalence of social workers and paraprofessional social service staff in NHs (Bern-Klug et al., 2009; Gammonley, Zhang, Frahm, & Paek, 2009; Vongxaiburana,

Thomas, Frahm, & Hyer, 2011), this study contributes to the literature by examining the determinants of staffing patterns over time for professionals, parapro- fessionals, and interprofessional teams. Furthermore, the existing literature uses older cross-sectional data and does not control for state variability that may arise due to more stringent state regulation.

METHOD CMS’s Certification and Survey Provider Enhanced Reports (CASPER) are the primary source of data. CASPER contains data that are collected and vali- dated onsite by state surveyors during the annual NH recertification process. These recertification surveys occur about once a year (at least every 15 months) and include data on social service staff and characteristics of the facility and residents, aggre- gated to the facility level. It is the most comprehen- sive source of descriptive facility-level information on NHs for Medicare- and Medicaid-certified NHs in the United States and is valid for research pur- poses (Feng, Katz, Intrator, Karuza, & Mor, 2005). Following the literature, we used CASPER survey inspections for NHs in the lower 48 states (exclud- ing Alaska and Hawaii), for the years 2009 to 2012 (N = 59,860), a sufficient time frame to run a series of random effects panel regressions.

Social Service Staffing CASPER reports information for two types of social service staff, QSWs and “other social service staff.” QSWs are individuals with a bachelor’s de- gree in social work or a human services field with at least one year of supervised social work experience in a health care setting working with the elderly. The second type of staff is “other social service staff.” We describe this role as a paraprofessional because they are not required to have bachelor’s degrees in a human services field, but are responsi- ble for providing some medical social services to residents. Although both QSWs and paraprofes- sionals have important roles in delivering psychoso- cial care in NHs (Simons et al., 2012), their roles are different. Compared with paraprofessionals, QSWs have higher education, role preparedness, salary, and licensure requirements (Bern-Klug et al., 2009). Using information from CASPER, three dichotomous dependent variables were created to describe whether the NH employs any QSWs, any paraprofessionals, or a combination of both, which we refer to as an interprofessional team.

16 Health & Social Work Volume 42, Number 1 February 2017

Regulatory, Contextual, and Structural Factors Following the Unruh and Wan (2004) adapted version of Donabedian’s (1966, 1988) structure- process-outcome framework, we examined the potential regulatory, contextual, and structural fac- tors that influence social service staffing. In this study, we conceptualized regulatory factors as fed- eral and state regulations. The contextual factors of rurality and market concentration are also thought to influence staffing. Structural factors encompass characteristics of the NH, such as ownership, part of multi-facility chain, location within a hospital or continuing care retirement community (CCRC), and presence of a special care unit, in addition to factors that influence an NH’s financial resources (that is, payer source and occupancy rate).

From a regulatory perspective, federal and state regulations should drive social service staffing deci- sions. At the federal level, social service staffing is dependent on facility size. All NHs with 121 beds or more are mandated to employ a full-time QSW (CMS, 2015, §483.15(g)). Therefore, larger NHs with more than 120 beds are likely to employ only QSWs or interprofessional social service teams. In contrast, NHs with 120 beds or fewer are required to provide medically related social services at the same quality level, but are not federally required to employ a full-time QSW. Individual states may either strengthen the federal requirements or exempt facilities from the requirement to hire a QSW. This creates significant state variation in social service pro- vider qualifications, training, and staffing (Allen, Nelson, & Netting, 2007; Bern-Klug, 2008).

We also expected to find that contextual factors such as rurality and market concentration will affect staffing decisions. Specifically, we anticipated that rural NHs will be less likely to hire QSWs and more likely to use paraprofessional staff. NHs in rural areas could have difficulty attracting QSWs, and there may also be a lack of a competitive incentive to hire more expensive QSW staff. Areas in which an NH provider makes up a larger por- tion of the total market are considered to be a more concentrated market (as measured by the Herfindahl-Hirschman Index) (Hirschman, 1964). We predicted that NHs will be less likely to hire QSW and more likely to hire paraprofessionals within higher concentrated markets, where there is less competition among providers and fewer NH options for consumers to choose from.

The structural determinants of social service staff- ing also were tested. NHs are likely constrained by their ability to pay as QSWs have higher wages than paraprofessional staff. This implies that structural characteristics, such as size of NH, will matter. NHs that are part of a health care complex or a multi- facility chain may be more likely to hire QSWs because costs can be shared across multiple units. Ownership status (that is, for-profit, nonprofit, or government) may also influence social service staff- ing. Because for-profit NHs tend to have a stronger emphasis on managing costs, they may potentially hire less expensive paraprofessionals instead of QSWs.

Hiring more expensive QSWs may also be more feasible when NHs have the potential to generate more resources or when social service staffing costs can be spread across multiple NHs or units within a health care complex (that is, hospital-based facili- ties or CCRCs). More financial resources from higher occupancy rates, better reimbursed payer mixes (that is, fewer residents funded by Medic- aid), the presence of special care units that draw additional demand, and the extent of health and psychosocial needs of residents due to a higher per- centage of mental illnesses or developmental dis- abilities were expected to be positively related to hiring QSWs.

Data Analysis Our empirical strategy used a series of random ef- fects linear panel models to examine the regulatory, contextual, and structural determinants of social service staffing decisions for NHs. First, models were estimated for NHs of all sizes, followed by models run for small NHs (120 beds or fewer) and then large NHs (121 beds or more). Last, we tested whether the determinants of social service staffing vary with NH size. As states may enforce stricter standards, the analytic model includes state fixed ef- fects to capture state differences in regulations as well as random effects at the NH level to account for unobserved differences across NHs. All standard errors were adjusted for clustering within states, and analyses were run in Stata version 14 (StataCorp, 2015).

RESULTS Descriptive information about prevalence of using QSWs and paraprofessional staff is provide in Table 1. Percentages for each category are presented for NHs

17Roberts and Bowblis / Structural and Contextual Determinants of Social Service Staffing in Nursing Homes

of all sizes, and then divided into subcategories for small (120 beds or fewer) and large (121 beds or more) NHs. Nationally, almost all NHs have some type of social service staff, with the vast majority of these NHs having a QSW (89 percent). Although 57 percent of all NHs do have paraprofessional staff, these staff are often part of interprofessional teams, with teams most prevalent in large facilities. Never- theless, 11 percent of all NHs relied exclusively on paraprofessionals. Smaller NHs were staffed exclu- sively with paraprofessionals considerably more often (14 percent versus 2 percent).

Determinants of Staffing Patterns in NHs of All Sizes Table 2 provides summary statistics and identifies the regulatory, contextual, and structural predic- tors of social service staffing decisions for NHs of all sizes. As shown, many of the factors that predict the use of QSWs are found to have opposite effects for the use of paraprofessionals and interprofes- sional social service teams.

Regulatory factors are a significant determinant of social service staffing. Our findings demonstrate that larger NHs were more likely to hire QSWs and use interprofessional teams. For the effect of bed size, we found that the use of QSWs increases until facility size reaches 121 beds. Thereafter, the

effect of size is very small because federal law re- quires these larger facilities to have a QSW on staff. In addition state regulations (not shown) had a strong effect in determining social service staffing.

Consistent with expectations, contextual factors contribute to differences in NH social service staff- ing. Compared with urban NHs, rural NHs were less likely to hire QSWs and more likely to hire paraprofessionals. This effect became stronger for NHs located in more isolated areas. In more con- centrated markets, where there is less competition among providers and consumers had fewer choices, NHs were less likely to hire QSWs and more likely to hire paraprofessionals. Interestingly, none of these factors predicted the use of interprofessional teams.

Our findings also support the importance of structural factors. Compared with for-profits, government- and nonprofit-operated NHs were more likely to employ QSWs, and nonprofits were also less likely to employ paraprofessionals or use interprofessional teams. A greater proportion of hospital-based NHs employ QSWs and were less likely to hire paraprofessionals. Paraprofessionals and the interprofessional team were less commonly found in CCRCs.

Factors influencing NH financial resources also mattered. NHs with higher occupancy rates were more likely to use QSWs and interprofessional teams. QSWs were less common among NHs with fewer residents funded by Medicaid. Yet, interpro- fessional teams were more common among NHs focused on rehabilitation, with a higher percentage of short-stay residents funded by Medicare. Sur- prisingly, other resident characteristics were not related to social service staffing.

Determinants of Staffing: Comparisons between Small and Large Facilities For a more nuanced understanding of the determi- nants of social service staffing, we separated small facilities (fewer than 121 beds, n = 11,220) from larger facilities (121 beds or more, n = 4,782) and re-ran the analyses for each sample. Next, we tested for statistically significant differences bet- ween small and large NHs. Table 3 reports the coefficient estimates for factors that were found to be statistically different between small and large NHs at the 10 percent, with differences at the 5 percent in italics and 1 percent level in bold.

Across all three staffing dependent variables, an increase in the number of beds in small facilities

Table 1: National Summary of Presence of Qualified Social Workers and

Paraprofessional Social Service Staff in All Nursing Homes and by Size of Facility

Paraprofessional Social Service Staff

Qualified Social Workers

None Present Total

Panel A: All facilities (%) None 0.48 42.37 42.85 Present 10.46 46.69 57.15 Total 10.94 89.06

Panel B: < 121 beds (%) None 0.62 43.98 44.60 Present 14.22 41.18 55.40 Total 14.84 85.16

Panel C: 121+ beds (%) None 0.17 38.70 38.87 Present 1.85 59.28 61.13 Total 2.02 97.98

Notes: Each panel of the table reports the percentage of observations calculated from a crosstab of the presence of qualified social workers and paraprofessional social ser- vice staff. The data to construct the table use all recertification surveys reported in the Centers for Medicare and Medicaid Services’ nursing facility Certification and Survey Provider Enhanced Reports for 2009 through 2012. The sample size is 59,855 observa- tions from 15,936 facilities.

18 Health & Social Work Volume 42, Number 1 February 2017

increased the probability of using a QSW, parapro- fessional, or interprofessional team. However, there was a statistically significant effect of bed size in large NHs for QSWs, but not for paraprofessionals or teams. Though not reported, we also found dif- ferences across states, suggesting that regulatory factors play an important role in determining staff- ing decisions.

Among the use of any QSW regressions, small NHs located in isolated small towns (the most rural category) were significantly less likely to hire QSWs compared with larger NHs in the same area. This may be due to inability of small NHs to find and attract QSWs. The effects of ownership by the government or a nonprofit organization, and hospital affiliation, were stronger for smaller NHs

compared with larger NHs. In addition, NHs with a higher portion of short-term residents funded by Medicare were also more likely to use QSWs.

In comparing the use of paraprofessionals in small and large NHs, facilities in micropolitan areas were more likely to use paraprofessionals in small NHs compared with large NHs. Although there was a statistically significant difference between small and large facilities, the effect of each coeffi- cient estimate was not statistically different from zero. As expected, nonprofits were less likely to use paraprofessionals, though the effect was greater in magnitude for larger NHs. Finally, larger NHs were less likely to use paraprofessionals if the NH had any non-dementia special care unit, compared with no effect for small NHs.

Table 2: Summary Statistics and Factors Influencing Social Service Staffing

Variables M (SD) Use of Any Qualified

Social Worker Use of Any

Paraprofessionals Use of Inter-

professional Teams

Regulatory factors Large size (121 + beds) 0.304 (0.460) 0.175** (0.028) 0.206** (0.039) 0.366** (0.033) Size (# of beds by 10 s) 10.865 (6.312) 0.009** (0.002) 0.019** (0.003) 0.026* (0.002) Interaction (# beds (10 s) x 121 + beds)

5.364 (8.922) −0.009** (0.002) −0.017** (0.003) −0.025** (0.002)

Contextual factors Micropolitan (Ref. = urban) 0.140 (0.347) −0.030** (0.010) 0.013 (0.017) −0.015 (0.015) Small rural town 0.116 (0.321) −0.029* (0.013) 0.032† (0.018) 0.007 (0.016) Isolated small town 0.094 (0.292) −0.065** (0.019) 0.056* (0.024) −0.006 (0.027) Concentration (HHI) 0.201 (0.238) −0.060** (0.020) 0.070** (0.027) 0.016 (0.027)

Structural factors Ownership: government (Ref. = for-profit)

0.058 (0.233) 0.027* (0.013) −0.031 (0.030) −0.002 (0.031)

Ownership: Nonprofit (Ref. = for-profit)

0.253 (0.435) 0.016† (0.010) −0.046** (0.011) −0.027* (0.013)

Multifacility chain 0.543 (0.498) −0.008 (0.009) 0.001 (0.011) −0.004 (0.011) Hospital based 0.062 (0.241) 0.036* (0.018) −0.079** (0.022) -0.039† (0.022) CCRC 0.091 (0.288) 0.011 (0.012) −0.059** (0.014) −0.041** (0.014) Occupancy rate (0–1) 0.814 (0.228) 0.037** (0.011) 0.002 (0.012) 0.036** (0.010) % Medicaid (0–1) 0.495 (0.248) −0.023* (0.010) 0.016 (0.013) −0.007 (0.011) % Medicare (0–1) 0.124 (0.122) 0.017 (0.012) 0.019 (0.016) 0.036* (0.016) Acuity index 10.072 (1.632) 0.001 (0.001) −0.001 (0.001) 0.000 (0.001) % Dementia 0.467 (0.189) 0.004 (0.006) −0.005 (0.007) −0.002 (0.007) % Psychiatric illness 0.255 (0.189) −0.004 (0.005) −0.003 (0.007) −0.007 (0.007) % Depression 0.493 (0.226) −0.002 (0.003) 0.002 (0.005) 0.002 (0.005) % Developmental disability 0.026 (0.057) 0.032† (0.018) −0.013 (0.025) 0.018 (0.026) Special Care Unit in NH (dementia)

0.162 (0.368) 0.003 (0.008) 0.004 (0.013) 0.007 (0.013)

Special Care Unit in NH (other)

0.056 (0.230) −0.019 (0.012) −0.023 (0.019) −0.037† (0.019)

Constant 0.769** (0.025) 0.543** (0.034) 0.322** (0.028) Notes: Coefficient estimates reflect the percentage point change in the presence of each type of staffing model. Standard errors are reported in parentheses below the coefficient and are robust to clustering within state. Results for the state indicator variables and year are not reported. The sample size is 59,855 observations from 15,936 facilities. HHI = Herfindahl-Hirschmann Index; CCRC = continuing care retirement community; NH = nursing home. †p < .1. *p < .05. **p < .01.

19Roberts and Bowblis / Structural and Contextual Determinants of Social Service Staffing in Nursing Homes

Interestingly, the differences in interprofessional teams between small and large NHs varied with size, presence of a non-dementia special care unit, and average resident acuity. Larger NHs with higher resident acuity were less likely to use these teams than smaller NHs. Similar to the use of any paraprofessionals, teams were also less common if the NH had a special care unit not related to dementia care among larger facilities, but no effect for smaller NHs.

DISCUSSION This study draws attention to the determinants of social service staffing decisions for NHs. Although most NHs have a QSW on-site (89 percent), nearly half of all NHs deliver social services through inter- professional teams. As expected, smaller NHs rely more on paraprofessionals to provide social services compared with larger NHs.

We found that state and federal regulations had the strongest influence on social service staffing. The high degree of variation in state NH regula- tions in QSW qualifications (Bern-Klug, 2008) contributes to differences in social service staffing decisions. Consistent with federal regulations and findings from other studies (for example, Gammonley et al., 2009), we found that larger NHs were most likely to hire QSWs. Although federal regulations increase the presence of QSWs, social service di- rectors recommend reducing high caseloads (Bern- Klug, Kramer, Sharr, & Cruz, 2010). To address this challenge, research could inform social service staffing guidelines by identifying a specific number of residents per staff member (as opposed to bed size), similar to nurse staffing ratios (Bowblis, 2011; Park & Stearns, 2009). However, regulation that increases the number and training requirements of staff will also increase the costs of providing care.

Table 3: Statistically Significant Differences in the Determinants of Social Service Staffing by Small and Large Facilities

Use of Any Qualified Social Worker

Use of Any Paraprofessionals

Use of Interdisciplinary Teams

Variable <121 Beds 121+ Beds <121 Beds 121+ Beds <121 Beds 121+ Beds

Regulatory factors # of beds (10 s) 0.009** 0.000* 0.018** 0.001 0.026** 0.002

(0.002) (0.000) (0.003) (0.002) (0.002) (0.001) Contextual factors Micropolitan (Ref. = urban) 0.022 −0.018

(0.018) (0.029) Isolated small town −0.061** 0.019

(0.017) (0.014) Structural factors Ownership: government (Ref. = for- profit)

0.038† 0.008†

(0.020) (0.004) Ownership: nonprofit (Ref. = for- profit)

0.023† −0.004 −0.042** −0.054* (0.013) (0.005) (0.014) (0.022)

Hospital based 0.038† 0.014** (0.020) (0.004)

% Medicare (0–1) 0.023† 0.001 (0.014) (0.009)

Acuity index 0.001 -0.004†

(0.001) (0.002) Special care unit in NH (other) 0.010 −0.067** −0.010 −0.070**

(0.022) (0.024) (0.023) (0.023) # Observations 41,651 18,204 41,651 18,204 41,651 18,204 # Facilities 11,220 4,782 11,220 4,782 11,220 4,782

Notes: Coefficient estimates and standard errors reported represent statistical difference between estimated effect sizes for large and small facilities (above and below 121 beds). Coefficient estimates that are statistically different between large and small facilities at the 10 percent level are reported, with statistical significance at the 5 percent level reported in italics and 1 percent level in bold. Coefficient estimates reflect the percentage point change in the presence of each type of staffing outcome. Standard errors are reported in paren- theses below the coefficient and are robust to clustering within state. Results for statistically different state indicator variables and year are not reported. NH = nursing home. †p < .1. *p < .05. **p < .01.

20 Health & Social Work Volume 42, Number 1 February 2017

The Bureau of Labor Statistics (2015) reported that the median hourly wage for QSWs in NHs was $22.86, compared with $15.74 per hour for para- professional social and human service assistants. Going forward, stricter regulations and guidelines may be the most effective policy tools for improv- ing psychosocial support and other quality out- comes (Walshe, 2001).

Our findings also illustrate rural–urban dispari- ties in social service staffing. With high rates of chronic diseases, older adults living in rural areas have substantial service needs as they experience more physical impairment over a longer period of time compared with urban elders, yet there are fewer social and health care services available (Hash, Krout, & Jurkowski, 2015; Laditka, Laditka, Olatosi, & Elder, 2007). Although NH utilization rates are high (Phillips, Hawes, & Leyk Williams, 2003), evidence suggests that rural NHs need to improve quality, as a number of studies have found lower quality of care in rural areas (Bowblis, Meng, & Hyer, 2013; Kang, Meng, & Miller, 2011; Lut- fiyya, Gessert, & Lipsky, 2013).

Future research is needed to fully explain these rural staffing trends. Beyond the increased likeli- hood of urban NHs using QSWs, our findings sug- gest that rural NHs may hire paraprofessionals to provide all psychosocial care. A role substitution of paraprofessionals for professionally trained staff is troubling because of concerns about the quality of dementia and mental health care in NHs (Grabowski, Aschbrenner, Rome, & Bartels, 2010; Kamble, Chen, Sherer, & Aparasu, 2009; Shea, Russo, & Smyer, 2000). It is possible that the lack of market competition places less emphasis on demonstrating professional qualifications of social service staff. Moreover, rural NHs may have more difficulty with recruiting QSWs. Personal and professional challenges are associated with working in a rural setting, such as lower pay, few resources for refer- rals, and the potential of dual relationships or other ethical issues (Daley & Hickman, 2011; Strom- Gottfried, 2005).

Increased regulation may be an option; how- ever, other structural factors within NHs also influ- ence social service staffing decisions. Nonprofit organizations consistently hired more professional staff, regardless of facility size. More notably, NHs with the ability to generate more financial re- sources through high occupancy, NHs with fewer residents relying on Medicaid to pay for care, and

NHs that could spread costs across multiple units were more likely to employ QSWs. Smaller NHs with more Medicare-reimbursed residents were also more likely to hire QSWs. Without increases in private pay fees or Medicaid reimbursement, hiring more expensive QSWs could potentially impose unfunded costs that may result in lower quality in other aspects of care.

IMPLICATIONS FOR SOCIAL WORK PRACTICE, EDUCATION, AND RESEARCH The structure and determinants of social service staffing decisions within NHs have many implica- tions for social work practice, social work educa- tion, and research. With approximately half of all NHs relying on interprofessional teams, QSWs and paraprofessionals are working together to enhance psychosocial care. Therefore, QSWs are in a position to educate, mentor, and supervise coworkers who may not have a background in social work or access to continuing education or specialized training in gerontology. To meet the growing demand for gerontological social workers (Institute of Medicine, 2008), social work education must ensure that a professionally trained workforce is developed and providers are fully informed about the value of this workforce. A comprehensive resource is available online at the CSWE Gero-Ed Center (National Center for Gerontological Social Work Education) for faculty, students, and practitioners interested in enhancing competencies for geronto- logical social work practice.

Findings from this study also show the power of federal and state regulations, which are driven by the size of the facility, in determining the hiring practices. Federal law requires a QSW in large NHs only, rather than consistent requirements for a QSW in NHs of all sizes. As reported earlier, a sub- stantial portion of NHs (11 percent) rely on para- professionals exclusively. This guideline based on bed size alone does not account for the level of psy- chosocial need among residents in smaller NHs, nor does it consider the caseload of practitioners. Advocacy at the state and federal levels is needed to influence policy around the qualifications of social workers in long-term care and appropriate roles and responsibilities of social service staff.

Understanding the structural and contextual fac- tors of NHs that influence social service staffing decisions highlights potential high-need areas for social work services, such as smaller NHs in rural

21Roberts and Bowblis / Structural and Contextual Determinants of Social Service Staffing in Nursing Homes

areas. It is particularly concerning that QSWs were less likely to be hired in rural NHs because these facilities have a higher percentage of residents with mental health problems compared with urban facilities (Bowblis et al., 2013). One of our major findings revealed that paraprofessionals may be ex- pected to provide all psychosocial care, as it appears that they function as a substitute for QSWs in some situations.

Moving forward, research must continue to examine workforce issues in long-term care; pro- vide leadership and identify best practices; and evaluate relationships between social service staff- ing, processes, and psychosocial quality outcomes. Additional research is needed to determine the appropriateness and efficiency of these staffing de- cisions for providing psychosocial care, addressing common practice challenges, and providing spe- cialized care for residents with mental illness and dementia.

CONCLUSION Although regulation at the state and federal levels is one of the biggest factors that influence social service staffing in NHs, additional structural and contextual factors of NHs determine staffing patterns. More research is needed to investigate the appropriateness and efficiency of these staffing decisions for providing psychosocial care and addressing practice challenges. Future studies should consider the optimal combina- tion of staff education, skills, and interprofessional social service responsibilities to inform policy and shape NH practice. HSW

REFERENCES Allen, P. D., Nelson, H. W., & Netting, F. E. (2007). Cur-

rent practice and policy realities revisited: Under- trained nursing home social workers in the U.S. Social Work in Health Care, 45(4), 1–22.

American Geriatrics Society and American Association for Geriatric Psychiatry. (2003). Consensus statement on improving the quality of mental health care in U.S. nursing homes: Management of depression and behav- ioral symptoms associated with dementia. Journal of the American Geriatrics Society, 51, 1287–1298.

Bern-Klug, M. (2008). State variations in nursing home social worker qualifications. Journal of Gerontological Social Work, 51, 379–409.

Bern-Klug, M., & Kramer, K.W.O. (2013). Core functions of nursing home social service departments in the United States. Journal of the American Medical Directors Association, 14(1), 75e1–75e7.

Bern-Klug, M., Kramer, K.W.O., Chan, G., Kane, R., Dorfman, L. T., & Saunders, J. B. (2009). Characteris- tics of nursing home social service directors: How

common is a degree in social work? Journal of the American Medical Directors Association, 10(1), 36–44.

Bern-Klug, M., Kramer, K.W.O., Sharr, P., & Cruz, I. (2010). Nursing home social service directors’ opi- nions about the number of residents they can serve. Journal of Aging & Social Policy, 22(1), 33–52.

Bowblis, J. R. (2011). Staffing ratios and quality: An analysis of minimum direct care staffing requirements for nurs- ing homes. Health Services Research, 46, 1495–1516.

Bowblis, J. R., Meng, H., & Hyer, K. (2013). The urban– rural disparity in nursing home quality indicators: The case of facility-acquired contractures. Health Services Research, 48, 47–69.

Bureau of Labor Statistics. (2015). May 2015 national industry-specific occupational employment and wage esti- mates: NAICS 623100—Nursing care facilities (skilled nursing facilities). Retrieved from http://www.bls.gov/ oes/current/naics4_623100.htm

Cassie, K. (2015). Continuing education practices and inter- est of nursing home social services staff. Professional Development: The International Journal of Continuing Social Work Education, 18(2), 47–54.

Centers for Medicare and Medicaid Services. (2015). United States Code of Federal Regulations (CRF) state operations manual (Section 42, part 483.15(g)). Baltimore: Author.

Council on Social Work Education. (2015). Accreditation: 2015 EPAS. Retrieved from http://www.cswe.org/ Accreditation/EPASRevision.aspx

Daley, M. R., & Hickman, S. (2011). Dual relationships and beyond: Understanding and addressing ethical challenges for rural social work. Journal of Social Work Values and Ethics, 8(1), 1–14.

Dellefield, M. E., Castle, N. G., McGilton, K. S., & Spilsbury, K. (2015). The relationship between regis- tered nurses and nursing home quality: An integrative review (2008–2014). Nursing Economics, 33(2), 95–108.

Donabedian, A. (1966). Evaluating the quality of medical care. Milbank Quarterly, 44, 166–203.

Donabedian, A. (1988). The quality of care: How can it be assessed? Journal of the American Medical Directors Associa- tion, 260, 1743–1748.

Feng, Z., Katz, P. R., Intrator, O., Karuza, J., & Mor, V. (2005). Physician and nurse staffing in nursing homes: The role and limitations of the Online Survey Certifi- cation and Reporting (OSCAR) system. Journal of the American Medical Directors Association, 6, 27–33.

Gammonley, D., Zhang, N. J., Frahm, K., & Paek, S. C. (2009). Social service staffing in U.S. nursing homes. Social Science Review, 83, 633–650.

Grabowski, D. C., Aschbrenner, K. A., Rome, V. F., & Bartels, S. J. (2010). Quality of mental health care for nursing home residents: A literature review. Medical Care Research and Review, 67, 627–656.

Harrington, C., Zimmerman, D., Karon, S. L., Robinson, J., & Beutel, P. (2000). Nursing home staffing and its rela- tionship to deficiencies. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 55B(5), S278–S287.

Harris-Kojetin, L., Sengupta, M., Park-Lee, E., & Valverde, R. (2013, December). Long-term care services in the United States: 2013 Overview. Vital and Health Statis- tics, Series 3(37), 1–107.

Hash, K. M., Krout, J. A., & Jurkowski, E. T. (2015). Aging in rural places: Policies, programs, and professional practice. New York: Springer.

Hirschman, A. O. (1964). The paternity of an index. Ameri- can Economic Review, 54, 761–762.

Institute of Medicine. (2008). Retooling for an aging America: Building the health care workforce. Washington, DC: National Academies Press.

22 Health & Social Work Volume 42, Number 1 February 2017

Kamble, P., Chen, H., Sherer, J. T., & Aparasu, R. R. (2009). Use of antipsychotics among elderly nursing home residents with dementia in the US: An analysis of national survey data. Drugs & Aging, 26, 483–492.

Kang, Y., Meng, H., & Miller, N. A. (2011). Rurality and nursing home quality: Evidence from the 2004 National Nursing Home Survey. Gerontologist, 51, 761–773.

Laditka, J. N., Laditka, S. B., Olatosi, B., & Elder, K. T. (2007). The health trade off of rural residence for impaired older adults: Longer life, more impairment. Journal of Rural Health, 23(2), 124–132.

Levinson, D. (2013). Skilled nursing facilities often fail to meet care planning and discharge planning requirements. Washington, DC: U.S. Department of Health and Human Services, Office of the Inspector General.

Lutfiyya, M. N., Gessert, C. E., & Lipsky, M. S. (2013). Nursing home quality: A comparative analysis using CMS Nursing Home Compare data to examine differ- ences between rural and nonrural facilities. Journal of the American Medical Directors Association, 14, 593–598.

National Association of Social Workers. (2003a). NASW standards for continuing professional education. Retrieved from https://www.socialworkers.org/practice/ standards/naswcontinuingedstandards.pdf

National Association of Social Workers. (2003b). NASW standards for social work services in long-term care facilities. Washington, DC: Author.

National Association of Social Workers. (2015). Code of ethics of the National Association of Social Workers. Retrieved from https://www.socialworkers.org/ pubs/code/code.asp

Park, J., & Stearns, S. C. (2009). Effects of state minimum staffing standards on nursing home staffing and quality of care. Health Services Research, 44(1), 56–78.

Phillips, C. D., Hawes, C., & Leyk Williams, M. (2003). Nursing homes in rural and urban areas, 2000. College Station: Texas A&M University System Health Sci- ence Center, School of Rural Public Health, South- west Rural Health Research Center.

Rehnquist, J. (2003). Psychosocial services in skilled nursing facil- ities. Washington, DC: U.S. Department of Health and Human Services, Office of the Inspector General.

Shea, D. G., Russo, P. A., & Smyer, M. A. (2000). Use of mental health services by persons with a mental illness in nursing facilities: Initial impacts of OBRA87. Journal of Aging & Health, 12, 560–578.

Simons, K. V. (2006). Organizational characteristics influ- encing nursing home social service directors’ qualifica- tions: A national study. Health & Social Work, 31, 266–274.

Simons, K. V., Bern-Klug, M., & An, S. (2012). Envision- ing quality psychosocial care in nursing homes: The role of social work. Journal of the American Medical Directors Association, 13, 800–805.

StataCorp. (2015). Stata statistical software: Release 14. Col- lege Station, TX: Author.

Strom-Gottfried, K. (2005). Ethical practice in rural envir- onments. In L. H. Ginsberg (Ed.), Social work in rural communities (4th ed., pp. 141–155). Alexandria, VA: CSWE Press.

Unruh, L., & Wan, T.T.H. (2004). A systems framework for evaluating nursing care quality in nursing homes. Journal of Medical Systems, 28(2), 197–214.

Vongxaiburana, E., Thomas, K. S., Frahm, K. A., & Hyer, K. (2011). The social worker in interdisciplinary care planning. Clinical Gerontologist, 34, 367–378.

Vourlekis, B., Zlotnik, J. L., & Simons, K. (2005). Evaluating social work services in nursing homes: Toward quality psycho- social care and its measurement—A report to the profession

and blueprint for action. Washington, DC: Institute for the Advancement of Social Work Research.

Walshe, K. (2001). Regulating U.S. nursing homes: Are we learning from experience? Health Affairs, 20(6), 128–144.

Zhang, N. J., Gammonley, D., Seung Chung, P., & Frahm, K. (2009). Facility service environments, staffing, and psychosocial care in nursing homes. Health Care Financing Review, 30(2), 5–17.

Amy Restorick Roberts, PhD, MSSA, LSW, is assistant professor of social work and research fellow, Scripps Gerontology Center, Miami University, 101J McGuffey Hall, 201 E. Spring Street, Oxford, OH 45056; e-mail: [email protected]. John R. Bowblis, PhD, is associate professor of economics, Farmer School of Business, and research fellow, Scripps Gerontol- ogy Center, Miami University, Oxford, OH.

Original manuscript received October 8, 2015 Final revision received December 29, 2015 Editorial decision February 22, 2016 Accepted March 2, 2016 Advance Access Publication December 7, 2016

23Roberts and Bowblis / Structural and Contextual Determinants of Social Service Staffing in Nursing Homes

Copyright of Health & Social Work is the property of Oxford University Press / USA and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.