Dementia Care Involvement and Training Needs of Social Services Directors in U.S. Nursing Homes

ABSTRACT This study describes social services directors’ involvement in dementia care in U.S. nursing homes, focusing on interest in and needs for dementia care training. Respondents were 924 social service directors from U.S. nursing homes. We found that 87% of social service departments engaged in cognitive assessment; 59% of social services directors were strongly interested in dementia care training, and 23% would need up to 10 hours of preparation time or would not be able to train staff on dementia-related care. Multinomial logistic regression analysis (n = 836) revealed that, in comparison to having no interest in dementia training, age, years of experience in nursing homes, outside mental health contracting, barriers to staffing, and hours needed to prepare dementia training predicted varying interest in dementia care training. These findings demonstrate social services directors’ active involvement in dementia care and need for training.


Background
In 2015-2016 in the U.S., 47.8% of nursing home residents were estimated to have Alzheimer's disease or other dementias; among long-term residents, dementia prevalence was 58.9% (Harris-Kojetin et al., 2019). When persons with dementia enter nursing homes, they often have advanced needs and dementia-related behavioral symptoms, including restlessness, agitation, aggression, impaired communication, and difficulties sleeping (Brasure et al., 2016;Kales et al., 2015). The primary recommended response to these symptoms comprises a range of non-pharmacologic interventions and social care that incorporates environmental and contextual influences (Zeisel et al., 2016).
Nursing home social services directly impact residents' quality of life, especially for those living with dementia, by addressing a wide range of emotional and psychosocial needs: assistance with admission, psychosocial assessment, interpersonal challenges, decision-making support, and ensuring residents' rights. In U.S. nursing homes, social service staff are lead providers of psychosocial assessment and intervention for residents (Simons et al., 2012). Their assessment of resident needs typically includes completing psychosocial items on the Minimum Data Set (MDS), a federally mandated, standardized assessment tool for all nursing homes certified for Medicare and Medicaid payments that informs treatment planning and psychosocial care for residents (Simons et al., 2012). While all staff need to be aware of residents' psychosocial needs, social service staff are responsible for assessment and care planning and provision of psychosocial services to both residents and their family members (Roberts et al., 2020;Simons et al., 2012).
Given that many residents have cognitive impairments associated with Alzheimer's disease and other dementias, interviewing and completing comprehensive assessment as well as developing care plans for residents with cognitive limitations is a necessary component of nursing home social work practice and requires special education, training and skills (Simons et al., 2012). Moreover, the qualifications of social service staff affect resident outcomes and quality of care. Two studies of nursing homes found direct relationship between the number of qualified social service staff and psychosocial care outcomes such as reduction in behavioral symptoms and use of antipsychotic medications (Roberts et al., 2020;Zhang, Gammonley, Paek, & Frahm, 2008-2009. Social staff with more years of nursing home experience also associated with better outcomes (R. P. Bonifas, 2011b).
Yet, the educational background and preparedness of social service staff varies in part due to limited federal regulations relative to the staffing of social service professionals and broad variation in the training and credentialing of such staff at the state levels (Roberts et al., 2020;Simons et al., 2012). Furthermore, there are no federal requirements that nursing home social services staff have dementia training. A primary federal regulation in Medicare-and/or Medicaid-certified nursing homes is that those with more than 120 beds are required to employ one full-time social services staff person with a college degree related to human services. Most nursing homes employ social services staff; roughly three-fourths employ a federally "qualified social worker" (Harris-Kojetin et al., 2019).
Development of a dementia care workforce in nursing homes with more thorough training is key to quality of care and life among residents living with dementia (Weiss et al., 2020). Nevertheless, few empirical studies have examined social service staff's need for and interest in dementia-care training. In a survey of nursing home social service staff in Missouri, Parker-Oliver and Kurzejeski (2003) found that one-third lacked necessary training and support to provide high-quality services for residents. Lacey (2006) reported that social service staff had varying degrees of comfort working in end-of-life dementia care and wanted more education and training in this area. Also, R. Bonifas (2011a) found dementia care the most common area of mental health knowledge among nursing home social service staff.
In this exploratory study, using nationally representative data for social services directors in U.S. nursing homes, we explored and described the role of social services departments in dementia-related care, focusing on interest in and needs for dementia-related training among social service directors. We asked the following questions: (a) Are social service departments involved in cognitive assessment relevant to dementia care? (b) To what extent do social service directors express a need for dementia-related training? (c) What social service director and facility characteristics are associated with interest in receiving dementia care training?

Data source
Data for this study are from a 2019 survey of social services directors from a sample of U.S. nursing homes. Of the 15,578 Medicare and/or Medicaid certified nursing homes in the December 2018 Centers for Medicare and Medicaid Services (CMS) Nursing Home Compare database, a random sample of 3,650 homes were contacted to determine whether they employed social services staff. Social service directors from 3,067 nursing homes that employed social services staff were then invited to participate in the study; 924 (30%) responded. Details of this procedure are provided elsewhere (Bern-Klug et al., 2021). The study was approved by the Institutional Review Board (IRB) of the University of Iowa (IRB # 201,810,727). All potential participants received informed consent documents that explained relevant information about the study including that consent was implied by completing and returning the survey.

Dementia care involvement and training interest and needs
Of 46 tasks describing various responsibilities for nursing home social services recommended by the National Association of Social Work (NASW) and Veteran's Affairs (VA) (Simons et al., 2012), we selected one item to assess staff's involvement in cognitive assessment of residents. This item asked whether social service staff were typically involved with completing Section C (Cognitive) of the CMS Minimum data set (MDS) assessment ("yes" or "no"). The cognitive section of the MDS assessment was chosen because this assessment is directly related dementia care needs assessment and care planning. For dementia care training interest, we selected an item that asked whether the respondent was interested in receiving training on how to instruct and support staff in psychosocial needs of persons with dementia. This item represented one of 15 topical areas for training interest in the survey. Response options were "no interest," "minor interest," "moderate interest," and "strong interest." For training needs in social service areas, respondents were asked how much time they would need to train someone else on how to compare and contrast dementia, depression, and delirium and anticipate common psychosocial needs related to cognitive loss. This item represented 1 of 27 areas for training needs assessed in the survey. Response options included "could do without prep time," "would need up to 2 hours of prep time," "would need up to 10 hours of prep time," and "not able to do at all."

Respondents' Background Characteristics
Background information Included age, gender, race (African American, Asian/Pacific Islander, Native American, White, Multi-racial, Other), educational attainment including social work degrees (bachelor's, master's, doctoral), and social work license or certification.

Barriers to care
Respondents were asked about barriers to providing social and emotional care. From 13 areas included in the survey, three items particularly relevant to social service departments' involvement in dementia care and training need were selected: "insufficient consultation or supervision," "not enough social services staff," and "not enough family involvement." Response options were "not a barrier," "minor barrier," "moderate barrier," and "major barrier."

Facilities' characteristics
In addition to survey responses, two CMS public data files provided additional data about facilities where respondents were employed (https://data.medicare. gov/data/nursing-home-compare). Facility characteristics included ownership (for profit, not-for-profit, government), location (metro vs. non-metro counties), U.S. census region (four regions), chain status (part of a chain; i.e., two or more facilities), and number of Medicare and/or Medicaid certified beds.

Data analysis
Descriptive statistics were used to assess variables for normality, collinearity, and missingness. All variables, except gender, met normality guidelines for skewness and kurtosis (e.g., normative ranges for skewness ±2 and kurtosis ±6 (Curran et al., 1996). A multinomial logistic regression model was used to determine likelihood differences between social services directors who reported no interest in dementia training and others who reported minor, moderate, or strong interest in dementia training. Collinearity was examined between study variables; no variable had a VIF above 5 (Craney & Surles, 2002). All tests for homogeneity of variance were non-significant, indicating no non-normal data within the model (Hosmer et al., 2013). Missingness was within acceptable limits; 9.5% (n = 88) of the original sample of 924 were missing for the multinomial logistic regression, leaving the total sample for regression analyses at 836. All analyses were completed with SPSS 27.

Sample Characteristics
The majority of nursing home social service directors were women (92%) and White (87%; Table 1). Almost half were licensed or certified in social work (47%), had a social work degree (46%), and had worked at least 10 years in nursing homes (47%). The majority of participants identified as female (92%) and White (88%). Most respondents were employed in nursing homes certified to accept both Medicare and Medicaid (96%) and located in metropolitan counties (67%). More respondents came from the Midwest region (41%) than from others. Data not shown in Table 1 also indicated that 54% of social services departments had only one staff person, 31% had two, and 9% had three. Nursing homes with a greater number of beds were significantly more likely to have more than one social service staff, X 2 (6, N = 911) = 210, p <.001. Only 15% of nursing homes with 60 or fewer beds employed more than one social service staff, whereas 41% of nursing homes with 60-120 beds and 76% of nursing homes with more than 120 beds employed more than one social services staff member.
Most participants (87%) responded that social service staff were involved in completing the cognitive section of the MDS assessment. The cognitive section was the third most frequently completed MDS section (after Section D, Mood, 95%, and Section Q, Participation assessment and goal setting, 91%). Twenty-three percent of respondents reported that they would need up to 10 hours of preparation time or would not be able to train other staff on residents' dementia, depression, delirium, and cognitively related psychosocial needs. Most indicated moderate (27%) or strong (59%) interest in receiving training related to the psychosocial needs of persons with dementia. Among 15 topic areas in our survey, this was the most frequently reported topic, with strong interest (data not shown). Respondents found moderate to major barriers to care to be insufficient consultation and supervision (29%), not enough social services staff (40%), and not enough family involvement (40%).

Factors associated with strong interest in dementia care
Given respondents' varying levels of interest in receiving dementia-related training, we conducted a multivariate analysis utilizing a multinomial logistic model to explore correlates of social service directors' interest in receiving training ( Table 2). Level of interest in dementia training was the dependent variable with each level being in reference to respondents with no interest in dementia training. All independent variables were entered simultaneously with categorical variables as factors and the only continuous variable (i.e., full-time equivalency) as a covariate in the model. The overall model was statistically significant, χ 2 (87) = 170.77, p < .001 with pseudo-R-squared values ranging between 10.1% and 21.3% of variance accounted by the model.
In the two comparisons of minor to no interest and moderate interest to no interest, the following were significant predictors. Respondents identifying as male, being 35-54 years old (only for moderate interest) and 55 years and older, having engaged in outside mental health contracting, facing minor barriers to supervision, having minor to major barriers to staffing, and needing 2 hours or more of preparation with training for dementia needs were more likely to report an interest in dementia training than were social service directors who were female, 18-34 years old, not contacting for outside mental health, facing no barriers in supervision or staffing, and needing no time for preparation of training for dementia needs. See Table 2 for the associated odds ratios and 95% confidence intervals.
In terms of the comparison between respondents with strong interest versus no interest, significant predictors as follows. Social service directors were more likely to endorse strong interest for training in dementia-related needs when identifying as male (OR = 0.01), 35 years or older compared to 18-34 years old (OR = 2.33 for 34-55 years old, OR = 3.36 for 55 and up years old), having 4-9 years of experience compared to 10+ years of experience (OR = 4.70), being in a rural part of the country (OR = 2.61), engaging in outside mental health  contracting (OR = 2.73), facing minor to major barriers to staffing compared to having no barriers (OR = 2.46 for minor barriers, OR = 11.56 for moderate to major barriers), and needing between 1 and 10+ hours, including not being able to prepare, compared to not needing any time to prepare to deliver dementia training (OR = 2.46 for up to 2 hours, OR = 10.69 for 10+ hours or unable to prep).

Discussion
Nursing home social services directly impact the quality of life of residents living with dementia, and the need for dementia care training for the nursing home workforce is a national priority (Weiss et al., 2020). In our survey, the vast majority of nursing home social services departments had completed the cognitive section of the MDS assessment, and social services directors reported greater interest in dementia care training than in any other area of training. This strong interest in dementia training was supported by almost a quarter of respondents, who reported that they would need up to 10 hours of preparation time or would not be able to train other staff on dementia, depression, delirium, and common cognitively related psychosocial needs. Competent dementia care requires knowledge of these conditions, because they can present in the same person at the same time or present separately (Simons et al., 2012). Nursing staff are keenly involved in physical care needs of residents with dementia, but social services staff must anticipate, assess, and address the psychosocial needs of persons with these multiple conditions. In terms of respondent characteristics, respondents who were 35 years or older compared to 18-34 years old, had 4-9 years of experience compared to 10+ years of experience, and needed 1-10 hours of preparation time to deliver dementia training or were not able to prepare dementia training at all were more likely to have a strong interest in dementia training than those without these characteristics. It is understandable that social services staff less experienced with nursing home care and those who need significant amount of time to training someone on dementia care would be more interested in dementia care training. However, we are uncertain why 35 years or older participants were more interested in training than younger participants. It is possible that younger participants may be more recent graduates who received education and training on aging in general and dementia specifically, given the increasing interest in and emphasis on delivering gerontological content in both undergraduate and graduate social work and social service curricula. Future studies are warranted to examine the role of professional experience, and gerontological education and training backgrounds among different cohorts of social services directors with training interest in dementia care. With regard to facility characteristics, the finding that a strong interest in training was associated with the facilities in a rural part of the country is consistent with the well-documented barriers to adequate staffing, funding, and training opportunities in rural health and long-term care settings. It may also be the case that in-house training opportunities may be limited in facilities that outsource mental health services and thus, explaining a strong interest in training among staff. For the association between the lack of sufficient social services staff and strong interest in training, it is plausible that nursing homes with insufficient social services staff are under-resourced for psychosocial care in general, and especially for dementia, which may lead to greater need for training and support for dementia care. Future studies may examine further and compare the availability and accessibility of and support for dementia care training for social service staff in both rural and urban areas.
In our study, we did not find any differences in the interest in dementia care training by educational attainment type and years of experience including whether the social service director had a social work degree or not. These findings suggest that while the formal dementia care education during undergraduate and or master's training is certainly important, continuing education and training for social service staff is critical to ensure high-quality dementia care delivered by social service staff. As such, more gerontological social work research is warranted to identify specific areas of knowledge and skills that social service staff with different educational and disciplinary backgrounds may have or need to develop and strengthen further in practice, and how they collaborate with other team members in delivering dementia care. Findings from such research may inform education and training efforts in development and dissemination of effective dementia care training programs for social services staff as well as development of policy to ensure adequate training and credentialing of social service staff.
Limitations of our study include its cross-sectional design, and the study was not theory-based. The survey targeted social service directors instead of all social service staff, although more than half of social services departments had only one staff person. The survey response rate was 30%, somewhat lower than the typical response rate for organizational surveys (35.7%; Baruch & Holtom, 2008). Additionally, although reasons are unknown, nursing homes in the Midwest were over-represented in our sample (41%). In the U.S., nursing homes in the Midwest represent 33% of all US nursing homes. The nature of skewness within the two predictor variables, gender and race/ethnicity, and wide confident intervals for two variables, experiencing moderate to major barriers to staffing and 10+ hours or unable to prepare to train others on dementia care suggest results should be interpreted with caution. In addition, our study finding is limited by using one-item indicators to measure dementia care involvement and training interest with limited reliability. Therefore, our results should be interpreted with caution.
To improve quality of life and effectively advocate for residents living with dementia, well-trained social services staff in essential care for persons with dementia is necessary ( (Gilster et al., 2018;Roberts et al., 2020;Simons et al., 2012;Weiss et al., 2020). Our study's limitations notwithstanding, its findings from the nationally representative data for social services directors in U.S. nursing homes provide clear evidence for social services staff's involvement in dementia care and interest in and need for more dementia training. There is a strong need for more systematic efforts to develop and widely disseminate dementia care education and training for social services staff in nursing homes in the U.S.