Author ORCID Identifier

0000-0002-7776-6341

Date of Award

8-31-2026

Document Type

Open Access Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Gerontology

First Advisor

Pamela Nadash

Second Advisor

Edward Alan Miller

Third Advisor

Third Advisor - Jeffrey Stokes Fourth Advisor - Tetyana Shippee

Abstract

ABSTRACT

THE EFFECT OF STATE POLICY ON INDIVIDUALS WITH MENTAL ILLNESS IN NURSING HOMES: A MULTILEVEL ANALYSIS

August 2026

Adrita Barooah, B.Sc., University of Delhi, India

M.S., University of Massachusetts Boston

Ph.D., University of Massachusetts Boston

Directed by Professor Pamela Nadash

The proportion of individuals living with a mental illness in U.S. nursing homes has grown substantially over the past few decades (Fashaw et al., 2020), raising serious concerns about care quality, resource adequacy, and costs to both nursing homes and government payers. The lack of adequate resources and knowledge has resulted in sub-par mental health care (O'Connor et al., 2009a), and states vary considerably in both the share of people with mental illness they serve in nursing homes and the quality of care provided (Grabowski et al., 2009; Linkins et al., 2006). Under the Nursing Home Reform Act, which was part of the Omnibus Budget Reconciliation Act (OBRA) 1987, nursing homes are required to screen all new residents for mental illness through the Pre-Admission Screening and Resident Review (PASRR) process, which ensures appropriate nursing home placement for people with mental illness by prioritizing community-based alternatives. Although all states are federally required to complete PASRR screenings, they are given considerable flexibility in implementing these requirements, resulting in variation in the prevalence of mental illness and the quality of mental health care across states (Linkins et al., 2001; O’Connor et al., 2009).

Research on mental illness in nursing homes is sparse and frequently out of date. Prior studies have not explored the role of policy with respect to mental health outcomes in nursing homes. Relatedly, no studies have looked at whether other factors -- nursing home and market factors -- moderate the relationship between state policies and the prevalence and quality of mental health care in nursing homes.

This dissertation draws on three theories -- the Socio-Ecological Model (SEM), Principal-Agent Theory (PAT), and Resource Dependence Theory (RDT) -- to examine the role of state policy in shaping mental health outcomes in nursing homes, specifically the prevalence of mental illness and the quality of mental health care. This study investigated the relationship between three policies -- PASRR, Medicaid reimbursement rates for nursing homes, and state mental health agency (SMHA) funding -- and mental health outcomes. Outcomes were assessed using two sets of variables: the prevalence of mental illness and the quality of mental health care. A systematic content analysis of states’ PASRR policy documents was conducted to quantify the robustness of a state’s PASRR policies. This measure, together with Medicaid nursing home reimbursement rates and state mental health agency expenditures, was utilized in a cluster analysis that produced a typology of states. These constructs served as the primary independent variables in a series of multilevel models examining the relationship between state policy context and the prevalence of mental illness and quality of mental health care in nursing homes, and the moderating effect of select nursing home and market factors.

Content analysis showed significant differences in state PASRR implementation, such as the thoroughness of Level I screenings, definitions of specialized services, approaches to community diversion, and support provided to nursing homes. Bivariate analysis indicated that policy features grouped together in clear, rather than random, categories, supporting the use of a composite score (ranging from 21.9 to 87) to represent overall policy coverage. Further cluster analysis revealed three unique state profiles: Lower Investment–Stronger PASRR (n=10), Lower Investment–Weaker PASRR (n=21), and Higher Investment–Moderate PASRR (n=12). These groups were identified based on combined PASRR scores, Medicaid reimbursement rates, and SMHA per-capita spending. In multilevel analysis, state policy measures generally did not show significant main effects on outcomes; however, moderation effects related to ownership-chain affiliation, Medicaid share, and market competition (HHI) emerged across several policy-outcome relationships.

The findings suggest that the link between state policy and nursing home mental health outcomes is not straightforward but is significantly shaped by conditions at both facility and market levels. Policymakers who want to improve mental health care in nursing homes should consider not only the content and robustness of PASRR policies and funding levels, but also the structural contexts in which those policies exist. Future research should continue to explore the intersection of state policy, facility characteristics, and local market dynamics on care quality for individuals with mental illness in nursing homes.

Comments

Free and open access to this work is made available to the UMass Boston community by ScholarWorks at UMass Boston. Those not on campus and those without a UMass Boston campus username and password may gain access to this work through Interlibrary Loan. If you have a UMass Boston campus username and password and would like to download this work from off-campus, click on the “Off-Campus Users” button.

Download

Share

COinS