Abstract: Identifying Patterns and Predictors of Health Behaviors Among LGBTQ+ Older Adults (Society for Social Work and Research 26th Annual Conference - Social Work Science for Racial, Social, and Political Justice)

Schedule:

Saturday, January 15, 2022

Marquis BR Salon 14, ML 2 (Marriott Marquis Washington, DC)

* noted as presenting author

Hailey Jung, PhD, Research Scientist, University of Washington, WA

Adam Davis, Doctoral Student, University of Washington, WA

Hyun-Jun Kim, PhD, Research Assistant Professor, University of Washington, Seattle, WA

Karen Fredriksen Goldsen, PhD, Professor, University of Washington, WA

Background and Purpose:

Health behaviors are significant predictors of health outcomes among LGBTQ+ older adults. However, analysis of the association of individual behaviors and outcomes may not reflect the multi-dimensionality and co-occurring patterns of these associations. Additionally, while LGBTQ+ older adults face disparities in health, many also display resiliencies that can protect against inequities in outcomes. Thus, we aim to (1) identify latent subgroups of LGBTQ+ older adults manifested by health behavior indicators; (2) to examine differences among the latent subgroups in general health as well as physical and psychological HRQOL (Health-Related Quality of Life); and (3) to examine risk and protective factors predicting the latent subgroups from the Health Equity Promotion Model.

Methods:

We conducted the latent class analysis with 14 health behavior indicators using the National Health, Aging, and Sexuality/Gender Study (NHAS) (N = 2,450), a national longitudinal study of LGBTQ+ adults aged 50 and older in the US. Indicators spanned four dimensions including risk-taking (e.g., smoking), health-promoting (e.g., physical activity), preventive care (e.g., routine checkups), and barriers to health care (e.g., distrust and health illiteracy). The three-step method was employed to examine the relationships between latent patterns of health behaviors and health outcomes as well as risk and protective factors (e.g., marginalization, identity stigma, and social support) with background characteristics and covariates (disability and depression) in the model.

Results:

Five classes model was determined to have empirical and substantive validity including C1: low risk overall (47.1%, n = 1097), C2: low risk overall but for LGBTQ+ specific barriers (19.8%, n = 461), C3: good self-managing behaviors with high barrier and low preventive care (15.7%, n = 366), C4: high preventive care with risks in drug use, physical activity, procrastination, financial, and health literacy with (10.3%, n = 241) and C5: high preventive care but high risk overall (7.0%, n = 164). C1, C2, and C3 did not differ in general health and both HRQOLs, and C4 and C5 did not differ between each other in the same outcomes. C4 and C5 had significantly lower scores on general health and physical and psychological HRQOL relative to C1, C2, and C3. Discrimination and victimization, identity stigma, and lower social support were associated with membership of C3 and C5, and microaggression with membership of C2. Being women, younger age, and having lower income, disability, and/or depression were associated with membership of higher risks classes.

Conclusions and Implications:

Findings elucidate different patterns of health behaviors among LGBTQ+ older adults: Two groups (C1 and C2) with lower risk, one group (C3) with good self-management and resilience to barriers, and two groups (C4 & C5) with higher risk despite good preventive care. Results suggest that programs and policies that promote social support, address needs of LGBTQ+ older adults with disabilities and depression, and reduce exposure to marginalization may promote health outcomes among older LGBTQ+ adults.