Marital Religious Homogamy and Dimensions of Well-Being in Later Life: Evidence From the United States

Abstract Past research points to the importance of couple-level religious similarity for multiple dimensions of older adults’ partnership quality and stability, but we have a limited understanding of whether religious homogamy matters for the well-being of seniors. This study uses dyadic data from the National Social Life, Health, and Aging Project (NSHAP), a representative sample of 953 individuals ages 62–91 plus their marital or cohabiting partners. Using actor-partner interdependence models in the general structural equation model framework (GSEM), we find that religious attendance homogamy is beneficial for the physical health of men and the mental health and self-reported happiness of women. There were no associations between religious homogamy for religious importance detected. Taken together, our results attest to the ongoing importance of religious similarity—service attendance, in particular—for mental and physical well-being in later life. Future research is needed to more fully examine which mechanisms account for these patterns.

PTSD affects 7-15% of older adults and is associated with elevated depressive symptoms, suicidal ideation, and poorer physical health. Individuals with sub-clinical PTSD often experience worsening symptoms, resulting in a full diagnosis of PTSD, albeit late onset. This evaluation informs the Veterans Health Administration (VHA)'s understanding of the developmental course of PTSD in older Veterans. VHA administrative data were used to examine health characteristics and service utilization in the five years before and five years after the initial VHA documentation of PTSD. We identified a cohort of Veterans (n=27,610), alive and with at least 1 encounter in all evaluation years, with a first PTSD diagnosis documented between ages 50-59, a cohort outside of the average age of diagnosis, but before Medicare eligibility. We compared periods before and after diagnosis across different ages of first diagnosis (50)(51)(52)(53)(54)(55)(56)(57)(58)(59). Veterans diagnosed at later ages (55-59) had a greater number of mental and physical health conditions. Increasing VHA use preceded a PTSD diagnosis in both groups, but the increase was steeper among those diagnosed at 55-59. Future analyses will compare these patterns to those of Veterans diagnosed at younger ages. Findings from this work may provide a profile of persons at risk for late-life PTSD for use in targeting interventions earlier to reduce the risk of developing worsening PTSD symptoms in late life.

IMPLEMENTING AN INTERDISCIPLINARY COMMITTEE TO REDUCE PSYCHOTROPIC DRUG AND INCREASE BEHAVIORAL INTERVENTION USE
Lauren Hess Conrad, 1 and David Portman, 2 1. Butler VA Health Care System,Pittsburgh,Pennsylvania,United States,2. VA Butler Healthcare System,Butler,Pennsylvania,United States In Fiscal Year (FY) 2018, the Butler VA Health Care System's Psychotropic Medication and Behavior Management Committee was identified as a Veterans Integrated Service Network 4 Best Practice. The goal of this committee is to reduce unnecessary psychotropic medication use and polypharmacy and to increase behavioral intervention implementation among Community Living Center (CLC) Veterans. This committee meets quarterly to review Psychotropic Drug Safety Initiative data, behaviors and behavior care plans, and all psychotropic medications prescribed to Veterans. Psychiatric diagnoses, changes to psychotropic medications, and appropriate behavioral interventions are discussed. Committee members take responsibility for action items in accordance with their discipline; documentation of recommendations are made in quarterly behavioral health assessments in CPRS; and follow-up on action items is completed at twice weekly interdisciplinary treatment team meetings, weekly behavior rounds, and/ or as needed. From the first quarter (Q1) of FY16 to Q1 FY20, the Butler VA CLC has seen decreased prescriptions of 2 or more anticholinergics (6.6% to 0.80%), antihistamines (12.5% to 5.9%), benzodiazepines (24.7% to 11.0%), and benzodiazepines or sedative hypnotics (23.2% to 9.0%). While prescription of antipsychotic use has increased (Q1 FY20 = 23.8%), the committee will follow Long Term Care Institute guidelines for gradual dose reductions, behavioral interventions, and as needed psychotropic medication PRN use. This committee provides an interdisciplinary forum to discuss and implement beneficial changes to pharmacological and non-pharmacological interventions among all CLC Veterans. The committee is a valuable process for monitoring and reinforcing best practices that may be easily replicated across VA CLCs nationwide.

MARITAL RELIGIOUS HOMOGAMY AND DIMENSIONS OF WELL-BEING IN LATER LIFE: EVIDENCE FROM THE UNITED STATES
Laura Upenieks, 1 Markus Schafer, 2 and Jeremy Uecker, 3 1. University of Texas at San Antonio, San Antonio, Texas, United States, 2. University of Toronto,Toronto,Ontario,Canada,3. Baylor University,Waco,Texas,United States Past research points to the importance of couple-level religious similarity for multiple dimensions of older adults' partnership quality and stability, but we have a limited understanding of whether religious homogamy matters for the well-being of seniors. This study uses dyadic data from the National Social Life, Health, and Aging Project (NSHAP), a representative sample of 953 individuals ages 62-91 plus their marital or cohabiting partners. Using actor-partner interdependence models in the general structural equation model framework (GSEM), we find that religious attendance homogamy is beneficial for the physical health of men and the mental health and self-reported happiness of women. There were no associations between religious homogamy for religious importance detected. Taken together, our results attest to the ongoing importance of religious similarity-service attendance, in particular-for mental and physical well-being in later life. Future research is needed to more fully examine which mechanisms account for these patterns. Objective and subjective cognitive function have been associated with decreased quality of life and increased psychological distress in older adults. The present study examined relations of objective and subjective cognition with quality-of-life and mental-health outcomes in individuals with amnestic mild cognitive impairment (aMCI). The sample included 98 older adults with aMCI (92.5% male, age = 70.9±9.2 years). Measures included objective cognition (i.e., attention, memory, language, visuospatial abilities, processing speed, executive function, and overall), subjective memory (Multifactorial Memory Questionnaire [MMQ]), quality of life (Dementia Quality of Life [DQoL]), and mental health (Geriatric Depression Scale, Geriatric Anxiety Inventory, and Penn State Worry Questionnaire). Objective and subjective cognition were weakly correlated (range |r| = .00-.23). Objective cognitive measures were largely uncorrelated with quality of life or mental health, with only two significant (p < .05) correlations between Processing Speed and Worry (r = -.24), and Overall Cognition and DQoL Aesthetics (r = .20). Subjective cognition was more strongly correlated with quality of life, including significant (p < .01) correlations between MMQ Abilities and DQoL Negative Affect (r = -.38), and MMQ Contentment and DQoL Positive Affect (r = .28). Additionally, MMQ Contentment and Abilities were significantly (p < .01) negatively correlated with all three mental-health outcomes (range |r| = .28-.43). This study demonstrated that subjective memory, particularly affect and self-appraisal regarding one's memory capabilities, is more closely related to quality-of-life and mental-health outcomes than objective cognitive performance in an aMCI sample, and, therefore, may represent important targets for intervention. Previous literature suggests that social factors (e.g., social cohesion, social support) are protective predictors of mental health problems. However, there might be a reciprocal relationship between social factors and mental health and the relationship changes over time. Therefore, this study examined the longitudinal relationship between community social cohesion and mental health using a latent growth curve model with 8 waves of the National Health and Aging Trends Study (NHATS;2011-2018, a nationally representative panel study of Medicare beneficiaries in the United States. Social cohesion measured the perceived level of mutual trust by three items (score range: 0-6) and mental health was measured by PHQ-4 (score range: 0-12). The final model including covariates (age, gender, functional disabilities) fit the data well: χ2=1036.383, p<.001; RMSEA=.037; CFI=.960; and SRMR=.070. Initial level of social cohesion was negatively associated with initial level of mental health problem (β=−.23, p< .001), suggesting that higher levels of social cohesion was associated with lower levels of mental health problems. The covariance between social cohesion slope and mental health slope was significant (β=−.16, p< .01), suggesting an increase in social cohesion was associated with a decrease in mental health problems over time. Functional disabilities significantly influenced mental health over time, while functional disabilities did not influence social cohesion consistently. This study adds to the growing literature on the ways mental health status and social connection have reciprocal relationships over time. Therefore, mental health status in later life could be decreased by improving social cohesion and connectedness with the community.

PATIENT-REPORTED OUTCOME MEASURES IN OLDER VETERANS INITIATING A NEW EPISODE OF MENTAL HEALTH CARE
Jenefer Jedele, 1 Karen Austin, 1 and Sandra Resnick, 2 1. Veterans Health Administration,Ann Arbor,Michigan,United States,2. Veterans Health Administration,West Haven,Connecticut,United States The VA Measurement Based Care (MBC) in Mental Health (MH) Initiative supports implementing patient reported outcome measures (PROMs) for MH treatment planning and shared decision-making as a routine aspect of care. Using VHA administrative data, we identified Veterans initiating a new MH treatment episode (index encounter), i.e. prior 6-months without VHA MH encounters. We compare MH diagnoses, medications, and encounters during the 6-months from and including the index encounter by age (50-64; 65-79; 80+) between Veterans receiving 1 or more measures (PROM) to those receiving none (noPROM). The percentage of PROM Veterans decreased with age: 26.7% (50-64); 18.5% (65-79); 12.5% (80+). Consistent across age, PROM Veterans had more encounters than noPROM Veterans. In the year before treatment initiation, a smaller percentage of PROM Veterans had multiple MH diagnoses (21.0% v. 29.1%). At treatment initiation, both groups were equally likely to have multiple diagnoses (20.7% v. 20.1%); a higher percentage of the noPROM group were diagnosed with schizophrenia (3.8% v. 1.0%), bipolar (4.5% v. 2.2%), or PTSD (29.2% v. 21.8%). Substance use disorder and major depression were more prevalent in the PROM group. These patterns held across age categories. A smaller percentage of PROM Veterans had been prescribed psychotropic medication during the index encounter (32.8% v. 42.8%). For PROM Veterans, an average of 3 measures were received 1.5 months apart. The number of measures declined and the interval between measures increased with age. Potential barriers and possible efforts to target the use of PROMs with older Veteran patients are discussed.

PERCEIVED PARTNER RESPONSIVENESS ATTENUATE THE LINK BETWEEN MENOPAUSAL SYMPTOMS AND DEPRESSED AFFECT
Kayeon Lee, Hey Jung Jun, Susanna Joo, and Sun Ah Lee, Yonsei University, Seoul, Republic of Korea The years around menopause are the time that associates not only with hormonal changes but with psychological and social transitions, and previous studies have consistently revealed the relationship between menopause and depression. The present study examined the moderating effect of perceived partner responsiveness (PPR) on the association between menopausal symptoms (MS) and depressed affect (DA). The sample was middle-aged climacteric women (N = 754, Age=49-60) from the second wave of Midlife in the United States (MIDUSⅡ). Measurement for MS consisted of the frequency of five symptoms in the past 30 days (insomnia, heavy sweating, painful intercourse, hot flashes, and irritability). PPR was assessed using three items matched the core components of responsiveness (understanding, validating, and caring). Results revealed that there were significant interactions between menopausal symptoms and PPR (b = 0.05, p < 0.039). Specifically, the level of elevation of DA in response to MS was smaller in women with higher levels of PPR (b = 2.93, p < 0.001) than in those with lower levels of PPR (b =3.05, p < 0.001). According to the region