Journals of Gerontology Series B-Psychological Sciences and Social Sciences最新文献

Objectives: Mortality prediction and the identification of mortality risks are central to social and biological sciences. Traditional models often assess linear associations between single risk factors and mortality. Transformer models, capable of capturing long-term dependencies across multiple variables, offer a novel approach to mortality prediction. This study introduces a transformer-based model applied to data from the Health and Retirement Study (HRS).

Methods: We analyzed data provided by 38,193 adults aged ≥ 50 years participating in the HRS, a longitudinal U.S. study surveyed biennially since 1992. Linked mortality data were obtained from the National Death Index and postmortem interviews. Using the transformer architecture, we modeled changes in 126 risk factors spanning financial, physical, and mental health domains manifesting over 29 years. Prediction performance was assessed across multiple settings, with traditional statistical and machine learning (ML) models serving as benchmarks.

Results: Over a median follow-up of 9 years, 17,448 deaths occurred (crude rate: 39.6 per 1,000 person-years). The transformer model consistently outperformed traditional and ML methods, achieving a 2-fold improvement in average precision scores for next-wave mortality prediction relative to the best benchmark model.

Discussion: Transformer-based models, such as BEHRT, significantly enhance mortality prediction compared with traditional approaches. These findings highlight the potential of transformer neural network models in social science-focused population health research on aging.

Objectives: Education is among the most robust predictors of cognitive health outcomes in later life. However, few studies have comprehensively evaluated whether and how much of this effect is explained by occupational exposures. This study aims to determine if and how much pre-retirement occupational exposures (occurring before age 60) mediate the association between education and cognitive function at age 65+.

Methods: We use data drawn from the Health and Retirement Study (HRS) and Occupation Information Network (O*NET) data. Informed by previous research and theory, we conducted Confirmatory Factor Analyses of occupation-level exposure measures using a longitudinal HRS-O*NET linked data set we created, and we identified 2 latent factors: occupational hazards and occupational complexity. Among initially employed adults (age 51-60 at baseline), we used Structural Equation Modeling (SEM) to evaluate the association between education and cognitive function at age 65+, and the role of our 2 occupational factors in mediating this association.

Results: The measurement and structural models both had good model fit (TLI, CFI ≥ 0.95, SRMR < 0.08). We found (a) that education remained a critical predictor of cognitive outcomes in later life even when accounting for occupational exposures, and (b) only hazardous exposures mediated the association between education and cognitive function in later life (a2b2=0.02, p = .01), explaining about 17% of the effect of education.

Discussion: These findings suggest interventions designed to decrease exposure to hazardous occupational exposures could reduce some of the cognitive disadvantages in later life associated with lower levels of education.

Objectives: Device-based sedentary time shows a nonlinear association with incident dementia among older adults. However, associations between sedentary time and cognitive performance have been inconsistent. We examined potential nonlinear associations between sedentary time and performance on cognitive tests among older adults.

Methods: We used data from the UK Biobank and included 32,875 adults aged 60-79. Sedentary time was estimated from a machine learning-based analysis of 1 week of wrist-worn accelerometer data. The primary outcomes were performance on 6 cognitive tests completed online (fluid intelligence test, short-term numeric memory test, symbol substitution test, visual-spatial memory test, alphanumeric, and numeric trail making tests), as well as a composite cognitive score.

Results: Except for the visual-spatial memory test, nonlinear approaches provided a better fit than linear methods to model the associations of sedentary time with other cognitive outcomes. For these outcomes, segmented regression models showed that, although effect sizes were small, higher sedentary time was associated with better cognitive performance up to a threshold of sedentary time that varied from 9.7 to 12.3 hr per day. Above this threshold, the association between sedentary time and cognitive performance was attenuated toward the null or became negative (for the symbol substitution test only).

Discussion: As accounted by our nonlinear approach, the association between sedentary time and cognitive performance may shift from positive to null or negative above a 10-12-hr threshold among older adults. A combination of device-based and self-report assessments of sedentary behavior is needed to better understand these nonlinear associations.

Objectives: This study examines how later-life perceptions of everyday discrimination vary as a function of cohort-level differences in exposure to 3 distinct racialized social systems (RSSs) across historical time (i.e., whether individuals experienced childhood and early adolescence during the Pre-Brown v Board Era, Protest Era, or Colorblind Era).

Methods: We used the Health and Retirement Study to estimate age-specific trajectories in 5-item Everyday Discrimination Scale (EDS-5) scores and counts of discriminatory situations among aging Black Americans. We then examined the likelihood of trajectory group membership as a function of cohort-level differences using multinomial logistic regression.

Results: Findings provide evidence of variation in EDS-5 score trajectories and in counts of discriminatory situation trajectories across birth cohorts. After adjustments, relative to members of the pre-Brown cohort, members of the Protest cohort are statistically significantly more likely to be members of the trajectory group characterized by high and rapidly declining levels of EDS-5 scores relative to a trajectory characterized by modest and declining EDS-5 scores. We found more evidence for variation in situational trajectory group membership across birth cohorts; however, this variation did not correspond with a clear pattern in terms of younger cohorts consistently experiencing either more or less discriminatory situations than their older counterparts.

Discussion: This study underscores the importance of examining perceived discrimination using multiple measurement approaches and efforts to disentangle the role of exposure to historically varying RSSs in contributing to perceptions of unfair treatment at older ages.

Objectives: Positive general views on and self-perceptions of aging (SPAs) are important for active and healthy aging. This cross-sectional study compared the general views on aging and SPAs of dementia caregivers with those of noncaregivers.

Methods: Data from the German Aging Study comprising 190 caregivers (mean age = 65.69 years; SD = 10.11) and 4,480 noncaregivers (mean age = 68.81 years; SD = 10.49) were used. The onset of old age was used to assess general views on aging. Estimated own reachable age, felt age, attitudes toward own aging, and aging-related cognitions were used to assess SPAs. Sociodemographic variables, health conditions, and depressive symptoms were used as covariates. Unadjusted and adjusted linear regression models were used.

Results: Onset of old age did not differ between dementia caregivers and noncaregivers. Dementia caregivers and noncaregivers had similar felt age, attitudes toward own aging, expected own reachable age, and aging-related cognitions on physical losses, ongoing development, and self-knowledge. Compared to noncaregivers, caregivers reported lower negative aging-related cognitions on social losses (adjusted B = -0.60; 95% CI: -0.93, -0.73; standardized beta = -0.05).

Discussion: Except from aging-related cognitions on social losses, dementia caregivers and noncaregivers had similar general views on aging and SPAs. Hence, despite close contact with people who have complex conditions such as dementia, dementia caregivers may not be at greater risk of negative general views of aging and SPAs.

Objectives: Little is known about how individual health conditions and residential characteristics jointly shape the risk of opioid use disorder (OUD) among older adults. This study examines hypotheses drawn from the collective resources and relative deprivation models to fill this gap.

Methods: Applying the cases/controls matching technique to the 2018-2021 Medicare Fee-for-Service Part A and Part B claims data, this study constructs a case-control data set containing 92,244 older (65+) beneficiaries with OUD and 372,310 older beneficiaries without OUD. The beneficiary-level information is merged with characteristics of residential counties drawn from the 2016-2020 American Community Survey. Interaction terms between the beneficiary's health conditions and county features, namely concentrated disadvantage and residential instability, are used to test the hypotheses.

Results: The multilevel logistic regression results support the relative deprivation hypothesis as gaps in the predicted probability of having OUD for older adults with more versus fewer health conditions narrow in areas with more concentrated disadvantage. Older beneficiaries with poor health have a lower risk of OUD when living in socioeconomically disadvantaged counties than in affluent counties. Results also show that poor mental health and physical conditions are positively associated with the risk of OUD, and including county-level characteristics does not alter the relationships found at the beneficiary level.

Discussion: Older adults' residential environment not only contributes to the risk of OUD but also moderates the relationships between health conditions and the risk of OUD. It is critical to consider the residential environment when developing interventions to reduce OUD among older adults.

Objectives: Greater perception of control over important life outcomes has been linked to better cognitive performance and greater benefits from cognitive training interventions among older adults. However, it is not yet known whether perceived control predicts incident Alzheimer's disease and related dementias (ADRD) or influences links between cognitive training and ADRD incidence.

Methods: Data were obtained from 2,021 participants in the Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE) trial. Baseline perceived control was measured with the Personality in Intellectual Aging Contexts Inventory. Incident diagnosed ADRD over 20 years was operationalized using Centers for Medicare and Medicaid Services claims data. Main effects of perceived control and interactions between perceived control and intervention group (reasoning, memory, speed, control) on diagnosed ADRD incidence were estimated using cause-specific hazard models.

Results: Greater perceived control at study baseline was associated with reduced hazard of ADRD. Perceived control did not moderate associations between intervention group and time to ADRD diagnosis.

Discussion: Strengthening perceived control through individual and/or environmental interventions may lower ADRD risk.

Objectives: Larger social networks are associated with a lower risk of dementia, but little is known about how social networks shift as someone with dementia approaches death. We investigate these shifts while giving special attention to race and ethnicity, which are related to different dementia patterns, social network sizes, and social network makeup.

Methods: This study included 2,301 deceased people with dementia from the Health and Retirement Study (2004-2018; waves = 8). Multilevel models estimated associations between dementia, race/ethnicity, time, and close family and friend network size while controlling for several variables, including instrumental activities of daily living, age, proxy status, and disease count, using retrospective and proxy data.

Results: Social networks shrank linearly as death approached. A decrease in close friends primarily drove shrinkage. However, when race/ethnicity was crossed with time, Hispanic/Latino persons with dementia showed the opposite pattern. As Hispanic/Latino persons with dementia approached death, the number of close extended family members increased dramatically: one additional person every 4 years.

Discussion: Dementia risk, social networks, and patterns of social network shrinking are unequal across people of different races and ethnicities. Adding nuance to known patterns, network shrinkage may be a phenomenon of White persons with dementia. In contrast, patterns of stable or even increasing numbers of network members may better describe Black and Hispanic/Latino networks, respectively. These findings may reveal unique strengths of Black and Hispanic/Latino networks that could be leveraged to develop care and support for individuals with dementia and those they leave behind.

Objectives: Cultural engagement (e.g., going to museums, the theater, and concerts) has been evidenced to support older adults' well-being. However, whether cultural engagement is associated with multiple well-being domains and whether associations vary by sociodemographics and health warrants further investigation.

Methods: Using 14 years of data from the English Longitudinal Study of Ageing, we tested the longitudinal associations between cultural engagement and 7 well-being outcomes among 6,932-10,428 individuals aged 50-99 years. We used fixed-effects regression to explore the longitudinal associations between cultural engagement and the outcomes, generalized method of moments estimators to assess directionality, and interactions to test for moderation effects of sociodemographic and health conditions.

Results: We found that increases in cultural engagement were associated with increases in life satisfaction, quality of life, happiness, and having a worthwhile life and decreases in depressive symptoms, anxiety, and loneliness. After assessing directionality, cultural engagement increases predicted decreases in depressive symptoms. Interactions suggest that older adults with a long-standing health condition, living without a partner, and who are female may experience greater well-being benefits from being culturally engaged.

Discussion: Our findings underscore the potential of cultural engagement to enhance multiple well-being domains for older adults. This emphasizes the need to ensure equitable access to cultural engagement for all older adults, particularly those facing barriers to participation and those with poorer health, who may benefit the most from such initiatives.

Objectives: Language deterioration is a marker of cognitive decline in late life. An emerging literature has examined features of language associated with executive functioning and working memory when older adults are cognitively healthy. This study aims to identify linguistic features that predict cognitive functioning in a sample of cognitively healthy individuals.

Method: Participants from the Daily Experiences and Well-being Study (DEWS; aged 65-89, N = 260) completed a battery of standard cognitive tests. They wore an Android device containing the Electronically Activated Recorder (EAR) app, which recorded ambient sound 30 seconds every 7 min for 5-6 days (N = 26,339 sound files with participant speech). Linguistic Inquiry and Word Count (LIWC) software generated linguistic features from transcriptions of recorded speech. Machine learning models (random forest classifier) were trained with the linguistic features (n = 29) to predict cognitive functioning.

Results: Principal component analysis revealed that the cognitive domains fit a single factor. The random forest classifier achieved robust model performance (accuracy = 0.72, precision = 0.74, recall = 0.91, F1-score = 0.81, and area under curve = 0.73). Linguistic features most strongly associated with cognitive functioning included: first-person singular pronouns (with worse cognitive functioning), articles, words indicating differentiation, first-person plural pronouns, and words per sentence (with better cognitive functioning).

Discussion: Findings suggest that language processes are evident across multiple domains of cognitive functioning when older adults remain within a cognitively healthy range. Use of complex language may indicate optimal cognitive functioning, a topic worthy of future investigation.