@article {12092, title = {Child{\textendash}Parent Relationships and Older Adults{\textquoteright} Health: A Cross-Cultural Comparison Between China and the United States}, journal = {Journal of Family Issues}, year = {Forthcoming}, abstract = {This study compared the associations of child?parent relationships with older adults? multidimensional health in the United States and China. Two waves of data from the US Health and Retirement Study and its sister study in China (2012?2015) were used (2174 non-Hispanic [NH] White Americans and 4467 Chinese). Linear regression models were conducted for cross-sectional and longitudinal comparisons. Results showed most child?parent relationships had nonsignificant associations with NH White Americans? health. In contrast, closer child?parent relationships were linked to fewer depressive symptoms and better cognitive function among Chinese. Co-residence was associated with poorer health among older Chinese. Over a 2-year period, living nearby was linked to poorer cognitive function among NH White Americans and fewer depressive symptoms among Chinese. Having weekly contact was predictive of better cognition among Chinese. This study revealed cross-cultural differences in the associations between child?parent relationships and older adults? health. Family relationships in accordance with Chinese culture could improve health.}, keywords = {CHARLS, child{\textendash}parent relationship, cognitive function, Cross-Cultural Comparison, Depressive symptoms, functional ability}, isbn = {0192-513X}, doi = {https://doi.org/10.1177/0192513X211064866}, author = {Lu, Peiyi and Kong, Dexia and Shelley, Mack} } @article {13695, title = {Memory trajectories and disability among older couples: the mediating role of depressive symptoms.}, journal = {Journal of Applied Gerontology, Series B, Psychological Sciences and social sciences}, year = {Forthcoming}, abstract = {

OBJECTIVE: Using a dyadic approach, this study examined the mediating effect of depressive symptoms on the longitudinal relationships between husbands{\textquoteright} and wives{\textquoteright} memory trajectories and their prospective disability status.

METHODS: Longitudinal data from the Health and Retirement Study 2004-2018 were used. Older (aged 50+) heterosexual couples who had no limitations in activity of daily living at the baseline (2004) were included (N = 1,310). Latent class growth analysis grouped wives and husbands into distinct memory trajectories in 2004-2014. A structural equation model examined the actor and partner effects of memory trajectories on depressive symptoms in 2016 and disability status in 2018. The mediating effect of depressive symptoms was tested.

RESULTS: Four distinct memory trajectories were found: persistently high, high and slow decline, moderate and slow decline, and low and rapid decline. Only the wife{\textquoteright}s low and rapid decline memory trajectory predicted her own more depressive symptoms (β = 0.588, 95\% CI = 0.209, 0.967) and her husband{\textquoteright}s more depressive symptoms (β = 0.326, 95\% CI = 0.004, 0.648). Meanwhile, depressive symptoms had strong and significant actor effects on disability (β = 0.046, 95\% CI = 0.036, 0.057 for wives; β = 0.060, 95\% CI = 0.046, 0.074 for husbands).

DISCUSSION: The wife{\textquoteright}s low and rapid decline trajectory was associated with her own and her husband{\textquoteright}s more depressive symptoms, which in turn increased the disability risk for both partners. Timely identification and treatment of memory decline among wives have the potential to mitigate couples{\textquoteright} depressive symptoms and, ultimately, disability risks.

}, keywords = {Dyad research, latent class growth analysis, Memory, Older couples, trajectory}, issn = {1758-5368}, doi = {10.1093/geronb/gbad163}, author = {Kong, Dexia and Lu, Peiyi and Jiang, Da and Chan, Helen Yue Lai} } @article {13789, title = {Racial/ethnic/gender-Based Differences in Health Trajectories Among American Older Adults: 10-Year Longitudinal Evidence from the Health and Retirement Study.}, journal = {Social Work Public Health}, year = {Forthcoming}, pages = {1-15}, abstract = {

Health disparity by race/ethnicity or gender has been well-documented. However, few researchers have examined health outcomes based on the intersection of individuals{\textquoteright} race, ethnicity, and gender or investigated various health dimensions. Guided by an intersectionality framework, this study explores racial/ethnic/gender-based differences in trajectories of multiple health outcomes over a ten-year period among American older adults. Longitudinal data from the Health and Retirement Study (2004-2014) were used ( = 16,654). Older adults (65+) were stratified into six mutually-exclusive groups based on their race, ethnicity, and gender: Non-Hispanic (NH) White Men, NH White Women, NH Black Men, NH Black Women, Hispanic Men, and Hispanic Women. Growth curve models examined the trajectories of three health measures, including cognitive function, physical function limitations (i.e. activities of daily living and instrumental activities of daily living), and depressive symptoms. NH White men and women reported significantly better outcomes in cognition and physical function trajectories than racial/ethnic minority groups. Women in all racial/ethnic groups had more depressive symptoms but better cognition than men. Hispanic women reported the most depressive symptoms. Hispanic women and NH Black women had the worst physical function limitations. NH Black men/women were the most disadvantaged in cognition. Racial/ethnic/gender-based differences were stable over time in all health trajectories. Study findings highlight the utility of an intersectional framework in understanding how multiple social identities intersect to generate protective and/or risk effects on cognitive, mental, and physical health. Multilevel intervention strategies are warranted to reduce the persistent health inequity gap.

}, keywords = {ethnicity, growth curve model, Health differences, Intersectionality, race}, issn = {1937-190X}, doi = {10.1080/19371918.2024.2320795}, author = {Kong, Dexia and Lu, Peiyi and Davitt, Joan and Shelley, Mack} } @article {12721, title = {Depressive Symptom Trajectories and Cognition Among Older American Couples: A Dyadic Perspective.}, journal = {Journal of Aging and Health}, year = {2023}, abstract = {

OBJECTIVES: This study examined whether trajectories of depressive symptoms of one spouse are associated with the other spouse{\textquoteright}s memory.

METHODS: Longitudinal data from the Health and Retirement Study (2004-2016) were used ( = 5690 heterosexual couples). Latent-class growth analysis and structural equation models examined the actor and partner effects of depressive symptom trajectories on memory.

RESULTS: Four depressive symptom trajectories were identified (i.e., persistently low, increasing, decreasing, and persistently high). Compared to the low trajectory group, the increasing and persistently high trajectories were associated with worse memory for both men and women. While none of the wives{\textquoteright} depressive symptom trajectories was significantly associated with husbands{\textquoteright} memory ( > .05), husbands{\textquoteright} decreasing trajectory was linked to wives{\textquoteright} better memory (β = 0.498, 95\% CI = 0.106, 0.890).

DISCUSSION: Older adults with increasing and persistently high depressive symptoms may experience worse memory. Psychosocial interventions targeting depressive symptoms among older men may be beneficial to their spouses{\textquoteright} memory.

}, keywords = {Couples, depressive symptom trajectory, dyadic data, latent class growth analysis, Memory}, issn = {1552-6887}, doi = {10.1177/08982643221125838}, author = {Kong, Dexia and Lu, Peiyi and Solomon, Phyllis and Woo, Jean and Shelley, Mack} } @article {13306, title = {Insomnia Symptom Trajectory of Spouse Caregivers of Older Adults with Functional Limitations.}, journal = {Clin Gerontol}, year = {2023}, pages = {1-12}, abstract = {

OBJECTIVES: This study examined the long-term impact of spouse caregiving on insomnia symptoms, compared to propensity-score matched non-caregivers.

METHODS: Health and Retirement Study data between 2006 and 2018 were used. Caregivers ( = 403) were respondents (aged 50+) who assisted their heterosexual spouses in performing (instrumental) activities of daily living at baseline. Non-caregivers were matched using a propensity score matching procedure based on baseline characteristics. Insomnia symptoms were measured every 4 years for both groups. Poisson mixed-effect models estimated the association between caregiver status and insomnia symptoms.

RESULTS: Compared to matched non-caregivers, caregivers had similar severity of insomnia symptoms at baseline ( = 0.018, 95\% CI = -0.089, 0.124) and reported a similar yearly change rate ( = -0.008, 95\% CI = -0.017, 0.001). No moderation effects of care-recipients{\textquoteright} dementia status and social support were significant.

CONCLUSIONS: In this study sample, there is no evidence that spouse caregivers, specifically those who performed light duties, experience more severe insomnia symptoms than non-caregivers.

CLINICAL IMPLICATIONS: Spouse caregiving, especially in a light-duty capacity, may not be detrimental to the caregivers{\textquoteright} sleep health. More data are needed regarding insomnia in spouse caregivers with heavy duties of care to fully assess the health impact of the caregiving experience.

}, issn = {1545-2301}, doi = {10.1080/07317115.2023.2211560}, author = {Kong, Dexia and Lu, Peiyi and Lou, Vivian W Q and Shelley, Mack} } @article {11613, title = {Advance Directives Completion among Older American Couples: A Dyadic Perspective on the Role of Cognitive Function and Other Factors.}, journal = {The Journals of Gerontology, Series B}, volume = {77}, year = {2022}, pages = {1144-1153}, abstract = {

OBJECTIVE: Dyadic perspective is scarce in existing advance directive (AD) literature. Particularly, the significance of one{\textquoteright}s own and/or one{\textquoteright}s partner{\textquoteright}s cognitive function on AD remains unknown. This study investigates the relationship of cognitive function and other factors with AD completion within the spousal context.

METHODS: Data from the Health and Retirement Study (2014-2015) were used. Older heterosexual couples (age >=65) married for 10+ years were asked if they had a living will and/or appointed a durable power of attorney for healthcare (DPAHC). Structural Equation Models examined the actor and partner effects of sociodemographic, health, and couple-level characteristics on wife{\textquoteright}s and husband{\textquoteright}s AD, respectively.

RESULTS: Moderate spousal interdependence was observed in living will (Kappa coefficient, k =0.60) and DPAHC (k=0.53). Older age and higher education were related to both spouses having AD. Less household wealth and being racial/ethnic minority were associated with a lower probability of having AD. Notable gender contrasts in actor and partner effects were found. Wife{\textquoteright}s higher level of cognitive function was associated with husband{\textquoteright}s AD completion. Contrarily, husband{\textquoteright}s lower level of cognitive function was associated with wife{\textquoteright}s AD completion. Retirement status had primarily actor effects for both husbands and wives. More chronic conditions were linked to husband{\textquoteright}s AD completion.

DISCUSSION: The spousal interdependence of AD warrants practitioner efforts to facilitate family-oriented end-of-life planning. Wives and husbands may have different thoughts regarding their spouse{\textquoteright}s cognitive capacity to surrogate. Facilitating couple-based discussions may be a feasible approach to promote engagement with AD among older adults.

}, keywords = {actor-partner interdependence model, advance directive, Couples, dyadic analysis}, issn = {1758-5368}, doi = {10.1093/geronb/gbab077}, author = {Lu, Peiyi and Kong, Dexia and Lee, Jeongeun and Shelley, Mack} } @article {11481, title = {Gender-based depression trajectories following heart disease onset: significant predictors and health outcomes.}, journal = {Aging \& Mental Health}, volume = {26}, year = {2022}, pages = {754-761}, abstract = {

BACKGROUND: Using a nationally representative sample of U.S. older adults (50+), this study investigates gender-based depression trajectories following heart disease onset and associated risk of disability and mortality over an 8-year period.

METHOD: Six waves of longitudinal data from the Health and Retirement Study (2006-2016) were used ( = 1787). Heart disease onset was defined as self-reporting no heart disease at baseline but reporting a positive diagnosis in a subsequent wave. Growth Mixture Modelling identified depression trajectories. Multinomial logistic regression models determined significant predictors of depression trajectories. Cox proportional-hazards models examined the associated disability and mortality risks.

RESULTS: Three distinct depression trajectories were identified, including persistent minimal depression (men: 68.65\%; women: 60.17\%), moderate depression (women: 29.70\%; men: 17.97\%), and chronic depression (women: 10.12\%) or emerging depression (men: 13.38\%). Younger age and depression status at baseline were associated with women{\textquoteright}s chronic depression and men{\textquoteright}s emerging depression. Chronic/emerging and moderate depression were associated with higher disability risks than was minimal depression among both women and men (hazard ratios [HR] ranged from 2.12 to 3.92, < 0.001). Only men{\textquoteright}s emerging depression was linked to higher mortality risk compared to minimal depression (HR = 2.03, < 0.001).

CONCLUSION: Longitudinal course of depression following onset of heart disease is heterogeneous in later life. Unfavorable depression trajectories (i.e. moderate, chronic, and emerging) were associated with higher disability risk compared to the minimal depression trajectory. Study findings characterize risk stratification regarding depression after heart disease onset, which can inform the development of interventions to improve health outcomes among older adults with heart conditions.

}, keywords = {Depression trajectory, Disability, gender comparison, growth mixture model, Mortality}, issn = {1364-6915}, doi = {10.1080/13607863.2021.1891202}, author = {Kong, Dexia and Lu, Peiyi and Solomon, Phyllis and Shelley, Mack} } @article {12073, title = {Intersectional Discrimination Attributions and Health Outcomes Among American Older Adults: A Latent Class Analysis.}, journal = {The International Journal of Aging and Human Development}, volume = {95}, year = {2022}, pages = {267-285}, abstract = {

Guided by an intersectionality framework, this study examined intersectional discrimination attributions and their associations with health outcomes. Older respondents (aged >=50) from the Health and Retirement Study in 2014-2015 were included ( = 6286). Their reasons for discrimination (age, gender, sexual orientation, race, national origin, religion, financial status, weight, physical appearance, disability, and others) were examined. Latent class analysis examined the subgroup profiles. Six classes were identified: class 1 (54.52\% of the sample) had no/minimal discrimination; Class 2 (21.89\%) experienced primarily ageism; class 3 (8.81\%) reported discrimination based on age/gender/national origin/race; class 4 (7.99\%) attributed discrimination to financial/other reasons; class 5 (5.87\%) experienced discrimination based on age/weight/physical appearance/disability; and class 6 (0.92\%) perceived high discrimination. Intersectional discrimination was associated with poorer self-rated health and higher depressive symptoms compared to the no/minimal discrimination group. Multiple marginalized identities co-occur and contribute to discrimination. An intersectional approach is recommended to understand discrimination in later life.

}, keywords = {Discrimination, Intersectionality, Latent Class Analysis}, issn = {1541-3535}, doi = {10.1177/00914150211066560}, author = {Lu, Peiyi and Kong, Dexia and Shelley, Mack and Joan Davitt} } @article {10996, title = {Making the Decision to Move to a Nursing Home: Longitudinal Evidence From the Health and Retirement Study}, journal = {Journal of Applied Gerontology}, volume = {40}, year = {2021}, pages = {1197-1205}, abstract = {This study applied the theory of planned behavior to examine predictors of U.S. older adults? (65+ years) intention and behavior of moving to a nursing home. Two waves of Health and Retirement Study data were used (N = 9,969). Moving intention was measured by respondents? self-reported probability to move in the next 5 years in Wave 1. Moving behavior was measured by whether they actually moved in Wave 2. Structural equation modeling was used to examine the relationships of demographic and health predictors with moving intention and behavior. For both genders, advanced age, poorer health, having fewer children, and long-term care insurance were associated with higher levels of moving intention and behavior. Men?s intention was a significant predictor of subsequent moving behavior. However, women?s intention was not associated with moving behavior, probably due to inadequate resources to support their preference. The findings provide meaningful personhood-centered insights into nursing home entry decisions.}, keywords = {decision-making, moving, nursing home, theory of planned behavior}, isbn = {0733-4648}, doi = {https://doi.org/10.1177/0733464820949042}, author = {Lu, Peiyi and Kong, Dexia and Shelley, Mack} } @article {11304, title = {Intersectional Discrimination Attributions and Health Outcome among American Older Adults}, journal = {Innovation in Aging}, volume = {4}, year = {2020}, pages = {871}, abstract = {Discrimination has been consistently documented to relate to adverse health outcomes. However, most existing research focused on a single discrimination attribution (e.g. ageism). Few studies considered multifaceted discrimination attributions. Guided by an intersectionality framework, this study examined intersectional discrimination attributions and their associations with health outcomes. Older respondents (aged \>50) from the Health and Retirement Study in 2014-2015 were included in the analysis (n=6,286). Their experiences and self-perceived reasons (age, gender, sexual orientation, race, national origin, religion, financial status, weight, physical appearance, disability, and others) for everyday discrimination were examined. Latent class analysis was employed to ascertain the profiles of subgroups characterized by their intersectional discrimination attributions. Regression models examined the correlates of the class memberships and the associations with health outcomes. Six classes were identified: class 1 (54.52\% of the sample) had no/minimal discrimination experience; Class 2 (21.89\%) experienced primarily ageism; class 3 (8.81\%) reported discrimination based on age/gender/national origin/race; class 4 (7.99\%) attributed discrimination to financial/other reasons; class 5 (5.87\%) experienced discrimination based on age/weight/physical appearance/disability; and class 6 (0.92\%) perceived discrimination from almost every aspect. Intersectional discrimination attributions were associated with poorer self-rated health, and greater levels of depressive symptoms and loneliness. The associations between intersectional discrimination and cognition were not statistically significant. This study found multiple marginalized identities co-occur and compound to contribute to perceived everyday discrimination among American older adults. Those experiencing discrimination due to multiple reasons warrant particular attention. Results underscore the utility of an intersectional approach in understanding discrimination in later life.}, keywords = {health outcomes, intersectional discrimination}, isbn = {2399-5300}, doi = {https://doi.org/10.1093/geroni/igaa057.3222}, author = {Lu, Peiyi and Kong, Dexia and Shelley, Mack and Joan Davitt} } @article {11308, title = {Racial/ethnic/gender-based Disparities in Health Trajectories among American Older Adults}, journal = {Innovation in Aging}, volume = {4}, year = {2020}, pages = {894}, abstract = {Numerous studies have examined racial/ethnic- or gender-based disparities in health. However, few examined health outcomes based on a combination of individuals{\textquoteright} race, ethnicity, and gender. Guided by an intersectionality framework, this study explores racial/ethnic/gender-based differences in older adults{\textquoteright} health trajectories over a ten-year period. Longitudinal data from the Health and Retirement Study (2004-2014) were used (n=16,654). Older adults (65+) were stratified into six groups based on their race, ethnicity, and gender, including (1) Non-Hispanic (NH) White Men; (2) NH White Women; (3) NH Black Men; (4) NH Black Women; (5) Hispanic Men; and (6) Hispanic Women. Growth curve models were used to examine the trajectories of three health indicators over time, including cognitive function, physical function (i.e. the sum of activities of daily living and instrumental activities of daily living), and depressive symptoms. The results indicated that NH White men and women outperformed racial/ethnic minority groups in cognition and physical function trajectories. Females in all racial/ethnic groups had more depressive symptoms but better cognition than their male counterparts. Hispanic women reported the most depressive symptoms. Hispanic women and NH Black women had the poorest physical function. NH Black men/women had the lowest cognition. Study findings highlighted the utility of an intersectionality framework in understanding health disparities in later life. Multiple social identities intersect with each other and generate protective and/or risk effects on cognitive, mental, and physical health status. Multilevel intervention strategies are warranted to close the health equity gap among various marginalized population groups.}, keywords = {Disparities, ethnicity, gender, Health Trajectories, race}, isbn = {2399-5300}, doi = {https://doi.org/10.1093/geroni/igaa057.3295}, author = {Kong, Dexia and Lu, Peiyi and Joan Davitt and Shelley, Mack} }