OBJECTIVE:To investigate the relationship between depressive symptoms and risk for COVID-19 hospitalization and death in a United States general population-based sample. METHODS:We studied participants enrolled into observational NIH-funded cohorts from 1971-2010 with ongoing follow-up through 2023. Pre-pandemic Centers for Epidemiologic Studies Depression (CES-D) 10-item scale scores ≥10 defined elevated depressive symptoms. Questionnaires, medical records, and death certificates classified incident COVID-19 as severe (hospitalized/fatal) or non-severe from April 2020 through February 2023. RESULTS:Of 33,565 participants, 633 (1.9%) had incident severe COVID-19 over a mean (SD) follow-up of 453 (207) days. Elevated pre-pandemic depressive symptoms were present in 4,922 (16.3%) of participants. Elevated pre-pandemic depressive symptoms increased the hazard of severe COVID-19 in fully-adjusted (aHR=1.27; 95%CI: 1.03-1.57) models. In sex-stratified models (p-interaction=0.03), elevated depressive symptoms increased the hazard in women (aHR=1.49; 95%CI: 1.17-1.90) but not in men (aHR=0.96; 95%CI: 0.67-1.39). CONCLUSIONS:Pre-pandemic depressive symptoms were associated with increased risk of severe COVID-19 among women in a large US general population-based study of adults, and associations in men were neither confirmed nor ruled out. These results support current CDC recommendations that depression be considered an underlying medical condition associated with higher risk for severe COVID-19 and suggest that increased clinical focus on risk mitigation for COVID-19 and other acute respiratory infections among patients with depressive symptoms is warranted.
OBJECTIVE:Observing a loved one undergo a stressful experience can be a stressor in and of itself. However, the extent to which an observer can experience or "catch" another person's stress may depend on the observer's available resources. Here, we test how personal resources shape empathic stress by examining the impact of sleep deprivation on people's biopsychosocial responses while observing a romantic partner undergo a stressful experience. METHODS:In the lab (Nfinal=136 individuals, 68 couples), one partner underwent the Trier Social Stress Test (TSST) while the other partner observed. Prior to the lab visit, half of observers were randomly assigned to undergo partial sleep deprivation for two nights. During the TSST, we assessed observers' and speakers' subjective stress, non-verbal behavior, and physiological reactivity (cortisol and autonomic nervous system reactivity). We then examined the extent to which observers showed stress reactivity during their partner's TSST (i.e., vicarious stress) and the extent to which observers' stress reactivity covaried with speakers' stress reactivity (i.e., stress contagion). RESULTS:Observers showed a vicarious stress response across self-report and physiological measures, but there was little evidence of stress contagion. That is, although watching one's partner undergo a stressor was associated with increased affective and physiological reactivity, those responses did not consistently align with their partners' responses. Sleep deprivation did not affect empathic stress. CONCLUSIONS:Overall, these results show robust, multi-measure evidence of a vicarious stress response for observers that was unaffected by two nights of partial sleep deprivation.
OBJECTIVE:To examine the association between body roundness index (BRI) and cardio-metabolic-depression multimorbidity. METHODS:This study used data of two prospective cohorts: China Health and Retirement Longitudinal Study (CHARLS) and English Longitudinal Study of Ageing (ELSA). Firstly, the relationship between baseline BRI and the incidence of cardio-metabolic-depression multimorbidity was analyzed using Cox regression models. Subsequently, stratified Cox models with interaction terms were conducted among different groups. To investigate the potential nonlinear relationships, restricted cubic spline (RCS) regression of hazard ratio (HR) was employed. Furthermore, longitudinal associations between BRI and cardio-metabolic-depression multimorbidity over time were estimated by using generalized estimating equations (GEE) model. RESULTS:A total of 12,840 participants from CHARLS (female: 52.7%, mean age: 58.39 y), 6962 from ELSA (female: 55.0%, mean age: 65.94 y) were included according to inclusion and exclusion criteria. Cox models revealed consistent and significant association between BRI (continuous and quartiles) and cardio-metabolic-depression multimorbidity risk. Fully adjusted HRs showed a progressive increase compared to Q1, indicating a strong graded association (P for trend <0.001). Stratified analyses showed no significant interactions by demographic and lifestyle factors. In both cohorts, RCS analyses revealed particularly steep risk gradients beyond BRI values of 4. GEE analyses confirmed significant longitudinal associations with cardio-metabolic-depression multimorbidity risk. Sensitivity analyses including landmark, outlier exclusion, and multiple imputation approach consistently supported these findings. CONCLUSIONS:Higher BRI consistently predicts cardio-metabolic-depression multimorbidity across populations in a dose-dependent patterns. These findings highlight BRI's potential for early risk stratification and integrated care approaches that target metabolic-mental health interactions.
OBJECTIVE:Lower socioeconomic status (SES) is associated with adverse health outcomes, including poor brain health and greater white matter hyperintensity (WMH) volume. Traditional SES measures may not fully capture the socioeconomic experience. This study examined associations between three SES indicators and WMH volume in a healthy, midlife adult sample. METHODS:In data from participants of the Neurobiology of Adult Health (NOAH) Project who underwent high-resolution 7T MRI, we estimated the association between three SES measures: household income; income relative to neighborhood average; and subjective SES with WMH volumes. We fit linear regression models estimating associations between SES and log-transformed WMHs, controlling for age, race, ethnicity, and household size. We tested effect measure modification by sex using its interaction with SES measures. We investigated whether subclinical cardiovascular disease or cardiovascular risk mediated these associations. RESULTS:Participants (N=189) were aged 28-56 (60% female). Neither absolute nor relative household income was associated with WMH volume. However, a significant interaction emerged between subjective SES and sex (β=-0.26, 95% CI: -0.44 to -0.08), indicating that lower subjective SES was associated with greater WMH volume in female but not male participants. We found no significant mediators. CONCLUSIONS:Subjective SES may be a more sensitive predictor of brain health in midlife than other SES measures, particularly for female participants. Sex differences in WMH burden and SES-health associations may reflect factors not captured in traditional models. This study highlights the value of incorporating multiple contextual SES measures and considering sex differences when assessing neurocognitive risk.
