Practical guidance for conducting mediation analysis with multiple mediators using inverse odds ratio weighting

Widening disparities in all-cause and despair-related mortality among Swedish youths: Disentangling selection and causation

Rising rates of "deaths of despair" - mortality from suicide, drug overdose, and alcohol-related causes - have contributed to widening educational disparities in mortality. It is not known to what extent the trends are due to selection effects (health causing education, or a third factor causing both) or social causation (education affecting health). This study investigated the relative contribution of selection and causation for these trends, focusing on the recently documented widening achievement-based disparities in mortality among Swedish youths. To this end, two cohorts of Swedish compulsory school graduates (graduating in 1992-1993 and 2009-2010, respectively) were followed for eight years after graduation using comprehensive administrative data (n = 424,715). Logistic regression models were used to assess the role of pre-graduation selection, while inverse odds ratio-weighting was used to assess mediation by post-graduation socioeconomic disadvantages. Roughly half of the association between low achievement and all-cause and despair-related mortality within cohorts was due to selection. However, selection effects did not explain the widening disparities over time. Socioeconomic mediators accounted for most of the remaining disparities within cohorts as well as of for most of the increase in these disparities over time. Overall, social causation was more important than selection in explaining the widening educational disparities in all-cause and despair-related mortality. We conclude that low academic achievement increasingly constrains life-course prospects of Swedish youths, amplifying its adverse health consequences. These findings highlight the need for lower barriers in the education system and for viable educational and employment pathways for low-achieving students in an increasingly knowledge-intensive labor market.

Omega-3 Fatty Acids and Risk of Ischemic Stroke in REGARDS

We examined associations between lipidomic profiles and incident ischemic stroke in the REasons for Geographic and Racial Differences in Stroke (REGARDS) cohort. Plasma lipids (n = 195) were measured from baseline blood samples, and lipids were consolidated into underlying factors using exploratory factor analysis. Cox proportional hazards models were used to test associations between lipid factors and incident stroke, linear regressions to determine associations between dietary intake and lipid factors, and the inverse odds ratio weighting (IORW) approach to test mediation. The study followed participants over a median (IQR) of 7 (3.4-11) years, and the case-cohort substudy included 1075 incident ischemic stroke and 968 non-stroke participants. One lipid factor, enriched for docosahexaenoic acid (DHA, an omega-3 fatty acid), was inversely associated with stroke risk in a base model (HR = 0.84; 95%CI 0.79-0.90; P = 8.33 × 10-8) and fully adjusted model (HR = 0.88; 95%CI 0.83-0.94; P = 2.79 × 10-4). This factor was associated with a healthy diet pattern (β = 0.21; 95%CI 0.12-0.30; P = 2.06 × 10-6), specifically with fish intake (β = 1.96; 95%CI 0.95-2.96; P = 1.36 × 10-4). DHA was a mediator between fish intake and incident ischemic stroke (30% P = 5.78 × 10-3). Taken together, DHA-containing plasma lipids were inversely associated with incident ischemic stroke and mediated the relationship between fish intake and stroke risk.

Factor IX and incident hypertension in Black and White adults: the REasons for Geographic and Racial Differences in Stroke cohort

BACKGROUND

Hypertension is a cardiovascular disease risk factor disproportionately affecting Black adults. Certain biomarkers of thrombosis and inflammation are associated with a greater risk of hypertension. Factor IX is a marker of thrombosis; Black adults have higher levels than others. Whether factor IX correlates with incident hypertension risk or explains some of the disproportionate burden faced by Black adults, is not known.

METHODS

REasons for Geographic and Racial Differences in Stroke (REGARDS) recruited 30 239 Black and White adults from the contiguous US in 2003-2007 (Visit 1) and had repeat assessment in 2013-2016 (Visit 2). Factor IX was measured in Visit 1 samples in a sex-race stratified sample of 4400 participants that attended both visits. Modified Poisson regression estimated adjusted risk ratios (RR) for incident hypertension at Visit 2 by factor IX tertiles. Inverse odds ratio weighting estimated the proportion of the excess burden of incident hypertension in Black adults due to factor IX levels.

RESULTS

Among 1824 participants (55% female and 24% Black race), 36% developed hypertension. The fully adjusted RR for the third vs. first tertile was 1.21; 95% confidence interval (CI) 1.03-1.43, and there was a significant linear trend across tertiles (P < 0.001). Factor IX did not mediate excess hypertension risk among Black adults in adjusted models.

CONCLUSIONS

In this prospective study of Black and White adults without prevalent hypertension, higher factor IX was associated with a greater risk of incident hypertension. This risk may relate to adverse thromboinflammation among persons in the 2nd and 3rd tertiles of factor IX.

Association between serum adiponectin and risk of incident hypertension: the REGARDS study

INTRODUCTION

Hypertension is a cardiovascular disease risk factor disproportionately affecting Black adults. Adiponectin is a cytokine secreted by adipocytes that improves insulin sensitivity, maintains vascular homeostasis, and is inversely associated with adiposity. We sought to determine the risk of incident hypertension by level of adiponectin.

METHODS

The REasons for Geographic And Racial Differences in Stroke (REGARDS) study recruited 30 239 adults from 2003 to 2007. We included REGARDS participants in the Biomarkers as Mediators of Racial Disparities in Risk Factors (BioMedioR) substudy. We estimated the risk ratio for incident hypertension in unadjusted and adjusted models for demographics factors, dietary patterns, measures of adiposity, and SBP.

RESULTS

Inverse odds ratio weighting estimated the excess hypertension incidence among Black participants that was explained by adiponectin. Of the 1498 BioMedioR participants, 35% developed incident hypertension in follow-up. White adults had higher baseline adiponectin levels than Black adults. For each 1-SD higher log adiponectin, the risk ratio of hypertension was 0.90 [95% confidence interval (95% CI) 0.84-0.96] in an unadjusted model, 0.92 (0.86-1.00) in a demographic adjusted model, and 0.99 (0.91-1.07) in a fully adjusted model. Lower adiponectin mediated 21-46% of the excess risk of incident hypertension among Black relative to White participants in models adjusting for just demographics and dietary patterns.

CONCLUSION

Among Black and White adults, lower adiponectin was associated with a greater risk of incident hypertension in unadjusted and minimally adjusted models. Future studies into how adiponectin changes in obesity could help to further explain its impact on hypertension risk.

Mediating effects of inequitable gender norms on intimate partner violence and contraceptive use in a cluster randomized control trial in Niger

Previous research has demonstrated that the Reaching Married Adolescents intervention (RMA) was associated with changes in inequitable gender norms, intimate partner violence (IPV), and modern contraceptive use. This study seeks to understand if changes in inequitable gender norms mediate the RMA intervention's effects on contraceptive use and intimate partner violence (IPV). A 4-arm cluster randomized control trial was conducted to evaluate effects of the RMA intervention (household visits, small groups, combination, control) on married adolescent girls and their husbands in Dosso, Niger (baseline: 1042 dyads; 24 months follow-up: 737 dyads; 2016-2019). Mediation was assessed using inverse odds ratio weighting. In the small group intervention, of the total effect on IPV prevalence (8% reduction), indirect effects via inequitable gender norms are associated with a 2% decrease (95% CI, -0.07 to 0.12) and direct effects with a 6% decrease (95% CI, -0.20 to -0.02). For household visits, of the total effect on contraceptive use (20% increase), the indirect effect accounts for an 11% decrease (95% CI, -0.18 to -0.01) and direct effect, a 32% increase (95% CI, 0.13-0.44); this is similar to findings for the combination arm. This experimental evidence informs the value of changing underlying social norms to reduce IPV and increase contraception use.

Childhood maltreatment and chronic "all over" body pain in adulthood: a counterfactual analysis using UK Biobank

ABSTRACT

Evidence linking adverse childhood experiences and chronic pain in adulthood is largely cross-sectional, potentially subject to recall bias and does not allow exploration of mediating pathways. We analysed a large population-based cohort (UK Biobank) using a causal framework, to determine if childhood maltreatment is related to chronic "all over" body pain in adulthood. We used doubly robust estimation with inverse probability weights to estimate the difference in risk of chronic pain "all over" between those exposed/not exposed to childhood maltreatment (abuse or neglect). In addition, we looked at interaction with adult stressful life events and examined mediation using inverse odds weighting in a generalized linear model. Using cases with complete data (n = 118,347), the risk of chronic "all over" body pain was higher in the exposed (6.3%, 95% confidence interval [CI] 6.0%-6.5%) than in the unexposed (4.0%; 95% CI 3.8%-4.2%). This difference remained in analyses stratified by sex. Conversely, when analyses were repeated with a negative control exposure, childhood sunburn, risk differences were 0.8% in women (95% CI 0.3%-1.3%) and 0.5% in men (95% CI 0.1%-0.9%). Childhood maltreatment and adult life events had similar effects, and there was a supra-additive risk (1.2%; 95% CI 0.6-1.7) when experiencing both. In mediation analyses, the total effect was a relative risk of 1.57 (95% CI 1.49-1.66), while the estimated indirect effect via all mediators was relative risk 1.16 (95% CI 1.14-1.18). Reducing childhood maltreatment would likely prevent cases of chronic widespread pain in adulthood. Stressful adult events and mediators may offer opportunities for intervention.

Social disadvantage accelerates aging

Social disadvantage, like advanced age, is a risk factor for a broad range of health conditions; however, whether it influences the aging process remains unclear. Here, using a multicohort approach, we investigated the associations of social disadvantage with age-related plasma proteins and age-related diseases. We found proteomic signatures of accelerated immune aging and 14 specific age-related proteins linked to social disadvantage during both early and later life. Individuals experiencing social disadvantage had an increased risk of 66 age-related diseases, with up to 39% of these associations mediated by the 14 age-related proteins (for example, DNAJB9, F2, HSPA1A, BGN). The main enriched pathway involved the upregulation of the pro-inflammatory regulator NF-κB24 and its downstream factor interleukin-8. Our findings support the hypothesis that social disadvantage throughout the life course may accelerate aging, a biological mechanism that could explain why social stratification plays such a fundamental role in determining human health.

Association of Historical Redlining With Gestational Diabetes Mellitus: The Mediating Role of BMI and Area Deprivation Index

OBJECTIVE

We investigated the association between historic redlining and risk of gestational diabetes mellitus (GDM) and whether this relationship is mediated by maternal obesity and area-level deprivation.

RESEARCH DESIGN AND METHODS

This retrospective study included 86,834 singleton pregnancies from Kaiser Permanente Southern California's health records (2008-2018). Redlining was assessed using digitized Home Owners' Loan Corporation (HOLC) maps, with patients' residential addresses geocoded and assigned HOLC grades (A, B, C, or D) based on their geographic location within HOLC-graded zones. For GDM case patients, exposure was assigned based on address at diagnosis date; for noncase patients, it was assigned based on address during the 24th to 28th gestational week. Health records were combined with area deprivation index (ADI) from 2011 to 2015 census data. Mixed-effect logistic regression models assessed associations between redlining and GDM, with mediation by BMI and ADI evaluated using inverse odds ratio weighting. Models were adjusted for maternal age, education, race and ethnicity, neighborhood-level income, and smoking status.

RESULTS

Among the 10,134 (11.67%) GDM case patients, we found increased risk of GDM in B ("still desirable," adjusted odds ratio [aOR] 1.20, 95% CI 0.99-1.44), C-graded ("definitely declining," aOR 1.22, 95% CI 1.02-1.47), and D-graded ("hazardous," i.e., redlined, aOR 1.30, 95% CI 1.08-1.57) neighborhoods compared with the "best"-graded zone. Prepregnancy BMI and ADI mediated 44.2% and 64.5% of the increased GDM risk among mothers in redlined areas.

CONCLUSIONS

Historic redlining is associated with an increased risk of GDM, mediated by maternal obesity and neighborhood deprivation. Future research is needed to explore the complex pathways linking redlining to pregnancy outcomes.

Greenspace proximity in relation to sleep health among a racially and ethnically diverse cohort of US women

Sleep is essential for overall health. Greenspace may contribute to sleep health through, for instance, improving mood, reducing sleep disruptors (e.g., poor air quality), and promoting physical activity. Although greenspace likely differs across populations, few studies have included diverse populations. To investigate greenspace-sleep health associations, overall and by age, race and ethnicity, and socioeconomic status, we used data collected at enrollment (2003-2009) from women in the Sister Study (n = 1612 Hispanic/Latina, n = 4421 non-Hispanic (NH)-Black, and n = 41,657 NH-White). Participants' geocoded home addresses were linked to NASA's Moderate Resolution Imaging Spectroradiometer Normalized Difference Vegetation Index data (250m resolution) to capture greenspace tertiles (further categorized as low/moderate vs. high). Participants reported seven sleep dimensions, which we assessed individually, along with a multidimensional sleep health measure (categories: favorable, moderate, poor). Adjusting for individual- and environmental/neighborhood-level characteristics, we used Poisson regression with robust variance to estimate prevalence ratios and 95 % confidence intervals (PR[CI]). We tested for interaction and estimated age-, race and ethnicity-, and educational attainment category-specific associations. Among participants (mean ± SD age = 55.7 ± 9.0 years), those with low/moderate vs. high greenspace had a lower prevalence of favorable sleep (58 % vs. 66 %). After adjustment, low/moderate vs. high greenspace was associated with a 32 % higher prevalence of moderate (PR = 1.32 [1.27-1.38]) and 12 % higher prevalence of poor (PR = 1.12 [1.07-1.16]) vs. favorable sleep health. Magnitudes of associations were higher among NH-White women vs. minoritized racial-ethnic groups and women with higher vs. lower educational attainment. Higher greenspace was associated with favorable sleep, with stronger associations among groups with more social advantages.

Body Mass Index, Comorbidities, and Ambulatory Care Visits: The REGARDS Study

BACKGROUND

Individuals with obesity have more ambulatory care usage than individuals with normal weight and overweight. There is limited information on whether this is consistent across provider specialties and whether comorbidities explain the associations.

METHODS AND RESULTS

Among REGARDS (Reasons for Geographic and Racial Differences in Stroke) cohort study participants with Medicare fee-for-service coverage (n=9648), we identified ambulatory visits over 5 years. We used marginalized zero-inflated Poisson models to calculate ratios of means by body mass index (BMI), adjusted for demographics and health behaviors, and inverse odds weighting to evaluate mediation by diabetes, hypertension, dyslipidemia, stroke, coronary heart disease, atrial fibrillation, heart failure, chronic kidney disease, depressive symptoms, cancer, arthritis, and sleep apnea. The mean age of participants was 71.7±7.3 years, 35.1% were Black individuals and 64.9% White individuals, and 51.1% were women. Participants had a mean of 37.8 total, 16.0 primary care, 3.4 cardiology, 1.9 orthopedics, 0.9 pulmonology, and 0.4 endocrinology visits. Compared with individuals with BMI 18.5 to <25 kg/m2 (n=2613), participants with BMI ≥35 kg/m2 (n=1259) had 23% (95% CI, 21%-24%) more ambulatory visits. Participants with BMI ≥35 kg/m2 had 26% more primary care, 20% more cardiology, 74% more orthopedics, 62% more pulmonology, and 85% more endocrinology visits. Comorbidities partly explained associations with overall, primary care, and orthopedics visits (39%, 38%, and 15%, respectively) and largely explained associations with cardiology, pulmonology, and endocrinology visits.

CONCLUSIONS

Understanding which specialty visits are associated with higher BMI can help with workforce planning and allocation of resources.

Differential Colorectal Cancer Mortality Across Racial and Ethnic Groups: Impact of Socioeconomic Status, Clinicopathology, and Treatment-Related Factors

INTRODUCTION

Non-Hispanic Black (Black) colorectal cancer (CRC) patients have a higher risk of mortality than most other racial/ethnic groups. Limited studies examine the contribution of socioeconomic (SES), clinicopathologic, or treatment variations to mortality disparities. This retrospective cohort investigation examined the extent to which SES, clinicopathologic, and treatment factors explain racial/ethnic differences in CRC mortality.

METHODS

We studied 146,515 individuals, 18+ years old, with a confirmed diagnosis of CRC within 2010-2017, identified from the Surveillance, Epidemiology, and End Results (SEER) database. We performed Cox regression analyses to examine the association of race and ethnicity, surgery type, and tumor site with all-cause mortality and CRC-specific mortality. We then performed mediation analysis to quantify the extent to which mortality differences were mediated by SES, clinicopathologic, and treatment factors.

RESULTS

Black patients had a significantly higher hazard of all-cause mortality than non-Hispanic White (White) patients. The White versus Black patients' comparison demonstrated that variations in SES and clinicopathologic factors significantly explained 46.63% (indirect effect HR: 0.92, 95% CI 0.91-0.93) and 10.87% (indirect effect HR: 0.98, 95% CI 0.97-0.99) of the excess all-cause mortality among Black patients, respectively. The Hispanic versus Black comparisons identified SES as the most influential mediator, explaining 19.68% of the excess all-cause mortality. The proportions mediating for CRC-specific mortality showed comparable outcomes to all-cause mortality.

CONCLUSION

Black patients had a greater risk for all-cause mortality and CRC-specific mortality attributed to SES and clinicopathologic variations compared to other racial/ethnic groups. Future studies should investigate equity in healthcare through interventions addressing SES-related disparities.

The long-term impact of childhood peer relationships on disability in later life: Causal mediation evidence from older Chinese adults

Despite the recognized importance of social connections in Chinese culture, research on how childhood peer relationship deficits impact health later in life has been limited. This study aimed to investigate the association between childhood peer relationship deficits and the odds of disability among older Chinese adults and to explore the potential mediating roles of social isolation, loneliness, and cognitive function. Using the longitudinal sample of respondents aged 60 years and older in the China Health and Retirement Longitudinal Study 2013-2018 (N = 7133), the link between peer relationship deficits in childhood and disability in late life was assessed using marginal structural models, and the potential mediating effects of social isolation, loneliness, and cognitive function were examined by the inverse odds ratio weighting technique. Participants who experienced greater childhood peer relationship deficits were more prone to disability (odds ratio: 1.19, 95% CI: 1.09, 1.29) than those with more positive childhood peer interactions. The inverse odds ratio weighting analysis indicated that social isolation, loneliness, and cognitive function individually played partial mediating roles in the association between childhood peer relationships and disability by 11.36% (95% CI: 4.04%-18.99%), 11.95 % (95% CI: 4.65%-19.23%), and 24.58% (95% CI: 17.01%-32.43%), respectively. The combined mediation effect of the three mediators was 30.57% (95% CI: 23.52%-39.91%). These findings suggest that interventions to enhance social connections and cognitive health in older adults may help mitigate the long-term impacts of childhood peer relationship deficits on disability among older Chinese adults.

Socioeconomic Status, Smoking, and Lung Cancer: Mediation and Bias Analysis in the SYNERGY Study

BACKGROUND

Increased lung cancer risks for low socioeconomic status (SES) groups are only partially attributable to smoking habits. Little effort has been made to investigate the persistent risks related to low SES by quantification of potential biases.

METHODS

Based on 12 case-control studies, including 18 centers of the international SYNERGY project (16,550 cases, 20,147 controls), we estimated controlled direct effects (CDE) of SES on lung cancer via multiple logistic regression, adjusted for age, study center, and smoking habits and stratified by sex. We conducted mediation analysis by inverse odds ratio weighting to estimate natural direct effects and natural indirect effects via smoking habits. We considered misclassification of smoking status, selection bias, and unmeasured mediator-outcome confounding by genetic risk, both separately and by multiple quantitative bias analyses, using bootstrap to create 95% simulation intervals (SI).

RESULTS

Mediation analysis of lung cancer risks for SES estimated mean proportions of 43% in men and 33% in women attributable to smoking. Bias analyses decreased the direct effects of SES on lung cancer, with selection bias showing the strongest reduction in lung cancer risk in the multiple bias analysis. Lung cancer risks remained increased for lower SES groups, with higher risks in men (fourth vs. first [highest] SES quartile: CDE, 1.50 [SI, 1.32, 1.69]) than women (CDE: 1.20 [SI: 1.01, 1.45]). Natural direct effects were similar to CDE, particularly in men.

CONCLUSIONS

Bias adjustment lowered direct lung cancer risk estimates of lower SES groups. However, risks for low SES remained elevated, likely attributable to occupational hazards or other environmental exposures.

An examination of acrylamide and cadmium as possible mediators of the association between cigarette smoking and chronic musculoskeletal pain

ABSTRACT

Chronic musculoskeletal pain (CMP) causes significant health loss worldwide and is one of the major public health issues of our time. Cigarette smoking is an independent risk factor of CMP. The present study examined the potential mediating role of 2 subproducts of cigarette smoke, acrylamide and cadmium, individually and combined, on the association between cigarette smoking and CMP, using the Inverse Odds Ratio Weighting (IORW) method. Analyses were conducted on data from 3670 adults who participated to National Health and Nutrition Examination Surveys 2003 to 2004. When smoking was measured with serum cotinine levels, there was an association of moderate and heavy smoking {adjusted Odds Ratio [aOR]>30 ng/mL = 1.99 (95% confidence interval [CI]: 1.44-2.74)} with CMP, but no association between light smoking and CMP (aOR1-30 ng/mL = 1.17 [95% CI: 0.75-1.80]) as compared to nonsmoking. Small indirect effects were identified through acrylamide (aOR = 1.24 [95% CI: 0.96-1.61]) and cadmium (aOR = 1.56 [95% CI: 0.92-2.63]) only among moderate and heavy smokers. When both biomarkers were considered together, their indirect effect was larger (aOR = 2.07 [95% CI: 1.32-3.23]). These results suggest that the association between cigarette smoking and CMP is mediated by acrylamide and cadmium and that these substances, also present in food and the environment, may serve as biomarkers of CMP.

Disparities in Thyroid Cancer Mortality Across Racial and Ethnic Groups: Assessing the Impact of Socioeconomic, Clinicopathologic, and Treatment Variations

BACKGROUND

Thyroid cancer remains a significant public health concern, with disparities in mortality rates observed across racial/ethnic groups. We quantified the extent to which socioeconomic, clinicopathologic, and treatment variations explain racial/ethnic disparities in thyroid cancer mortality.

METHODS

We studied a cohort of 109,981 thyroid cancer patients diagnosed from 2006 to 2018 using the United States Surveillance, Epidemiology, and End Results database. We used multivariable logistic regression to assess the association of race/ethnicity with treatment status. We also performed mediation analyses to estimate how much the racial/ethnic differences in thyroid cancer-specific mortality were explained by variations in treatment and clinicopathologic and socioeconomic factors.

RESULTS

Non-hispanic (NH) Black patients were more likely to not receive the recommended surgical resection than NH White patients (adjusted odds ratio [aOR] 1.10, 95% confidence interval [CI] 1.02-1.20). NH Black patients had a significantly higher risk of all-cause mortality compared with NH White patients (adjusted hazards ratio [aHR] 1.19, 95% CI 1.07-1.31). Mediation analysis showed that socioeconomic status significantly explained 48.7% (indirect effect HR 1.07, 95% CI 1.01-1.14) of the difference in thyroid cancer-specific mortality between NH Black and NH White patients.

CONCLUSIONS

This study found that race/ethnicity was associated with treatment status and the risk of mortality among patients diagnosed with thyroid cancer. Moreover, clinicopathologic and socioeconomic factors were identified as the most crucial mediators that explained the excess mortality among minority groups. These findings provide insight into the pathways through which disparities in thyroid cancer mortality in NH Black and Hispanic thyroid patients could operate.

Number of children and dementia risk: a causal mediation analysis using data from the HUNT study linked with national registries in Norway

BACKGROUND

Childlessness, as well as having a high number of children, has been reported to be associated with an elevated risk of dementia compared to having 2-3 children. The mechanisms underlying these relationships are not well understood and may be mediated by different midlife risk factors. We examined the mediating role of various factors on the relationship between the number of children and dementia risk. These factors include socioeconomic factors (e.g., occupational complexity), psychosocial (e.g.., social activities, loneliness, life satisfaction), lifestyle (e.g., smoking, physical inactivity, alcohol intake), and chronic diseases (e.g., obesity, diabetes, depression, hearing impairment and hypertension).

METHODS

Using a historic cohort design, we included 9,745 participants born between 1931-48, with a mean age of 78.2 (SD = 6.4) years at the time of cognitive testing in the HUNT4 70 + sub-study (2017-2019). Further measures were obtained through data linkage between information from Statistics Norway and the HUNT1(1984-86), and HUNT2 (1995-97) Surveys. Causal mediation analyses using an inverse odd weighting approach were conducted to decompose the total effect of the number of children (0, 1, or 4 + children vs. 2-3) on the risk of dementia at age 70 + years into direct and indirect effects with mediators assessed at a mean age of 50.7 (SD = 6.4) years. The analyses were adjusted for age, sex, marital status at age 25 years, educational status, and religion assessed during HUNT3 (2006-2008).

RESULTS

Overall, 15.7% were diagnosed with dementia. The proportions with dementia by the number of children were 22.3% among those with no children, 21.4% for those with one child, 13% for those with 2-3 children (specifically, 12.6% for those with 2 children and 13.4% for those with 3 children), and 19.9% for those with 4 + children. Compared to the reference group of individuals with 2-3 children, the dementia risk was higher among the groups with no children (relative risk (RR): 1.30, 95% confidence interval (CI) (1.12, 1.51)), those with one child (RR: 1.30, 95% CI (1.14, 1.47)) and those with 4 + children (RR: 1.12, 95% CI (1.01, 1.24)). The elevated risks of dementia were not mediated by the socioeconomic, psychosocial, lifestyle, or chronic diseases related factors that we tested. Sex-stratified analysis showed higher dementia risk for men without children and women with one or 4 + children compared to those with 2-3 children, with similar patterns across sexes. None of the mediators contributed to mediation in either group. None of the mediators appeared to contribute through mediation in either group.

CONCLUSIONS

Our findings suggest that the number of children-specifically being childless, having one child, or having four or more children-may influence the risk of dementia. These relationships were not mediated by psychosocial, lifestyle, and socioeconomic factors, or markers of chronic diseases in adulthood considered in this study.

Mediation of the effect of prenatal maternal smoking on time to natural menopause in daughters by birthweight-for-gestational-age z-score and breastfeeding duration: analysis of two UK birth cohorts born in 1958 and 1970

BACKGROUND

Prenatal maternal smoking, lower birthweight, and shorter breastfeeding duration have all been associated with an earlier age at menopause in daughters. We estimated the extent to which birthweight-for-gestational-age z-score and breastfeeding duration mediate the effect of prenatal maternal smoking on time to natural menopause in daughters.

METHODS

Using pooled data from two prospective birth cohort studies - the 1970 British Cohort Study (n = 3,878) followed-up to age 46 years and the 1958 National Child Development Study (n = 4,822) followed-up to age 50 years - we perform mediation analysis with inverse odds weighting implemented in Cox proportional-hazards models.

RESULTS

Prenatal maternal smoking was associated with lower birthweight z-scores [β: -0.29; 95% CI -0.34, -0.24] and reduced breastfeeding duration [RRR< 1month: 0.90; 95% CI 0.79, 1.02; RRR≥ 1 month: 0.66; 95% CI 0.59, 0.73 relative to women who were never breastfed]. Greater z-score for birthweight [HR: 0.96; 95% CI 0.91, 1.01] and longer breastfeeding duration [HR≥ 1 month: 0.84; 95% CI 0.74, 0.96] were associated with lower hazards for earlier age at natural menopause. The total effect of prenatal maternal smoking on the time to natural menopause in daughters was estimated as a HR of 1.13 [95% CI 1.02, 1.24]. Birthweight z-score and breastfeeding duration jointly explained an estimated 14% of the total effect [HRNIE: 1.02; 95% CI 0.99, 1.05].

CONCLUSIONS

The consequences of smoking during pregnancy on the earlier experience of natural menopause in daughters may partly be offset by intrauterine growth and longer breastfeeding duration to the extent that they mediate the risk of earlier menopause. However, since the extent of mediation by birthweight z-score and breastfeeding duration is small, other factors, including the direct effect of maternal smoking in utero, may play a more important role.

Childhood Peer Relationships and Dementia Risk in Chinese Older Adults: A Mediation Analysis

BACKGROUND

Early-life social experiences significantly influence later-life health, yet the association between childhood peer relationships and dementia, as well as the underlying mechanisms, remains underexplored. This study aimed to investigate this association and the mediating roles of social disengagement and loneliness.

METHODS

Leveraging data from 7574 adults aged ≥ 60 in the China Health and Retirement Longitudinal Study (2011-2018), we employed marginal structural models to assess the associations between childhood peer relationships and dementia risk in later life. Inverse odds ratio weighting was used to examine the mediating roles of formal and informal social disengagement and loneliness.

RESULTS

Individuals with deficits in childhood peer relationships had a higher risk of dementia (odds ratio [OR], 1.21; 95% confidence interval [CI], 1.10-1.34) compared with those with more positive experiences. This association was partially mediated by formal social disengagement (proportion mediated, 21.44%; 95% CI, 12.20%-40.94%), loneliness (proportion mediated, 22.00%; 95% CI, 13.42%-33.82%), and their combination with informal social disengagement (proportion mediated, 41.50%; 95% CI, 30.76%-66.07%). Informal social disengagement alone did not show a significant mediating effect.

CONCLUSIONS

In this cohort study of older Chinese adults, negative childhood peer relationship experiences were associated with an elevated risk of dementia in later life. Formal social disengagement and loneliness partially mediated this association. These findings underscore the importance of fostering positive social relationships in early life and suggest potential psychosocial strategies to mitigate dementia risk in older adults due to childhood peer relationship deficits.

The role of multiple birth and birth complications in the association between assisted reproductive technology conception and autism diagnosis

In recent decades, the use of assisted reproductive technology (ART) has increased rapidly. To assess the relationship between ART and autism diagnosis, we linked California birth records from 2000 through 2016 with contemporaneous records from the National ART Surveillance System (NASS) and autism caseload records from California's Department of Developmental Services from 2000 through November 2019. All 95 149 birth records that were successfully linked to a NASS record, indicating an ART birth, were matched 1:1 using propensity scores to non-ART births. We calculated the hazard risk ratio for autism diagnosis and the proportions of the relationship between ART conception and autism diagnosis mediated by multiple birth pregnancy and related birth complications. The hazard risk ratio for autism diagnosis following ART compared with non-ART conception is 1.26 (95% CI, 1.17-1.35). Multiple birth, preterm birth, and cesarean delivery jointly mediate 77.9% of the relationship between ART conception and autism diagnosis. Thus, increased use of single embryo transfer in the United States to reduce multiple births and related birth complications may be a strategy to address the risk of autism diagnosis among ART-conceived children.

Dietary Restriction, Socioeconomic Factors, Access to Kidney Transplantation, and Waitlist Mortality

INTRODUCTION

Dietary restrictions for patients with end-stage kidney disease (ESKD) are burdensome. Kidney transplantation (KT) candidates who lack neighborhood resources and are burdened by dietary restrictions may have decreased access to KT.

METHODS

In our two-center prospective cohort study (2014-2023), 2471 ESKD patients who were evaluated for KT (candidates) reported their perceived burden of dietary restrictions (not at all, somewhat/moderately, or extremely bothered). Neighborhood-level socioeconomic factors were derived from residential ZIP codes. We quantified the association of perceived burden of the dietary restrictions with a chance of listing using Cox models and risk of waitlist mortality using competing risks models. Then we tested whether these associations differed by neighborhood-level socioeconomic factors.

RESULTS

At evaluation, 18% of KT candidates felt extremely bothered by dietary restrictions. Those who felt extremely bothered were less likely to be listed for KT (adjusted hazard ratio [aHR] = 0.75, 95% confidence interval [CI]: 0.64-0.87); this association did not differ by neighborhood-level socioeconomic factors. Overall, the burden of dietary restrictions was not associated with waitlist mortality (p = 0.62). However, among candidates living in high food insecurity neighborhoods, those who felt extremely bothered had higher waitlist mortality (adjusted subhazard ratio [aSHR] = 2.07, 95% CI: 1.14-3.75, p[interaction] = 0.02). The association between dietary burden and waitlist mortality did not differ by neighborhood-level healthy food access.

CONCLUSION

The perceived burden of dietary restrictions is associated with a lower chance of listing for KT, and higher waitlist mortality only among candidates residing in neighborhoods with high food insecurity. Transplant centers should identify vulnerable patients and support them with nutrition education and access to food assistance programs.

The role of influenza Hemagglutination-Inhibition antibody as a vaccine mediator in children

BACKGROUND

Influenza vaccination may protect through the humoral immune response, cellular immune response, or possibly both. Immunity after vaccination can be mediated through antibodies that may be detected by the rise of serum hemagglutination inhibition (HAI) titers. Our objective was to investigate the proportion of protection against influenza mediated through antibodies by measuring the rise of HAI titer (indirect effect) compared to that induced through other immune mechanisms (direct effect) for influenza A and B.

METHODS

We analysed data from a cluster randomized trial conducted during the 2008-2009 season in which Canadian Hutterite children were vaccinated against influenza. We used inverse probability weighting to calculate the indirect and direct effect of vaccination against influenza A/H3N2 and influenza B/Brisbane using HAI titres and overall vaccine efficacy.

RESULTS

We included data on 617 children from 46 Hutterite colonies, aged between 3 and 15 years who were vaccinated with either inactivated trivalent influenza vaccine or hepatitis A vaccine. Vaccine efficacy was 63 % for influenza A (H3N2) and 28 % for influenza B. The hazard ratio for protection against influenza A/H3N2 due to an indirect effect of vaccination was 0.96 (95 % confidence interval (CI) of 0.00 to 2.89) while for the direct effect it was 0.38 (95 % CI of 0.00 to 5.47). The hazard ratio for influenza B indirect effect was 0.75 (95 % CI of 0.07 to 1) and for the direct effect 0.96 (95 % CI of 0.00 to 12.02). In contrast, repeating the analysis using microneutralization in a subgroup of 488 children revealed that the protective effect for vaccination for A/H3N2 was entirely mediated by antibodies but only for 13 % for influenza B.

CONCLUSIONS

Although vaccination provided higher protective effectiveness against influenza A than B, most of the influenza A vaccine efficacy likely occurred through antibodies other than what could be detected by HAI titres. In contrast, for influenza B, while the HAI titres appeared to mediate most of the vaccine effectiveness, this was not confirmed by microneutralization analysis.

Race, Social Determinants of Health, and Comorbidity Patterns Among Participants with Heart Failure in the REasons for Geographic and Racial Differences in Stroke (REGARDS) Study

Background

Among individuals with heart failure (HF), racial differences in comorbidities may be mediated by social determinants of health (SDOH).

Methods

Black and White US community-dwelling participants in the REasons for Geographic and Racial Differences in Stroke (REGARDS) study aged ≥ 45 years with an adjudicated HF hospitalization between 2003 and 2017 were included in this cross-sectional analysis. We assessed whether higher prevalence of comorbidities in Black participants compared to White participants were mediated by SDOH in socioeconomic, environment/housing, social support, and healthcare access domains, using the inverse odds weighting method.

Results

Black (n = 240) compared to White (n = 293) participants with HF with preserved ejection fraction (HFpEF) had higher prevalence of diabetes [1.38 (95% CI: 1.18 - 1.61)], chronic kidney disease [1.21 (95% CI: 1.01 - 1.45)], and anemia [1.33 (95% CI: 1.02 - 1.75)] and lower prevalence of atrial fibrillation [0.80 (95% CI: (0.65 - 0.98)]. Black (n = 314) compared to White (n = 367) participants with HF with reduced ejection fraction (HFrEF) had higher prevalence of hypertension [1.04 (95% CI: 1.02 - 1.07)] and diabetes [1.26 (95% CI: 1.09 - 1.45)] and lower prevalence of coronary artery disease [0.86 (95% CI: 0.78 - 0.94)] and atrial fibrillation [0.70 (95% CI: 0.58 - 0.83)]. Socioeconomic status explained 14.5%, 26.5% and 40% of excess diabetes, anemia, and chronic kidney disease among Black adults with HFpEF; however; mediation was not statistically significant and no other SDOH substantially mediated differences in comorbidity prevalence.

Conclusions

Socioeconomic status partially mediated excess diabetes, anemia, and chronic kidney disease experienced by Black adults with HFpEF, but differences in other comorbidities were not explained by other SDOH examined.

On the ability of the SF-6D to capture the consequences of chronic illnesses on subjective well-being: Evidence from France

Using cross-sectional data from a representative sample of the French population (the 2008 Disability Health survey), this paper examines whether the SF-6D, a widely used preference-based measure of health-related quality of life in economic evaluations, fully captures the variation in subjective well-being (SWB) due to chronic illnesses. We conduct a mediation analysis to disentangle the direct and indirect, through the SF-6D, effects of various chronic conditions on SWB (happiness). Our results show that the SF-6D reflects changes in happiness due to most illnesses except mental illness. Changes in SWB mediated by the SF-6D account for 74% of the total effect. The variation unexplained by the SF-6D is significant and increases substantially in the presence of multimorbidity when a chronic illness is combined with anxiety or depression. Overall, our results suggest that the SF-6D incompletely captures the subjective experience of chronically ill patients, especially those with comorbid conditions.

High adherence to the French dietary guidelines decreases type 2 diabetes risk in females through pathways of obesity markers: Evidence from the E3N-EPIC prospective cohort study

OBJECTIVE

Obesity and type 2 diabetes (T2D) have been associated with low adherence to the 2017 French food-based dietary guidelines, as assessed by the Programme National Nutrition Santé - guidelines score 2 (PNNS-GS2). Whether the association between T2D and PNNS-GS2 is direct or mediated by obesity has been little investigated.

RESEARCH METHODS

The study included 71,450 women from the E3N-EPIC cohort, mean age of 52.9 y (SD 6.7). The simplified PNNS-GS2 was derived via food history questionnaire. Multivariable Cox regression models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) of T2D. Causal mediation analyses were used to decompose the total effect of sPNNS-GS2 on T2D into a direct effect and indirect effect mediated by body mass index (BMI) or the waist-hip ratio (WHR).

RESULTS

During a mean follow-up of 19 y, 3679 incident T2D cases were identified and validated. There was a linear association between adherence to sPNNS-GS2 and T2D (P-nonlinearity = 0.92). In the fully adjusted model, each 1-SD increase in the sPNNS-GS2 was associated with a lower T2D risk [HR (95% CI), 0.92 (0.89, 0.95)]. The overall associations were mainly explained by sPNNS-GS2-associated excess weight, with BMI and WHR mediating 52% and 58% of the associations, respectively.

CONCLUSIONS

Higher adherence to French food-based dietary guidelines was associated with a lower risk of T2D in women, and a significant portion of this effect could be attributed to excess weight measured by BMI or WHR. This finding helps better understand the mechanisms underlying the diet-T2D association.

Racial and ethnic disparities in mortality among World Trade Center Health Registry enrollees with post-9/11 cancer

INTRODUCTION

There are well-documented racial and ethnic disparities in mortality after cancer in the general population, but less is known about whether disparities also exist in disaster-exposed populations.

METHODS

We conducted a longitudinal cohort study of 4341 enrollees in the World Trade Center Health Registry (WTCHR) with a first-ever primary invasive cancer diagnosis after 9/11/2001 and followed through 2020. We examined associations of race and ethnicity with all-cause mortality risk and cause-specific mortality risk using multivariable Cox proportional hazards regression models and Fine and Gray's proportional sub-distribution hazards models, respectively. Models were adjusted for baseline characteristics and tumor characteristics. We also examined models further adjusted for socioeconomic status (SES), and we used inverse odds weighting to formally test for mediation by SES.

RESULTS

Compared to non-Hispanic White enrollees with cancer, non-Hispanic Blacks had higher risks for all-cause mortality (adjusted hazard ratio (aHR) = 1.20, 95% CI = 1.02-1.41) and non-cancer mortality (aHR = 1.48, 95% CI = 1.09-2.01) in the full model. In the model without SES, Hispanic enrollees with cancer had higher risks for all-cause mortality (aHR = 1.32, 95% CI = 1.09-1.60) and cancer mortality (aHR = 1.31, 95% CI = 1.05-1.64) compared to non-Hispanic Whites; these associations became not statistically significant in the full model. In the inverse odds weighting analysis, SES explained 24% and 29% of the disparity in all-cause mortality risk observed in non-Hispanic Blacks and Hispanics, respectively, compared to non-Hispanic Whites.

CONCLUSION

This study found that there are racial and ethnic disparities in mortality after cancer in the WTCHR. Additional studies are needed to further explore the factors mediating these disparities.

Mothers' parental divorce experience in childhood and their children's mental health: Mediating role of maternal education

BACKGROUND

Adverse childhood experiences (ACEs) of mothers have been shown to be associated with children's mental health and behavior problems. It is important to identify effective intervention points to prevent negative consequences among children. This study aimed to investigate whether mothers' education is protective against children's depressive symptoms in families with mothers who experienced parental divorce in childhood.

METHODS

Data from a cross-sectional study of 5th and 8th grade children and their caregivers in a prefecture in Japan were used. Final analytic sample consisted of 9666 child-caregiver pairs. Mediation analyses using inverse odds weighting were performed where the exposure was maternal experience of parental divorce in childhood, outcome was child depressive symptoms, and potential mediators were maternal education beyond high school, maternal mental health, and household income.

RESULTS

Maternal experience of parental divorce was associated with an elevated risk of child depressive symptoms (risk ratio: 1.22, 95 % confidence interval (CI): 1.07-1.39). Mediation analyses indicated that the combination of maternal education, maternal depression and household income mediated about half of the total effect. In the model where maternal education was the sole mediator, maternal education mediated nearly half of the total effect (risk ratio: 1.10, 95 % CI: 1.01-1.20).

LIMITATIONS

Results should be cautiously interpreted given observational and cross-sectional nature of the data.

CONCLUSIONS

Maternal education beyond high school may be protective against negative mental health consequences among children due to mothers' experience of parental divorce. Further studies are needed regarding potential mechanisms and the roles of other potential mediators.

Hemagglutination Inhibition Antibody Titers as Mediators of Influenza Vaccine Efficacy Against Symptomatic Influenza A(H1N1), A(H3N2), and B/Victoria Virus Infections

BACKGROUND

The hemagglutination inhibition antibody (HAI) titer contributes only a part of vaccine-induced protection against influenza virus infections. Using causal mediation analysis, we quantified the proportion of vaccine efficacy mediated by postvaccination HAI titers.

METHODS

We conducted causal mediation analyses using data from a randomized, active-comparator controlled, phase III, trial of an inactivated, split-virion seasonal quadrivalent influenza vaccine in children conducted from October 2010 to December 2011 in 8 countries. Vaccine efficacy was estimated using a weighted Cox proportional hazards model. Estimates were decomposed into the direct and indirect effects mediated by postvaccination HAI titers.

RESULTS

The proportions of vaccine efficacy mediated by postvaccination HAI titers were estimated to be 22% (95% confidence interval, 18%--47%) for influenza A(H1N1), 20% (16%-39%) for influenza A(H3N2), and 37% (26%-85%) for influenza B/Victoria.

CONCLUSIONS

HAI titers partially mediate influenza vaccine efficacy against influenza A(H1N1), A(H3N2), and B/Victoria. Our estimates were lower than in previous studies, possibly reflecting expected heterogeneity in antigenic similarity between vaccine and circulating viruses across seasons.

Increased risk of preterm birth due to heat exposure during pregnancy: Exploring the mechanism of fetal physiology

BACKGROUND

Heat exposure during pregnancy can increase the risk of preterm birth (PTB) through a range of potential mechanisms including pregnancy complications, hormone secretion and infections. However, current research mainly focuses on the effect of heat exposure on pathophysiological pathways of pregnant women, but ignore that maternal heat exposure can also cause physiological changes to the fetus, which will affect the risk of PTB.

OBJECTIVE

In this study, we aimed to explore the mediating role of fetal heart rate (FHR) in the relationship between maternal heat exposure and PTB incidence.

METHODS

We assigned heat exposure to a multi-center birth cohort in China during 2015-2018, which included all 162,407 singleton live births with several times FHR measurements during the second and third trimesters. We examined the associations between heat exposure, FHR and PTB in the entire pregnancy, each trimester and the last gestational month. The inverse odds ratio-weighted approach applied to the Cox regression was used to identify the mediation effect of heat exposure on PTB and its clinical subtypes via FHR.

FINDINGS

Exposure to heat significantly increased the risk of PTB during the third trimester and the entire pregnancy, hazard ratios and 95 % CIs were 1.266 (1.161, 1.379) and 1.328 (1.218, 1.447). Heat exposure during the third trimester and entire pregnancy increased FHR in the third trimester by 0.24 bpm and 0.14 bpm. The proportion of heat exposure mediated by FHR elevation on PTB and its subtype ranged from 3.68 % to 24.06 %, with the significant mediation effect found for both medically indicated PTB and spontaneous PTB.

CONCLUSIONS

This study suggests that heat exposure during pregnancy has an important impact on fetal health, and FHR, as a surrogate marker of fetal physiology, may mediate the increased risk of PTB caused by extreme heat. Monitoring and managing physiological changes in the fetus would constitute a promising avenue to reduce adverse birth outcomes associated with maternal heat exposure.

Do health literacy, physical health and past rehabilitation utilization explain educational differences in the subjective need for medical rehabilitation? Results of the lidA cohort study

BACKGROUND

Medical rehabilitation can be helpful for maintaining workers' health and work ability. Its contribution to longer working lives is of high economic relevance in aging populations. In Germany, individuals must apply for rehabilitative measures themselves. Therefore, the subjective need for rehabilitation (SNR) is a prerequisite for rehabilitation access. A low education level is associated with poor health, lower health literacy and more frequent utilization of health services. In the present study, we investigated whether lower educational levels are also associated with a greater SNR and whether health literacy, past rehabilitation utilization and physical health play a mediating role in this path in older employees.

METHODS

3,130 socially insured older employees (born in 1959 or 1965) who participated in the German prospective lidA (leben in der Arbeit) cohort-study in 2011, 2014 and 2018 were included. A causal mediation analysis with an inverse odds weighting approach was performed with the SNR as the dependent variable; educational level as the independent variable; and health, health literacy and past rehabilitation utilization as the mediating variables. Sociodemographic variables were adjusted for.

RESULTS

The SNR was significantly greater in subjects with a low education level, poor physical health, inadequate health literacy and those who had utilized rehabilitation in the past. For health literacy, past rehabilitation utilization and physical health, a significant partial mediating effect on the SNR was found for employees with low compared to those with high education levels. However, the combined mediating effect of all the mediators was lower than the sum of their individual effects. Among those with medium or high education levels, none of the variables constituted a significant mediator.

CONCLUSIONS

The path between a low education level and a high SNR is mediated by inadequate health literacy, past rehabilitation utilization and poor physical health; these factors do not act independently of each other. Promoting health education may lower the SNR by improving physical health and health literacy. While improving physical health is beneficial for individuals, improved health literacy can be economically advantageous for the health system by reducing inappropriate expectations of rehabilitation benefits and subsequent applications for rehabilitation.

Differences in metabolomic profiles between Black and White women in the U.S.: Analyses from two prospective cohorts

There is growing interest in incorporating metabolomics into public health practice. However, Black women are under-represented in many metabolomics studies. If metabolomic profiles differ between Black and White women, this under-representation may exacerbate existing Black-White health disparities. We therefore aimed to estimate metabolomic differences between Black and White women in the U.S. We leveraged data from two prospective cohorts: the Nurses' Health Study (NHS; n = 2077) and Women's Health Initiative (WHI; n = 2128). The WHI served as the replication cohort. Plasma metabolites (n = 334) were measured via liquid chromatography-tandem mass spectrometry. Observed metabolomic differences were estimated using linear regression and metabolite set enrichment analyses. Residual metabolomic differences in a hypothetical population in which the distributions of 14 risk factors were equalized across racial groups were estimated using inverse odds ratio weighting. In the NHS, Black-White differences were observed for most metabolites (75 metabolites with observed differences ≥ |0.50| standard deviations). Black women had lower average levels than White women for most metabolites (e.g., for N6, N6-dimethlylysine, mean Black-White difference = - 0.98 standard deviations; 95% CI: - 1.11, - 0.84). In metabolite set enrichment analyses, Black women had lower levels of triglycerides, phosphatidylcholines, lysophosphatidylethanolamines, phosphatidylethanolamines, and organoheterocyclic compounds, but higher levels of phosphatidylethanolamine plasmalogens, phosphatidylcholine plasmalogens, cholesteryl esters, and carnitines. In a hypothetical population in which distributions of 14 risk factors were equalized, Black-White metabolomic differences persisted. Most results replicated in the WHI (88% of 272 metabolites available for replication). Substantial differences in metabolomic profiles exist between Black and White women. Future studies should prioritize racial representation.

Social Determinants of Health and Incident Apparent Treatment-Resistant Hypertension Among White and Black US Adults: The REGARDS Study

BACKGROUND

We examined the association of multilevel social determinants of health with incident apparent treatment-resistant hypertension (aTRH).

METHODS AND RESULTS

We analyzed data from 2774 White and 2257 Black US adults from the REGARDS (Reasons for Geographic and Racial Differences in Stroke) study taking antihypertensive medication without aTRH at baseline to estimate the association of social determinants of health with incident aTRH. Selection of social determinants of health was guided by the Healthy People 2030 domains of education, economic stability, social context, neighborhood environment, and health care access. Blood pressure (BP) was measured during study visits, and antihypertensive medication classes were identified through a pill bottle review. Incident aTRH was defined as (1) systolic BP ≥140 mm Hg or diastolic BP ≥90 mm Hg, or systolic BP ≥130 mm Hg or diastolic BP ≥80 mm Hg for those with diabetes or chronic kidney disease while taking ≥3 classes of antihypertensive medication or (2) taking ≥4 classes of antihypertensive medication regardless of BP level, at the follow-up visit. Over a median 9.5 years of follow-up, 15.9% of White and 24.0% of Black adults developed aTRH. A percent of the excess aTRH risk among Black versus White adults was mediated by low education (14.2%), low income (16.0%), not seeing a friend or relative in the past month (8.1%), not having someone to care for them if ill or disabled (7.6%), lack of health insurance (10.6%), living in a disadvantaged neighborhood (18.0%), and living in states with poor public health infrastructure (6.0%).

CONCLUSIONS

Part of the association between race and incident aTRH risk was mediated by social determinants of health.

The role of cardiovascular health and vascular events in the relationship between excessive daytime sleepiness and dementia risk

Many studies suggest a relationship between excessive daytime sleepiness (EDS) and dementia incidence, but the underlying mechanisms remain uncertain. The study aimed to investigate the role of cardiovascular burden in the relationship between EDS and dementia incidence over a 12-year follow-up in community-dwelling older adults. We performed analyses on 6171 subjects (aged ≥65 years) free of dementia and vascular disease at baseline. Participants self-reported EDS at baseline and an expert committee validated both prevalent and incident dementia. We defined cardiovascular burden by a low Cardiovascular Health score, constructed using the American Heart Association metrics, and incident vascular events. To explore the potential role of the cardiovascular burden in the relationship between EDS and dementia, we conducted mediation analyses with inverse odds ratio-weighted estimation, using multivariable-adjusted proportional hazard Cox and logistic regression models. Subjects with EDS had a higher risk of all-cause dementia (hazard ratio [HR] 1.39, 95% confidence interval [CI] 1.13-1.69) and dementia with vascular component (DVC) (HR 2.14, 95% CI 1.30-3.51), but not Alzheimer's disease (HR 1.18, 95% CI 0.93-1.51). Cardiovascular burden explained 5% (95% CI 4.1-5.2) and 11% (95% CI 9.7-11.3) of the relationship between EDS and all-cause dementia and DVC, respectively. These findings confirm that EDS may be implicated in the development of dementia and indicate a weaker than expected role of cardiovascular burden in the relationship between EDS and DVC.

Plasma Metabolites and Life's Simple 7 in REGARDS

BACKGROUND

The American Heart Association's Life's Simple 7 (LS7) is a health metric that captures important factors associated with cardiovascular and cerebrovascular health. Previous studies highlight the potential of plasma metabolites to serve as a marker for lifestyle and health behavior that could be a target for stroke prevention. The objectives of this study were to identify metabolites that were associated with LS7 and incident ischemic stroke and mediate the relationship between the two.

METHODS

Targeted metabolomic profiling of 162 metabolites by liquid chromatography-tandem mass spectrometry was used to identify candidate metabolites in a stroke case-cohort nested within the REGARDS study (Reasons for Geographic and Racial Differences in Stroke). Weighted linear regression and weighted Cox proportional hazard models were used to identify metabolites that were associated with LS7 and incident ischemic stroke, respectively. Effect measures were based on a 1-SD change in metabolite level. Metabolite mediators were examined using inverse odds ratio weighting mediation analysis.

RESULTS

The study comprised 1075 ischemic stroke cases and 968 participants in the random cohort sample. Three out of 162 metabolites were associated with the overall LS7 score including guanosine (β, -0.46 [95% CI, -0.65 to -0.27]; P=2.87×10-6), cotinine (β, -0.49 [95% CI, -0.70 to -0.28]; P=7.74×10-6), and acetylneuraminic acid (β, -0.59 [95% CI, -0.77 to -0.42]; P=4.29×10-11). Guanosine (hazard ratio, 1.47 [95% CI, 1.31-1.65]; P=6.97×10-11), cotinine (hazard ratio, 1.30 [95% CI, 1.16-1.44]; P=2.09×10-6), and acetylneuraminic acid (hazard ratio, 1.29 [95% CI, 1.15-1.45]; P=9.24×10-6) were associated with incident ischemic stroke. The mediation analysis identified guanosine (27% mediation, indirect effect; P=0.002), cotinine (30% mediation, indirect effect; P=0.004), and acetylneurminic acid (22% mediation, indirect effect; P=0.041) partially mediated the relationship between LS7 and ischemic stroke.

CONCLUSIONS

We identified guanosine, cotinine, and acetylneuraminic acid that were associated with LS7, incident ischemic stroke, and mediated the relationship between LS7 and ischemic stroke.

Understanding the relationship between type-2 diabetes, MRI markers of neurodegeneration and small vessel disease, and dementia risk: a mediation analysis

To explore to which extent neurodegeneration and cerebral small vessel disease (SVD) could mediate the association between type-2 diabetes and higher dementia risk. The analytical sample consisted in 2228 participants, out of the Three-City study, aged 65 and older, free of dementia at baseline who underwent brain MRI. Diabetes was defined by medication intake or fasting or non-fasting elevated glucose levels. Dementia status was assessed every 2 to 3 years, during up to 12 years of follow-up. Brain parenchymal fraction (BPF) and white matter hyperintensities volume (WMHV) were selected as markers of neurodegeneration and cerebral SVD respectively. We performed a mediation analysis of the effect of baseline BPF and WMHV (mediators) on the association between diabetes and dementia risk using linear and Cox models adjusted for age, sex, education level, hypertension, hypercholesterolemia, BMI, smoking and alcohol drinking status, APOE-ε4 status, and study site. At baseline, 8.8% of the participants had diabetes. Diabetes (yes vs. no) was associated with higher WMHV (βdiab = 0.193, 95% CI 0.040; 0.346) and lower BPF (βdiab = -0.342, 95% CI -0.474; -0.210), as well as with an increased risk of dementia over 12 years of follow-up (HRdiab = 1.65, 95% CI 1.04; 2.60). The association between diabetes status and dementia risk was statistically mediated by higher WMHV (HRdiab=1.05, 95% CI 1.01; 1.11, mediated part = 10.8%) and lower BPF (HRdiab = 1.12, 95% CI 1.05; 1.20, mediated part = 22.9%). This study showed that both neurodegeneration and cerebral SVD statistically explained almost 30% of the association between diabetes and dementia.

Mediation of the association between depression and coronary heart disease by metabolic syndrome components

BACKGROUND

Depression is associated with incident coronary heart disease (CHD) via a pathway that may be causal, but the mechanisms underlying this association are unclear. We assessed the extent to which metabolic syndrome (MetS) and its components (i.e., elevated waist circumference, low high-density lipoprotein [HDL] cholesterol, elevated triglycerides, elevated blood pressure, and elevated fasting plasma glucose) may mediate this association.

METHODS

Data were Framingham Heart Study Research Materials obtained from the National Heart, Lung, and Blood Institute (NHLBI) Biologic Specimen and Data Repository Information Coordinating Center. We used Cox proportional hazards regression to estimate adjusted hazard ratios (aHR) representing the total effect (aHRTE) of probable depression, measured via the Centers for Epidemiological Studies - Depression scale, on incident CHD over approximately 18 years. Using inverse odds ratio weighting, we decomposed this estimate into natural direct effects (aHRNDE) and natural indirect effects (aHRNIE) through potential mediators (measured approximately three years after depression).

RESULTS

Probable depression was associated with incident CHD (aHRTE=1.45, 95% confidence interval [CI]: 0.93, 2.25), and elevated waist circumference partially mediated this association (aHRNDE=1.34, 95% CI: 0.76-2.32; aHRNIE=1.08, 95% CI: 0.63-1.91). We did not find evidence of additional mediation by additional MetS components.

CONCLUSIONS

Elevated waist circumference appears to play a role in the association between depression and CHD.

Quantifying the association of individual-level characteristics with disparities in kidney transplant waitlist addition among people with HIV

BACKGROUND

Over 45% of people with HIV (PWH) in the United States at least 50 years old and are at heightened risk of aging-related comorbidities including end-stage kidney disease (ESKD), for which kidney transplant is the optimal treatment. Among ESKD patients, PWH have lower likelihood of waitlisting, a requisite step in the transplant process, than individuals without HIV. It is unknown what proportion of the inequity by HIV status can be explained by demographics, medical characteristics, substance use history, and geography.

METHODS

The United States Renal Data System, a national database of all individuals ESKD, was used to create a cohort of people with and without HIV through Medicare claims linkage (2007-2017). The primary outcome was waitlisting. Inverse odds ratio weighting was conducted to assess what proportion of the disparity by HIV status could be explained by individual characteristics.

RESULTS

Six thousand two hundred and fifty PWH were significantly younger at ESKD diagnosis and more commonly Black with fewer comorbidities. PWH were more frequently characterized as using tobacco, alcohol and drugs. Positive HIV-status was associated with 57% lower likelihood of waitlisting [adjusted hazard ratio (aHR): 0.43, 95% confidence interval (CI): 0.46-0.48, P  < 0.001]. Controlling for demographics, medical characteristics, substance use and geography explained 39.8% of this observed disparity (aHR: 0.69, 95% CI: 0.59-0.79, P  < 0.001).

CONCLUSION

PWH were significantly less likely to be waitlisted, and 60.2% of that disparity remained unexplained. HIV characteristics such as CD4 + counts, viral loads, antiretroviral therapy adherence, as well as patient preferences and provider decision-making warrant further study.

The estimated mediating roles of anemia-related variables in the association between kidney function and mortality: a National Health and Nutrition Examination Survey (NHANES) study

Anemia is a common complication of chronic kidney disease (CKD), impacting long-term outcomes such as mortality and morbidity. Analyzing NHANES data from 1999 through 2016 for adults aged ≥ 20 years, we assessed the mediating effects of anemia biomarkers (hemoglobin, hematocrit, red cell distribution width [RDW], and mean corpuscular hemoglobin concentration [MCHC]) on CKD-related outcomes by using hazard ratios from a biomarker-adjusted model. Of 44,099 participants, 7463 experienced all-cause death. Cox proportional hazard models revealed a higher all-cause mortality risk in the > 45 years and CKD groups than in the early CKD group. Hemoglobin, hematocrit and MCHC were inversely related to all-cause mortality; RDW was related to mortality. Single mediation analysis showed greater mediating effects of anemia indicators on CKD and mortality in the elderly (> 65 years) population than those in the general population. In the multimediation analysis, the combined mediating effect of anemia was higher in the CKD population than in the general population. This study showed a proportional increase in the mediating effect of anemia with CKD stage, suggesting potential therapeutic avenues. However, further exploration of other mediating factors on kidney outcomes is necessary.

Childhood physical activity and pubertal timing: findings from the LEGACY girls study

BACKGROUND

There is limited research on whether physical activity (PA) in early childhood is associated with the timing of pubertal events in girls.

METHODS

We used data collected over 2011-16 from the LEGACY Girls Study (n = 984; primarily aged 6-13 years at study enrolment), a multicentre North American cohort enriched for girls with a breast cancer family history (BCFH), to evaluate if PA is associated with age at thelarche, pubarche and menarche. Maternal-reported questionnaire data measured puberty outcomes, PA in early childhood (ages 3-5 years) and total metabolic equivalents of organized PA in middle childhood (ages 7-9 years). We used interval-censored Weibull parametric survival regression models with age as the time scale and adjusted for sociodemographic factors, and we tested for effect modification by BCFH. We used inverse odds weighting to test for mediation by body mass index-for-age z-score (BMIZ) measured at study enrolment.

RESULTS

Being highly active vs inactive in early childhood was associated with later thelarche in girls with a BCFH [adjusted hazard ratio (aHR) = 0.39, 95% CI = 0.26-0.59), but not in girls without a BCFH. In all girls, irrespective of BCFH, being in the highest vs lowest quartile of organized PA in middle childhood was associated with later menarche (aHR = 0.70, 95% CI = 0.50-0.97). These associations remained after accounting for potential mediation by BMIZ.

CONCLUSION

This study provides new data that PA in early childhood may be associated with later thelarche in girls with a BCFH, also further supporting an overall association between PA in middle childhood and later menarche.

Investigation of the role of perceived access to primary care in mediating and moderating racial and ethnic disparities in chronic disease control in the veterans health administration

OBJECTIVE

To examine the role of patient-perceived access to primary care in mediating and moderating racial and ethnic disparities in hypertension control and diabetes control among Veterans Health Administration (VA) users.

DATA SOURCE AND STUDY SETTING

We performed a secondary analysis of national VA user administrative data for fiscal years 2016-2019.

STUDY DESIGN

Our primary exposure was race or ethnicity and primary outcomes were binary indicators of hypertension control (<140/90 mmHg) and diabetes control (HgbA1c < 9%) among patients with known disease. We used the inverse odds-weighting method to test for mediation and logistic regression with race and ethnicity-by-perceived access interaction product terms to test moderation. All models were adjusted for age, sex, socioeconomic status, rurality, education, self-rated physical and mental health, and comorbidities.

DATA COLLECTION/EXTRACTION METHODS

We included VA users with hypertension and diabetes control data from the External Peer Review Program who had contemporaneously completed the Survey of Healthcare Experience of Patients-Patient-Centered Medical Home. Hypertension (34,233 patients) and diabetes (23,039 patients) samples were analyzed separately.

PRINCIPAL FINDINGS

After adjustment, Black patients had significantly lower rates of hypertension control than White patients (75.5% vs. 78.8%, p < 0.01); both Black (81.8%) and Hispanic (80.4%) patients had significantly lower rates of diabetes control than White patients (85.9%, p < 0.01 for both differences). Perceived access was lower among Black, Multi-Race and Native Hawaiian and Other Pacific Islanders compared to White patients in both samples. There was no evidence that perceived access mediated or moderated associations between Black race, Hispanic ethnicity, and hypertension or diabetes control.

CONCLUSIONS

We observed disparities in hypertension and diabetes control among minoritized patients. There was no evidence that patients' perception of access to primary care mediated or moderated these disparities. Reducing racial and ethnic disparities within VA in hypertension and diabetes control may require interventions beyond those focused on improving patient access.

Paternal involvement in infant care and developmental milestone outcomes at age 3 years: the Japan Environment and Children's Study (JECS)

BACKGROUND

In developed countries, the time fathers spend on childcare has increased steadily in recent decades. However, studies on the relationship between paternal care and child outcomes remain scarce. Thus, we examined the association between paternal involvement in childcare and children's developmental outcomes.

METHODS

We used Japan's largest birth cohort data, the Japan Environment and Children's Study, to examine the relationship between paternal involvement in childcare at the child's age of 6 months and developmental milestone outcomes at the child's age of 3 years (n = 28,050). Developmental delays were assessed with Ages and Stages Questionnaire. Potential mediation by maternal parenting stress at the child's age of 1.5 years was also examined. We used log-binomial regression analyses to estimate risk ratios.

RESULTS

Fathers' high involvement in childcare was associated with a lower risk of developmental delay in gross-motor, fine-motor, problem solving, and personal-social domains compared with low involvement, adjusting for potential confounders. For example, the risk ratio with 95% confidence intervals was 0.76 [0.67, 0.86] for the gross-motor domain. We also observed that the associations were partially mediated by maternal parenting stress.

CONCLUSIONS

Fathers' active involvement in childcare during infancy may promote young children's development, partially by reducing maternal parenting stress.

IMPACT

Using Japan's largest birth cohort data (Japan Environment and Children's Study), we showed that paternal involvement in infant care might benefit young children's development. Fathers' active involvement in infant care was associated with a lower risk of developmental delays in gross-motor, fine-motor, problem solving, and personal-social domains. Maternal parenting stress may mediate the association between paternal involvement in infant care and child development outcomes at 3 years.

Inpatient multimodal rehabilitation and the role of pain intensity and mental distress on return-to-work: causal mediation analyses of a randomized controlled trial

OBJECTIVE

Studies suggest that symptom reduction is not necessary for improved return-to-work after occupational rehabilitation programmes. This secondary analysis of a randomized controlled trial examined whether pain intensity and mental distress mediate the effect of an inpatient programme on sustainable return-to-work.

METHODS

The randomized controlled trial compared inpatient multimodal occupational rehabilitation (n = 82) with outpatient acceptance and commitment therapy (n = 79) in patients sick-listed due to musculoskeletal and mental health complaints. Pain and mental distress were measured at the end of each programme, and patients were followed up on sick-leave for 12 months. Cox regression with an inverse odds weighted approach was used to assess causal mediation.

RESULTS

The total effect on return-to-work was in favour of the inpatient programme compared with the control (hazard ratio (HR) 1.96; 95% confidence interval (95% CI) 1.15-3.35). There was no evidence of mediation by pain intensity (indirect effect HR, 0.98; 95% CI, 0.61-1.57, direct effect HR, 2.00; 95% CI, 1.02-3.90), but mental distress had a weak suppression effect (indirect effect HR, 0.89; 95% CI, 0.59-1.36, direct effect HR, 2.19; 95% CI, 1.13-4.26).

CONCLUSION

These data suggest that symptom reduction is not necessary for sustainable return-to-work after an inpatient multimodal occupational rehabilitation intervention.

Exploring the effect of early menarche on sexual violence among adolescent girls and young women in southeastern Ghana: a longitudinal mediation analysis

Research from several high-income countries links early menarche with an increased risk for sexual violence. However, the role of early menarche in adolescent girls' and young women's sexual violence risk in sub-Saharan Africa, where sexual violence rates are high, is not well understood. The current study explores the association between early menarche and sexual violence in Ghana with secondary analysis of data collected from 700 adolescent girls and young women followed over three years. Logistic regressions were used to assess the cross-sectional association between early menarche and sexual violence. Generalised estimating equations were used to assess whether the association between early menarche and sexual violence persisted over time. Inverse odds weighting was used to test potential mediators of the association between early menarche and sexual violence. Sexual violence was fairly common in the study sample, with 27% reporting having experienced sexual violence at baseline, and approximately 50% at year three. Early menarche was associated with 72% greater odds of having experienced sexual violence at baseline (95% confidence interval: 1.01-2.93). However, the odds ratio attenuated and lost significance over the three-year study period, with a lower risk of sexual violence among girls with early menarche at year three. Neither child marriage nor early sexual initiation significantly mediated the association between early menarche and sexual violence. The findings suggest that early-maturing girls may be particularly vulnerable to sexual violence in early adolescence, thus necessitating prevention interventions around the time of menarche to reduce the risk for sexual violence.

Treatment Disparities Partially Mediate Socioeconomic- and Race/Ethnicity-Based Survival Disparities in Stage I-II Hepatocellular Carcinoma

BACKGROUND

Low socioeconomic status (SES) patients with early-stage hepatocellular carcinoma (HCC) receive procedural treatments less often and have shorter survival. Little is known about the extent to which these survival disparities result from treatment-related disparities versus other causal pathways. We aimed to estimate the proportion of SES-based survival disparities that are mediated by treatment- and facility-related factors among patients with stage I-II HCC.

METHODS

We analyzed patients aged 18-75 years diagnosed with stage I-II HCC in 2008-2016 using the National Cancer Database. Inverse odds weighting mediation analysis was used to calculate the proportion mediated by three mediators: procedure type, facility volume, and facility procedural interventions offered. Intersectional analyses were performed to determine whether treatment disparities played a larger role in survival disparities among Black and Hispanic patients.

RESULTS

Among 46,003 patients, 15.0% had low SES, 71.6% had middle SES, and 13.4% had high SES. Five-year overall survival was 46.9%, 39.9%, and 35.7% among high, middle, and low SES patients, respectively. Procedure type mediated 45.9% (95% confidence interval [CI] 31.1-60.7%) and 36.7% (95% CI 25.7-47.7%) of overall survival disparities for low and middle SES patients, respectively, which was more than was mediated by the two facility-level mediators. Procedure type mediated a larger proportion of survival disparities among low-middle SES Black (46.6-48.2%) and Hispanic patients (92.9-93.7%) than in White patients (29.5-29.7%).

CONCLUSIONS

SES-based disparities in use of procedural interventions mediate a large proportion of survival disparities, particularly among Black and Hispanic patients. Initiatives aimed at attenuating these treatment disparities should be pursued.

Does disaster-related relocation impact mental health via changes in group participation among older adults? Causal mediation analysis of a pre-post disaster study of the 2016 Kumamoto earthquake

BACKGROUND

Disaster-related relocation is associated with depression and post-traumatic stress disorder, especially in older adults. Disaster-related relocation often deprives survivors of opportunities for social group participation, potentially deteriorating their mental health. On the contrary, the relocation could also be an opportunity for optimizing social relationships, ending/reducing unwanted participation. This study examined the potential mediation effects of changing participation for the link of disaster-related relocation to mental health.

METHODS

We analyzed a pre-post disaster dataset of functionally independent older adults from the Japan Gerontological Evaluation Study. Following the 2013 survey, a follow-up survey was conducted seven months after the 2016 Kumamoto earthquake (n = 828).

RESULTS

The causal mediation analyses indicated that compared to no relocation, the relative risk for experiencing major depressive episodes among those relocating to temporary housing was 3.79 [95% confidence interval: 1.70-6.64] (natural direct effect). By contrast, the relative risk for those renewing (either ceased or started) group participation was 0.60 [95% CI: 0.34-0.94] (natural indirect effect).

CONCLUSIONS

Optimization of social ties according to a renewal of group participation status might have protected older adults in temporary housing against depression.

Socio-economic disparities in exposure to and endorsement of COVID-19 vaccine misinformation and the associations with vaccine hesitancy and vaccination

OBJECTIVES

We examined disparities in vaccine misinformation exposure and endorsement and the associations with vaccine hesitancy and vaccination uptake.

STUDY DESIGN

Population-based survey.

METHODS

A population-based survey was conducted on 5,002 Hong Kong adults oversampling low socio-economic status (SES, n = 2,200). Information on exposure (13 misinformation statements, total 0-13, median = 2), endorsement (13 statements, score 0-10, high scores indicate higher levels of endorsement, median = 5.75) of misinformation, vaccine hesitancy (14 items, score 1-5), and vaccination (two doses) were collected. Multivariable regression (adjusted β [aβ]) and Poisson regression (adjusted risk ratio [aRR]) adjusting for demographic characteristics were used to examine the associations of exposure to and endorsement of misinformation with vaccine hesitancy and vaccination.

RESULTS

A total of 71.8% of respondents had at least exposure to one vaccine misinformation, and 35.7% had a high level of endorsement (median or above). Respondents with lower SES had a lower exposure (≤2 statements, 57.1% vs 50.1%, P < 0.001) but a higher level of endorsement (36.6% vs 34.9%, P = 0.01) of misinformation. Overall, 72.9% had been vaccinated for two or more doses, with a lower proportion in respondents with lower SES (83.6% vs 61.1%; P < 0.001). Compared with no exposure to misinformation, high levels of exposure and endorsement were associated with vaccine hesitancy (aβ = 0.44, 95% confidence interval 0.40-0.48; aβ = 0.50, 0.47-0.54, respectively) and lower vaccination rates (aRR = 0.98, 0.97-0.99; aRR = 0.92, 0.88-0.96, respectively). Vaccine hesitancy mediated the associations of exposure (fully, 100%) and endorsement (partially, 73%) with vaccination uptake.

CONCLUSION

Endorsement of vaccine misinformation in respondents with lower SES was associated with low vaccination uptake.

Racial and Ethnic Inequities in Development of Type 2 Diabetes After Gestational Diabetes Mellitus

OBJECTIVE

To estimate racial and ethnic disparities in type 2 diabetes mellitus after gestational diabetes mellitus (GDM) and to investigate baseline pregnancy clinical and social or structural characteristics as mediators.

METHODS

We conducted a retrospective cohort of individuals with GDM using linked 2009-2011 New York City birth and hospital data and 2009-2017 New York City A1c Registry data. We ascertained GDM and pregnancy characteristics from birth and hospital records. We classified type 2 diabetes as two hemoglobin A 1c test results of 6.5% or higher. We grouped pregnancy characteristics into clinical (body mass index [BMI], chronic hypertension, gestational hypertension, preeclampsia, preterm delivery, caesarean, breastfeeding, macrosomia, shoulder dystocia) and social or structural (education, Medicaid insurance, prenatal care, and WIC [Special Supplemental Nutrition Program for Women, Infants, and Children] participation). We used Cox proportional hazards models to estimate associations between race and ethnicity and 8-year type 2 diabetes incidence, and we tested mediation of pregnancy characteristics, additionally adjusting for age and nativity (U.S.-born vs foreign-born).

RESULTS

The analytic data set included 22,338 patients with GDM. The 8-year type 2 diabetes incidence was 11.7% overall and 18.5% in Black, 16.8% in South and Southeast Asian, 14.6% in Hispanic, 5.5% in East and Central Asian, and 5.4% in White individuals with adjusted hazard ratios of 4.0 (95% CI 2.4-3.9), 2.9 (95% CI 2.4-3.3), 3.3 (95% CI 2.7-4.2), and 1.0 (95% CI 0.9-1.4) for each group compared with White individuals. Clinical and social or structural pregnancy characteristics explained 9.3% and 23.8% of Black, 31.2% and 24.7% of Hispanic, and 7.6% and 16.3% of South and Southeast Asian compared with White disparities. Associations between education, Medicaid insurance, WIC participation, and BMI and type 2 diabetes incidence were more pronounced among White than Black, Hispanic, and South and Southeast Asian individuals.

CONCLUSION

Population-based racial and ethnic inequities are substantial in type 2 diabetes after GDM. Characteristics at the time of delivery partially explain disparities, creating an opportunity to intervene on life-course cardiometabolic inequities, whereas weak associations of common social or structural measures and BMI in Black, Hispanic and South and Southeast Asian individuals demonstrate the need for greater understanding of how structural racism influences postpartum cardiometabolic risk in these groups.

To what extent do job insecurity, income satisfaction, and self-rated health mediate the association between low income and presenteeism in Germany? A gendered analysis

PURPOSE

Low income is considered a possible determinant of presenteeism, explained by poor working and living conditions, increased levels of uncertainties and anxiety, and poor health status. We aimed to examine the association between low income and presenteeism by gender and to explain their association using different mediators.

METHODS

A total of 14,299 employees aged 18-65 from the 6th BIBB/BAuA Employment Survey 2012 were used, and mediation analyses with inverse odds weighting stratified by gender were conducted.

RESULTS

Low income was significantly associated with presenteeism for men at a significant level of α < .05 (β: 0.376; 95%-CI 0.148-0.604) and for women at a significant level of α < .10 (β: 0.120; 95%-CI - 0.015-0.255). The total effect (TE) was fully and significantly mediated for women when all mediator-weights were considered, whereas for men the consideration of single mediator-weights led to a full and significant mediation of the association between low income and presenteeism. Self-rated health status and income satisfaction contributed the most to the differences in presenteeism by low income with a proportion mediated of 96.3% (men) and 169.2% (women) for self-rated health and 101.6% (men) and 162.5% (women) for income satisfaction.

DISCUSSION

The results indicated a strong association between low income and presenteeism, in particular for men. Self-rated health and income satisfaction were the most important mediators of this association. The results underline not only the relevance of occupational health management and preventive measures, but also the need of a public debate about employment traditions, possibly resulting in role conflicts among men, and wage equality to prevent presenteeism of low-income earners.

The contribution of work and health-related lifestyle to educational inequalities in physical health among older workers in Germany. A causal mediation analysis with data from the lidA cohort study

OBJECTIVES

The objective of the study was to investigate the contribution of work factors and health-related lifestyle to educational inequalities in physical health among older workers in Germany by applying causal mediation analysis with longitudinal data.

METHODS

Data from the German lidA study was used. 2653 persons (53% female, 47% male) aged 46 (born 1965) and 52 (born 1959) at baseline were followed up for seven years with exposure and outcome assessments in 2011 (t0), 2014 (t1) and 2018 (t2). The total effect of education on physical health was decomposed into a natural direct effect (NDE) and a natural indirect effect (NIE) by using a sex-stratified causal mediation analysis with an inverse odds weighting approach. Baseline health, partner status and working hours were entered as a first set of mediators preceding the putative mediators of interest. All analyses were adjusted for age and migrant status.

RESULTS

Independent of the first set of mediators, work factors explained 21% of educational inequalities in physical health between low and high educated women and 0% comparing moderate versus high educated women. The addition of health behaviors explained further 26% (low vs. high education) and 20% (moderate vs. high education), respectively. Among men, net of the first set of mediators, work factors explained 5% of educational inequalities in physical health between low and high educated and 6% comparing moderate versus high educated persons. Additional 24% (low vs. high education) and 27% (moderate vs. high education) were explained by adding health behaviors to the models.

CONCLUSIONS

To reduce educational inequalities in physical health among older workers in Germany, interventions to promote healthy behaviors are promising. Improving working conditions is likely an important prerequisite.