Bayesian Kernel Machine Regression for Social Epidemiologic Research.

BACKGROUND: Little attention has been devoted to framing multiple continuous social variables as a "mixture" for social epidemiologic analysis. We propose using the Bayesian kernel machine regression analytic framework that yields univariate, bivariate, and overall exposure mixture effects. METHODS: Using data from the 2023 Survey of Racism and Public Health, we conducted a Bayesian kernel machine regression analysis to study several individual, social, and structural factors as an exposure mixture and their relationships with psychological distress among individuals with at least one police arrest. Factors included racial and economic polarization, neighborhood deprivation, perceived discrimination, police perception, subjective social status, and substance use. We complemented this analysis with a series of unadjusted and adjusted models for each exposure mixture variable. RESULTS: We found that more self-reported discrimination experiences in the past year (posterior inclusion probability = 1.00) and greater substance use (posterior inclusion probability = 1.00) correlated with higher psychological distress. These associations were consistent with the findings from the unadjusted and adjusted linear regression analyses: past year perceived discrimination (unadjusted b = 2.58, 95% CI: 1.86, 3.30; adjusted b = 2.20, 95% CI: 1.45, 2.94) and substance use (unadjusted b = 2.92, 95% CI: 2.21, 3.62; adjusted b = 2.59, 95% CI: 1.87, 3.31). CONCLUSIONS: With the rise of big data and the expansion of variables in long-standing cohort and census studies, novel applications of methods from adjacent disciplines are a step forward in identifying exposure mixture associations in social epidemiology and addressing the health needs of socially vulnerable populations.

Enhanced family history-based algorithms increase the identification of individuals meeting criteria for genetic testing of hereditary cancer syndromes but would not reduce disparities on their own.

OBJECTIVE: This study aimed to 1) investigate algorithm enhancements for identifying patients eligible for genetic testing of hereditary cancer syndromes using family history data from electronic health records (EHRs); and 2) assess their impact on relative differences across sex, race, ethnicity, and language preference. MATERIALS AND METHODS: The study used EHR data from a tertiary academic medical center. A baseline rule-base algorithm, relying on structured family history data (structured data; SD), was enhanced using a natural language processing (NLP) component and a relaxed criteria algorithm (partial match [PM]). The identification rates and differences were analyzed considering sex, race, ethnicity, and language preference. RESULTS: Among 120,007 patients aged 25-60, detection rate differences were found across all groups using the SD (all P < 0.001). Both enhancements increased identification rates; NLP led to a 1.9 % increase and the relaxed criteria algorithm (PM) led to an 18.5 % increase (both P < 0.001). Combining SD with NLP and PM yielded a 20.4 % increase (P < 0.001). Similar increases were observed within subgroups. Relative differences persisted across most categories for the enhanced algorithms, with disproportionately higher identification of patients who are White, Female, non-Hispanic, and whose preferred language is English. CONCLUSION: Algorithm enhancements increased identification rates for patients eligible for genetic testing of hereditary cancer syndromes, regardless of sex, race, ethnicity, and language preference. However, differences in identification rates persisted, emphasizing the need for additional strategies to reduce disparities such as addressing underlying biases in EHR family health information and selectively applying algorithm enhancements for disadvantaged populations. Systematic assessment of differences in algorithm performance across population subgroups should be incorporated into algorithm development processes.

Detecting univariate, bivariate, and overall effects of drug mixtures using Bayesian kernel machine regression.

Background: Innovative analytic approaches to drug studies are needed to understand better the co-use of opioids with non-opioids among people using illicit drugs. One approach is the Bayesian kernel machine regression (BKMR), widely applied in environmental epidemiology to study exposure mixtures but has received far less attention in substance use research.Objective: To describe the utility of the BKMR approach to study the effects of drug substance mixtures on health outcomes.Methods: We simulated data for 200 individuals. Using the Vale and Maurelli method, we simulated multivariate non-normal drug exposure data: xylazine (mean = 300 ng/mL, SD = 100 ng/mL), fentanyl (mean = 200 ng/mL, SD = 71 ng/mL), benzodiazepine (mean = 300 ng/mL, SD = 55 ng/mL), and nitazene (mean = 200 ng/mL, SD = 141 ng/mL) concentrations. We performed 10,000 MCMC sampling iterations with three Markov chains. Model diagnostics included trace plots, r-hat values, and effective sample sizes. We also provided visual relationships of the univariate and bivariate exposure-response and the overall mixture effect.Results: Higher levels of fentanyl and nitazene concentrations were associated with higher levels of the simulated health outcome, controlling for age. Trace plots, r-hat values, and effective sample size statistics demonstrated BKMR stability across multiple Markov chains.Conclusions: Our understanding of drug mixtures tends to be limited to studies of single-drug models. BKMR offers an innovative way to discern which substances pose a greater health risk than other substances and can be applied to assess univariate, bivariate, and cumulative drug effects on health outcomes.

Single-world intervention graphs for defining, identifying, and communicating estimands in clinical trials.

Confusion often arises when attempting to articulate target estimand(s) of a clinical trial in plain language. We aim to rectify this confusion by using a type of causal graph called the Single-World Intervention Graph (SWIG) to provide a visual representation of the estimand that can be effectively communicated to interdisciplinary stakeholders. These graphs not only display estimands, but also illustrate the assumptions under which a causal estimand is identifiable by presenting the graphical relationships between the treatment, intercurrent events, and clinical outcomes. To demonstrate its usefulness in pharmaceutical research, we present examples of SWIGs for various intercurrent event strategies specified in the ICH E9(R1) addendum, as well as an example from a real-world clinical trial for chronic pain. code to generate all the SWIGs shown is this paper is made available. We advocate clinical trialists adopt the use of SWIGs in their estimand discussions during the planning stages of their studies.

The impact of correlated exposures and missing data on multiple informant models used to identify critical exposure windows.

There has been heightened interest in identifying critical windows of exposure for adverse health outcomes; that is, time points during which exposures have the greatest impact on a person's health. Multiple informant models implemented using generalized estimating equations (MIM GEEs) have been applied to address this research question because they enable statistical comparisons of differences in associations across exposure windows. As interest rises in using MIMs, the feasibility and appropriateness of their application under settings of correlated exposures and partially missing exposure measurements requires further examination. We evaluated the impact of correlation between exposure measurements and missing exposure data on the power and differences in association estimated by the MIM GEE and an inverse probability weighted extension to account for informatively missing exposures. We assessed these operating characteristics under a variety of correlation structures, sample sizes, and missing data mechanisms considering various exposure-outcome scenarios. We showed that applying MIM GEEs maintains higher power when there is a single critical window of exposure and exposure measures are not highly correlated, but may result in low power and bias under other settings. We applied these methods to a study of pregnant women living with HIV to explore differences in association between trimester-specific viral load and infant neurodevelopment.

Racial Composition of Social Environments Over the Life Course Using the Pictorial Racial Composition Measure: Development and Validation Study.

BACKGROUND: Studies investigating the impact of racial segregation on health have reported mixed findings and tended to focus on the racial composition of neighborhoods. These studies use varying racial composition measures, such as census data or investigator-adapted questions, which are currently limited to assessing one dimension of neighborhood racial composition. OBJECTIVE: This study aims to develop and validate a novel racial segregation measure, the Pictorial Racial Composition Measure (PRCM). METHODS: The PRCM is a 10-item questionnaire of pictures representing social environments across adolescence and adulthood: neighborhoods and blocks (adolescent and current), schools and classrooms (junior high and high school), workplace, and place of worship. Cognitive interviews (n=13) and surveys (N=549) were administered to medically underserved patients at a primary care clinic at the Barnes-Jewish Hospital. Development of the PRCM occurred across pilot and main phases. For each social environment and survey phase (pilot and main), we computed positive versus negative pairwise comparisons: mostly Black versus all other categories, half Black versus all other categories, and mostly White versus all other categories. We calculated the following validity metrics for each pairwise comparison: sensitivity, specificity, correct classification rate, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, false positive rate, and false negative rate. RESULTS: For each social environment, the mostly Black and mostly White dichotomizations generated better validity metrics relative to the half Black dichotomization. Across all 10 social environments in the pilot and main phases, mostly Black and mostly White dichotomizations exhibited a moderate-to-high sensitivity, specificity, correct classification rate, positive predictive value, and negative predictive value. The positive likelihood ratio values were >1, and the negative likelihood ratio values were close to 0. The false positive and negative rates were low to moderate. CONCLUSIONS: These findings support that using either the mostly Black versus other categories or the mostly White versus other categories dichotomizations may provide accurate and reliable measures of racial composition across the 10 social environments. The PRCM can serve as a uniform measure across disciplines, capture multiple social environments over the life course, and be administered during one study visit. The PRCM also provides an added window into understanding how structural racism has impacted minoritized communities and may inform equitable intervention and prevention efforts to improve lives.

Racial Composition of Past and Current Social Environments and Health Literacy.

BACKGROUND: Research is needed to understand the impact of social determinants of health on health literacy throughout the life course. This study examined how racial composition of multiple past and current social environments was related to adults' health literacy. METHODS: In this study, 546 adult patients at a primary care clinic in St. Louis, Missouri, completed a self-administered written questionnaire that assessed demographic characteristics and a verbally administered component that assessed health literacy with the Rapid Estimate of Adult Literacy in Medicine - Revised (REALM-R) and Newest Vital Sign (NVS), and self-reported racial composition of six past and four current social environments. Multilevel logistic regression models were built to examine the relationships between racial composition of past and current social environments and health literacy. RESULTS: Most participants identified as Black or multiracial (61%), had a high school diploma or less (54%), and household income <$20,000 (72%). About 56% had adequate health literacy based on REALM-R and 38% based on NVS. In regression models, participants with multiple past white environments (e.g., locations/conditions in which most of the people who live, go to school, work, and have leisure time are White) and (vs. 0 or 1) were more likely to have adequate health literacy based on REALM-R (adjusted odds ratio [aOR] = 1.79; 95% confidence interval [CI]: 1.04-3.07). Similarly, participants who had multiple past white social environments were more likely (aOR = 1.94, 95% CI: 1.15-3.27) to have adequate health literacy based on NVS than those who had not. The racial composition of current social environments was not significantly associated with health literacy in either model. CONCLUSIONS: Racial composition of past, but not current, educational and residential social environments was significantly associated with adult health literacy. The results highlight the importance of examining the impact of social determinants over the life course on health literacy. The findings suggest that policies ensuring equitable access to educational resources in school and community contexts is critical to improving equitable health literacy. [HLRP: Health Literacy Research and Practice. 2024;8(3):e130-e139.].

PLAIN LANGUAGE SUMMARY: We studied how the racial make-up of past and current places where people live, work, and go to school were related to their health literacy as adults. We found that the racial make-up of past places, but not current places, was related to health literacy. Our results show the need to study the impact of childhood places on health literacy.

Associations between perceived discrimination over the life course, subjective social status, and health literacy: A racial/ethnic stratification analysis.

Objective: To analyze the relationship between perceived discrimination over the life course, social status, and limited health literacy (HL). Methods: 5040 adults who participated in the 2023 Survey of Racism and Public Health. We applied stratified multilevel models adjusted for sociodemographic characteristics. Results: The average age was 47 years, 48% identified as White, 20% as Latinx, and 17% as Black. In the overall sample, we observed associations of perceived discrimination (b = 0.05, 95% CI: 0.01, 0.09), subjective social status (b = -0.16, 95% CI: -0.23, -0.10), and their interaction (b = 0.02, 95% CI: 0.01, 0.03). More perceived discrimination was associated with lower HL in the White and Multiracial participants. Higher subjective social status was associated with higher HL in the White and Latinx participants. There was a statistically significant interaction between perceived discrimination and subjective social status on HL among the White, Latinx, and Multiracial participants. Conclusion: This analysis has implications for public health practice, indicating that multi-level interventions are needed to address limited HL. Innovation: Our findings provide novel insights for identifying key SDOH indicators to assess in clinical settings to provide health literate care.

Love after lockup: examining the role of marriage, social status, and financial stress among formerly incarcerated individuals.

BACKGROUND: Upon reintegration into society, formerly incarcerated individuals (FIIs) experience chronic financial stress due to prolonged unemployment, strained social relationships, and financial obligations. This study examined whether marriage and perceived social status can mitigate financial stress, which is deleterious to the well-being of FIIs. We also assessed whether sociodemographic factors influenced financial stress across marital status. We used cross-sectional data from 588 FIIs, collected in the 2023 Survey of Racism and Public Health. The financial stress outcome (Cronbach's [Formula: see text] = 0.86) comprised of five constructs: psychological distress, financial anxiety, job insecurity, life satisfaction, and financial well-being. Independent variables included marital and social status, age, race/ethnicity, gender identity, educational attainment, employment status, and number of dependents. Multivariable models tested whether financial stress levels differed by marital and perceived social status (individual and interaction effects). Stratified multivariable models assessed whether social status and sociodemographic associations varied by marital status. RESULTS: We found that being married/living with a partner (M/LWP, b = -5.2) or having higher social status (b = -2.4) were protective against financial stress. Additionally, the social status effect was more protective among divorced, separated, or widowed participants (b = -2.5) compared to never married (NM, b = -2.2) and M/LWP (b = -1.7) participants. Lower financial stress correlated with Black race and older age, with the age effect being more pronounced among M/LWP participants (b = -9.7) compared to NM participants (b = -7.3). Higher financial stress was associated with woman gender identity (overall sample b = 2.9, NM sample b = 5.1), higher education (M/LWP sample b = 4.4), and having two or more dependents (overall sample b = 2.3, M/LWP sample b = 3.4). CONCLUSIONS: We provide novel insights into the interrelationship between marriage, perceived social status, and financial stress among FIIs. Our findings indicate the need for policies and programs which may target the family unit, and not only the individual, to help alleviate the financial burden of FIIs. Finally, programs that offer legal aid to assist in expungement or sealing of criminal records or those offering opportunities for community volunteer work in exchange for vouchers specific to legal debt among FIIs could serve to reduce financial stress and improve social standing.

Lower social vulnerability is associated with a higher prevalence of social media-involved violent crimes in Prince George's County, Maryland, 2018-2023.

BACKGROUND: Social vulnerability may play a role in social media-involved crime, but few studies have investigated this issue. We investigated associations between social vulnerability and social media-involved violent crimes. METHODS: We analyzed 22,801 violent crimes occurring between 2018 and 2023 in Prince George's County, Maryland. Social media involvement was obtained from crime reports at the Prince George's County Police Department. Social media application types included social networking, advertising/selling, ridesharing, dating, image/video hosting, mobile payment, instant messaging/Voice over Internet Protocol, and other. We used the Centers for Disease Control and Prevention's Social Vulnerability Index to assess socioeconomic status (SES), household characteristics, racial and ethnic minority status, housing type and transportation, and overall vulnerability. Modified Poisson models estimated adjusted prevalence ratios (aPRs) among the overall sample and stratified by crime type (assault and homicide, robbery, and sexual offense). Covariates included year and crime type. RESULTS: Relative to high tertile areas, we observed a higher prevalence of social media-involved violent crimes in areas with low SES vulnerability (aPR: 1.82, 95% CI: 1.37-2.43), low housing type and transportation vulnerability (aPR: 1.53, 95% CI: 1.17-2.02), and low overall vulnerability (aPR: 1.63, 95% CI: 1.23-2.17). Low SES vulnerability areas were significantly associated with higher prevalences of social media-involved assaults and homicides (aPR: 1.64, 95% CI: 1.02-2.62), robberies (aPR: 2.00, 95% CI: 1.28-3.12), and sexual offenses (aPR: 2.07, 95% CI: 1.02-4.19) compared to high SES vulnerability areas. Low housing type and transportation vulnerability (vs. high) was significantly associated with a higher prevalence of social media-involved robberies (aPR: 1.54, 95% CI:1.01-2.37). Modified Poisson models also indicated that low overall vulnerability areas had higher prevalences of social media-involved robberies (aPR: 1.71, 95% CI: 1.10-2.67) and sexual offenses (aPR: 2.14, 95% CI: 1.05-4.39) than high overall vulnerability areas. CONCLUSIONS: We quantified the prevalence of social media-involved violent crimes across social vulnerability levels. These insights underscore the need for collecting incident-based social media involvement in crime reports among law enforcement agencies across the United States and internationally. Comprehensive data collection at the national and international levels provides the capacity to elucidate the relationships between neighborhoods, social media, and population health.