Evaluating Data Product Exposure Metrics for Use in Epidemiologic Studies of Dust Storms.

Dust storms are increasing in frequency and correlate with adverse health outcomes but remain understudied in the United States (U.S.), partially due to the limited spatio-temporal coverage, resolution, and accuracy of current data sets. In this work, dust-related metrics from four public areal data products were compared to a monitor-based "gold standard" dust data set. The data products included the National Weather Service (NWS) storm event database, the Modern-Era Retrospective analysis for Research and Applications-Version 2, the EPA's Air QUAlity TimE Series (EQUATES) Project using the Community Multiscale Air Quality Modeling System (CMAQ), and the Copernicus Atmosphere Monitoring Service global reanalysis product. California, Nevada, Utah, and Arizona, which account for most dust storms reported in the U.S., were examined. Dichotomous and continuous metrics based on reported dust storms, particulate matter concentrations (PM10 and PM2.5), and aerosol-type variables were extracted or derived from the data products. Associations between these metrics and a validated dust storm detection method utilizing Interagency Monitoring of Protected Visual Environments monitors were estimated via quasi-binomial regression. In general, metrics from CAMS yielded the strongest associations with the "gold standard," followed by the NWS storm database metric. Dust aerosol (0.9-20 µm) mixing ratio, vertically integrated mass of dust aerosol (9-20 µm), and dust aerosol optical depth at 550 nm from CAMS generated the highest standardized odds ratios among all metrics. Future work will apply machine-learning methods to the best-performing metrics to create a public dust storm database suitable for long-term epidemiologic studies.

Household profiles of neglected tropical disease symptoms among children: A latent class analysis of built-environment features of Tanzanian households using the Demographic and Health Survey.

Background: While malaria and neglected tropical disease (NTD) morbidity and mortality rates among children <5 years old have decreased through public health efforts in the United Republic of Tanzania, associations between household environments and disease outcomes are relatively unknown. Methods: We conducted latent class analysis (LCA) on 2015-2016 Tanzania Demographic Health Survey data from mothers with children <5 years old (N = 10 233) to identify NTD household risk profiles. The outcome of child NTD was assessed by mothers' reports of recent diarrhoea, cough, treatment for enteric parasites, and fever symptoms. Household-built environment indicators included urban/rural designation, electricity access, water source, cooking fuel, flooring, wall, and roofing materials. External environmental covariates were considered to further differentiate profiles. Results: Five profiles were identified in the sample: rural finished walls households (40.2%) with the lowest NTD risk; rural rudimentary households (20.9%) with intermediate-low NTD risk; finished material households (22.5%) with intermediate NTD risk; urban households (14.4%) with intermediate-high NTD risk and high likelihood of enteric parasites; rural finished roof/walls households (2.1%) with the highest overall NTD risk. Conclusions: This study is among the first to use LCA to examine household environment characteristics to assess child NTD risk in Tanzania. This paper serves as a framework for community-level rapid NTD risk assessment for targeted health promotion interventions.

A Time-Varying Effect Model (TVEM) of the Complex Association of Tobacco Use and Smoke Exposure on Mean Telomere Length: Differences between Racial and Ethnic Groups Assessed in the National Health and Nutrition Examination Survey.

Telomere length is affected by lifestyle and environmental factors and varies between racial and ethnic groups; however, studies are limited, with mixed findings. This study examined the effects of tobacco use and smoke exposure on mean telomere length to identify critical age periods by race/ethnicity. We used time-varying effect modeling on the National Health and Nutrition Examination Survey for continuous years 1999-2002 to observe the effects of active tobacco use and environmental tobacco smoke-measured through serum cotinine-and mean telomere length for adults 19 to 85 and older (N = 7826). Models were run for Mexican American, other Hispanic, non-Hispanic White, non-Hispanic Black, and other/multi-race categories to allow for time-varying group differences, and controlled for biological sex, socioeconomic status, education, and ever-smoker status. Serum cotinine was found to have an increasing effect on telomere length from age 37 to approximately age 74 among Mexican Americans. Among other/multi-race individuals serum cotinine was found to have a decreasing effect at approximately age 42, and among Blacks, it had an overall decreasing effect from age 61 to 78. Findings reveal a further need to focus additional support and resources to intervene regarding disparate health effects from tobacco use and environmental smoke exposure for already vulnerable groups at particular ages.

Intimate partner violence and HIV testing during antenatal care: A latent class analysis to identify risk factors for HIV infection in mothers and their children in the United Republic of Tanzania.

Intimate partner violence has adverse effects on mother's overall health and prevention of mother to child HIV transmission. To identify and examine subgroups of mothers experiencing intimate partner violence and the likelihood of HIV testing during antenatal care, we conducted a latent class analysis using data from the Tanzania Demographic and Health Survey 2010 (N = 2,809). Intimate partner violence included mother's experiences with partners' controlling behaviors, as well as emotional, physical, and sexual violence. The outcome was mother's accepting HIV testing offered during their antenatal care visit. Covariates included mother's level of education, rural/urban residence, and prevention of mother to child HIV transmission talk during antenatal care visit. The latent class analysis indicated a three-class solution was the best model and identified the following profiles: mothers with no experience of intimate partner violence (61% of sample) with a 90.5% likelihood of HIV testing; mothers with moderate levels of intimate partner violence (26%) with an 84.7% likelihood of testing; and mothers with extreme levels of intimate partner violence (13%) with an 82% likelihood of testing. An auxiliary multinomial logistic regression with selected covariates was conducted to further differentiate IPV profiles, where mothers with extreme levels of intimate partner violence had 57% increased odds [95%CI:1.06-2.33, p = .023] of living in rural areas compared to mothers with no experience of intimate partner violence. Our person-centered methodological approach provided a novel model to understand the impact of multiple intimate partner violence risk factors on antenatal care HIV testing to identify mothers in need of interventions and their children at highest for parent to child HIV transmission. Our model allows person-centered interventional designs tailored for the most at-risk subgroups within a population.

Sociodemographic and Geographic Disparities of Prostate Cancer Treatment Delay in Tennessee: A Population-Based Study.

The relationship of social determinants of health, Appalachian residence, and prostate cancer treatment delay among Tennessee adults is relatively unknown. We used multivariate logistic regression on 2005-2015 Tennessee Cancer Registry data of adults aged ≥18 diagnosed with prostate cancer. The outcome of treatment delay was more than 90 days without surgical or nonsurgical intervention from date of diagnosis. Social determinants in the population-based registry were race (White, Black, Other) and marital status (single, married, divorced/separated, widow/widower). Tennessee residence was classified as Appalachian versus non-Appalachian (urban/rural). Covariates include age at diagnosis (18-54, 54-69, ≥70), health insurance type (none, public, private), derived staging of cancer (localized, regional, distant), and treatment type (non-surgical/surgical). We found that Black and divorced/separated patients had 32% (95% confidence interval [CI]: 1.22-1.42) and 15% (95% CI: 1.01-1.31) increased odds to delay prostate cancer treatment. Patients were at decreased odds of treatment delay when living in an Appalachian county, both urban (odds ratio [OR] = 0.89, 95% CI: 0.82-0.95) and rural (OR = 0.83, 95% CI: 0.78-0.89), diagnosed at ≥70 (OR = 0.59, 95% CI: 0.53-0.66), and received surgical intervention (OR = 0.72, 95% CI: 0.68-0.76). Our study was among the first to comprehensively examine prostate cancer treatment delay in Tennessee, and while we do not make clinical recommendations, there is a critical need to further explore the unique factors that may propagate disparities. Prostate cancer treatment delay in Black patients may be indicative of ongoing health and access disparities in Tennessee, which may further affect quality of life and survivorship among this racial group. Divorced/separated patients may need tailored interventions to improve social support.

Identifying Risk Profiles of Malignant Prostate Cancer Surgical Delay Using a Person-Centered Approach to Understand Prostate Cancer Disparities: The Constellation of Health Determinants Using Latent Class Analysis on Cancer Registry Data.

Surgical prostate cancer (PCa) treatment delay (TD) may increase the likelihood of recurrence of disease, and influence quality of life as well as survival disparities between Black and White men. We used latent class analysis (LCA) to identify risk profiles in localized, malignant PCa surgical treatment delays while assessing co-occurring social determinants of health. Profiles were identified by age, marital status, race, county of residence (non-Appalachian or Appalachian), and health insurance type (none/self-pay, public, or private) reported in the Tennessee Department of Health cancer registry from 2005 to 2015 for adults ≥18 years (N = 18,088). We identified three risk profiles. The highest surgical delay profile (11% of the sample) with a 30% likelihood of delaying surgery >90 days were young Black men, <55 years old, living in a non-Appalachian county, and single/never married, with a high probability of having private health insurance. The medium surgical delay profile (46% of the sample) with a 21% likelihood of delay were 55-69 years old, White, married, and having private health insurance. The lowest surgical delay profile (42% of the sample) with a 14% likelihood of delay were ≥70 years with public health insurance as well as had a high probability of being White and married. We identified that even with health insurance coverage, Blacks living in non-Appalachian counties had the highest surgical delay, which was almost double that of Whites in the lowest delay profile. These disparities in PCa surgical delay may explain differences in health outcomes in Blacks who are most at-risk.