Epidemiology

The Increasing Exposure of the Global Population to Ionizing Radiation No abstract available

Estimating Causal Effects of Particulate Matter Regulation on Mortality Background: Estimating the causal effect of pollution on human health is integral for evaluating returns to pollution regulation, yet separating out confounding factors remains a perennial challenge. Methods: We use a quasi-experimental design to investigate the causal relationship between regulation of particulate matter smaller than 2.5 micrograms per cubic meter (PM2.5) and mortality among those 65 years of age and older. We exploit regulatory changes in the Clean Air Act Amendments (CAAA). Regulation in 2005 impacted areas of the United States differentially based on pre-regulation air quality levels for PM2.5. We use county-level mortality data, extracted from claims data managed by the Centers for Medicare & Medicaid Services, merged to county-level average PM2.5 readings and attainment status as classified by the Environmental Protection Agency. Results: Based on estimates from log-linear difference-in-differences models, our results indicate after the CAAA designation for PM2.5 in 2005, PM2.5 levels decreased 1.59 micrograms per cubic meter (95% CI = 1.39, 1.80) and mortality rates among those 65 and older decreased by 0.93% (95% CI = 0.10%, 1.77%) in nonattainment counties, relative to attainment ones. Results are robust to a series of alternate models, including nearest-neighbor matching based on propensity score estimates. Conclusion: This analysis suggests large health returns to the 2005 PM2.5 designations, and provides evidence of a causal association between pollution and mortality among the Medicare population.

Evaluating the Sensitivity of PM2.5–Mortality Associations to the Spatial and Temporal Scale of Exposure Assessment Background: The temporal and spatial scales of exposure assessment may influence observed associations between fine particulate air pollution (PM2.5) and mortality, but few studies have systematically examined this question. Methods: We followed 2.4 million adults in the 2001 Canadian Census Health and Environment Cohort for nonaccidental and cause-specific mortality between 2001 and 2011. We assigned PM2.5 exposures to residential locations using satellite-based estimates and compared three different temporal moving averages (1, 3, and 8 years) and three spatial scales (1, 5, and 10 km) of exposure assignment. In addition, we examined different spatial scales based on age, employment status, and urban/rural location, and adjustment for O3, NO2, or their combined oxidant capacity (Ox). Results: In general, longer moving averages resulted in stronger associations between PM2.5 and mortality. For nonaccidental mortality, we observed a hazard ratio of 1.11 (95% CI = 1.08, 1.13) for the 1-year moving average compared with 1.23 (95% CI = 1.20, 1.27) for the 8-year moving average. Respiratory and lung cancer mortality were most sensitive to the spatial scale of exposure assessment with stronger associations observed at smaller spatial scales. Adjustment for oxidant gases attenuated associations between PM2.5 and cardiovascular mortality and strengthened associations with lung cancer. Despite these variations, PM2.5 was associated with increased mortality in nearly all of the models examined. Conclusions: These findings support a relationship between outdoor PM2.5 and mortality at low concentrations and highlight the importance of longer-exposure windows, more spatially resolved exposure metrics, and adjustment for oxidant gases in characterizing this relationship.

Within-city Spatial Variations in Ambient Ultrafine Particle Concentrations and Incident Brain Tumors in Adults Background: Ambient ultrafine particles (UFPs, <0.1 µm) can reach the human brain, but to our knowledge, epidemiologic studies have yet to evaluate the relation between UFPs and incident brain tumors. Methods: We conducted a cohort study of within-city spatial variations in ambient UFPs across Montreal and Toronto, Canada, among 1.9 million adults included in multiple cycles of the Canadian Census Health and Environment Cohorts (1991, 1996, 2001, and 2006). UFP exposures (3-year moving averages) were assigned to residential locations using land-use regression models with exposures updated to account for residential mobility within and between cities. We followed cohort members for malignant brain tumors (ICD-10 codes C71.0–C71.9) between 2001 and 2016; Cox proportional hazards models (stratified by age, sex, immigration status, and census cycle) were used to estimate hazard ratios (HRs) adjusting for fine particle mass concentrations (PM2.5), nitrogen dioxide (NO2), and various sociodemographic factors. Results: In total, we identified 1,400 incident brain tumors during the follow-up period. Each 10,000/cm3 increase in UFPs was positively associated with brain tumor incidence (HR = 1.112, 95% CI = 1.042, 1.188) after adjusting for PM2.5, NO2, and sociodemographic factors. Applying an indirect adjustment for cigarette smoking and body mass index strengthened this relation (HR = 1.133, 95% CI = 1.032, 1.245). PM2.5 and NO2 were not associated with an increased incidence of brain tumors. Conclusions: Ambient UFPs may represent a previously unrecognized risk factor for incident brain tumors in adults. Future studies should aim to replicate these results given the high prevalence of UFP exposures in urban areas.

Are Descriptions of Methods Alone Sufficient for Study Reproducibility? An Example From the Cardiovascular Literature No abstract available

Positive Epidemiology? No abstract available

Methods to Account for Uncertainty in Latent Class Assignments When Using Latent Classes as Predictors in Regression Models, with Application to Acculturation Strategy Measures Latent class models have become a popular means of summarizing survey questionnaires and other large sets of categorical variables. Often these classes are of primary interest to better understand complex patterns in data. Increasingly, these latent classes are reified into predictors of other outcomes of interests, treating the most likely class as the true class to which an individual belongs even though there is uncertainty in class membership. This uncertainty can be viewed as a form of measurement error in predictors, leading to bias in the estimates of the regression parameters associated with the latent classes. Despite this fact, there is very limited literature treating latent class predictors as measurement error models. Most applications ignore this issue and fit a two-stage model that treats the modal class prediction as truth. Here, we develop two approaches—one likelihood-based, the other Bayesian—to implement a joint model for latent class analysis and outcome prediction. We apply these methods to an analysis of how acculturation behaviors predict depression in South Asian immigrants to the United States. A simulation study gives guidance for when a two-stage model can be safely implemented and when the joint model may be required.

Trends in Cancer Incidence Among American Indians and Alaska Natives and Non-Hispanic Whites in the United States, 1999–2015 Background: Female breast, prostate, lung, and colorectal cancers are the leading incident cancers among American Indian and Alaska Native (AI/AN) and non-Hispanic White (NHW) persons in the United States. To understand racial differences, we assessed incidence rates, analyzed trends, and examined geographic variation in incidence by Indian Health Service regions. Methods: To assess differences in incidence, we used age-adjusted incidence rates to calculate rate ratios (RRs) and 95% confidence intervals (CIs). Using joinpoint regression, we analyzed incidence trends over time for the four leading cancers from 1999 to 2015. Results: For all four cancers, overall and age-specific incidence rates were lower among AI/ANs than NHWs. By Indian Health Service regions, incidence rates for lung cancer were higher among AI/ANs than NHWs in Alaska (RR: 1.46; 95% CI: 1.37, 1.56) and Northern (RR: 1.29; 95% CI: 1.25, 1.33) and Southern (RR: 1.06; 95% CI: 1.03, 1.09) Plains. Similarly, colorectal cancer incidence rates were higher in AI/ANs than NHWs in Alaska (RR: 2.29; 95% CI: 2.14, 2.45) and Northern (RR: 1.04; 95% CI: 1.00, 1.09) and Southern (RR: 1.11; 95% CI: 1.07, 1.15) Plains. Also, AI/AN women in Alaska had a higher incidence rate for breast cancer than NHW women (RR: 1.05; 95% CI: 1.05, 1.20). From 1999 to 2015, incidence rates for all four cancers decreased in NHWs, but only rates for prostate (average annual percent change: –4.70) and colorectal (average annual percent change: –1.80) cancers decreased considerably in AI/ANs. Conclusion: Findings from this study highlight the racial and regional differences in cancer incidence.

Spatial–Temporal Cluster Analysis of Childhood Cancer in California Background: The observance of nonrandom space–time groupings of childhood cancer has been a concern of health professionals and the general public for decades. Many childhood cancers are suspected to have initiated in utero; therefore, we examined the spatial–temporal randomness of the birthplace of children who later developed cancer. Methods: We performed a space–time cluster analysis using birth addresses of 5,896 cases and 23,369 population-based, age-, sex-, and race/ethnicity-matched controls in California from 1997 to 2007, evaluating 20 types of childhood cancer and three a priori designated subgroups of childhood acute lymphoblastic leukemia (ALL). We analyzed data using a newly designed semiparametric analysis program, ClustR, and a common algorithm, SaTScan. Results: We observed evidence for nonrandom space–time clustering for ALL diagnosed at 2–6 years of age in the South San Francisco Bay Area (ClustR P = 0.04, SaTScan P = 0.07), and malignant gonadal germ cell tumors in a region of Los Angeles (ClustR P = 0.03, SaTScan P = 0.06). ClustR did not identify evidence of clustering for other childhood cancers, although SaTScan suggested some clustering for Hodgkin lymphoma (P = 0.09), astrocytoma (P = 0.06), and retinoblastoma (P = 0.06). Conclusions: Our study provides evidence that childhood ALL diagnosed at 2–6 years and malignant gonadal germ cell tumors sporadically occurs in nonrandom space–time clusters. Further research is warranted to identify epidemiologic features that may inform the underlying etiology.

ClustR: A Space–Time Cluster Analysis R Package for Individual-level Data Background: Until recently, large individual-level longitudinal data were unavailable to investigate clusters of disease, driving a need for suitable statistical tools. We introduce a robust, efficient, intuitive R package, ClustR, for space–time cluster analysis of individual-level data. Methods: We developed ClustR and evaluated the tool using a simulated dataset mirroring the population of California with constructed clusters. We assessed Cluster’s performance under various conditions and compared it with another space–time clustering algorithm: SaTScan. Results: ClustR mostly exhibited high sensitivity for urban clusters and low sensitivity for rural clusters. Specificity was generally high. Compared with SaTScan, ClustR ran faster and demonstrated similar sensitivity, but had lower specificity. Select cluster types were detected better by ClustR than SaTScan and vice versa. Conclusion: ClustR is a user-friendly, publicly available tool designed to perform efficient cluster analysis on individual-level data, filling a gap among current tools. ClustR and SaTScan exhibited different strengths and may be useful in conjunction.