Integrated causal-predictive machine learning models for tropical cyclone epidemiology.

Strategic preparedness reduces the adverse health impacts of hurricanes and tropical storms, referred to collectively as tropical cyclones (TCs), but its protective impact could be enhanced by a more comprehensive and rigorous characterization of TC epidemiology. To generate the insights and tools necessary for high-precision TC preparedness, we introduce a machine learning approach that standardizes estimation of historic TC health impacts, discovers common patterns and sources of heterogeneity in those health impacts, and enables identification of communities at highest health risk for future TCs. The model integrates (i) a causal inference component to quantify the immediate health impacts of recent historic TCs at high spatial resolution and (ii) a predictive component that captures how TC meteorological features and socioeconomic/demographic characteristics of impacted communities are associated with health impacts. We apply it to a rich data platform containing detailed historic TC exposure information and records of all-cause mortality and cardiovascular- and respiratory-related hospitalization among Medicare recipients. We report a high degree of heterogeneity in the acute health impacts of historic TCs, both within and across TCs, and, on average, substantial TC-attributable increases in respiratory hospitalizations. TC-sustained windspeeds are found to be the primary driver of mortality and respiratory risks.

Exposure Assessment for Tropical Cyclone Epidemiology.

PURPOSE OF REVIEW: Tropical cyclones impact human health, sometimes catastrophically. Epidemiological research characterizes these health impacts and uncovers pathways between storm hazards and health, helping to mitigate the health impacts of future storms. These studies, however, require researchers to identify people and areas exposed to tropical cyclones, which is often challenging. Here we review approaches, tools, and data products that can be useful in this exposure assessment. RECENT FINDINGS: Epidemiological studies have used various operational measures to characterize exposure to tropical cyclones, including measures of physical hazards (e.g., wind, rain, flooding), measures related to human impacts (e.g., damage, stressors from the storm), and proxy measures of distance from the storm's central track. The choice of metric depends on the research question asked by the study, but there are numerous resources available that can help in capturing any of these metrics of exposure. Each has strengths and weaknesses that may influence their utility for a specific study. Here we have highlighted key tools and data products that can be useful for exposure assessment for tropical cyclone epidemiology. These results can guide epidemiologists as they design studies to explore how tropical cyclones influence human health.

Tropical Cyclone Exposures and Risks of Emergency Medicare Hospital Admission for Cardiorespiratory Diseases in 175 Urban United States Counties, 1999-2010.

BACKGROUND: Although injuries experienced during hurricanes and other tropical cyclones have been relatively well-characterized through traditional surveillance, less is known about tropical cyclones' impacts on noninjury morbidity, which can be triggered through pathways that include psychosocial stress or interruption in medical treatment. METHODS: We investigated daily emergency Medicare hospitalizations (1999-2010) in 180 US counties, drawing on an existing cohort of high-population counties. We classified counties as exposed to tropical cyclones when storm-associated peak sustained winds were ≥21 m/s at the county center; secondary analyses considered other wind thresholds and hazards. We matched storm-exposed days to unexposed days by county and seasonality. We estimated change in tropical cyclone-associated hospitalizations over a storm period from 2 days before to 7 days after the storm's closest approach, compared to unexposed days, using generalized linear mixed-effect models. RESULTS: For 1999-2010, 175 study counties had at least one tropical cyclone exposure. Cardiovascular hospitalizations decreased on the storm day, then increased following the storm, while respiratory hospitalizations were elevated throughout the storm period. Over the 10-day storm period, cardiovascular hospitalizations increased 3% (95% confidence interval = 2%, 5%) and respiratory hospitalizations increased 16% (95% confidence interval = 13%, 20%) compared to matched unexposed periods. Relative risks varied across tropical cyclone exposures, with strongest association for the most restrictive wind-based exposure metric. CONCLUSIONS: In this study, tropical cyclone exposures were associated with a short-term increase in cardiorespiratory hospitalization risk among the elderly, based on a multi-year/multi-site investigation of US Medicare beneficiaries ≥65 years.

The association between socio-economic status and diabetes care and outcome in children with diabetes type 1 in Germany: The DIAS study (diabetes and social disparities).

OBJECTIVE: To evaluate the association between socioeconomic status (SES) and diabetes outcomes in German children and adolescents. METHODS: A total of 1829 subjects <18 years old with type 1 diabetes mellitus from 13 German diabetes centers were included from June 2013 until June 2014. Data were collected within the multicenter DPV (Diabetes Prospective Follow-up) registry. SES was measured with a composite index. Multivariable regression models were applied to analyze the association of SES and outcomes adjusted for age, sex, diabetes duration, and migration status. RESULTS: Low SES was significantly associated with worse diabetes outcomes: higher hemoglobin A1C (HbA1c) (64.3 mmol/mol), lower proportion of insulin pump therapy (43.6%), fewer daily self-monitored blood glucose (SMBG) measurements (5.7), more inpatient days per patient-year (5.8) compared to patients with medium/high SES (HbA1c: 61.3 mmol/mol, P < 0.001/59.8 mmol/mol, P < 0.0001; proportion of pump therapy: 54.5%, P < 0.01/ 54.9%, P < 0.01; SMBG: 6.0, P < 0.01/ 6.1, P < 0.01; inpatient days: 4.5, P < 0.0001/3.4, P < 0.0001). The inclusion of migration status in the models resulted in only minor changes in the outcomes. CONCLUSION: Despite free health care, low SES is associated with unfavorable diabetes outcomes in Germany. The poorer diabetes outcomes of children with diabetes have been attributed to their migration status and may be partly explained by low SES. Both factors must become part of targeted diabetes care in children and adolescents with type 1 diabetes.

Direct diabetes-related costs in young patients with early-onset, long-lasting type 1 diabetes.

OBJECTIVE: To estimate diabetes-related direct health care costs in pediatric patients with early-onset type 1 diabetes of long duration in Germany. RESEARCH DESIGN AND METHODS: Data of a population-based cohort of 1,473 subjects with type 1 diabetes onset at 0-4 years of age within the years 1993-1999 were included (mean age 13.9 (SD 2.2) years, mean diabetes duration 10.9 (SD 1.9) years, as of 31.12.2007). Diabetes-related health care services utilized in 2007 were derived from a nationwide prospective documentation system (DPV). Health care utilization was valued in monetary terms based on inpatient and outpatient medical fees and retail prices (perspective of statutory health insurance). Multiple regression models were applied to assess associations between direct diabetes-related health care costs per patient-year and demographic and clinical predictors. RESULTS: Mean direct diabetes-related health care costs per patient-year were €3,745 (inter-quartile range: 1,943-4,881). Costs for glucose self-monitoring were the main cost category (28.5%), followed by costs for continuous subcutaneous insulin infusion (25.0%), diabetes-related hospitalizations (22.1%) and insulin (18.4%). Female gender, pubertal age and poor glycemic control were associated with higher and migration background with lower total costs. CONCLUSIONS: Main cost categories in patients with on average 11 years of diabetes duration were costs for glucose self-monitoring, insulin pump therapy, hospitalization and insulin. Optimization of glycemic control in particular in pubertal age through intensified care with improved diabetes education and tailored insulin regimen, can contribute to the reduction of direct diabetes-related costs in this patient group.