Abstract:Background: Exposure to environmental factors has a high burden on human health, with millions of premature annual deaths associated with the short-term health effects of ambient temperatures and air pollution. However, direct estimations of exposure-related mortality from real data are still not available in most parts of the world, especially in low-resource settings, due to the unavailability of daily health records to calibrate epidemiological models. Methods: In this study, we have filled the crucial gap in available direct estimations by developing a method to make valid inference for the relationship between exposure and response data that uses only exposure and temporally aggregated response data. We provided the mathematical derivation of the method, and compared the results by using simulations applied to daily temperature and daily, weekly, and monthly mortality data. The method was then applied to the newly created database of the EARLY-ADAPT project. Findings: The daily and weekly models produced similar and unbiased estimates of the temperature-related relative risks and attributable mortality, with only slightly more imprecision in the weekly model. Even the estimates of the monthly model were unbiased when using enough data, although at the expense of a substantial increase in variability. The real data analysis showed that the similarity between the regional values of two aggregation models increased with the number of years and regions of the dataset, and decreased with the difference in their degree of temporal aggregation. Interpretation: Our method opens the door to conducting epidemiological studies in low-resource settings, where access to daily health data is not possible. Moreover, it allows accurate estimation of the short-term health effects of environmental exposures in near-real time, when daily health data are still not available, such as in the estimation of the mortality burden of recent record-breaking heat episodes. Overall, our method represents an important new approach to how the public health community can use data to create new evidence for research, translation and policy making. Funding: European Research Council (ERC).

A penalized distributed-lag non-linear model for modeling the joint delayed effect of two predictors: impact of minimum and maximum temperature on mortality

Spatial Bayesian distributed lag non-linear models (SB-DLNM) for small-area exposure-lag-response epidemiological modelling

A unifying modelling approach for hierarchical distributed lag models

Modeling of Correlated Cognitive Function and Functional Disability Outcomes with Bounded and Missing Data in a Longitudinal Aging Study

Treed distributed lag nonlinear models

Penalized distributed lag interaction model: Air pollution, birth weight, and neighborhood vulnerability

Multiple exposure distributed lag models with variable selection

The Impact of Temperature on Mortality in Tianjin, China: A Case-Crossover Design with a Distributed Lag Nonlinear Model

The effect of temporal data aggregation to assess the impact of changing temperatures in Europe: an epidemiological modelling study

Aggregating the response in time series regression models, applied to weather-related cardiovascular mortality

The short-term association between environmental variables and mortality: evidence from Europe

Unbiased temperature-related mortality estimates using weekly and monthly health data: a new method for environmental epidemiology and climate impact studies

Bayesian Distributed Lag Interaction Models to Identify Perinatal Windows of Vulnerability in Children's Health

EMD-regression for modelling multi-scale relationships, and application to weather-related cardiovascular mortality

Using a generalized additive model with autoregressive terms to study the effects of daily temperature on mortality

Spatiotemporal distribution and lag effect of extreme temperature exposure on mortality of residents in Jiangsu, China

Bayesian Generalized Distributed Lag Regression with Variable Selection

Targeted Maximum Likelihood Estimation for Dynamic and Static Longitudinal Marginal Structural Working Models

Revisiting the importance of temperature, weather and air pollution variables in heat-mortality relationships with machine learning

Prediction of heatwave related mortality magnitude, duration and frequency with climate variability and climate change information

[Distributed Lag Effects in the Relationship Between Daily Mean Temperature and Mortality in Shanghai].