The pitfalls of incidence-based time series regression for inferring the effects of weather on infectious diseases
- 1Max Planck Institute for Infection Biology, Infectious Disease Epidemiology group, Berlin, Germany
- 2Institute of Public Health, Charité – Universitätsmedizin Berlin, Berlin, Germany
- 3Department of Biology, New York University, New York, New York, United States of America
Abstract
A central question in environmental epidemiology is how the weather affects infectious diseases. Time-series regression (TSR) on population level case-counts data is widely used to estimate weather effects; however, this design may be biased due to the complexities of infectious disease dynamics, including nonlinear feedback, various types of noise, and latent, dynamic variables such as population immunity. Here, we assess the reliability of incidence-based TSR through a controlled simulation study across four different climates and fifty scenarios representing different pathogens. For each scenario, we simulated 10 years of weekly case-count data using a transmission model that included real-world weather data on temperature and relative humidity. We then examined whether the ground-truth weather effects could be recovered from simulated data using negative binomial generalized additive models, a flexible class of TSR models commonly used in empirical applications. We find that these models frequently fail to yield accurate and precise estimates of weather effects, even under favorable conditions such as no process noise and low observation noise (overdispersion). Hence, our results suggest that more causally elaborated formulations of TSR models, or more mechanistic approaches like transmission models, are needed for statistical inference of weather effects from population data. Author summary: Time-series regression on case-count data at the population level is a common technique for examining how weather affects infectious diseases. However, this analytical approach overlooks the fact that these data come from complex epidemic dynamics with hidden factors, such as population immunity. In this study, we developed a parsimonious mathematical model of disease dynamics in which temperature and relative humidity directly influence transmission rates, and used real weather data from four climates to simulate 10 years of weekly case counts across a range of scenarios representing different pathogens. In a numerical experiment, we then asked whether time-series regression techniques could recover the true weather effects from these simulated case counts. We find that these methods often produce biased, imprecise estimates, even under favorable experimental conditions with minimal noise. Our results suggest that regression analyses of case-count time series can provide a misleading picture of how weather affects infectious diseases. More broadly, our findings show that mechanistic models and system-specific simulation studies are powerful strategies for testing statistical methods in infectious disease epidemiology.
Related
- Hospital-treated infectious diseases, infection burden, and subsequent risk of demyelinating diseases: a prospective cohort study
- Digital Footprint and Audience Engagement Among Infectious Diseases Medical Educators on TikTok
- Reported Cases and Deaths of National Notifiable Infectious Diseases — China, July 2026*
- Benchmarking time series foundation models for multi-country infectious disease forecasting: a retrospective evaluation across over 70 diseases
- Current Advances in Imaging Diagnosis of Bacterial Infectious Diseases: From Structural to Nano Imaging Modalities
BibTeX
@article{gemo2026pitfalls,
title = {{The pitfalls of incidence-based time series regression for inferring the effects of weather on infectious diseases}},
author = {Pietro Gemo and Laura Andrea Barrero Guevara and Cana Kussmaul and Sarah C. Kramer and Matthieu Domenech de Cellès},
journal = {PLOS Computational Biology},
year = {2026},
month = sep,
doi = {10.1371/journal.pcbi.1014797},
url = {https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014797},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - The pitfalls of incidence-based time series regression for inferring the effects of weather on infectious diseases
AU - Pietro Gemo
AU - Laura Andrea Barrero Guevara
AU - Cana Kussmaul
AU - Sarah C. Kramer
AU - Matthieu Domenech de Cellès
PY - 2026
DA - 2026/09/10
JO - PLOS Computational Biology
DO - 10.1371/journal.pcbi.1014797
UR - https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014797
AB - A central question in environmental epidemiology is how the weather affects infectious diseases. Time-series regression (TSR) on population level case-counts data is widely used to estimate weather effects; however, this design may be biased due to the complexities of infectious disease dynamics, including nonlinear feedback, various types of noise, and latent, dynamic variables such as population immunity. Here, we assess the reliability of incidence-based TSR through a controlled simulation study across four different climates and fifty scenarios representing different pathogens. For each scenario, we simulated 10 years of weekly case-count data using a transmission model that included real-world weather data on temperature and relative humidity. We then examined whether the ground-truth weather effects could be recovered from simulated data using negative binomial generalized additive models, a flexible class of TSR models commonly used in empirical applications. We find that these models frequently fail to yield accurate and precise estimates of weather effects, even under favorable conditions such as no process noise and low observation noise (overdispersion). Hence, our results suggest that more causally elaborated formulations of TSR models, or more mechanistic approaches like transmission models, are needed for statistical inference of weather effects from population data. Author summary: Time-series regression on case-count data at the population level is a common technique for examining how weather affects infectious diseases. However, this analytical approach overlooks the fact that these data come from complex epidemic dynamics with hidden factors, such as population immunity. In this study, we developed a parsimonious mathematical model of disease dynamics in which temperature and relative humidity directly influence transmission rates, and used real weather data from four climates to simulate 10 years of weekly case counts across a range of scenarios representing different pathogens. In a numerical experiment, we then asked whether time-series regression techniques could recover the true weather effects from these simulated case counts. We find that these methods often produce biased, imprecise estimates, even under favorable experimental conditions with minimal noise. Our results suggest that regression analyses of case-count time series can provide a misleading picture of how weather affects infectious diseases. More broadly, our findings show that mechanistic models and system-specific simulation studies are powerful strategies for testing statistical methods in infectious disease epidemiology.
SN - 1553-7358
ER -