# Differentiating the impact of ambient temperature on hospitalization due to cause-specific pneumonias: an individual-level, case-crossover study

## TL;DR

This study examines how extreme heat affects hospitalizations for different types of pneumonia in Shanghai, finding that high temperatures significantly increase the risk, especially for infectious pneumonia.

## Contribution

The study provides new insights into the differential impacts of ambient temperature on hospitalization for cause-specific pneumonias.

## Key findings

- High temperatures significantly increase hospitalization risks for infectious pneumonia, with a relative risk of 2.55.
- The hot-related risk appears within 0–1 day and is more pronounced in females.
- Effect estimates remain robust after adjusting for air pollutants and model parameters.

## Abstract

Associations between exposure to ambient temperature and the risks of hospital admission for overall or a specific type of pneumonia were reported in prior studies. However, few studies have systematically explored the potentially differential impacts of ambient temperature on hospitalization for cause-specific pneumonias.

We performed a time-stratified, case-crossover study based on individual-level pneumonia-related hospital data with pathogen identification results in Shanghai, the largest metropolis in China from 2013 to 2019. Conditional logistic regression combined with distributed lag non-linear model was used to estimate effects of extreme temperature.

We included a total of 6,277 hospitalized cases of pneumonias with various causes. The observed exposure-response curve for the association between temperature and pneumonias was nonlinear, with significantly elevated risks in high temperature. The hot-related risk appeared on lag 0–1 day, and became non-significant on lag 5–10 day, with differential lag patterns in various subtypes of pneumonias. Extreme high temperature exhibited the largest effect estimate in hospitalization from infectious pneumonia [relative risk: 2.55, 95% confidence interval (CI): 1.58, 4.13], followed by bacterial pneumonia (2.16, 95% CI, 1.15, 4.04), and total pneumonia (1.96, 95% CI, 1.40, 2.74). Stronger relationships were observed in females in stratified analyses. The effect estimates remain robust after adjusting for air pollutants and various model parameters.

As pneumonias cause a huge disease burden worldwide, our study adds heat-related risk assessments on hospitalization due to cause-specific pneumonias, which is beneficial for prevention and management of patients with specific subtypes of pneumonia in urban areas.