# Air temperature and all-cause emergency hospital admissions in people with and without diabetes in Germany (2005–2022): a time-series analysis

## TL;DR

This study found that extreme temperatures increase emergency hospital admissions in Germany, with similar effects in people with and without diabetes.

## Contribution

The study provides new evidence on the comparable vulnerability of people with and without diabetes to temperature extremes in Germany.

## Key findings

- Both extreme heat and cold increased emergency hospital admissions in all groups.
- Relative risks for heat and cold were similar between people with and without diabetes.
- Type 1 diabetes showed higher vulnerability in some age and sex subgroups.

## Abstract

People with diabetes may be more vulnerable to temperature extremes due to impaired thermoregulation and higher prevalence of comorbidities, but evidence is limited. We aimed to compare short-term effects of extreme heat and cold on all-cause emergency hospital admissions in Germany among people with and without diabetes.

We applied space- and time-stratified conditional quasi-Poisson regression with distributed lag non-linear models (up to 21 days) to estimate short-term effects of daily average temperature. We chose the reference temperature (20 °C) such that it approximates the minimum morbidity temperature in most subgroups. Analyses were stratified by sex, age, and diabetes status using data from all emergency hospital admissions in Germany, 2005–2022 (N = 132,243,083) at the level of 400 administrative districts, enabling an ecological study.

Both heat and cold increased hospital admissions. Heat-related relative risks (RR) were broadly similar between people with and without diabetes. Considering all ages, heat-related RRs (95% confidence interval) were 1.03 (1.03–1.04), 1.07 (1.00–1.13), and 1.02 (1.01–1.03) in males without, with type 1, and with type 2 diabetes. Age-specific RRs for heat and cold were similar between people without and with type 2 diabetes but higher for type 1 diabetes in some subgroups; e.g. cold-related RRs were 1.13 (1.12–1.15) and 1.51 (1.14–2.01) in men aged ≥80 years without and with type 1 diabetes.

Contrary to prior hypotheses, diabetes was not associated with greater vulnerability. This may reflect good healthcare access and increased awareness of heat and cold-related risks among people with diabetes. Nevertheless, given the higher baseline risk of hospital admission in diabetes, similar RR may still translate into larger absolute effects of extreme temperatures. Hence, clinical practice and policies aimed at mitigating temperature-related effects should continue to consider diabetes as a potential vulnerability factor.

None.