The effect of weather on unscheduled healthcare utilisation for mental health conditions in England, 2014–2022

Background: Weather conditions have been associated with adverse mental health outcomes, and concern about climate change has increased interest in these relationships. However, much of the existing literature focuses on extreme weather events such as heatwaves or acute clinical outcomes such as suicide. There is less research on the population-level variations in mental health–related healthcare utilisation across the full range of weather conditions.

Objective: To examine associations between daily weather conditions and mental health–related healthcare contacts in England using large-scale national surveillance data.

Methods: We conducted a retrospective observational study across nine English regions from 1 January 2014 to 31 December 2022. Outcomes were daily counts of unscheduled mental health–related contacts to emergency departments (EDs), general practice out-of-hours (GP OOH), and the NHS 111 telephone advice service. Weather exposures were mean daily temperature (°C), hours of full sunshine, and total daily rainfall (mm). Associations were estimated using distributed lag non-linear models at regional level and synthesised using two-stage multivariate meta-analysis, adjusting for seasonality, long-term trends, day of week, public holidays, and population size.

Results: Unscheduled mental health–related healthcare contacts showed modest but consistent associations with temperature and sunshine. Across services, relative risks were higher at increasing temperatures up to approximately 18 °C and on days with fewer hours of sunshine. Sunshine showed the most consistent pattern, with increased healthcare utilisation on days with low sunshine across all services. Rainfall was not consistently associated with healthcare contacts. Age-stratified analyses indicated a U-shaped association between temperature and ED attendances among adults aged over 64 years, with higher utilisation during both colder and warmer conditions. Overall variations in healthcare demand were modest, generally within ±10–20% of typical daily levels.

Conclusion: In England, short-term variations in temperature and sunshine are associated with changes in unscheduled mental health–related healthcare utilisation, while rainfall shows little consistent effect. Although effect sizes were modest, these findings have implications for understanding weather-related fluctuations in healthcare demand and for planning mental health services under current and future climate conditions.