Background: Cold weather is associated with increases in mortality and demands on hospital services in the UK, particularly by the elderly. Less is known about the relationship with patterns of consultation in primary care. We wished to determine the magnitude and consistency of associations between cold temperature and consultations for respiratory conditions in primary care settings at different sites in the UK.
Methods: Time series analysis of any short-term effects of temperature on daily general practitioner (GP) consultations made by elderly people (65+ years) for lower and upper respiratory tract infections (LRTI, URTI) over a 10-year period, 1992--2001. Practices were situated in 16 urban locations across the UK where a Met Office monitoring station was in operation.
Results: An association between low temperatures and an increase in LRTI consultations was observed in all 16 locations studied. The biggest increase was estimated for the Norwich practices for which a 19.0% increase in LRTI consultations (95% CI 13.6, 24.7) was associated with every 1 degrees C drop in mean temperature below 5 degrees C observed 0-20 days before the day of consultation. Slightly weaker relationships were observed in the case of URTI consultations. A north/south gradient, with larger temperature effects in the north, was in evidence for both LRTI and URTI consultations.
Conclusions: An effect that was consistent and generally strongest in populations in the north was observed between cold temperature and respiratory consultations. Better understanding of the mechanisms by which cold weather is associated with increases in consultations for respiratory infections could lead to improved strategies for prevention and reduced burdens for health services.