Temperature-related mortality and associated vulnerabilities: evidence from Scotland using extended time-series datasets

BACKGROUND: Adverse health impacts have been found under extreme temperatures in many parts of the world. The majority of such research to date for the UK has been conducted on populations in England, whilst the impacts of ambient temperature on health outcomes in Scottish populations remain largely unknown. METHODS: This study uses time-series regression analysis with distributed lag non-linear models to characterise acute relationships between daily mean ambient temperature and mortality in Scotland including the four largest cities (Aberdeen, Dundee, Edinburgh and Glasgow) and three regions during 1974–2018. Increases in mortality risk under extreme cold and heat in individual cities and regions were aggregated using multivariate meta-analysis. Cold results are summarised by comparing the relative risk (RR) of death at the 1(st) percentile of localised temperature distributions compared to the 10(th) percentile, and heat effects as the RR at the 99(th) compared to the 90(th) percentile. RESULTS: Adverse cold effects were observed in all cities and regions, and heat effects were apparent in all cities and regions. Aggregate all-cause mortality risk in Scotland was estimated to increase by 9% (95% confidence interval, CI: 8%, 11%) under extreme cold and 4% (CI: 3%, 5%) under extreme heat. The elderly had the highest RR under both extreme cold and heat. Males experienced greater cold effects than females, whereas the reverse was true with heat effects, particularly among the elderly. Those who were unmarried had higher RR than those married under extreme heat, and the effect remained after controlling for age. The younger population living in the most deprived areas experienced higher cold and heat effects than in less deprived areas. Deaths from respiratory diseases were most sensitive to both cold and heat exposures, although mortality risk for cardiovascular diseases was also heightened, particularly in the elderly. Cold effects were lower in the most recent 15 years, which may be linked to policies and actions in preventing the vulnerable population from cold impacts. No temporal trend was found with the heat effect. CONCLUSIONS: This study assesses mortality risk associated with extreme temperatures in Scotland and identifies those groups who would benefit most from targeted actions to reduce cold- and heat-related mortalities. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12940-022-00912-5.