Improved socio-economic and health-care conditions over the past few decades in England and Wales may be a better explanation for the decreasing trend of excess winter mortality, rather than milder winters due to climate change.These results, reported this week in Nature Climate Change, show for the first time that winter temperature variations no longer predict changes in excess winter deaths (EWDs).

Past research has attributed EWDs - calculated from the number of deaths from December to March minus the average number of deaths in the preceding and following months - to the direct effect of very cold days, such as hypothermia, and to increased vulnerability to diseases, including influenza strains that are common in winter. Therefore it has been expected that winter warming due to climate change would decrease EWDs.

To determine if these causes are supported by available information, Philip Staddon and colleagues collated data from the past 60 years to identify the key factors - ranging from temperature changes to housing quality, heating costs and flu activity - associated with the decreasing trend in EWDs in England and Wales. The team confirm that EWDs have decreased over this period. However, they find that the association of yearly variation in EWDs with the number of cold days (below 5 °C), which was strong until the mid-1970s, has disappeared in recent decades. The annual variation in EWDs since the 1970s is associated only with the incidence of influenza-like illnesses.The authors conclude that climate change is unlikely to offer a winter health dividend.