May 21, 2015

expert reaction to study and linked comment piece investigating deaths from cold weather and hot weather

Both hot and cold weather can contribute to premature deaths in different contexts, and scientists publishing in the Lancet journal have attempted to determine the relative contribution of each temperature type. The researchers report that more deaths were caused by cold rather than hot weather, and that extremes had less of an impact than “milder but non-optimal” weather.

Prof. Sir David Spiegelhalter, Winton Professor of the Public Understanding of Risk, University of Cambridge, said:

“This is a fascinating study, which estimates that 8.5% of deaths in the UK are ‘due to cold’.  But it’s important to note that by ‘cold’, they mean any temperature less than the safest in the UK, which they estimate to be 19 degrees Centigrade – the risk only really rises substantially at less than 1 degree.  So perhaps they are really saying that the UK climate is killing people!”

Prof. Philip Staddon, Associate Professor in Environmental Science, Xi’an Jiaotong – Liverpool University, China, said:

“The research by Gasparrini et al. is certainly interesting, however I wonder on the validity of using a single temperature (different for each site) as the optimum temperature. I think an alternative analysis of the data would show an optimal temperature range (different at each site). Because a single temperature is used, which for example for London, UK is 19 °C, it is completely unsurprising that most deaths are cold related rather than warm related – very few days will be above this value with most days below this value.

“Nonetheless the conclusion that cold related deaths are substantially more numerous than heat related is important and confirms previous work. In their linked comment, Dear and Wang note that the supposed lack of importance of extreme cold temperature events in accounting for mortality implies that climate change will not decrease winter mortality in temperate countries such as the UK, a conclusion also reached recently by several independent research groups (e.g. Staddon et al. 2014).”

*Article: ‘Mortality risk attributable to high and low ambient temperature: a multicountry observational study’ by Antonio Gasparrini et al. published in the Lancet on Thursday 21 May 2015. 

^ǂ Comment: ‘Climate and health: mortality attributable to heat and cold’ by Keith Dear and Zhan Wang published in the Lancet on Thursday 21 May 2015.

Declared interests

Prof. David Spiegelhalter: “No conflicts of interest.”

Prof. Philip Staddon: “I declare that I have no conflict of interest in providing this quote. I am employed by Xi’an Jiaotong – Liverpool University as Associate Professor in Environmental Science and I am a Honorary Senior Research Fellow at the University of Exeter Medical School.”