A report from the Centers for Disease Control and Prevention (CDC) analyzing deaths from U.S. weather events between 2006 and 2010 finds cold weather was responsible for the majority of weather related fatalities. The conclusion challenges a widely-held view that heat is the top weather-related killer.  But some researchers argue – depending on how a weather-related death is defined – heat can still be considered more deadly.

What CDC found

CDC’s National Center for Health Statistics, examining death certificates from coroners or medical examiners, computed approximately 2,000 people died from weather-related causes each year.  Of those 2,000 deaths, the report  says 63 percent were cold-related, 31 percent heat-related, and just 6 percent due to floods, storms, or lightning.

For the temperature-related fatalities, the report found much higher death rates for older adults. “Among adults, heat-related and cold-related death rates increased with age, particularly for those aged 75 and over,” the report says.  This result is not all surprising as past studies have shown older adults – especially those who are socially isolated, sick and/or poor – are most vulnerable to severe heat and cold.

Income  also played an important role in the death rates.  The weather-related death rate more than doubled for household incomes under $36,000 versus incomes over $49,000.  “The finding that weather-related mortality increased as median county income quartiles decreased may indicate that lower-income areas have fewer resources to prepare for and adapt to extreme weather events,” the report says.

Men and non-Hispanic blacks were found to have higher temperature-related death rates than women and other race and ethnic groups. (But for floods, storms and lightning, non-Hispanic white death rates were almost twice as high as non-Hispanic black persons, the report says.)

The report’s regional analysis identified elevated death rates due to the cold in the rural West, while heat-related deaths rates were highest in the South and West – in the most urbanized but also most rural areas.

Is this analysis misleading?

CDC’s finding that cold is the top weather-related killer in the U.S. contrasts with some alternative analyses.  For example, the National Weather Service’s National Hazard Statistics – which examines different data – shows heat rather than cold was the leading weather-related killer between 2004-2013.

The National Weather Service calculates weather-related mortality by counting deaths associated with specific hazardous weather events.  CDC’s data – in addition to specific severe weather events – capture deaths that may have occurred during more routine weather when National Weather Service advisories may not have been in effect.   In addition to counting cases in which weather was the underlying cause of death, CDC also counted cases in which weather was a contributing cause.

In an email, University of Miami climatologist Larry Kalkstein, who has published numerous research papers on weather-related mortality – explained that the differences in weather-related death estimates stem from different definitions of what constitutes a weather-related death.  CDC’s analysis – which estimates more cold-related deaths than heat – is counting more “indirect” deaths, he suggests:

….a cold-related death might not be a direct death, while virtually all heat-related deaths are.  So if we try to compare apples to apples….i.e. direct cold and heat-related deaths, we have found that heat-related deaths exceed cold-related deaths.
The problem is defining a cold-related death.  We know that, in winter, general mortality is often 10-15 percent higher than in summer during typical daily periods.  Are these considered cold-related deaths?  They are happening during the cold season, but are not directly related to cold.  If you use this definition of cold-related deaths, then yes, cold would exceed heat-related.
So in summary, it is all dependent upon definition, in my opinion.  Comparing apples to apples, which would be to evaluate acute or short-term responses to weather, I would always give the nod to heat-related deaths.  However, if you are considering the seasonal differences in daily mortality, rather than just the “spikes” that we find with acute deaths, I can see why one can argue that winter (or cold-related) mortality is greater.

A 2005 study in the Bulletin of the American Meteorological Society advised against leaping to conclusions about whether heat or cold is more deadly, unless one is careful to specify how estimates were made to reach a particular conclusion:

Depending on the compilation nature of the dataset, the numbers of heat- or cold-related mortality are quite divergent. Consequently, in general, these separate mortality datasets should not be combined or compared in policy determination, and the specific dataset used in a given study should be clearly identified. All of the datasets suffer from some major limitations, such as the potential incompleteness of source information, long compilation time, limited quality control, and subjective determination of the direct versus indirect cause of death. These factors must be considered if the data are used in policy determination or resource allocation.