Excess winter deaths are higher in milder-winter countries: what the data show

A fair guess is that people in warm countries get sick less in winter, because there is less winter to get sick in. The data on deaths point the other way. Excess winter deaths, the surplus of deaths in the cold months over the rest of the year, are higher in several southern European countries than in the colder north. The studies that look for the reason keep landing on the same place: how cold it is inside the home.

This article goes through six studies on that. It is informational material, not medical advice. If symptoms are severe, getting worse, or not going away, that is a question for a doctor.

Do people in warm countries get sick less?

The data here are about deaths, not about how often people fall ill. None of the sources below counts colds, flu or visits to a doctor, so the question "do people in warm countries get sick more often?" cannot be answered from them. What they measure is the winter surplus of deaths from all causes. That is a much coarser number than illness, and it is the one with national statistics behind it.

For the infections side of winter, see why we get sick more in winter. For the effect of day-to-day swings, see sudden temperature changes. This article is about something else: the gap between countries, and the home.

Excess winter deaths: where the surplus is largest

A 2003 analysis compared 14 EU countries over 1988 to 1997, using deaths from all causes. Excess winter mortality was highest in Portugal at 28% (CI 25% to 31%), followed by Spain at 21% (19% to 23%) and Ireland at 21% (18% to 24%) (Healy, 2003).

A later study covered 31 European countries across the seasons 2002/03 to 2010/11. It used an Excess Winter Deaths Index, the ratio of deaths in December to March to deaths in the other months. The index rose from northern to southern Europe and with mean winter temperature (r² = 0.50). Malta, Portugal, Spain, Cyprus and Belgium were each significantly higher than the average of the other 30 countries. There was no clear link with how much a country's winter temperature varies from year to year (Fowler et al., 2015).

Two things to keep in view. First, both are comparisons of countries, not of people. They show that countries with warmer winters have a larger winter surplus, not that a warmer winter does harm to anyone. Second, the Fowler authors do not claim to know why. They write that many of these deaths may be avoidable, and that the causes of the differences between countries need to be understood better.

The cold that counts is indoors

Healy tested a long list of candidates against the winter surplus. Mean winter temperature, mean winter relative humidity, national income, health spending per person, fuel poverty, inequality, deprivation and several measures of how well homes keep heat in were all significantly related at the 5% level. Income and poverty measures were significant too; this article does not interpret their direction, which an abstract alone does not let us do.

The author's reading: the strong positive link with outdoor temperature and the strong negative link with thermal efficiency suggest that housing standards in southern and western Europe play a large part in that seasonality. The paper's proposed remedy was better protection from the cold indoors, more public spending on health care and better socioeconomic conditions. That is the author's conclusion from a cross-country comparison, and it is quoted here as such.

Britain gives a closer look, at the level of electoral wards. A study of people aged 65 and over in 1986 to 1996 counted about 30,000 excess winter deaths a year. The odds of dying in winter were 1.5% higher for every 1 °C drop in daily mean winter temperature, and the excess grew with age. Lack of central heating was linked to a higher risk of dying in winter, with an odds ratio of 1.016 (95% CI 1.009 to 1.022). That is a small association. For deaths from all causes, the link with deprivation was weak, and the authors said more work was needed to untangle housing quality from deprivation (Aylin et al., 2001).

A third British study looked at money for heating. Winter fuel payments go to people aged 65 and over. A time-series model found a structural break in the excess winter mortality series for England and Wales in 1999/2000, and attributed almost half of the reduction since then to these payments (Iparraguirre, 2015). It is a statistical model fitted to a trend over time, not an experiment, and the author's affiliation is Age UK. Nobody assigned the payments to some households and withheld them from others.

Mild cold, not frost

The Lancet study took a different cut. It pooled 74,225,200 deaths from 384 locations in 13 countries (Australia, Brazil, Canada, China, Italy, Japan, South Korea, Spain, Sweden, Taiwan, Thailand, the UK and the USA) between 1985 and 2012. Non-optimum temperature, meaning any day above or below the temperature of lowest mortality, accounted for 7.71% of deaths in the sample. By country the share ran from 3.37% in Thailand to 11.00% in China (Gasparrini et al., 2015).

Cold accounted for 7.29% of deaths and heat for 0.42%. Extreme cold and extreme heat together accounted for 0.86%. The authors wrote that the effect of extreme days was substantially smaller than that of milder but non-optimum weather. Most of the burden came from ordinary, moderately cool days.

This study does not measure homes, heating or excess winter deaths as such, and nothing in it explains why one country's share is higher than another's. The countries are named here only because they are in the sample. It does fit the picture above in one respect: the harm is not concentrated in the coldest days of the year.

Not only the body

One more finding sits apart from mortality. A study in the UK Household Longitudinal Study followed households over time and found that moving into a home that is not suitably warm was associated with nearly double the odds of severe mental distress for people who had none at the start, and more than three times the odds for people who had been on the borderline. The authors controlled for mental distress and housing temperature at entry to the survey, to account for the link running in both directions. They also noted that cold homes are a significant problem in the UK despite a comparatively mild climate (Clair & Baker, 2022). It is an observational study. It shows an association within one country.

What nobody knows yet

• Whether warmer homes would lower the surplus in a given country. All six sources are observational. Healy and Fowler compare countries, Aylin compares wards, Iparraguirre fits a trend. None of them heated a single home and watched what happened.
• What explains the gap between countries. Fowler's authors say outright that the causes are not well understood. Healy points to housing standards, but a country-level correlation cannot separate housing from everything else that differs between southern and northern Europe.
• Whether any of this tells you about your own risk. The data are about populations and about deaths, and most of the British work concerns people aged 65 and over.
• Whether the same pattern holds outside Europe and Britain. The Lancet study covers 13 countries but says nothing about housing, and the studies of excess winter deaths here are European. This article has no data on homes or heating in Japan, China or any other country in that sample.
• What is a "cold home" in numbers. The abstracts use proxies such as lack of central heating and indicators of thermal standards, and the UK study uses "not suitably warm". They do not give a common temperature.

What you can look at in your own data

The app records local outdoor weather (pressure, temperature, relative humidity, wind speed, UV index) next to your own symptom entries and Apple Health data, and looks for correlations in your own history with Pearson correlation, ANOVA and false-discovery-rate correction. Its nine fixed symptoms are headache, fatigue, drowsiness, dizziness, anxiety, irritability, nausea, joint pain and insomnia. It does not diagnose or predict illness.

The limit that matters here is that the app uses outdoor weather. It does not measure or know the temperature in your room, which is the thing this article is about. If you feel worse on cold days, the app can show whether that lines up with outdoor temperature in your entries. It cannot see how cold your home was. Why a pattern like that needs care before it is called a cause is covered in weather sensitivity: what the science says.