Sudden temperature changes: what they may do to your nose, heart and head

You step out of a warm office into a cold evening, or a mild week turns hot overnight, and your nose starts to run or your head feels off. Sudden temperature changes get blamed for a lot, and some of that blame has data behind it. Some of it points in a direction people don't expect.

This article goes through what studies have measured: the nose, the heart and mortality statistics, colds, and headaches. It is informational material, not medical advice. If symptoms are severe, getting worse, or not going away, that is a question for a doctor.

What sudden temperature changes are

Researchers use two different measures, and mixing them up causes most of the confusion.

• Diurnal temperature range (DTR): the temperature difference between day and night within the same day.
• Temperature change between neighboring days (TCN): today's temperature minus yesterday's. Positive means it got warmer, negative means it got colder.

A day can have a large DTR and a small TCN, and the other way round. Several studies below find that the two measures behave differently.

A runny nose after a temperature change is usually not an allergy

Many people call this "an allergy to temperature change". A 2024 review of nonallergic rhinitis (the nose reacting without an allergic reaction) says the most common form of chronic nonallergic rhinitis is vasomotor rhinitis, about 80% of patients. Expert panels have recommended replacing that name with "nonallergic rhinopathy". The review describes symptoms triggered by chemical irritants and weather changes, through temperature, chemical, mechanical and osmotic sensors that act through TRP channels (Baroody et al., 2024).

A 2016 review defines nonallergic rhinitis as nasal symptoms without evidence of IgE-mediated sensitization to relevant airborne allergens. Its largest subgroup is idiopathic: nasal hyperreactivity to nonspecific triggers, including temperature, humidity and chemical exposures, often without inflammation of the lining (Shusterman, 2016).

The two can overlap. People with allergic rhinitis also often get symptoms after irritants and changes in temperature and humidity, just like people with nonallergic rhinitis, according to a 2011 review (an English abstract of a paper in Croatian). Itch and eye symptoms are more common in allergic rhinitis. Allergy is confirmed by skin prick tests and specific IgE in the blood (Kalogjera, 2011). The 2024 review also describes mixed rhinitis, where both mechanisms are present, and lists sinusitis, headaches, asthma, chronic cough and sleep disturbances as common companions.

Why bother telling them apart? The 2024 review puts it directly: getting the type right matters because inappropriate therapies lead to poor outcomes. A 2019 review notes that skin testing helps separate allergic from nonallergic rhinitis (Agnihotri & McGrath, 2019). Sorting out which one you have is a job for a doctor, based on your symptom history and allergy tests, not for a weather article.

Sudden temperature changes and deaths: it depends on the direction

The largest body of data here is population statistics: how many people died on days with a given temperature pattern.

A systematic review of 25 studies on DTR (11 on mortality, 14 on illness) found that most reported a significant link with mortality and illness, especially cardiovascular and respiratory. Elderly people and children were more vulnerable than adults. The authors also list inconsistencies about who is vulnerable, how long the effect lags and what DTR threshold matters, and say the effect may be confounded or modified by season, socioeconomic status and education (Cheng et al., 2014).

A study of 308 cities in 10 countries over 1972 to 2013 attributed 2.5% of total mortality to DTR (95% interval 2.3 to 2.7%). Across all countries the share rose from 2.4% to 2.7% between the first and last study years, with a significant rise in the USA, the UK, Spain and South Korea (Lee et al., 2018). Another, covering 372 locations in 12 countries and regions, measured temperature variability as the standard deviation of minimum and maximum temperatures over the exposure days. It found a significant link with mortality in every country, even after accounting for the average temperature, in cold, hot and moderate seasons alike. Over short exposures (0 to 1 days) the risk was higher in hot areas; over longer ones (0 to 7 days) it was higher in moderate areas (Guo et al., 2016).

Now the twist. Intuition says a sudden cold snap is the dangerous part. Two studies that looked at the day-to-day change found the opposite for mortality.

• In 106 US communities (1987 to 2000), a temperature drop from the previous day was associated with lower mortality, and a rise with higher mortality. For lags up to 21 days, the relative risk was 0.63 (0.59 to 0.68) at an extreme drop (1st percentile) and 1.46 (1.39 to 1.54) at an extreme rise (99th percentile). People aged 75 and older and people with respiratory disease, especially those who died of pneumonia, were the most vulnerable. The link depended on season and region (Zhan et al., 2017).
• In one subtropical Chinese city (2008 to 2012), a rise in temperature from the previous day was linked with more deaths from non-accidental and cardiovascular causes, and a fall with a protective effect. Men and people aged 65 and older appeared more vulnerable (Cheng et al., 2014).

Children's data adds another piece. In Brisbane (2001 to 2010), the change between neighboring days was associated with emergency visits for childhood pneumonia, while DTR was not. Children aged 5 to 14, girls and Indigenous children were more vulnerable (Xu et al., 2014). The abstract does not say whether the rise or the fall carried the risk, so this article won't guess.

Put together, these results are not a contradiction. Different outcomes, different places and different measures can behave differently. Every study here is an observational time series, which finds links, not causes, and describes populations, not you. "A sharp warm-up is worse than a cold snap" is not a rule these papers establish for everyone.

Colds and headaches, briefly

Colds. A study of Finnish conscripts with 147 PCR-confirmed rhinovirus cases found that each 1 °C drop in temperature was associated with about 8% higher risk. In the authors' words, "a decrease rather than low temperature and humidity per se during the preceding few days increases the risk" (Ikäheimo et al., 2016). So for infection, the fall is the signal there. The full picture, including the limits of that cohort, is in does getting cold give you a cold.

Headaches. One result is about the level of temperature, not the change. In 7,054 patients at a single emergency department (May 2000 to December 2007), higher mean temperature over the previous 24 hours increased the risk of headache, with an odds ratio of 1.075 per 5 °C (1.021 to 1.133). Lower pressure 48 to 72 hours before raised the risk of non-migraine headache, and air pollutants did not change the risk (Mukamal et al., 2009). A diary study of 238 migraine patients in Vienna over 90 days found that a day-to-day change in minimum temperature decreased the risk that a migraine would persist, but after correction for multiple testing none of the findings remained significant. Subjective weather perception did not match attacks, and the authors called the influence of weather on migraine and headache "small and questionable" (Zebenholzer et al., 2011). For the pressure side of the story, see barometric pressure and headaches.

What nobody knows yet

• Whether any of this applies to you. The mortality studies describe populations over years. They cannot say that a sharp temperature swing will make a given person ill.
• Why the direction differs. The papers report that a rise and a fall behave differently across outcomes; their abstracts do not settle why.
• Which measure matters. DTR, change from yesterday and variability over several days are different numbers. The studies disagree on thresholds and on how long the effect lags.
• Whether a runny nose after a temperature change is your allergy or not. The studies above can't tell you.
• Whether warming up, cooling down or dressing differently changes anything. No source here tested it.

What you can look at in your own data

The app has a notification called Critical Weather Alerts, which you can switch on or off in settings and which is on by default. It fires when the outdoor temperature differs by more than 10 °C from a reading taken 2 to 4 hours earlier, and when pressure falls by 8 hPa or more. The same alert comes at most once every 6 hours. Outdoor weather comes from OpenWeatherMap, and the app does not measure indoor air.

Among the factors the app compares with your own ratings of how you feel and your energy is the temperature change over the day: the difference between the last and the first reading of that day. Headache and fatigue are among the nine fixed symptoms you can log (headache, fatigue, drowsiness, dizziness, anxiety, irritability, nausea, joint pain, insomnia). A runny nose, sneezing and cough are not on the list. The app does not diagnose, recognize or predict illness, and it cannot tell a vasomotor reaction from an allergy.

What it can show is whether, in your own entries, headaches or tiredness cluster on days with big swings. That is a pattern about you and the weather. Weather sensitivity: what the science says explains why such a pattern needs care before you call it a cause.