Mild winter temperatures offer surprising mental health benefits

As winters warm, Swiss researchers reveal how higher temperatures can lift mood, lower stress, and support well-being, offering new insights into climate change’s psychological effects.

Study: Cold season air temperature as predictor of psychological well-being and mental health. Image Credit: NStafeeva / Shutterstock

In a recent study published in the journal Scientific Reports, researchers investigated the effects of warm ambient temperatures on mental health and psychological well-being during the cold season.

According to the World Meteorological Organization, 2023 was the warmest on record. Higher temperatures have manifold consequences and can occur directly (e.g., acute weather events like floods or wildfires) and indirectly (e.g., agricultural losses and migration). Besides, high temperatures are reportedly a risk factor for vector-borne, waterborne, nutrition-related, respiratory, and cardiovascular diseases.

In addition, various studies have observed associations between elevated ambient temperatures and higher suicides and mental health-related hospitalizations. However, only a few studies have investigated direct mental health measures. One study reported associations between temperatures above 30 °C and more mental health difficulties than at 10–15 °C. Another study noted that higher temperatures were associated with more days of poor mental health.

Notably, these findings contrast with reports of positive psychological effects of heat and warmth. A study found that countries with warmer temperatures in the coldest month were happier than those with colder temperatures. Another study observed that brief exposure to indoor temperatures of 22 °C to 24 °C was followed by more relational focus than cooler temperatures. This raises questions about whether high ambient temperatures have positive psychological effects in cooler conditions.

About the study

The present study investigated the associations between ambient temperature in cold seasons and mental health and psychological well-being. Participants were recruited from the Swiss Household Panel, an annual survey conducted since 1999. The researchers used data from the survey wave 21 from 2019 and obtained relevant meteorological data from MeteoSwiss, the Federal Office of Meteorology and Climatology.

Survey items assessed stress, loneliness, mood, life satisfaction, self-reported health, optimism, depression, blues, anxiety, strength, and energy. Multiple polynomial regression models were created for each dependent variable, integrating meteorological variables at three time points: the survey day, the average of the seven days, and the average of the 14 days.

Meteorological data for the 30 days before the survey were also included for the stress variable. Various covariates, including lifestyle and sociodemographic characteristics such as age, body mass index (BMI), and gender, were incorporated into models to control for confounding. Besides, bootstrapping was applied to control for multiple testing.

Findings

Overall, 13,151 individuals participated in the survey, and 8,017 were included for analyses. The study population was 53.2% female and 46.8% male, with a mean age and BMI of 50.9 years and 24.68 kg/m², respectively. The meteorological data encompassed daily records from September 2, 2019, to March 3, 2020, during which the temperature ranged from -6.9 °C to 22.6 °C, averaging 7.5 °C.

The mean relative humidity was 81.4%, ranging between 26.1% and 100%. On average, there were 3.6 hours of sunshine per day, with a maximum of 12.6 hours and a minimum of 0 hours. Precipitation ranged between 0 and 86.4 mm per day, averaging 2.96 mm per day. The regression models built to predict stress revealed that higher mean temperatures over the past 14 and 30 days were associated with reduced stress levels, which was confirmed by bootstrapping.

The models for predicting loneliness showed that intermediate mean temperatures, with the paper's discussion suggesting protective effects for outdoor temperatures above 0 °C and up to 20 °C, over the past 14 days, were associated with less loneliness, which bootstrapping confirmed. Notably, models built to predict sadness, anger, joy, and worry, and those for strength, optimism, and energy were not significant. Similarly, models for predicting anxiety, blues, and depression were not significant. The paper suggests that these null findings, for instance, regarding affect and optimism, might be partly because these aspects were measured as more stable, trait-like characteristics rather than immediate feelings related to recent temperatures.

The models to predict self-reported health indicated that intermediate mean temperatures, with the paper's discussion suggesting beneficial effects for temperatures above 0 °C and up to 20 °C, on the day of the survey and over the past 14 days, were associated with better health. Moreover, higher average temperatures over the past 7 days were also associated with better self-reported health. Finally, models on life satisfaction revealed that intermediate mean temperatures, which the paper's discussion indicated were between 0 °C and 10 °C during the cold season, over the past 7 and 14 days were associated with greater life satisfaction.

Conclusions

The findings illustrate that higher ambient temperatures were associated with lower stress, and intermediate temperatures were associated with better life satisfaction, self-reported health, and less loneliness. Temperature was not associated with optimism, strength, energy, or affect.

The paper also speculates on several potential mechanisms for these observed positive effects, including the influence of skin temperature changes on μ-opioid receptors (linked to social connection and thermoregulation), the deactivation of brain regions associated with unpleasant sensations by warm stimuli, and the impact of temperature on cognitive function, energy expenditure, sleep quality, and engagement in leisure activities.

The study also acknowledges certain limitations, such as the use of single-item measures for some psychological variables and the operationalization of some constructs. Although the identified effects were described in the paper as small, they are relevant at the population level and shed more nuanced light on the impact of climate change on mental health.