Temperature-related mortality: a systematic review and investigation of effect modifiers

Abstract

Background: Understanding which populations are vulnerable and which factors affect vulnerability to temperature-mortality associations is important to reduce the health burden from current day weather extremes and climate change.

Objectives: We reviewed population-based studies on the impact of temperature on mortality and assessed the vulnerability to temperature-mortality associations systematically.

Methods: We identified 207 studies published between 1980 and 2017 and summarized findings on effect modification based on individual- and community-level characteristics.

Results: In our assessment of vulnerability to temperature-mortality associations, we found strong evidence of effect modification for several individual-level factors such as age and sex. We also found limited or suggestive evidence for other individual-level factors such as education, place of death, occupation, race, marital status, and chronic conditions. Evidence on effect modification by community-level characteristics for temperature-mortality associations is limited. We found weak evidence of effect modification for population density, heating system, healthcare facilities, proximity to water, housing quality, and air pollution level. We found limited or suggestive evidence for community-level socio-economic status, latitude, urban/rural, air conditioning, climatic condition, green space, and previous winter mortality.

Conclusions: Our findings provide scientific evidence on which populations could be targeted for establishing appropriate strategies to reduce the health burden from extreme temperatures, and for policies on climate change.

Keywords: effect modification; mortality; temperature; vulnerability.

Figure 1.

Flow diagram of literature selection process.

Figure 2.

Number of studies on the relationship between ambient temperature and mortality, by country. Note: In addition, there were 16 multi-country studies(US 3; China 3; Greece 5; Spain 8; Hungary 6; Finland 5; United Kingdom 10; France 5; Italy 11; Sweden 5; Germany 2; Ireland 4; Slovenia 3; Czech Republic 3; Switzerland 2; India 1; Mexico 1; Canada 2; Thailand 2; Australia 3; Brazil 4; Taiwan 5; South Korea 5; Japan 5; Netherlands 1; Chile 1; Vietnam 1; Colombia 1; Iran 1; Philippines 1; Moldova 1).

Figure 3.

Summary of study characteristics across 207 studies of ambient temperature and mortality. Note: ^a e.g. different study period by city/country, two separate periods; ^b e.g. risk of mortality during heat waves/cold spells days compared to normal days; ^c e.g. temperature deviation index, spatial synoptic classification, universal thermal climate index.