Cold ambient temperature in utero and birth outcomes in Uppsala, Sweden, 1915-1929

Although the literature reports adverse birth outcomes following ambient heat, less work focuses on cold. We, moreover, know of no studies of cold that examine stillbirth. We tested the relation between cold ambient temperature during pregnancy in Sweden and four outcomes: stillbirth, preterm, birth weight for gestational age, and birth length. We examined births from 1915 to 1929 in Uppsala, Sweden, which-unlike most societies today-experienced substandard indoor-heating and fewer amenities to provide shelter from cold.

We retrieved data on almost 14,000 deliveries from the Uppsala Birth Cohort Study. We linked a validated, daily ambient temperature series to all pregnancies and applied Cox proportional hazards (stillbirth and preterm) and linear regression models (birth weight and length). We tested for nonlinearity using quadratic splines.

The risk of stillbirth rose as ambient temperature during pregnancy fell (hazard ratio for a 1°C decrease in temperature, 1.08; 95% confidence interval, 1.00 to 1.17). Cold extremes adversely affected preterm and birth length, whereas warm extremes increased preterm risk. We observed no relation between cold and birth weight for gestational age.

In historical Sweden, cold temperatures during pregnancy increased stillbirth and preterm risk and reduced birth length among live births.