ObjectivePrior studies have reported that higher body temperature correlates with greater depressive symptoms, but did not adjust for potential confounders.MethodsWe used linear regression models and linear mixed models to assess associations between self-reported depressive symptoms and body temperature (once-daily self-reported temperature from a thermometer and minute-level distal temperature from a commercially available wearable device, worn on the finger). Adjusted models covaried for biological sex, age, daylight length, ambient temperature, and number of medical conditions. Distal body temperature metrics included awake and asleep temperatures, the asleep–awake temperature difference, and the diurnal temperature amplitude. Distal body temperature models also covaried for daylight length and ambient temperature over time. Finally, we used adjusted logistic regression models to assess associations between all body temperature metrics and four categories of depressive symptom severity.ResultsIn an adjusted linear regression model, the positive association between depressive symptoms and self-reported body temperature remained significant (b = 0.940, 95% CI [0.686, 1.193], p = 1.7 × 10–12). Adjusted associations with awake distal body temperature (b = 0.300, 95% CI [0.229, 0.371], p = 1.2 × 10–15), diurnal distal body temperature amplitude (b = -0.204, 95% CI [-0.248, -0.159], p = 6.1 × 10–18), and asleep-awake distal body temperature difference (b = -0.308, 95% CI [-0.375, -0.240], p = 6.1 × 10–18) were stronger than in unadjusted models. Standardized effect estimates from the adjusted linear regression models indicated that the magnitude of these associations was robust but small in magnitude, with standardized coefficients ranging from -0.067 to 0.061 and partial R² values ranging from 0.0006 to 0.0041. Adjusted linear mixed models using repeated measurements showed similar significant associations between distal body temperature metrics and depressive symptom severity. Adjusted logistic regression models showed that increases in self-reported body temperature and awake distal body temperature conferred significantly higher odds of a more severe depressive symptom category; increases in diurnal distal body temperature amplitude and asleep-awake temperature difference conferred significantly lower odds.ConclusionHigher body temperature was associated with increased depressive symptoms, and this association was robust to adjustment for potential confounders. These findings support body temperature as a physiological correlate of depressive symptom severity and motivate further investigation of thermoregulatory measures as part of multimodal approaches for monitoring depressive symptoms.