Abstract
BACKGROUND: Differences in multiple sclerosis (MS) risk by latitude have been observed worldwide; however, the exposures driving these associations are unknown. Ultraviolet radiation (UV) has been explored as a risk factor, and ambient temperature has been correlated with disease progression. However, no study has examined the impact of all three exposures. We examined the association between these exposures and incidence of MS within two nationwide prospective cohorts of women, the Nurses' Health Study (NHS) and Nurses' Health Study II (NHSII). METHODS: Both cohorts were followed with biennial questionnaires to ascertain new diagnoses and risk factors. Time-varying exposures to latitude, cumulative average July temperature (°C), and cumulative average July erythemal UV (mW/m(2)) were predicted at each participant's biennially updated residential addresses. Using Cox proportional hazards models adjusted for MS risk factors, we calculated hazard ratios (HR) and 95% confidence intervals (CIs) within each cohort and pooled via meta-analyses. RESULTS: In multivariable models, there were suggestions that decreasing latitude (meta-analysis multivariable-adjusted HR = 0.72; 95% CI 0.55, 0.94 for women living <35.73° compared with those ≥42.15°, P-for-trend = 0.007) and increasing cumulative average July temperature (meta-analysis multivariable-adjusted HR = 0.81; 95% CI 0.72, 0.91 for each interquartile range increase [3.91°]) were associated with decreasing risk of MS. There was no evidence of heterogeneity between cohorts. We did not observe consistent associations with cumulative average UV. CONCLUSION: Our results suggest that adult exposures to decreasing latitude and increasing temperature, but not UV, were associated with reduced MS risk in these two cohorts of women. Studies of MS incidence may want to consider temperature as a risk factor.