Climatic and geographic temporal patterns of pain in the Multicenter Study of Hydroxyurea

No multi-site comparisons have tested whether seasonally cold temperature or climate exacerbate pain intensity in sickle cell disease (SCD). We examined seasonal SCD pain intensity and frequency patterns and compared them with concurrent climate conditions (temperature and barometric pressure) and geography of patient residence in the Multicenter Study of Hydroxyurea (MSH). We conducted a time series analysis of the monthly average daily pain intensity (0-9 scale) and pain frequency of the 299 MSH patients from December 1991 to December 1994. We used both an unobserved component model (UCM) and a nonparametric local regression (LOESS) to probe for a cycle and/or trend associated with the time series. We also examined base mixed regression models of season, monthly average temperature and barometric pressure, and geographic region as stand-alone predictors of pain intensity and frequency. Expanded models included additional predictor variables. UCM and LOESS analyses showed a cyclic pattern of pain intensity and frequency with peaks in late Fall/early Winter and troughs in Spring. Base regression models showed colder seasons were significantly associated with greater pain intensity (p = .0035) but not frequency (p = .07); higher monthly temperatures were significantly associated with both lower pain intensity and pain frequency, but higher monthly barometric pressures were significantly associated with greater pain intensity and frequency (all p's < .0001); and northern sites had nonsignificantly higher pain intensity (p = .40) and frequency (p = .07) than southern sites. This pattern of results did not change in expanded models including other predictors. Our results suggest that seasonably colder temperatures exacerbate sickle cell-related pain, but low barometric pressure does not, and geographic region of residence is not significantly related to pain in this sample.