Relationship between weather and seasonal factors and trauma admission volume at a Level I trauma center

Background: We sought to determine whether trauma patient admission volume to our Level I trauma center was correlated with observable weather or seas onal phenomena. Methods: Trauma registry data and national weather service data for the per iod between September 1, 1992, and August 31, 1998, were combined into a co mmon data set containing trauma admission data and weather data for each da y. SequentiaI linear regression models were constructed to determine relati onships between variables in the data set. Results: There is a highly significant relationship (p < 0.00001) between m aximum daily temperature and trauma admissions (R = 0,22). Rain is associat ed with a decrease in overall trauma volume. Rain had no effect on the numb er of admissions because of motor vehicle crash, however. Neither humidity nor snowfall affects trauma admission volume. Trauma admissions are signifi cantly more frequent in July and August, and on Saturdays and Sundays (p < 0.05), Linear regression analysis identified maximum temperature, precipita tion, day of week, and month as independent predictors of trauma admission volume (p < 0.001, R = 0.328). Conclusion: There is a significant relationship between weather and trauma center activity; temperature and precipitation are independently associated with trauma admission volume at our institution. Statistical models of tra uma incidence should consider these phenomena. Evaluation of a larger, popu lation-based data set is needed to confirm these relationships.