Multiple rewarming methods have been recommended for the treatment of hypothermia in the ED. Because the hypothermic patient population is heterogenous, a method for stratifying mortality risk when comparing therapies is desired. We used univariable and multivariable statistical analyses to identify variables which discriminated between patient death or survival in the 24 h after arrival in the ED. Prehospital cardiac arrest, a low or absent presenting BP, elevated BUN, and the need for either tracheal intubation or NG tube placement in the ED were found to be significant predictors of patient demise in a large database (n = 428). The likelihood ratio was used to develop and validate an empiric hypothermia outcome score that can be used in future hypothermia treatment studies to account for differences of patient presentation.