A statistical method to evaluate dental classification systems used by military dental services.

UNLABELLED

The aim of this research project was to develop a systematic approach to evaluate the effectiveness of dental classifications used by military dental services to predict dental emergencies.

DESIGN

A retrospective cohort study of U.S. Navy recruits were followed for an average of 3.5 years. Dental records of 1,114 recruits who had their initial dental examination during the calendar year 1997 were collected in 2001. Four hundred sixty-two dental emergencies (DE) were recorded, of which 50 were class 3 patients. The incidence density ratio (IDR) of DE was 2.2 (p < 0.001; class 3 compared to class 1 or 2 patients). The mean times to DE in patients who experienced DE were 92 days (class 3 patients) and 418 days (class 1 and 2 patients; p < 0.001), while the median times to DE were 59 days and 295 days (p < 0.001) for class 3 and class 1 or 2, respectively. The adjusted hazard ratio for DE in class 3 versus class 1 and 2 patients was 1.519 (p = 0.0149). We have concluded that examination of the following statistics is necessary to assess the effectiveness of military dental classification systems: (1) the percentage of DE that occur in class 3 patients, (2) the IDR of DE, (3) the mean time to DE, (4) the median time to DE, and (5) the hazard ratio of DE.