LAUSR

Identifying a population at risk of risks is imperative for primary disease prevention and allows the promotion of health maintenance in a healthy population. Previous studies on hypertension and dyslipidemia using data envelopment analysis (DEA) showed that populations at risk of risks could be identified using this method. In this study, we extended DEA to include pre-diabetes. A retrospective cohort study was conducted using specific health check-up data from 2008 to 2013. DEA efficiency scores were calculated for healthy subjects with baseline glycated hemoglobin (HbA1c) <5.7%. Odds ratios (ORs) for pre-diabetes onset within 3 years were analyzed. Among 1,501 subjects, with 373 cases of disease onset (24.9%), the OR for the incidence of pre-diabetes (on the basis of a 0.1-point increase in the efficiency score) was 0.77 (90% confidence interval [CI] 0.68–0.86, p < 0.0002). After adjusting for age and sex, the OR was 0.66 (90% CI 0.58–0.75, p < 0.0001). Furthermore, for the subgroup with no conventional diabetes risk factors, the adjusted OR was 0.50 (90% CI 0.38–0.67, p = 0.0001). We showed that the DEA efficiency score can help identify a pre-diabetes population at risk of risks. Further studies to validate these findings would be worthwhile for optimizing primary preventative measures. DEA was assessed for its ability to evaluate the risk of pre-diabetes. Results showed that an efficiency score could predict pre-diabetes onset and demonstrated the feasibility of applying this analysis in primary preventive healthcare.