JMIR Publications

ABSTRACT

Many tools have been developed to predict the risk of diabetes for non-diabetic population; however, these tools have shortcomings that include the omission of race, inclusion of variables that are not readily available to patients, low sensitivity or specificity. We collected the data from the NAfld in the Gifu Area Longitudinal Analysis (NGALA) database. The Lasso model was used to select potentially relevant features. Multiple Cox proportional hazard analysis was used to develop a model based on the training set. The final study population of 15464 participants had a mean (range) age of 42 (18-79) years; 54.5% were males. The mean (SD) follow-up duration was 6.05 years (3.78 years). 373 (2.41%) participants were progressed to diabetes during the follow-up period. Then, we established a novel parameter, FHTHWA index, to evaluate the incidence of diabetes for non-diabetic population, which was comprised by six parameters based on the training set. After multivariable adjustment, individuals in Tertile 3 had a significantly higher rate of diabetes compared with those in Tertile 1 cohort (hazard ratio, 32.141; 95% CI, 11.545-89.476). TimeROC analyses showed that FHTHWA index had high accuracy, with the AUC value greater than 0.9 during the more than 12 years followed-up. This study established a diabetes risk assessment index based on a large sample and many indices, and classified the diabetes risk of the population, thus providing a new forecast tool for the screening of patients and providing information on diabetes prevention for healthy populations.