After exclusion of ineligible subjects (10,028 with prevalent diabetes, 8,797 of other ethnic groups, 6,202 with missing covariates, 2,537 with missing dietary information, 812 subjects with unconfirmed diabetes, and 10 with missing information on follow-up or diabetes at baseline), 36,256 men and 39,256 women were part of this analysis.
Of the potential 28,128 person-examinations that included four repeat measures for each of the 7,032 eligible participants, we excluded 6,665 observations with unavailable glucose and insulin values or missing covariates, 5,919 observations with non-fasting sampling (<8 h) and 2,481 observations due to afternoon sampling.
To avoid casewise deletion of records with missing covariates, we employed multiple imputation with five imputation data sets via predictive mean matching.
These missing covariates were imputed with a multiple imputation model.
However, distributions of OAB symptom severity, TUG and depressive symptoms are similar between individuals with complete data and those with missing covariates.
For models with missing covariates, we tested three different methods (missing indicator, imputed with median, and assuming no change from before to after MI periods) with similar results, and therefore we present only the results using the missing indicator method.
The 498 persons excluded included 324 pregnant women, 1 person with unknown diabetes status, and 173 people with missing covariates.
The generalized semiparametric regression models for the cumulative incidence functions with missing covariates are investigated.
Patients with missing covariates were removed from the analysis.
