Two-stage community-based screening model for estimating prevalence of diabetic polyneuropathy (KCIS no. 6)

The aim of this study was to apply a calibrated two-stage community-based design to estimate the prevalence of diabetic distal sensorimotor polyneuropathy (DPN) in Taiwan. Type 2 diabetics were identified from a population-based screening program in Keelung, Taiwan. Administration of the Neurological Symptom Score (NSS) questionnaire identified 210 DPN-positive cases of the 587 diabetic subjects tested. The accuracy of this NSS screening was tested with sensitive electrophysiological DPN diagnosis. A smaller validation study was also conducted to estimate the sensitivity and specificity of the NSS questionnaire. In the validation study, the overall prevalence rate for DPN among type 2 diabetics was 26.79%. The sensitivity, specificity, positive predictive value and negative predictive values were 73.33, 30.49, 24.72 and 75.76%, respectively. The estimates of sensitivity, specificity and positive predictive values were 75.00, 33.33 and 42.86% for old cases of diabetes, and 72.22, 29.69 and 22.41% for newly diagnosed cases. The range of prevalence rate in the main study was adjusted to 28.46-36.30% after calibration for sensitivity and specificity with the validation study data. A two-stage community-based screening model with calibration of prevalence rate was developed and enabled a cost-effective DPN prevalence rate estimate in a study with a large number of subjects.