Reliability and validity of the National Institutes of Health: Chronic Prostatitis Symptom Index in a Malaysian population

The objective of the study is to determine the short- and long-term utility of the Chinese, Malay and English versions of the National Institutes of Health--Chronic Prostatitis Symptom Index (NIH-CPSI) in our ethnically diverse population. The NIH-CPSI was translated into Chinese and Malay, and then verified by back translation into English. Subjects included 100 new chronic prostatitis/chronic pelvic pain (CP/CPPS) patients, 71 new benign prostatic hyperplasia patients and 97 healthy individuals. Reliability was evaluated with test-retest reproducibility (TR) by calculating intraclass correlation coefficients (ICC). Internal consistency was evaluated by calculating Cronbach's alpha (alpha). Validity assessments included discriminant and construct validity. (Presented in the order of Chinese, Malay then English). ICC values for short-term (1 week) TR were 0.90, 0.80 and 0.89, while ICC values for long-term (14 weeks) TR were 0.54, 0.61 and 0.61. Cronbach's alpha values were 0.63, 0.62 and 0.57. The NIH-CPSI total score discriminated CP/CPPS patients (P<0.001) from the control groups with receiver operating curve values of 0.95, 0.98 and 0.94, respectively. Construct validity, reflected by the correlation coefficient values between the International Prostate Symptom Score and the NIH-CPSI of CP/CPPS patients were 0.72, 0.49 and 0.63 (all P<0.05). The Chinese, Malay and English versions of the NIH-CPSI each proved effective in our population. Short-term TR and discriminant validity were excellent for all three versions. However, long-term TR was only moderate, which might reflect variation in patients' perceptions of symptoms over time.