Objectives: To determine which case-definition of clinical benign prostatic hyperplasia (BPH) has the best predictive value for general practitioner visits for lower urinary tract symptoms (LUTS) suggestive of BPH. The incidence and prevalence rates of general practitioner visits for LUTS were also determined.
Methods: A longitudinal, population-based study from 1995 to 2003 was conducted among 1688 men aged 50 to 78 years old. Data were collected on physical urologic parameters, quality of life, and symptom severity as determined from the International Prostate Symptom Score. Information on health-care-seeking behavior of all participants was collected from the general practitioner (GP) record using a computerized search engine and an additional manual check of the electronically selected files.
Results: The incidence and prevalence rate of the men at risk was 19.6% and 14.0%, respectively, and these rates increased with age. For sensitivity and the positive predictive value, the case-definition of clinical BPH as an International Prostate Symptom Score greater than 7 had the best predictive value for GP visits for LUTS within 2 years after baseline.
Conclusions: Because only marginal improvement (greater specificity but lower sensitivity) in the prediction of GP visits for LUTS was possible by adding information on prostate volume and flow, for the prediction of future GP visits for LUTS suggestive of BPH, we suggest that the International Prostate Symptom Score questionnaire be used and that estimation of the prostate volume and flow is not required.