Limitations of using outcomes in the placebo arm of a clinical trial of benign prostatic hyperplasia to quantify those in the community

OBJECTIVES: To quantify potential biases that may occur when placebo arms of clinical trials are used to characterize the natural history of disease and to compare incidence rates of benign prostatic hyperplasia (BPH) outcomes in community-dwelling men with outcomes in the placebo arm of a clinical trial of BPH. SUBJECTS AND METHODS: White men aged 50 years or older at baseline were selected randomly from the Olmsted County, Minnesota, community in 1990 and were monitored biennially through 1996 for urologic outcomes. Symptom progression, acute urinary retention, and minimally invasive or surgical treatment of BPH were assessed from a validated questionnaire and a review of community medical records. Findings from the Olmsted County Study (N=1193) and a selected subcohort (n=238) were compared with those from the placebo arm of the Medical Therapy of Prostatic Symptoms trial (N=737). RESULTS: During more than 5088 person-years of follow-up (mean, 4.9 years) in the Olmsted County Study, incidence rates per 1000 person-years were 8.5 (95% confidence interval [CI], 6.4-11.2) for acute urinary retention, 97.1 (95% CI, 88.7-106.0) for symptom progression, 6.6 (95% CI, 4.8-9.0) for surgery or minimally invasive treatment, and 105.1 (95% CI, 96.4-114.4) for any outcomes for all men. For those meeting trial inclusion criteria (selected subcohort, n=238), incidence rates were 18.3, 86.5, 16.8, and 109.4, respectively. By comparison, incidence rates per 1000 person-years for the placebo arm of the Medical Therapy of Prostatic Symptoms clinical trial for BPH (mean follow-up, 4.5 years) were 6 for acute urinary retention, 36 for symptom progression, and 45 for any outcome, but the estimate of 13 for surgery or minimally invasive treatment was higher than for men in the Olmsted County Study. CONCLUSIONS: Compared with community-dwelling men, men in the placebo arm of this clinical trial of BPH treatments had a substantially lower risk of BPH-related outcomes. Extrapolation of findings from the placebo arm of clinical trials to describe the natural history of disease in community-dwelling men should be done with caution and appropriate recognition of limitations.