Variation in urologists' opinions on treatment choice in benign prostatic hyperplasia

The aim of this study was to examine the extent of agreement in a panel of urologists with respect to therapy choice in patients with benign prostatic hyperplasia (BPH). From a prospective study on treatment choice for BPH (n=670), a stratified sample of 16 patients was drawn, 4 each from the following actual treatment groups: surgery, 5-a-reductase-inhibitor, a-blocker and watchful waiting. Detailed information on the results of the various diagnostic procedures of these patients was presented to a panel of 58 urologists. Subsequently, they were asked to give their individual opinions on the most suitable therapy for each patient. When the information presented was considered as insufficient to make a choice, each urologist was asked to write down which additional data were required. In 23% of all cases, the information was considered to be insufficient for making an adequate choice. For these cases, urodynamic study (44%) and cystoscopy (24%) were most frequently mentioned as necessary additional procedures. Agreement between the panel members was low. The k value for agreement amongst multiple raters was 0.29 (SE 0.003), when taking into account all categories and 0.48 (SE 0.005) when the category ‘no choice’ was excluded. Analyses on agreement by the extent of the urologist's experience and work setting showed no large differences between the subgroups. The overall agreement between actual treatment decisions and panel opinions was highest for surgery (86%), followed by watchful waiting (66%) and medical therapies (53% and 28% respectively). Variation in opinions on therapy choice for BPH among urologists is significant. Given the differing effectiveness, side-effects and clinical sequelae of the various treatment options, consensus development and subsequent quality assurance are obviously desirable.