Interexaminer variability of digital rectal examination in detecting prostate cancer.

OBJECTIVES Digital rectal examination (DRE) is an important method of prostate cancer detection used by primary care physicians and medical specialists to identify patients in whom a prostatic biopsy is warranted. However, there is little empirical evidence assessing the degree of interexaminer variability in the use of DRE for the detection of prostate cancer. We addressed this issue within the framework of a prostate cancer screening study. METHODS We performed DRE examinations in 116 consecutive volunteers twice on the same day, with different urologists performing the examinations. The urologists were blinded to the results of the other examination and to the subject's serum prostate-specific antigen (PSA) level. DRE results were coded as being benign or sufficiently suspicious for cancer to warrant a prostatic biopsy. RESULTS Examiners independently agreed on the DRE findings in 98 of the 116 (84%) subjects. However, when adjusted for chance agreement, the interexaminer agreement among urologists was only fair (ie, kappa = 0.22, P = 0.009). Interexaminer variability was greater between faculty and resident examiners than among faculty examiners. CONCLUSIONS Our results suggest that the reproducibility of DRE for detecting prostate cancer is only fair among urologists. Further studies are indicated to evaluate interexaminer variability between primary care physicians and urologists.