Chun T Y
Department of Urology, New York University Medical Center, New York, USA.
J Urol. 1997 Jan;157(1):65-7.
Diagnostic bias can be a problem when determining the coincidence of prostate and bladder cancer. The objective of this study was to calculate the coincidence of bladder and prostate cancer after correcting for diagnostic bias.
The cancer registry at 1 medical institution provided the number of patients diagnosed with prostate or bladder cancer between 1989 and 1994. The expected incidences of bladder and prostate cancer were calculated using 1987 to 1991 Surveillance, Epidemiology and End Results data. Diagnostic bias was determined by reviewing the charts of all patients with both types of cancer.
Of 100 patients with a diagnosis of bladder cancer 25 (25%) also had prostate cancer and 25 of 651 (3.8%) with prostate cancer also had bladder cancer. Diagnostic bias was involved in diagnosing bladder cancer in 3 of 25 patients and prostate cancer in 7 of 25. When these patients are eliminated from the incidence data, the rate of prostate cancer in those with bladder cancer is 17% and the rate of bladder cancer in those patients with prostate cancer is 3.2%. In an age, sex and race matched general population the expected rates of prostate and bladder cancer were 0.9 and 0.18%, respectively.
The rate of bladder cancer in patients with prostate cancer is 18 times higher (p < 0.01) and the rate of prostate cancer in those with bladder cancer is 19 times higher (p < 0.01) than expected. These rates do not include cases involving diagnostic bias.
