Is it reporting bias doubled the risk of prostate cancer in vasectomised men in Mumbai, India?

ABSTRACT

BACKGROUND: Vasectomy is a common method of family planning in India and worldwide. The objective of the present study was to assess the association of vasectomy with prostate cancer in a low risk population of a developing country. A population based case control study was conducted in Mumbai, India, for this purpose. METHODS: Included in this study were microscopically proved cases of prostate cancer diagnosed during 1998 to 2000 and registered by Bombay Population Based Cancer Registry (n=594). The controls were healthy men belonging to the resident general population of Mumbai, India. Two controls for each case matched by age and place of residence were selected as the comparison group. Data on vasectomy and potential confounding factors were obtained by structured face to face interviews. After exclusions, 390 cases and 780 controls were available for final analysis and confounding was controlled by multiple logistic regression. RESULTS: Overall 14.9% of cases and 10.0% of controls had undergone vasectomy. Compared with no vasectomy the OR with ever having undergone vasectomy was 1.9 (95% CI 1.3-2.9), after controlling for age and other possible confounding factors. The risk for those who had had a vasectomy before the age of 45 years was 2.1 fold (95% CI 1.2-3.9) and for those who underwent the procedure at a later age was 1.8 fold (95% CI 1.1-2.9). The linear trend for an increase in risk with a decrease in age at vasectomy was statistically significant (p for trend = 0.01). The risk for those who completed 25 years or more time since undergoing vasectomy was 3.8 fold (95% CI 1.9-7.6) and for those who completed less than 25 years it was 1.2 fold (95% CI 0.7-2.1). The linear trend for an increase in risk with an increase in time since vasectomy was highly significant (p for trend = 0.001). CONCLUSION: There are major public health and birth control implications on vasectomy increases the risk for prostate cancer. It is likely, however, that biases identified in this study result in high estimates of risk and the true risk due to vasectomy is substantially less than the estimated one. Due to the several limitations and possibilities for reporting biases in this study, the evidence for the estimates of the higher odds ratio for prostate cancer in vasectomised men may not be a strong one. In view of the importance of vasectomy for fertility control, further studies with good design and conduct (the information on vasectomy need to be collected with better reliability) are required to clarify the issue of vasectomy associations with prostate cancer.