Abstract
A case-control study in Somerset and east Devon was undertaken to investigate possible risk factors for prostatic cancer. A total of 159 cases, diagnosed at Taunton. Yeovil and Exeter hospitals between May 1989 and May 1991, were identified prospectively and interviewed with a structured questionnaire. A total of 161 men diagnosed with benign prostatic hypertrophy and 164 non-urological hospital controls were given identical questionnaires. The questionnaire covered a wide range of factors identified from previous studies, but the central hypotheses for this study related to diet (fat and green vegetables), sexual activity and farming as an occupation. This study found no association between farming and risk of prostatic cancer (odds ratio = 0.74, 95% confidence interval 0.46-1.18), nor with sexual activity as measured by number of sexual partners (chi-squared test for trend P = 0.52). A history of sexually transmitted disease was not significantly associated with prostatic cancer, but the numbers involved were very small and the odds ratio of 2.06 (0.38-11.2) is consistent with the hypothesis. A range of questions aimed at eliciting dietary fat intake produced no significant associations, although meat consumption showed increasing risk with increasing consumption (test for trend P = 0.005). Increased consumption of leafy green vegetables was associated with lower risk, but not significantly so (test for trend P = 0.16). As expected with so many factors investigated, some statistically significant associations were found, although these can only be viewed as hypothesis generating in this context. These included apparent protective effects of circumcision and high fish consumption.
Highlights
In 1991, there were 8570 deaths from prostate cancer in England and Wales, second only to lung cancer as the commonest single site for neoplasm deaths among males (Office of Population Censuses and Surveys; OPCS, 1991)
This paper reports on a case-control study undertaken in Somerset and east Devon to identify possible causes
The study took the form of a hospital-based case-control study of patients diagnosed at Taunton, Yeovil and Exeter Hospitals
Summary
Histologically proven cases arising in the three hospitals in the study period were considered eligible, including those patients whose diagnoses were made incidentally. The two controls were: (i) a patient with histologically confirmed benign enlargement of the prostate (BEP); (ii) a hospital patient being treated for a non-urological condition, excluding any condition that might share a common aetiology with prostate cancer. This meant avoiding other cancers and cardiovascular disorders in which diet might be a factor. A list of potentially eligible patients in the relevant age group, in a randomised order, was used by the interviewer to select the control. Whenever possible it was intended to conduct interviews in hospital, but this generally proved impossible for most cases and BEP controls, and most these patients were interviewed at home. Multivariate analyses were conducted by the use of logistic regression (Breslow and Day, 1982) using the GLIM package (Francis et al, 1993)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
