Objective To quantify incidence of erectile dysfunction (ED) and the associated risk factors in men attending community clinics in a large population in Wales, UK. Subjects and methods Of 4060 men who were invited to attend 11 community clinics, primarily to check for prostate disease, 2025 (aged 55±70 years) attended. Of these, 2002 men answered a questionnaire about personal details, medical, family and sexual history, and detailed alcohol and smoking habits. All had their serum prostate-speci®c antigen (PSA) analysed and those referred for investigation of prostatic disease underwent serum testosterone analysis. Results Complete ED was reported by 265 men (13.2%), and was closely related to age (r=0.19, P<0.001) and medication (r=0.2, P<0.001). ED occurred in 6.9% of men aged 55±60 years, 12.5% aged 61±65and 22.2% of those aged 66±70. Patients taking diabetic medication had the highest relative risk for ED and 11.3% of men with ED were taking nitrates. The numbers of years of smoking had the third closest correlation with impotence (r=0.16, P<0.001). A low serum testosterone level was a poor predictor of ED and increasing serum PSA levels did not in¯uence the distribution of ED. Conclusion About 13% of these men aged 55±70 years had complete ED; if this value is extrapolated to the whole of the UK, this equates to almost half a million men being unable to achieve any erections. The estimate would be much greater if those with milder forms of ED are included.

This study sought to identify whether a true relationship exists between benign prostatatic hyperplasia (BPH) and erectile dysfunction (ED). In a community-based study, 427 men underwent transrectal ultrasound (TRUS), uroflow studies and a questionnaire concerning erectile function. ED had a significant correlation to age (r ¼ 0.19, P < 0.001). But comparisons of prostate volume and analysis of maximum flow rate showed no significant difference between three erectile functional groups; ranging from no ED to complete ED, (one way analysis of variance). However when these two parameters were correlated to age a significant association was found to exist (log prostate volume; r ¼ 0.26, P < 0.001, log maximum flow rate; r ¼ 70.13, P ¼ 0.02). Prostate size and uroflow studies show no correlation with ED, but ED and BPH had a significant correlation with ageing. This makes a direct association between male ED and BPH unlikely but supports the theory that the association between the two pathologies could be due instead to the common link of ageing.