With evidence, albeit limited, that oxytocin released at the time of orgasm is associated with a reduction in the risk of breast cancer in women [1], I was disturbed by the unwarranted misinterpretation published by Fox [2], the scientific correspondent of New Scientist, of the data on the relationship between ejaculation frequency and risk of prostate cancer in the report by Giles et al.[3]. This led to worldwide media headlines suggesting that masturbation prevented prostate cancer. Having now had the chance to read the original data, and found no analysis between frequency of intercourse and masturbatory-related ejaculation, I am even more disturbed and wondered why Giles et al. did not consider oxytocin as a possible explanation of their data.

Furthermore, although the authors decided prospectively not to collect data on sexually transmitted diseases, they have concluded somewhat unscientifically that the data argued against infection having any role in the cause of prostate cancer. The data in Table 2 of their report show two divergent trends, i.e. that two to four partners reduces the risk but that any increase beyond five or more partners reduces that benefit. It would be interesting to see the risk for the proportion of partners analysed on the basis of frequency of ejaculation, i.e. are the promiscuous low- ejaculation group more likely to get prostate cancer while those not promiscuous but in the high-ejaculation group are at an even lower risk?

Possible support for this view comes from the data in Table 3 of the paper, where a high frequency of ejaculation in men in their fourth decade is associated with low ejaculation frequency of the second decade. Dividing the data on the basis of the number of partners, i.e. five or less vs more than five, and re-examining the effect of frequency of ejaculation in second and fourth decade would better assess any possible effect of an infectious element in the cause of prostate cancer.