Ronald L. Poland, professor and chair of the department of pediatrics at the Pennsylvania State University College of Medicine, responds:

"You have asked an interesting question, one that continues to puzzle the medical profession as well as the general public. It is a difficult question to answer, because one could not design a definitive yet ethical human study that would randomly select whether or not a group of enrolled neonatal subjects were circumcised. So the information that we do have is culled from studies of boys or men, circumcised or not, who differ from one another in nonrandom ways--that is, they differ in other characteristics that might have led to the original family decision about whether or not to circumcise. These potentially confounding characteristics include religion (which may modify behavior), ethnic group or tribal membership, and economic status, among others. Therefore, all studies of the medical effects of circumcision have inherent flaws that reduce their power to provide convincing evidence.

"Researchers have published studies to show that vaginal or cervical cancer and penile cancer are more prevalent among couples in which the man is uncircumcised. But all of these cancers are strongly associated with, if not caused by, a virus (the human papillomavirus), which is transmitted through sexual contact. Even if circumcision does reduce the spread of this virus--or any virus for that matter--it could not be a reliable form of prevention. Many studies show that cervical and penile cancers are associated with sexual activity that starts at a young age and that involves many partners. The type of sexual activity may correlate with the social and cultural factors that control decisions about circumcision, producing a possible bias.

"The same demographic limitations apply to the spread of human immunodeficiency virus (HIV). A study from a venereal disease clinic in Africa reported that circumcision was less common among HIV-infected males as compared with HIV-negative males who attended the same clinic. This clinic served two different tribes, each having who have different religions and mores. Again, the prevalence of circumcision was but one difference between the groups and so cannot be considered the only reason for the discrepancy in their infection levels--and circumcision certainly cannot be depended on for protection against a deadly virus.

"There are several published studies that conclude that circumcision prevents urinary tract infection in infant boys. These studies focus on infants who were examined for fever, were hospitalized and were diagnosed as having discharge from a urinary tract infection. These studies may be biased in another way. For years, physicians have heard that uncircumcised boys may be more prone to urinary tract infections. Circumcised boys, therefore, are more likely to be checked for signs of infection than are their uncircumcised friends. Unfortunately, there have been no studies designed to test boys (circumcised and not) prospectively for urinary tract infection.

"A 1996 statement of the Canadian Pediatric Society concluded that there are no medical reasons to perform a routine circumcision on a newborn infant. An earlier American Academy of Pediatrics Task Force on Neonatal Circumcision noted some potential risks and benefits associated with the procedure but did not see a compelling medical reason for recommending routine circumcision either. So the short answer to the question is no. There are no conclusive medical studies documenting the health benefits of circumcision, although there are suggestive studies on both sides of the issue.

"The following references may be helpful:"

Decreased Incidence of Urinary Tract Infections in Circumcised Male Infants. T. E. Wiswell, F. R. Smith and J. W. Bass in Pediatrics, Vol. 75, pages 901-903; 1985.

Human Immunodeficiency Virus Infection among Men with Sexually Transmitted Diseases: Experience from a Center in Africa. J. N. Simonsen et al. in New England Journal of Medicine, Vol. 319, pages 274-278; 1988.

Comments