Background. Controversy exists regarding the efficacy of routine neona
tal circumcision of male infants. Little is known about parental or pr
ovider characteristics or the use of medical resources associated with
this procedure. Methods. Records of 3703 male infants born during 199
0 and 1991 at four US sites were analyzed to discern associations betw
een circumcision and the above factors. Analyses were limited to healt
hy infants. Results. Eighty-five percent of the infants in the study p
opulation were circumcised. White and African-American male infants we
re much more likely to be circumcised than those of other races (odds
ratios [ORs], 7.3 and 7.1, respectively, P<.001). Compared with self-p
ay patients, those covered by private insurance were 2.5 times more li
kely to be circumcised (P<.001). Logistic regression showed that rates
for obstetricians and family physicians were not significantly differ
ent. Increased odds of circumcision were found if the mother received
an episiotomy (OR=1.9, P<.001) or cesarean section (OR=2.1, P<.001). C
ircumcised infants stayed in the hospital an average of one fourth of
a day longer than did those who were not circumcised (mean difference,
0.26 days; 95% confidence interval, 0.16 to 0.36). Conclusions. Mothe
r's insurance status and race as well as surgical interventions during
delivery are related to circumcision. Associations with episiotomy an
d cesarean section suggest physician and/or parental preference for in
terventional approaches to health care. Generalizing the difference in
hospital length of stay to the United States suggests an annual cost
between $234 million and $527 million beyond charges for the procedure
it self.