Barrier and spermicidal contraceptive methods and risk of invasive cervical cancer
Hildesheim, A.; Brinton, L.A.; Mallin, K.; Lehman, H.F.; Stolley, P.; Savitz, D.A.; Levine, R.
Epidemiology 1(4): 266-272
1990
ISSN/ISBN: 1044-3983 PMID: 2083303 DOI: 10.1097/00001648-199007000-00003Document Number: 376395
A case-control study of cervical cancer and confounding factors barrier and spermicidal contraceptives, smoking, socio-economic status, and screening found that barrier methods confer no protection against cervical cancer, but spermicides may offer some protection. 479 cases diagnosed with invasive cervical cancer at 5 centers (Birmingham, Chicago, Denver, Miami and Philadelphia) and 788 community controls selected by random dialing were interviewed at home. Questions covered contraception, sociodemographics, pregnancies, menstruation, hygiene, sexual behavior, medical history, smoking, diet, marriage, and family history of cancer. 7 contraceptive methods, spermicides, diaphragm, condom, pill, IUD, female and male sterilization were analyzed. In unconditional logistic regression, at first barrier methods appeared to lower cancer risk. After controlling for age, race, education, income, last Pap smear, number of sex partners, barrier methods were not significant (odds ratio 1.2), with no significance in trend. Neither did duration of oral contraception or smoking alter the results. A slight insignificant decrease in risk was seen over time with diaphragm use. The odds ratio for vaginal spermicides was 1.0 with no trend for duration of use. For the small number of women reporting use of vaginal spermicides solely (11 cases and 29 controls) for 5 or more years, however, the odds ratio was 0.7 indicating protection against cervical cancer. The test for trend was not significant. Long-term users of spermicides were also less likely to develop cancer when ever-pill users were excluded from analysis. Analyzing only high socioeconomic groups, spermicide use gave an odds ratio of 0.5 for long-term use, and a significant trend for duration of use. Some explanations of these results were the possible protection of the whole vagina with spermicides compared to diaphragm or condom, and the potential use of larger quantities, or more consistent use of spermicides, by high socioeconomic groups of women. It is known that human papilloma virus may infects the penile shaft and that it can spread from the vagina to the cervix.