Oral contraceptives and reproductive factors in multiple sclerosis incidence

Contraception. 1993 Feb;47(2):161-8. doi: 10.1016/0010-7824(93)90088-o.


Data from the Oxford.FPA prospective study show that oral contraceptive use and pregnancy have no discernible effect on the risk of developing multiple sclerosis (MS). Women of parity 0-2 developed MS twice as often as women of parity 3 or more but the difference did not reach statistical significance. Smoking may be a risk factor for developing MS. A nested case-control analysis did not identify any associations between MS onset and preceding illnesses.

PIP: Public health researchers analyzed data on 63 women who were 25-39 years old in 1968-1974 and had been followed up until at least December 1991 (the prospective Oxford Family Planning Association study in England) to study the effects of pregnancy, parity, and oral contraceptive (OC) use on the risk of developing multiple sclerosis (MS). MS onset was highest among 40-44 year olds (relative risk [RR], 1.7) and lowest among those less than 45 years old (RR, 0.4), but MS was not significantly related to age. Only 21% had developed MS symptoms by 34 years, while medical textbooks claim the peak ages to be 30-35. Women who had ever used OCs had a lower RR than nonusers (RR range, 0.5-0.8), but no trend with duration of OC use or time since last use existed. Women who had at least 3 children had a lower risk of developing MS than those of parity 2 or less and nulliparity (RR, 0.4); yet this was not significant. Further, pregnancy did not significantly affect MS onset, but there was a slight excess of low-birth-weight infants and a small deficit of miscarriages and terminations in women who later developed MS. The authors urged colleagues to conduct further research to examine the relationship between low birth weight and MS. Women who had ever smoked had a higher RR than those who had never smoked (RR for ex-smoker = 1.5, RR for 1-14 cigarettes/day = 1.6, and RR for at least 15 cigarettes/day = 1.8) and the association was almost statistically significant (p = .054). The nested case control analysis did not find any link between MS onset and preceding illnesses, including those identified by the literature as being linked with MS (bowel dysfunction, menstrual problems, endometriosis, preeclampsia/eclampsia, sinusitis, catarrh, and tonsillitis). In conclusion, the data did not provide new insights for understanding the etiology of MS.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Age Factors
  • Contraceptives, Oral / adverse effects*
  • Family Planning Services
  • Female
  • Humans
  • Incidence
  • Middle Aged
  • Multiple Sclerosis / epidemiology*
  • Multiple Sclerosis / etiology
  • Multiple Sclerosis / physiopathology
  • Parity / physiology*
  • Pregnancy / physiology*
  • Prospective Studies
  • Risk Factors
  • Smoking / adverse effects


  • Contraceptives, Oral