Oxford University home page UHCE home page Epidembase


  • To provide information about the benefits and risks of oral contraceptives (OC's) as measured by:-
    • effectiveness in preventing pregnancy.
    • return of fertility after cessation of use.
    • outcome of planned and accidental pregnancies.
    • patterns of morbidity (as measured by hospital visits) and mortality in users and ex-users.
  • To provide similar information about other methods of contraception, especially intrauterine devices (IUD's) and the diaphragm.
  • To provide information about the epidemiology of diseases occurring with reasonable frequency in women of childbearing age and beyond.


  • A total of 17032 married white British women aged 25-39 years were recruited at 17 family planning centres in England and Scotland between 1968 and 1974.
  • Of these women, 9653 were using OC’s, 4217 were using a diaphragm and 3162 were using an IUD. Those using the latter methods had never used OC’s.
  • Information collected at recruitment included date of birth, social class, smoking habits, height and weight, contraceptive history, childbearing history, past medical history and detailed identification information including NHS number.
  • The women were followed-up at least annually at clinic visits or by post, telephone or home visit. Follow-up in each centre was organised by a part-time research assistant.
  • Information collected from the women at follow-up included all pregnancies and their outcome, all changes in contraceptive methods and the reasons for the changes, all hospital referrals and all clinic cervical smears.
  • Discharge summaries were obtained for women reporting in-patient spells. All diagnostic coding was done by MPV.


  • Detailed information was collected from individual women (with certain complex exceptions) up to mid-1994. Losses to follow-up for reasons which might lead to bias were kept to about 4 women per 1000 per annum.
  • Analyses of morbidity are based on up to 300,000 woman-years of observation.
  • Most of the oral contraceptive (OC) data refer to pills containing 50 mcg oestrogen.
  • Data are still being collected on deaths and cancer registrations.
  • In total, 89 original scientific papers presenting findings in the study have been published. A copy of the publications list is here.
  • There was some delay in return of fertility after stopping oral contraceptives, but no increase in infertility.
  • With regard to morbidity, diseases occurring significantly more often in OC users than in other women (after adjusting for the effects of confounding variables) were:-
    • Venous thrombosis and embolism
    • Thrombotic stroke
    • Myocardial infarction
    • Cervical neoplasia
    • Chronic inflammatory bowel disease
    • Carpal tunnel syndrome
  • Diseases occurring significantly less often in OC users than in other women were:-
    • Benign breast lumps
    • Uterine fibroids
    • Anaemia
    • Menstrual disorders
    • Endometriosis
    • Pelvic inflammatory disease
    • Functional ovarian cysts
    • Ovarian cancer
    • Endometrial cancer
    • Undiagnosed abdominal pain
  • Diseases with no significant relationship with OC use were:-
    • Liver disease
    • Diabetes mellitus
    • Pituitary tumours
    • Breast cancer
    • Psychiatric illness
    • Thyroid disease
    • Benign ovarian tumours
    • Rheumatoid arthritis
    • Headache & migraine
    • Malignant melanoma
    • Peptic ulcer
    • Gallbladder disease
    • Eye disease
    • Ear disease
    • Fractures
    • Back disorders
    • Epilepsy
    • Multiple sclerosis
  • Diseases occurring significantly more often in IUD users than in other women were:-
    • Menstrual disorders
    • Anaemia
    • Pelvic inflammatory disease
    • Undiagnosed abdominal pain
    • Impairment of fertility after long term use in nulliparous women
    • Pregnancies occurring with IUD in situ, showed an increased proportion of ectopics and abortions.
  • For diaphragm users, significantly more UTI and less cervical neoplasia and PID.
  • In mortality analysis, pill users died significantly less from endometrial and ovarian cancer and more from cervical cancer. No other significant differences.


Unit of Health Care Epidemiology

Nuffield Department of Population Health

The content of this website was designed and produced by Michael Goldacre, Nick Hall, Hong Harper, Leicester Gill, Myfanwy Griffith, and Glenys Bettley.

The work was funded at various times by the Department of Health, the NHS National Co-ordinating Centre for Research Capacity Development, the Department of Health's National Institute for Health Research, the Department of Health's National Centre for Health Outcomes Development, and the Nuffield Department of Population Health.

© 2016 Unit of Health Care Epidemiology, Nuffield Department of Population Health, Oxford University, Old Road Campus, Headington, Oxford, OX3 7LF
Tel: +44 (0)1865 289377/8