OC use and risk of breast cancer: summary of study findings

StudyDatabases searched
Coverage
Included studies
Quality assessment
HeterogeneityAge ranges/menopausal statusOutcome(s)Main results
Romieu et al (1990)

Meta-analysis
MEDLINE

1966–1989
27 case-control and 5 cohort studies

No quality assessment
Random effects model used to adjust for between study varianceVarious age ranges according to individual studies. Some analyses by pre/postmenopausal status.Breast cancer (BC) risk by OC ever-use, duration of use, time since 1st use, duration of use before 1st full-term pregnancy. BC risk by OC use in women with a family history (FH) of BC.No increase in risk with ever-use or long duration of use. 46% increase in risk of premenopausal BC with long duration of OC use (P=0.001), esp. in women using OCs before 1st full-term pregnancy (RR=1.72; 95% CI, 1.36–2.19). No increased risk in women with a FH of BC.
Delgado-Rodriguez et al (1991)

Meta-analysis
MEDLINE

1966–1990
26 case-control and 6 cohort studies

Quality assessment of studies using published guidelines
Woolf’s chi-squared used to test for heterogeneity among studiesVarious age ranges according to individual studies. Some analyses by pre/postmenopausal status.BC risk by OC ever- use, duration of use, ever-use according to age at diagnosis, use before age 25, use before 1st full-term pregnancy. OC use according to parity and FH of BCIncreased risk with ever-use for premenopausal women (RR=1.14; 95% CI, 1.05–1.24). Also increased risk in women who used OCs before 1st full-term pregnancy (RR=1.17, 95% CI, 1.06–1.30). No increased risk in women with FH of BC.
Rushton et al (1992)

Meta-analysis
Databases not specified

1980–1989
21 case-control and 6 cohort studies

No quality assessment
Substantial heterogeneity of studies discussedAnalyses by 8 categories of age from <25–55 yrs, and by <45 and ≥45 yrs.BC risk by OC use, age at diagnosis, parity and total duration of useBC risk increased by 16% in women <45, greatest in 30–34 age group. No increase in older women. Steady increase in risk with duration of OC use: 27% increase for >8 yrs’ use.
Hawley et al (1995

Meta-analysis
MEDLINE

1966–1990
38 case-control studies.

Quality assessment and quality scored, all studies included. Analysis by higher/ lower scoring studies
Test for homogeneity of RR estimates carried outVarious age ranges according to individual studiesBC risk by OC ever-use, duration of use, use before 1st full-term pregnancyEver-use: no statistically significant increase in BC risk (RR=1.08; 95% CI, 0.55–1.61). Also no association with long-term use (P=0.386). OC use before 1st full-term pregnancy significantly associated with increased BC risk (P<0.001).
Schlesselman (1995)

Meta-analysis
MEDLINE

1980–1994
79 epidemiological studies, 25 of which relating to BC risk.

No quality assessment
Heterogeneity between studies not discussedWomen aged ≥45 yrs and less than 60 yrsRisk of BC by duration/recency of OC use (also risk of cervical, endometrial, ovarian and liver cancer)Duration of OC use: no statistically significant association with increased BC risk in this age group of women
Collaborative Group on Hormonal Factors in Breast Cancer (1996)

Re-analysis
Databases not specified

Included studies date between 1976–1992
54 case-control and cohort studies

No systematic quality assessment
Heterogeneity assessed using chi-squared testsVarious ages according to individual studiesBC risk by OC ever-use, duration of use, age at 1st use, time since 1st/last use. Subgroup analysis including women with FH of BC.Statistically significant increase in risk in current users (RR=1.24 [SD 0.04], 2p<0.00001) and in 10 yrs after use stops. No significant increase in risk 10 or more yrs after stopping, or in women with FH of BC.
Van Hoften et al (2000)

Case-control study
Not applicableNot applicableNot applicableWomen aged ≤55 yrs and >55 yrsBC risk by OC ever-use and duration of useNo statistically significant increase in risk with ever-use of OCs for combined/separate groups. No increase in risk for duration of use of 1–10 yrs, but doubling of risk in women >55 yrs for >10 yrs use (OR=2.05; 95% CI, 1.07–3.95). No increase in younger age group.
Marchbanks et al (2002)

Case-control study
Not applicableNot applicableNot applicableWomen aged 35–64 yrsBC risk by OC current/former use, duration of use, age at 1st use, time since last use, type/dose. Subgroup analysis, including women with FH of BC.No statistically significant increases in risk for ever-use, current/former use, duration of use, age at 1st use, interval since last use, estrogen dose. No association between OC use and BC risk in women with FH of BC.
Narod et al (2002)

Case-control study
Not applicableNot applicableNot applicableWomen aged 46–47 BRCA1 or BRCA2 mutation carriers onlyBC risk by OC ever use, duration of use, use before age 30, age at BC diagnosis, year of 1st use.BC risk increased in BRCA1 mutation carriers (OR=1.20; 95% CI, 1.02–1.40), but not in BRCA2 mutation carriers. BRCA1 mutation carriers only: 29% increased BC risk in women with ever use <30 years; 38% increased risk with ever use where BC diagnosed at <40 years; 42% increased risk with 1st use before 1975.

From: Appendix 14, Hormonal contraceptives

Cover of Clinical Guidelines for the Classification and Care of Women at Risk of Familial Breast Cancer in Primary, Secondary and Tertiary Care
Clinical Guidelines for the Classification and Care of Women at Risk of Familial Breast Cancer in Primary, Secondary and Tertiary Care [Internet].
NICE Clinical Guidelines, No. 14.
School of Health and Related Research (ScHARR), University of Sheffield.
Sheffield (UK): University of Sheffield; 2004 May.
Copyright © 2004, School of Health and Related Research (ScHARR), University of Sheffield.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.