To evaluate the association between the use of oral contraceptives (OC) or hormone replacement therapy (HRT) and an increased risk of venous thromboembolism (VTE).

MEDLINE was searched from 1966 to October 1996 for relevant studies published in the English language; the keywords are provided in the article. Other sources searched include bibliographies, Current Contents and other reference lists used in clinical practice.

Study designs of evaluations included in the review

The criteria for inclusion were: case-control studies that determined OC or HRT exposure by direct inquiry or a validated alternative method; prospective or retrospective cohort studies with a follow-up of more than 1000 person-years and a contemporaneous control group; any relevant randomised controlled trial (RCT).

Specific interventions included in the review

OC or HRT.

Participants included in the review

Details of the patients' characteristics were not reported.

Outcomes assessed in the review

The outcomes were deep vein thrombosis or pulmonary embolism.

How were decisions on the relevance of primary studies made?

The authors do not state how the papers were selected for the review, or how many of the authors performed the selection.

The validity of a study was assessed according to 4 criteria: (1) avoidance of diagnostic suspicion bias due to prior knowledge of OC or HRT use; (2) accurate determination of the disease of interest (deep vein thrombosis or pulmonary embolism); (3) accurate determination of the exposure of interest (OC or HRT); and (4) matching or stratification of factors that could bias OC or HRT exposure or baseline clinical susceptibility for VTE. Study quality was rated from 1 to 4 according to the number of criteria met. Each study was rated independently by two reviewers, and any disagreements were resolved by consensus.

The authors do not state how the data were extracted for the review, or how many of the authors performed the data extraction.

How were the studies combined?

In studies that were heterogeneous, the pooled risk ratio was estimated as the unweighted average of natural, logarithmic-transformed unadjusted risk ratio from each study. In studies that were homogeneous, the pooled risk ratios were estimated by the Mantel-Haenszel method.

How were differences between studies investigated?

The chi-squared test for homogeneity was performed across studies.

Studies of OC use and nonfatal VTE: 13 case-control studies, 3 retrospective cohort studies, 5 prospective cohort studies and 1 RCT. Studies of OC use and fatal VTE: 2 case-control studies, 2 prospective cohort studies and 1 RCT. Studies of HRT use and nonfatal VTE: 6 case-control studies, 1 prospective cohort study and 1 RCT.

OC studies: the pooled risk ratio was 3.0 (95% confidence interval, CI: 2.6, 3.4) in case-control studies, 4.8 (95% CI: 2.5, 7.7) in retrospective cohort studies, 2.4 (95% CI: 1.6, 3.5) in prospective cohort studies, and 1.1 (95% CI: 0.4, 2.9) in RCTs. In users of third-generation OCs, the pooled risk ratio for VTE was 5.0 (95% CI: 2.5, 7.5).

HRT studies: the pooled risk ratio was 2.4 (95% CI: 1.7, 3.5) in case-control studies, 1.7 (95% CI: 1.0, 2.9) in prospective cohort studies, and 0.7 (95% CI: 0.3, 1.6) in RCTs.

For the methodological quality of OC studies, 6 were rated as level 2 and 16 as level 3. For the methodological quality of HRT studies, 6 were rated as level 2 and 3 as level 3. No study was rated as level 1.

An association between OC use and VTE is likely to be valid, but the reported risks are probably exaggerated. Users of non-third-generation OCs have a less than 3-fold increase in the risk of VTE compared with nonusers; the risk for VTE is possibly higher with the use of third-generation OCs. In addition, an association between HRT and VTE might exist;, although further investigation is required before definitive conclusions can be made.

The review's research question is clear and the inclusion criteria were well documented. The validity of the included trials is assessed and the methodological limitations are discussed. Only studies published in English were included in the review, but the potential publication bias was not discussed. The number of patients in individual studies was not reported.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.