The transnational study on oral contraceptives and the health of young women. Methods, results, new analyses and the healthy user effect

• Published in: Human reproduction update Vol. 5; no. 6; pp. 707 - 720
• Main Author: Lewis, M.
• Format: Journal Article
• Language: English
• Published: England Oxford University Press 01.11.1999; Oxford Publishing Limited (England)
• Subjects: Adolescent; Adult; Brain Ischemia - chemically induced; Case-Control Studies; Contraceptives, Oral - adverse effects; Female; Humans; Models, Theoretical; Myocardial Infarction - chemically induced; Proportional Hazards Models; Reference Values; Risk Factors; Stroke - chemically induced; Thromboembolism - chemically induced; United Kingdom; Venous Thrombosis - chemically induced; Women's Health
• ISSN: 1355-4786; 1362-4946; 1460-2369; 1460-2369
• DOI: 10.1093/humupd/5.6.707

The results of the Transnational study on Oral Contraceptives and the Health of Young Women studies are reviewed, potential biases shown, and new analyses are presented. The initial finding was a slight increase in the risk of venous thromboembolism (VTE) when users of third generation OC (containing the progestins desogestrel and gestodene) were compared with users of older (second generation) OC [odds ratio (OR) of 1.5; 95% confidence interval (CI) 1.0, 2.1]. In order to account for potential biases in the study, new analyses were performed. A quadratic spline technique comparing first users of OC with never-users found that risk of VTE is highest in the first year of OC use, but showed no difference between the two generations of OC upon adjustment. An analysis of new dataset which now included full information on the past exposure history of all study subjects with a Cox regression model with time-dependent covariates produced an OR of 0.8 (0.5, 1.3) for the comparison of current users of third generation OC with current users of second generation OC. The results of the new analyses with more refined techniques and with an enhanced dataset which includes data collected on the past exposures of all study subjects indicates that bias, particularly that introduced by healthy user cohorts, accounts for the results initially shown for VTE within the Transnational Study. Key words: case-control study/epidemiology/oral contraceptives/Transnational study/venous thromboembolism