The increased risk of venous thromboembolism and the use of third generation progestagens: Role of bias in observational research

摘要: Abstract A matched case-control study was undertaken in 10 centers Germany and the United Kingdom to explore association of current use major combination oral contraceptives with occurrence venous thromboembolism. The cases recruited were 505 women aged 16–44 years thromboembolism, controls 1877 (at least 3 per case) for 5-year age group region without VTE. main outcome measures odds ratios derived by comparing a specific contraceptive or OC against other groups no OC. (95% confidence intervals) thromboembolism were: third generation products (low dose ethinyloestradiol, gestodene desogestrel) versus second desogestrel, 1.5 (1.1 2.0), norgestimate included generation, 1.6 (1.2 2.2). progastagens levonorgestrel-containing compounds 1.7 2.6) gestodene, 1.8 1.9 (1.0 3.6) 1.3 (0.7 2.5) progestagen-only pills. For 25 44 likely be exposed any these progestagens, comparison progestagens levonorgestrel showed successive increase market introduction ranging from (0.9 desogestrel 30 μg oestrogen content (introduced 1981) 2.8 (1.3 6.5) 20 1992) significant linear trend analysis (p = 0.00012). influence classification as product does not significantly alter results regarding direct (which is primarily metabolized levonorgestrel) shows an increased ratio. increasing risk recency when compared strongly indicative existence external bias due attrition susceptibles.