Pill-stroke link: case not closed.

Although 30 years of research have pointed to a link between birth control pill use and the risk of stroke, the association is "tenuous at best and perhaps nonexistent," according to a meta-analysis including 36 studies. (1) Overall, the pooled data showed that pill users had nearly twice nonusers' stroke risk (odds ratio, 1.9). The association was significant in case-control studies (2.1), for stroke caused by blood clots (2.7) and for current pill users (2.0). However, the data suggest a negative association in cohort studies, show no association for stroke caused by bleeding in the brain or for stroke-related deaths, and indicate no elevation in risk for ever-users. Increases in risk were similar regardless of estrogen dose (1.8 for high- and low-dose pills) or type of progestin (2.4-2.9). Women's risk factors were key to the relationship: Increases in risk were more marked for pill users who were 35 or older (2.3), smoked (3.5) or had high blood pressure (9.8) than for users without these risk factors (1.3-2.1). The analysts contend that the inconsistencies in the data, the low absolute values of the odds ratios and "severe methodological limitations" (which they outline in their report) cast doubt on the association between pill use and stroke risk, particularly in young women who do not smoke or have high blood pressure.

(1.) Chan W-S et al., Risk of stroke in women exposed to low-dose oral contraceptives: a critical evaluation of the evidence, Archives of Internal Medicine, 2004, 164(7):741-747.