Study Questions NSAID–Contraceptive Blood Clot Link

BLOOD CLOT associations reported during combined use of prescription painkillers and hormonal contraceptives may partly reflect the conditions prompting treatment, a large European study suggests.

The findings highlight the difficulty of separating a medicine’s effects from the health problems for which it is prescribed, with implications for how clinicians interpret observational safety signals.

Comparing Treatment Periods

Researchers examined records covering more than 3 million women aged 15–49 years using hormonal contraceptives in Denmark, Norway, Spain and the UK.

Their self-controlled case series compared periods within women who experienced venous thromboembolism (VTE), rather than simply comparing different groups of medicine users.

VTE includes deep vein thrombosis and pulmonary embolism. The analysis assessed prescription non-steroidal anti-inflammatory drugs (NSAIDs), with contraceptives grouped by clot risk.

Researchers found that clot incidence rose before NSAID initiation, peaked around treatment initiation and subsequently declined.

In Danish users of high-risk contraceptives, the incidence rate ratio was 3.70 before exposure and 12.29 during the first week, compared with contraceptive use alone.

Prescribed paracetamol, used as a negative control, showed a similar pattern. This supports the possibility that illness or symptoms leading to painkiller use contributed to the association. The authors nevertheless could not exclude a direct NSAID effect.

Interpreting the Safety Signal

The sequence matters clinically as if an association is already present before a prescription begins, attributing the entire subsequent increase to that prescription risks overlooking other explanations.

Painkiller prescribing may coincide with an evolving health problem, making treatment timing particularly important when assessing possible adverse effects.

Equally, these results should not be read as proof that every combination is safe. Relative incidence estimates do not describe an individual woman’s absolute likelihood of a clot, and a large underlying population does not guarantee precise estimates for every exposure subgroup.

The practical interpretation is therefore a more careful assessment of the original signal. Decisions about analgesia and contraception require attention to the patient’s circumstances and the indication for treatment, rather than reliance on a single headline risk estimate.

Further research that distinguishes prescribing indications, symptoms preceding diagnosis and medicine exposure would help clarify whether an additional drug-related effect remains. For now, the study offers a reason to scrutinise causal explanations before translating an observational association into broad prescribing restrictions.

Reference:

Li X et al. Prescription non-steroidal anti-inflammatory drugs and venous thromboembolism among women aged 15–49 years using hormonal contraceptives: population-based self-controlled case series study from four European countries. Lancet Obstet Gynaecol Womens Health. 2026. https://doi.org/10.1016/S3050-5038(26)00289-X