Women were less than half of the people tested for almost all of these common medicines.

How we did this

Until 1993, U.S. rules kept most women who could get pregnant out of early medicine studies. Today women are still often a minority in studies, and results are rarely shown separately for women — even though women report side effects more often. InvisibleWoman shows those gaps, one medicine at a time.

Why these medicines

Each one is widely prescribed to women. In every one of them, women were under half of the study participants, or fewer than 3 in 100 studies looked at women's results separately.

Who was in the studies

We collected finished studies that posted their results on ClinicalTrials.gov (the U.S. government's study registry) and counted who took part: women and men, race, Hispanic or Latina background, age and weight.

Which women were left out

Every study has rules about who can join. We read those rules and flagged ones that turned women away — for example, being pregnant, breastfeeding, on the pill, over a certain age or above a weight limit. You can always see the exact rule.

Group by group

For pregnant women, breastfeeding moms, older women and other groups, we also searched for studies that focused on them, and we summarized what each medicine's official FDA label says, in plain words.

Side effects reported more by women

These come from a 2020 study by Chandak and Tatonetti (Patterns journal), which used millions of reports sent to the FDA and compared women and men taking the same medicine. We left out problems with body parts only one sex has (like the uterus or prostate).

Checking your medicines

We look up your medicines in the NIH RxNorm list and then search each medicine's FDA label for warnings about the others.

What this can't tell you

  • Reports to the FDA are voluntary. "Reported more often" is a warning sign, not proof the medicine causes more harm in women.
  • Only studies that posted results online are included. Many older studies are missing.
  • A study can appear here if the medicine was only used for comparison.
  • Our rule-reading is automatic and can miss unusual wording.
  • No warning found doesn't mean a combination is safe.
  • Nothing here is medical advice.

Chandak P, Tatonetti NP. Using Machine Learning to Identify Adverse Drug Effects Posing Increased Risk to Women. Patterns, 2020.

How we did this · InvisibleWoman