Aim: Pregnant women are hypothesized to have low adherence to prescribed medication, because of concerns about toxicity and harmful effects on the unborn child. However, very little is known about the actual adherence to prescribed medication during pregnancy. We determined to what extent women follow treatment recommendations regarding prescribed medication use in mid-pregnancy. Methods: Dutch women participating in the PRIDE Study completed a six-week diary on medication use. Additionally, pharmacy records were obtained. For each medication dispensed, we determined three measures of adherence: 1) whether use was reported in the diary (actual use), 2) difference between dispensing date and date of first reported use (initiation time), and 3) proportion of days with at least the correct number of doses taken (implementation adherence). Results: During the six-weeks study period, 235 of 816 women (29%) were prescribed medication. Actual use was highest for medications used for chronic conditions (88%; 95% confidence interval [95% CI] 81-93), followed by medication for pregnancy-related conditions (79%; 95% CI 71-86) and medication for occasional and short-time use (69%; 95% CI 60-77). We observed a ≥1 day delay in treatment initiation for 42% of medications dispensed for the first time in the study period. For medications that were actually used, mean implementation adherence was 74.2% (95% CI 69.3-79.2). Conclusion: Although actual use of medications dispensed was high, many pregnant women did not adhere to treatment recommendations. This non-adherence may impact maternal and child health and lead to exposure misclassification in studies in perinatal pharmacoepidemiology relying on administrative databases.