Abstract
Importance
No recent study has examined whether method of abortion is associated with psychotropic medication use, an indicator of mild mental health problems.
Objective
To examine whether medication or procedural abortions were associated with increased risk of psychotropic medication use.
Design, Setting, and Participants
This population-based Danish registry cohort study followed the psychotropic medication redemptions of females aged 12 to 38 years having elective, first first-trimester abortions between January 1, 2000, and December 31, 2018, from 1 year before their medication or procedural abortion until their first psychotropic medication prescription claim (redemption), December 31, 2018, emigration from Denmark, or death, whichever came first. Analyses were conducted between June 2023 and June 2025.
Exposure
First abortion method (first medication abortion or first procedural abortion).
Main Outcomes and Measures
Any first psychotropic medication prescription redemption, first antidepressant medication redemption, and first antianxiety medication redemption during the study period. Incidence rate ratios (IRRs) were calculated in adjusted regression models. Bonferroni-corrected
P
values were used to account for multiple tests.
Results
Of 67 390 females included in this study (mean [SD] age, 21.8 [4.6] years), 4575 (6.8%) had a prior psychiatric diagnosis; 33 793 had a record of a first medication abortion, 33 597 had a record of a first procedural abortion, and 19 979 (29.6%) had a first psychotropic medication redemption during the study period. In fully adjusted models, using the conventional
P
value level of .05, compared with the year before a medication or procedural abortion, there were small increased risks of any first psychotropic medication redemption in the first year after medication abortion (IRR, 1.10; 95% CI, 1.02-1.19;
P
= .01) and procedural abortion (IRR, 1.09; 95% CI, 1.01-1.16;
P
= .02). There were no statistically significant associations in 1 to 2 years and 2 to 5 years after abortion relative to the year before an abortion for each abortion method. There were decreased risks of any first psychotropic medication redemption more than 5 years after medication abortion (IRR, 0.85; 95% CI, 0.78-0.91;
P
< .001) and more than 5 years after procedural abortion (IRR, 0.83; 95% CI, 0.78-0.88;
P
< .001). Using the Bonferroni-corrected
P
value of .002, the small increased risks of psychotropic medication redemption observed in the first year after an abortion no longer met the criteria for statistical significance, while the lower risks for more than 5 years after abortion continued to meet the criteria for statistical significance.
Conclusions and Relevance
In this Danish population-based cohort study, compared with the year before a medication or procedural abortion, small increased risks of psychotropic medication prescription redemptions were observed during the first year after abortion, and decreases in risk of psychotropic medication prescription redemptions were observed more than 5 years after abortion when using conventional levels of statistical significance. The small increased risks observed in the first year after an abortion no longer met the criteria for statistical significance following Bonferroni adjustment for multiple statistical tests. Additional research may be needed to replicate and understand these results.