Article Title

Patterns of Prednisone Use During Pregnancy: Daily and Cumulative Dose

Publication Date



observational studies, maternal and perinatal care, biostatistics, epidemiology, pharmaceuticals, costs


Background: Descriptions of oral corticosteroid use during pregnancy are usually limited to trimester prevalence estimates. We sought to better characterize the use during pregnancy.

Methods: Data were collected from MotherToBaby Pregnancy Studies (2005–2014) that enrolled pregnant women with asthma and autoimmune diseases before gestational week 20. Information on medication use and pregnancy outcomes was collected by telephone interview at enrollment, 24 and 32 weeks gestation, and after delivery, plus by medical record review. Women were included if they had a live or still birth and reported dates of oral prednisone use and dose during pregnancy. Women were classified by diagnosis: asthma (n = 15), rheumatoid arthritis (n = 254), Crohn’s disease (n = 39), ankylosing spondylitis (n = 15), and > 1 autoimmune disease (n = 29). Prednisone daily dose and cumulative dose by gestational day were plotted using a heatmap for each individual and stratified by disease. To summarize the many observations for rheumatoid arthritis, we used k-means clustering method to identify and plot group trajectories for prednisone dose. The associations between trajectory group and maternal age, race/ethnicity, socioeconomic status, obesity, rheumatoid arthritis severity as measured by the Health Assessment Questionnaire-Disability Index (HAQ) (range: 0–3) before gestational week 20, and gestational age at delivery were evaluated.

Results: Women used prednisone on 1 to 292 days total during pregnancy. Daily doses ranged from 2 to 60 mg for asthma; < 1 to 60 mg for rheumatoid arthritis, ankylosing spondylitis and > 1 autoimmune disease; and < 1 to 70 mg for Crohn’s disease. Total cumulative dose ranged from 15 to 1,325 mg for asthma, 8 to 6,225 mg for rheumatoid arthritis, 10 to 9,105 mg for Crohn’s disease, 60 to 2,720 mg for ankylosing spondylitis, and 20 to 5,120 mg for > 1 autoimmune disease. High-dose prednisone for a short duration was more common for asthma, whereas lower doses for longer was more common for rheumatoid arthritis. For women with rheumatoid arthritis, the highest versus lowest cumulative dose trajectory groups had significantly different disease severity (median HAQ: 0.7 vs 0.0) and gestational length (gestational weeks: 35.9 vs 39.0).

Conclusion: The individual-level plots illustrate variability in prednisone dosage (amount and pattern), which may impact pregnancy outcomes. Group trajectories, used to summarize these factors, provide an alternative to the typical trimester exposure approach when studying risks of corticosteroids and other medications during pregnancy.




June 23rd, 2017


August 10th, 2017