Background: Conference presentations are a valuable means of early dissemination of research findings. However, a complete understanding of the study methods, results, and implications is achieved only with publication of a full‐text article in a scientific journal.
Objectives: To explore the characteristics and publication outcome (ie, full‐text article in a scientific journal) of abstracts on pregnancy‐related topics presented at the International Society for Pharmacoepidemiology (ISPE) annual conferences (ICPE).
Methods: Abstracts presented at the ICPE in 2014 and 2015 were identified through the conference program and book of abstracts. Workshops and symposia were excluded. Publications were identified from PubMed (year of abstract presentation through 31 December 2017); letters to the editor were excluded. Abstract characteristics (first author's affiliation, countries of authors' affiliations, type of presentation, methods, results, and conclusions), publication outcome, time of first publication, journal category, and impact factor (IF) were evaluated. Kaplan‐Meier analysis was used to estimate time to first publication, and multivariable logistic regression was used to identify factors associated with publication.
Results: A total of 119 abstracts were presented (28.6% oral, 71.4% poster), 68% with academic affiliation of first author, 30.3% with authors' affiliation from >1 country (17.6% with intercontinental author's affiliations). Drug safety (57.1%) and drug utilization (16.0%) were the most frequent study types. Of all abstracts, 84.0% used secondary data, 18.5% used multiple data sources, and 63.9% used analytical methods. The primary focus, in 82.4% of abstracts, was medications, most frequently, nervous system medications (30.3%). Outcomes or complications of pregnancy, fetus, or infant, or combinations thereof, were evaluated in 58.8%. A total of 65 abstracts (54.6%) led to publications during the follow‐up period. Median time to publication was 1.5 years (95% CI, 1.3‐1.7 years), and mean (SD) of the journal IF was 6.3 (8.9). A strong association with publication was observed for international collaborations (OR = 3.0; 95% CI, 1.2‐7.3), and marginal associations were observed for type of presentation, medications as a focus of study, and year of abstract presentation.
Conclusions: More than half of the abstracts on pregnancy‐related topics presented at the ISPE annual conferences resulted in published papers within 2 years. The factor most strongly associated with publication was international collaboration.