Factors Predictive of Oral Abstract Being Published: Is Gender Disparity Playing a Role?

Introduction: It remains unclear what percentage of abstracts proceed to manuscript publication and the characteristics that predict successful publication. This study aimed to determine factors associated with successful peer-reviewed publications following oral presentation at the American Society of Hematology (ASH) annual meeting.

Methods: All oral abstract presentations (n=621) in the hematological malignancy category from 2016 ASH annual meeting were included in the study. Abstract publication was confirmed by searching for the publicly listed abstract on PubMed by title, first, and last author names, and institutional matching. We recorded time to online publication, US versus foreign journal publication, and journal impact factor by 3.5 years from 2016 ASH annual meeting. Abstracts characteristics that were analyzed also included number of authors, gender of first author, gender of last author, and single vs multi-institution studies. Gender of the first and last author was confirmed by looking at their biography details on their institutional website. Descriptive analysis was performed and an association between presenter’s or last author’s gender and publication matrix was analyzed using Chi-square tests.

Results: Of the 621 abstracts, 350 (56%) were published in full text by three and a half years since the 2016 ASH annual meeting. The abstracts’ average time to journal publication was 17.46 months (SD +/- 11.32) (Table 1). Of the published articles, 64% (223/350) were published in U.S. journals; mean impact factor for all publications was 14.46 (SD+/- 11.47).The median number of authors for published and unpublished abstracts were similar. Females presented 37% (228/621) of the abstracts and 35% (123/350) of the journal publications had female first author and 22% (77/350) had female last author. A total of 53.9% (123/228) abstracts presented by a female author were published versus 57.7% (227/393) abstracts presented by a male author (p=<0.001). Males were more often involved in multi-institutional trials (p=0.045) and were more likely to have senior authorship (p=0.005). There was no correlation between the gender of the first author to journal impact factor (p=0.109) or time to publication (p=0.091).

Conclusion: More than half of the oral abstracts were successfully published regardless of gender and number of authors. The rate of successful publication is higher for male authors even though there was no correlation between the gender of the first author to journal impact factor or time to publication. Our study highlights gender disparity in senior authorship, however this difference is not as wide in first authorship.