Conference presentation in palliative medicine: predictors of subsequent publication

Concerns have been raised about poor-quality palliative care research and low publication rate from conference abstracts. The study objectives: to estimate the publication rate for European Association for Palliative Care research conference abstracts (2008) and explore associated characteristics and to understand reasons for non-publication.

Full published papers were searched to March 2015 (Medline; Pubmed; Google Scholar) and data extracted: country of origin, study design/population/topic. Multivariate logistic regression was used to identify predictors of publication.Members of two different palliative care associations were surveyed to understand reasons for non-publication. χ² statistic was used to explore associations with publication.

Overall publication rate of the 445 proffered abstracts was 57%. In the final model, publication was more likely for oral presentations (OR 2.13; 95% CI 1.28 to 3.55; P=0.003), those from Europe (3.24; 1.09 to 9.56; P=0.033) and much less likely for non-cancer topics (0.21; 0.07 to 0.64; P=0.006). Funding status, academic unit or study design were not associated with publication.

407/1546 (26.3%) physicians responded of whom 254 (62%) had submitted a conference abstract. Full publication was associated with: oral presentation (P<0.001), international conference abstracts (P=0.01) and academic clinicians versus clinicians (P<0.001). Reasons for non-publication included: low priority for workload (53%) and time constraints (43%).

The publication rate was similar to 2005 clinical conference. Probable quality markers were associated with publication: oral presentations selected by conference committee, international conference abstracts and abstracts from those with an academic appointment. Publication was given a low priority among clinical time pressures.