Article type
Abstract
Background
There has been widespread debate within the academic community regarding the quality and reliability of academic conference abstracts and their resulting inclusion in systematic reviews (SR). Literature arguing against the inclusion of abstracts highlights that obtaining the unpublished abstract is time and resource-intensive. In addition to this, it is thought that conference abstract results may be unreliable as they have not undergone the same rigorous peer review process as published journal articles and may include only interim data with a high likelihood of change. Alternatively, other SR authors argue for their inclusion to minimise publication bias and reduce the occurrence of ‘research waste’. This project aimed to investigate if results were consistent between conference abstracts and their corresponding publications.
Objectives
To determine if results from conference abstracts presented at the American Society for Reproductive Medicine (ASRM) and the European Society of Human Reproduction and Embryology (ESHRE) were consistent with their resulting publications.
Methods
Conference abstracts presented at ASRM and ESHRE between the years 2015-2019 were identified using the Cochrane Gynaecology and Fertility specialised register of controlled trials. PubMed was then used to identify full publications by searching on authors' names and corresponding keywords, trials were matched by comparing the methods from the two sources (including trial registration number, location etc). Results were extracted for the following outcomes live birth rate, clinical pregnancy rate, miscarriage rate and multiple pregnancy rate from both the abstract and the full text document. We then compared these and noted if numbers were identical, numbers differed slightly but significance and conclusions were the same, or numbers differed and significance and conclusions had changed.
Results and Conclusions
606 conference presentations were identified and from these 216 corresponding publications were found. These 216 matching pairs were then analysed. We will present results and discuss the implications for those conducting SRs, or other types of evidence summary, on the inclusion or exclusion of conference abstracts.
Relevance and importance to patients
This work contributes to methods used in evidence synthesis by addressing the question of the exclusion of conference abstracts which may result in publication bias.
There has been widespread debate within the academic community regarding the quality and reliability of academic conference abstracts and their resulting inclusion in systematic reviews (SR). Literature arguing against the inclusion of abstracts highlights that obtaining the unpublished abstract is time and resource-intensive. In addition to this, it is thought that conference abstract results may be unreliable as they have not undergone the same rigorous peer review process as published journal articles and may include only interim data with a high likelihood of change. Alternatively, other SR authors argue for their inclusion to minimise publication bias and reduce the occurrence of ‘research waste’. This project aimed to investigate if results were consistent between conference abstracts and their corresponding publications.
Objectives
To determine if results from conference abstracts presented at the American Society for Reproductive Medicine (ASRM) and the European Society of Human Reproduction and Embryology (ESHRE) were consistent with their resulting publications.
Methods
Conference abstracts presented at ASRM and ESHRE between the years 2015-2019 were identified using the Cochrane Gynaecology and Fertility specialised register of controlled trials. PubMed was then used to identify full publications by searching on authors' names and corresponding keywords, trials were matched by comparing the methods from the two sources (including trial registration number, location etc). Results were extracted for the following outcomes live birth rate, clinical pregnancy rate, miscarriage rate and multiple pregnancy rate from both the abstract and the full text document. We then compared these and noted if numbers were identical, numbers differed slightly but significance and conclusions were the same, or numbers differed and significance and conclusions had changed.
Results and Conclusions
606 conference presentations were identified and from these 216 corresponding publications were found. These 216 matching pairs were then analysed. We will present results and discuss the implications for those conducting SRs, or other types of evidence summary, on the inclusion or exclusion of conference abstracts.
Relevance and importance to patients
This work contributes to methods used in evidence synthesis by addressing the question of the exclusion of conference abstracts which may result in publication bias.