Globally, women account for more than half of persons living with HIV (PWLH), yet remain underrepresented in research. Starting in 2018, CROI guidelines specifically recommended reporting of sex distribution and sex-adjusted analyses, but formal review showed rates of such reporting among CROI 2018 oral abstracts were low. Members of the Women’s Health Inter-Network Scientific Committee (WHISC) of the ACTG and IMPAACT networks reviewed adherence to these guidelines among oral presentations from CROI 2019.
Each CROI 2019 webcast of oral abstract presentations was reviewed by at least 2 WHISC members. For presentations of research relevant to both sexes, reviewers assessed whether sex distribution and sex-adjusted analyses were presented.
Overall, 85 oral abstracts (79 clinical studies and 6 pre-clinical studies) addressed a scientific question relevant to both sexes. 85% (67 of 79) of clinical studies included sex distribution compared to 76% at CROI 2018. In 10% of these (7 out of 67), sex was incorrectly referred to as gender, which was similar to CROI 2018. 51% (34 of 67) of clinical studies included ≤25% women; 7 included no women and 3 included no men, but only 5 of these 10 communicated this restriction in the abstract title. Reporting of sex of animals in pre-clinical studies remained low (0 of 6 in 2019 and 1 of 13 in 2018). Finally, 49% (39 of 79) of clinical studies at CROI 2019 included sex-adjusted results, which was slightly higher than the 30% reporting at CROI 2018. Sex-adjusted analyses were higher among observational studies (64%, 29 of 45) than in clinical trials (29%, 10 of 34); see Table.
While there was some improvement in reporting sex-specific analyses among oral presentations at CROI 2019 compared to 2018, there is still much room for improvement. Consistent failure to report sex distribution in pre-clinical studies needs to be addressed. Reporting of sex distribution in clinical studies needs more emphasis since 15% of oral presentations failed to include this. Education regarding the difference between sex and gender is necessary and titles should indicate whether findings are restricted to one sex. Finally, enrolling adequate numbers of women to perform meaningful sex-stratified analyses and performing such analyses require additional guidance and even mandates given that over half of PLWH worldwide are women.