Increasing rates of sexually transmitted infections (STIs) in the United States among men who have sex with men (MSM) have raised concerns that pre-exposure prophylaxis (PrEP) has been associated with higher engagement in condomless anal intercourse (CAI). While partnership characteristics have previously been found to influence condom use, the extent to which PrEP use may modify their effect on CAI remains unknown. A secondary analysis of 535 sexual partnerships from a cross-sectional study in San Francisco was conducted to evaluate interactions between PrEP use and partnership characteristics on CAI. Bivariate and multivariate generalized estimating equation (GEE) logistic regression models were used to estimate relative measures of association, adjusted for confounding by seroconcordance and partnership type, as well as account for repeated partnerships per respondent. Partnerships where both partners used biomedical prevention had significantly greater odds of CAI [odds ratio (OR) = 5.19, 95% confidence interval (CI): 2.27-11.9] compared to those where only one partner used biomedical prevention, while those where neither partner used biomedical prevention had significantly lower odds of CAI (OR = 0.61, 95% CI: 0.40-0.93). There was no significant association between meeting place (online vs. offline) and sexual risk taking (OR = 1.03, p = 0.894). Having one partner disclose their HIV status (compared to neither partner having disclosed) was associated with significantly higher odds of CAI among partnerships of PrEP-using MSM [adjusted OR (aOR) = 5.28, 95% CI: 1.91-14.61], while the association was not significant among the partnerships of non-PrEP-using MSM (aOR = 1.29, 95% CI: 0.75-2.21). Differences in condom use among MSM using PrEP may not be well explained by differences in the effect of partnership characteristics. MSM using PrEP appear to commonly practice biomedical matching and high engagement in CAI with other biomedical prevention users, which could indicate relatively concentrated sexual networks and partly explain their disproportionate risk for STIs. Future studies should further investigate biomedical matching to develop interventions that further promote the sexual health of those using PrEP.

Objectives

To describe heavy alcohol use patterns and correlates in a diverse sample of MSM.

Methods

We used respondent-driven sampling (RDS) to enroll 252 alcohol-using MSM in San Francisco from March 2015-July 2017. We examined heavy alcohol use patterns and conducted RDS-adjusted multivariable analyses to characterize correlates of hazardous alcohol consumption and binge drinking.

Results

RDS-adjusted prevalence of weekly and at least weekly binge drinking was 24.9% and 19.3%, respectively. Hazardous consumption was common; prevalence of mid- and high-levels of hazardous drinking was 11.4% and 29.9%, respectively. In multivariable analyses, identifying as Hispanic/Latino or mixed/other race; being moderately or extremely interested in reducing alcohol use; ever receiving alcohol treatment; using ecstasy; reporting syphilis diagnosis; and having more than 5 male partners were independently associated with hazardous alcohol consumption. Less hazardous consumption was associated with having a bachelor's degree or completing post-graduate studies; and not being in a relationship. Reporting chlamydia infection; being somewhat, moderately or extremely interested in reducing alcohol use; and having multiple male sex partners were associated with higher odds of at least weekly binge drinking. Lower odds of binge drinking were associated with completing post-graduate studies. Moreover, for the outcomes of hazardous alcohol consumption and binge-drinking, we observed significant interaction effects between race/ethnicity and interest in reducing alcohol, past receipt of alcohol treatment, use of ecstasy, syphilis diagnosis, and number of male partners.

Conclusion

Among alcohol-using MSM in San Francisco, heavy drinking patterns were common and independently associated with greater number of male sexual partners and sexually transmitted infections (STI). Moreover, significant racial/ethnic and socioeconomic disparities related to heavy alcohol use were observed and race/ethnicity modified the effect of the risk factors associated with these outcomes. These findings underscore the need to develop more MSM-specific interventions that jointly address heavy alcohol use and HIV/STI risk, as well as culturally-tailored and targeted strategies to alleviate health disparities.

BACKGROUND: Qualitative methods are incredibly beneficial to the dissemination and implementation of new digital health interventions; however, these methods can be time intensive and slow down dissemination when timely knowledge from the data sources is needed in ever-changing health systems. Recent advancements in generative artificial intelligence (GenAI) and their underlying large language models (LLMs) may provide a promising opportunity to expedite the qualitative analysis of textual data, but their efficacy and reliability remain unknown. OBJECTIVE: The primary objectives of our study were to evaluate the consistency in themes, reliability of coding, and time needed for inductive and deductive thematic analyses between GenAI (ie, ChatGPT and Bard) and human coders. METHODS: The qualitative data for this study consisted of 40 brief SMS text message reminder prompts used in a digital health intervention for promoting antiretroviral medication adherence among people with HIV who use methamphetamine. Inductive and deductive thematic analyses of these SMS text messages were conducted by 2 independent teams of human coders. An independent human analyst conducted analyses following both approaches using ChatGPT and Bard. The consistency in themes (or the extent to which the themes were the same) and reliability (or agreement in coding of themes) between methods were compared. RESULTS: The themes generated by GenAI (both ChatGPT and Bard) were consistent with 71% (5/7) of the themes identified by human analysts following inductive thematic analysis. The consistency in themes was lower between humans and GenAI following a deductive thematic analysis procedure (ChatGPT: 6/12, 50%; Bard: 7/12, 58%). The percentage agreement (or intercoder reliability) for these congruent themes between human coders and GenAI ranged from fair to moderate (ChatGPT, inductive: 31/66, 47%; ChatGPT, deductive: 22/59, 37%; Bard, inductive: 20/54, 37%; Bard, deductive: 21/58, 36%). In general, ChatGPT and Bard performed similarly to each other across both types of qualitative analyses in terms of consistency of themes (inductive: 6/6, 100%; deductive: 5/6, 83%) and reliability of coding (inductive: 23/62, 37%; deductive: 22/47, 47%). On average, GenAI required significantly less overall time than human coders when conducting qualitative analysis (20, SD 3.5 min vs 567, SD 106.5 min). CONCLUSIONS: The promising consistency in the themes generated by human coders and GenAI suggests that these technologies hold promise in reducing the resource intensiveness of qualitative thematic analysis; however, the relatively lower reliability in coding between them suggests that hybrid approaches are necessary. Human coders appeared to be better than GenAI at identifying nuanced and interpretative themes. Future studies should consider how these powerful technologies can be best used in collaboration with human coders to improve the efficiency of qualitative research in hybrid approaches while also mitigating potential ethical risks that they may pose.