Inadequate sexual history screening in US primary care limits identification of candidates for HIV pre-exposure prophylaxis (PrEP) and exacerbates inequities in PrEP uptake [1,2]. Many electronic health record (EHR) systems lack standardized prompts for documenting sexual risk, which contributes to missed opportunities for sexual health discussions and HIV prevention. This documentation gap is compounded by limited provider knowledge of PrEP and the absence of validated tools to guide prescribing decisions [3,4].

EHR-based clinical decision support (CDS) tools may address these gaps by facilitating sexual risk assessment and supporting PrEP prescribing [5]. Prior studies suggest that CDS tools are acceptable and feasible [6,7] and may improve PrEP prescribing [5]. Nevertheless, clinician experiences using CDS tools for sexual risk assessment and PrEP prescribing in primary care remain poorly understood. Clinician feedback about such CDS tools may inform efforts to optimize use of these tools. Therefore, we aimed to describe primary care clinicians’ experiences using CDS tools implemented in urban community health clinics to support sexual risk assessment and PrEP prescribing.

Our study population comprised 33 clinicians (12% of 284 eligible). Participant characteristics are summarized in Multimedia Appendix 5. Most respondents were physicians (MD/DO, n=20, 61%), followed by nurse practitioners (n=10, 30%). The majority were older than 30 years (n=29, 88%), identified as female (n=23, 70%), and were non-Hispanic White (n=19, 58%).

Fewer than half of respondents (n=15, 45%) were aware of the updated sexual history questionnaire. Table 1 summarizes awareness, use, and perceptions of the CDS tool among respondents who reported awareness of the tool. Most clinicians who reported awareness or use of the tool agreed that the questionnaire was important to implement (n=10, 67%), helpful for identifying PrEP candidates (n=9, 60%), and appropriate for primary care (n=9, 60%). In contrast, clinicians reported concerns about increased patient interaction time (n=9, 60%) and workflow fit (n=7, 47%). Most clinicians agreed that the PrEP advisory alert was appropriate for primary care (n=12, 78%) and supported guideline-concordant care (n=11, 73%), although fewer reported that the alert fit within clinical workflows (n=6, 40%).

Qualitative analysis of open-ended responses from 10 clinicians identified 5 themes: workflow disruption and time burden, alert fatigue, discomfort with the sexual history questionnaire, limited understanding of CDS functionality, and perceived usefulness for identifying at-risk patients. Table 2 summarizes illustrative quotes from these themes.

Our findings suggest suboptimal awareness of the CDS tool among primary care clinicians, and uptake was limited even among clinicians who reported awareness. Clinicians who used the tool generally endorsed the value of the sexual history questionnaire and PrEP advisory alert but reported concerns regarding usability and workflow integration, particularly increased time burden during patient visits. Open-ended responses revealed workflow disruption, alert fatigue, discomfort with the sexual history questionnaire, and limited understanding of tool functionality.

Our findings should be considered in the context of a key limitation. Despite a structured reminder strategy, financial incentives, and peer-led outreach, only 12% of eligible clinicians completed the survey, and thus our findings are likely sensitive to selection bias. Low response proportions are a persistent challenge when using provider surveys and may reflect competing clinical demands or survey fatigue [8]. We were unable to apply quantitative approaches to explore the effects of this selection bias [9] because we lacked data about clinicians who did not respond to the survey. Nevertheless, we likely overestimated CDS tool awareness among our primary care clinicians, and responses among tool users likely reflect the perspectives of a highly select group of early adopters, which could overestimate favorable impressions of the CDS tool compared with the overall eligible population.

Despite the key limitation, our findings align with prior studies of CDS implementation in primary care [10] that report low awareness and barriers related to workflow integration, time constraints, alert fatigue, and limited stakeholder engagement. Clinicians in our study endorsed the potential value of the CDS tool for identifying patients who may benefit from PrEP, but perceived workflow burdens may limit adoption. These findings reinforce the importance of aligning CDS tools with clinical workflows and eliciting postimplementation clinician feedback to improve the use of CDS tools.