Optimizing HIV PrEP Through Shared Decision Making

The purpose of this study is to assess if a clinical Decision Aid (DA) that integrates a decision-support tool for patients and a risk-prediction tool for providers can facilitate shared decision making (SDM) and reduce decisional conflict regarding pre-exposure prophylaxis (PrEP) use. The patient tool was developed through an iterative process using the Ottawa Decision Support Framework (ODSF). The provider tool is a validated model that uses patient-reported data to estimate HIV risk for individual men who have sex with men (MSM). The DA includes information on the harms associated with HIV, a list of treatment options, which includes doing nothing, a patient's individualized risk of acquiring HIV, positive and negative features of PrEP, exercises to clarify patient values, and interactive exercises to help patients weight the risks and benefits of PrEP use. The Health Communications Core at the Dana-Farber Cancer Center, a Core that receives NIH funding to provide web-design services to Harvard-affiliated investigators, is building a web-based version of the Decision Aid under the investigator's direction, as the investigators have learned from qualitative research that patients and providers would prefer a web-based version. Presentation of the DA's content will be done in a manner to enhance comprehension and minimize bias.

The DA will be tested with 20 patient-provider dyads to determine acceptability, and then an additional 60 provider-dyad visits will be conducted. The study will involve Boston-area MSM and primary care providers at Beth Israel Deaconess Medical Center (BIDMC) and Fenway Health. The investigators believe it will be important to capture the perspectives of clinicians at Fenway (which specializes in the care of sexual and gender minorities), BIDMC (a large, academic medical center), as well as those who practice at other clinical sites in the community who may be less knowledgeable about and experienced with PrEP.

To pilot test the DA, MSM-provider dyads will be scheduled for a simulated 20 minute clinic visit at BIDMC or Fenway Health. Providers will complete a verbal consent procedure prior to the visit. Providers will be emailed the Decision Aid for review, as well as a pre-study questionnaire to complete on their own time.MSM will arrive 60 minutes before the visit. At this time, the MSM participant will complete an informed consent procedure followed by a brief demographic and behavioral survey. The survey will be administered using either pen and paper or a web-based format. This survey will be reviewed by a study staff member to confirm eligibility. MSM Participants will then have the opportunity to view the DA before meeting with providers. Using a computer or touch screen tablet, both the participant and provider will complete the DA, which will be online, and enter the patient-reported data that are needed for the risk-prediction tool. The website will not store any answers to the questions in the DA. The questions that are used to obtain these sensitive data will be presented in a culturally-sensitive manner. Both MSM participant and provider will be able to explore the DA website, which will include information about HIV, PrEP and next steps for providers that are interested in prescribing PrEP. After the visit, MSM and providers will complete acceptability surveys and brief exit interviews with a member of the study staff. Acceptability will be pre-specified as >70% of participants and providers indicating that the DA is acceptable for use. Based on the surveys and feedback, the investigators will refine the DA, website content, and visit process in preparation for testing with an additional 60 dyads. If the MSM participant is interested in learning more about the study, staff will provide instructions to access information about PrEP on the Fenway Institute's website. If a participant would also like to speak to a healthcare professional about obtaining a prescription for PrEP, the study staff will recommend scheduling a visit with a provider at Fenway Health or the BIDMC Infectious Disease clinic.

Once acceptability has been established, and additional 60 MSM-provider dyads will be scheduled either for a simulated or in-clinic 20 minute visit at BIDMC or Fenway Health, depending on availability of providers during clinic hours. These visits will follow the same protocol as the pilot testing visits, with possible adjustments made based on pilot testing feedback. In addition to the initial study visit, there will be a 3 month follow up survey provided to MSM participants. This survey will be available online through RedCap, and the link to complete the survey will be emailed to MSM participants. Alternatively, participants can complete these surveys in person or by phone according to their preferences.

Assuming a standard deviation of 12 for both pre- and post-DA primary outcome measures of Decisional Conflict Scale scores, a null hypothesis of no difference in pairs pre- and post-DA, a 2-sided significance level of 0.05, and correlation of 0.23 between pre- and post-DA scores, and adjusting sample size for non-parametric test (using worst case of 86.5% efficiency) in case of non-normal data, a sample size of 60 MSM (each with a pair of evaluations pre- and post-DA) will provide approximately 80% power to detect a difference between pre- and post-DA scores of 6 (~1/2 SD).

Acceptability surveys will be analyzed with descriptive statistics to determine the percentage of participants who find the DA to be acceptable. The paired t-test (or Wilcoxon Signed Rank test if data are non-normal) will be used to determine if there are changes in the following measures pre- and post-use of the DA: (1) MSM Decisional Conflict Scale scores; (2) MSM and provider knowledge; (3) intentions to initiate PrEP; (4) decisional regret; and (5) sexual risk behaviors. The proportion of MSM reporting HIV or sexually transmitted disease testing in the 3 months prior to use of the DA will be compared to the proportion who report testing in the 3 months after use of the DA using a paired (1-sample) exact test of rate/proportion against the null value of 0. Kruskal-Wallis tests (multi-group extension to Wilcoxon Rank Sum Test) will be used to determine if the distribution of post-DA Decisional Conflict Scale scores differs between/among groups defined by level of adherence.