This project was approved by the University of Pennsylvania, the Philadelphia Department of Public Health, and the School District of Philadelphia. The primary study site was the University of Pennsylvania. Recruitment took place at public locations in the city of Philadelphia (e.g., recreational centers, community events), Community Behavioral Health (CBH) provider sites, and schools in the School District of Philadelphia.
Phase 1: Team-Building and Elicitation Research (Study aim #1; N = 53). Team building and community engagement were an ongoing process throughout the course of the study. The investigators initially focused on team-building and strengthening ties with collaborators, securing buy-in from key community stakeholders and providers, recruiting 8 members for a youth CAB, and hiring and training project staff. The investigators were committed to the development of an acceptable, practical intervention which could easily be implemented and sustained in the targeted community setting. Therefore, continual feedback from Black adolescents with mental illnesses, CBH providers and research staff was pivotal. Further, Black adolescents in the target demographic and young adults from the community were hired as members of the study team in a capacity building approach to create a pipeline for future HIV/AIDS research investigators and community workers. The investigators also built upon their existent elicitation research with members of the study population. Four mixed gender focus groups were conducted with concurrent open-ended questionnaire administration with Black adolescent males (n= 18) and females (n= 15). The team held meetings with CBH providers and the youth CAB to determine factors associated with HIV/STI risk related sexual behaviors in the population, and the role of psychiatric symptoms, the social determinants of health, gender and culture in sexual risk taking among Black adolescents with mental illnesses. The baseline intervention questionnaire was also piloted with Black adolescent males (n= 10) and females (n= 10) for accuracy; revisions were made based on participant suggestions. Psychometrics of the revised instrument were assessed and results compared to the original version. The qualitative and quantitative data were examined concurrently to explore participants' behavioral, normative and control beliefs, as well as contextual factors, as they relate to HIV/STI risk. Subsequently, this knowledge was used in the design of the targeted intervention.
Phase 2: Intervention Development and Pre-Pilot Testing (Study aim #2; N = 29). The investigators used an iterative process to develop the targeted intervention. They sought feedback from their elicitation participants, youth CAB, and CBH providers at each stage of intervention development. This input was used to ensure accurate interpretation of the study findings, as well as to refine the intervention for relevance and sustainability. The primary outcome of these processes was the development of "Project GOLD", a theoretically-driven, gender and culturally relevant, developmentally and psychologically appropriate behavioral intervention to reduce HIV/STI risk among heterosexually-active Black adolescents receiving outpatient mental health treatment. The intervention name was created by the youth CAB, along with the project logo. The intent in this branding was to foster gender and cultural pride through visually and verbally reinforcing the intervention's tagline: "We are Kings and Queens who make confident and royal decisions". The first pre-pilot testing, or dress rehearsal, of the interventions and all study materials and instruments was conducted with a sample of Black adolescent males (n= 4) and females (n= 4). Participants completed pre- and post-intervention assessments, and the team debriefed with them to solicit their feedback on ways to improve the project. Revisions were made and a new, independent sample of Black adolescent males (n= 3) and females (n= 4) completed the second pre-intervention pilot. The investigators completed the same process for the general health intervention (N = 14). The goals of the intervention pilots were to evaluate whether the interventions were likely to affect the conceptual variables, to evaluate whether the participants viewed the interventions as valuable and engaging, and to ensure the study protocols were relevant to achieve the aims.
Phase 3: Intervention Pilot (Study aims #3 \& 4; N = 108). Based on the Phase 2 findings, the final versions of the intervention protocols, materials and instruments were developed and sent for printing. The intervention was piloted with 108 youth; 109 enrolled but one participant was terminated at the baseline visit for eminent suicide risk. Participants were randomized to either Project GOLD (n = 52) or the general health promotion control arm (n = 56). There were five RCT visits: baseline, immediate post, and 3-, 6-, and 12-month follow-up.
RCT Sample, Recruitment and Enrollment. Provider referrals, waiting room encounters, flyers, social media, and teen-focused community events and locations (e.g., talent showcases and game nights; recreational centers) were used to inform potential participants about the study. As part of the investigators' community engagement mission, the research team hosted programs and/or partnered with local community-based organizations to promote HIV/STI awareness and encourage testing among youth. In an effort to increase study enrollment, staff members also screened interested youth for eligibility at these events. The programs took place in reserved campus space at the University of Pennsylvania, or a public community space as approved (e.g., local YMCA or recreation center). Eligibility was determined using a structured screening script either in person or by telephone. All study visits took place at one of the sites where participants were recruited, at the University of Pennsylvania in a private conference room, or in a private conference room at a community-based organization near where participants are recruited (e.g., recreation center). No data collection occurred at any School District schools or property as a condition of their IRB approval. For the RCT, participants registered with the study coordinator, were assessed by a research team member using an electronic version of the Patient Health Questionnaire-9 (PHQ-9), provided a urine specimen, had their mouths swabbed for the rapid HIV test, and completed the self-administered sexual risk behavior questionnaire in a private, quiet room.
Assent versus Consent. In the state of Pennsylvania, adolescents between the ages of 14 to 17 can autonomously make decisions about their sexual and mental health assessment and treatment. "…\[I\]f a minor is able to consent to medical, dental and health services under state law, the minor is also able to consent to take part in medical research" (pg. 9). Therefore, youth aged 14 to 17 provided consent to participate in the research without parental permission. Initial disagreements among the approving IRBs, however, lead to substantial protocol delays. The investigators agreed to exclude (or withdraw from participation) any adolescent whose parent/guardian is aware of/finds out about the study and indicates that they do not want their son or daughter to participate in the research. This was honored if a participant shared that his or her parent/guardian expressed to them that they did not want them in the study, or if the parent/guardian directly contacted a member of the research team to state they did not want their son or daughter to be involved.
Procedures for Consent/Assent \& Diagnostic Assessment. Adolescents who were preliminarily eligible were scheduled for their first study visit to undergo a 30-minute diagnostic assessment using the MINI to ascertain psychiatric diagnosis and rule out exclusionary disorders. Trained study personnel conducted all of the diagnostic assessments. Based on the diagnostic assessment, adolescents were excluded if they: 1) had a psychotic disorder, or disorder with psychosis, 2) were cognitively impaired, or 3) their current treatment required hospitalization. Eligible participants immediately participated in a brief, 10-minute structured demographic interview. The interview was used to obtain locator information to follow participants over the course of the study, as well as to elicit information regarding referral source, demographics, and clinical data (i.e. psychiatric and medical history). The structured diagnostic and demographic interviews took approximately 40 minutes. At the end of the structured interviews, participants were scheduled for their second study visit which included the baseline assessment and first intervention session.
Procedures for Enrollment, Randomization \& Intervention Trial. At the second study visit, participants completed the electronic baseline assessments (Intervention Baseline Questionnaire which includes the Background Information Questionnaire and PHQ-9) and provided urine specimens for STI testing. When the urine specimens were obtained, all participants signed a medical release form so that the study's Nurse Practitioner could receive the STI test results and confirm treatment for positive results with the health centers or chosen providers. They then immediately attended the first intervention session following these procedures. Randomization to either the targeted intervention or treatment control group occurred at the diagnostic visit. Where feasible, intervention and control group sessions ran concurrently at the same site; if the sessions could not be run on the same day, intervention and control group sessions were delivered within the same week at each site, or as close as possible thereafter. Participants attended the second intervention session (their third study visit) two weeks after the first intervention session. Immediately following the second/final intervention session they completed post-intervention measures, had their mouths swabbed for a rapid HIV test, and received their HIV/STI results. In accordance with the CDC's Revised Recommendations for HIV Testing of Adults, Adolescents, and Pregnant Women in Health Care Settings, HIV screening was provided for all participants unless a participant expressly refused to take an HIV test (59.3% consented to HIV testing). Thus, study participants were offered HIV testing but not required to consent to testing to participate in the study. Participants also returned for post-intervention follow-up visits 3-, 6-, and 12-months after the date of their final intervention session. At each study visit, participants registered with the study coordinator, were screened with the electronic PHQ-9 and had all appropriate labs collected. They were then seated in a quiet room and provided with the electronic, self-administered Background Information Questionnaire.
No adolescents tested positive for HIV over the course of the study. Those who tested positive for another STI were permitted to stay in the study. The study Nurse Practitioner notified participants of their baseline STI results at the end of the final intervention session, and participants received subsequent notifications by phone within 2-weeks after each of the 3 follow-up time points. Individuals who tested positive for an STI were referred to receive free treatment through the Philadelphia Health Department at the designated STI control health centers (Health Centers #1 \[500 S. Broad St.\] and #5 \[1900 N. 20th St.\]). Individuals who tested positive for an STI and fail to be located, did not comply with returning to the study site, or who failed to seek clinical treatment through an alternative provider, were reported to the City of Philadelphia Department of Public Health, as required by the City of Philadelphia.
