Given the continued high human immunodeficiency virus (HIV) prevalence rates among gay, bisexual, and other men who have sex with men (MSM) in North American cities, there is a critical need for HIV prevention interventions for MSM in Canada. Social anxiety, or anxiety about being evaluated in interpersonal and performance situations, is a reliable risk factor for condomless anal sex (CAS) among MSM. Social anxiety may also increase substance use in sexual situations, which is another risk factor for HIV among MSM. As such, an empirically-based social anxiety treatment may also reduce HIV risk behaviours among MSM. The present study will provide the first efficacy data for a novel and innovative HIV prevention intervention for MSM. This intervention will build upon empirically supported interventions to reduce HIV risk among MSM and therapies to reduce social anxiety. The investigators propose to test the efficacy of a novel integrated HIV prevention intervention that combines the most empirically supported treatment for social anxiety disorder, cognitive-behavioural therapy, with HIV risk reduction counselling in order to simultaneously treat social anxiety disorder, substance use disorders, and HIV sexual risk behaviour. This study will be a randomized controlled trial comparing the study intervention relative to applied relaxation, a behavioural intervention that is efficacious in treating social anxiety disorder but that does not address substance use problems or HIV sexual risk behaviours. For this trial, 176 participants will be randomized to either 12 sessions of cognitive-behavioural therapy with HIV risk reduction counselling or 12 sessions of applied relaxation.

Participants will be eligible for the trial if they are HIV-negative, report clinically significant symptoms of social anxiety disorder, substance use 2 hours before or during sexual activity, and CAS without the use of pre-exposure prophylaxis (PrEP) with a male partner who was not known to be HIV-negative. PrEP is a biomedical prevention approach in which HIV-negative individuals are provided with daily oral antiretroviral medication for the primary prevention of HIV.126 The present intervention, if found to be efficacious, is innovative in that mental health clinicians will be able to not only extend empirically supported therapies tested primarily with heterosexual populations to MSM, but they will also be able to prevent HIV through empirically supported psychotherapy practice.

Gay, bisexual, and other men who have sex with men (MSM) continue to bear a disproportionate burden of the sexually transmitted infections (STI), such as HIV in Canada, largely attributable to efficient transmission during condomless anal sex (CAS). Despite overall declines in HIV incidence since 2005, HIV incidence among MSM remains stable. In 2016, MSM accounted for 52.5% of new HIV diagnoses in Canada, compared with 56.8% in 2014. MSM also represent over half of all prevalent HIV cases. The HIV epidemic among MSM has failed to decline despite biomedically-driven efforts to reduce infectiousness through the proliferation of antiretroviral Treatment as Prevention, and the emergence of HIV Pre-Exposure Prophylaxis (PrEP) to reduce HIV susceptibility. The epidemiologic data underscore the need for efficacious interventions for reducing sexual risk behaviour among MSM. The Public Health Agency of Canada (PHAC) has therefore identified MSM as a high priority population to reach with prevention initiatives.

Social anxiety is a risk factor for CAS among MSM. It is typically defined as the experience of fear of and apprehension about being evaluated in interpersonal situations, including romantic or sexual relationships. Social anxiety is higher among gay men and other sexual minority populations than among heterosexuals on both dimensional measures and categorical measures, with recent studies reporting the prevalence of social anxiety disorders among MSM as 22.3-22.9%.

Social anxiety predicts risky sex among MSM. Research suggests a direct relation between social anxiety and CAS among MSM. Social anxiety was associated with a greater proportion of receptive and insertive CAS even after adjusting for condom use negotiation, depression, and club drug use. Social anxiety may also increase substance use in sexual situations, which is another risk factor for HIV among MSM. However, no studies have examined the efficacy of social anxiety treatment on CAS among MSM. Social anxiety is highly modifiable via cognitive-behavioural therapy (CBT).

Fears of being sexually rejected by one's partner fully accounted for the association between social anxiety at baseline and CAS at 6-month follow-up. This study is consistent with previous findings that 32% of gay men reported being less likely to use condoms when they were concerned their partner would react negatively. Substance use may be another mediator. Social anxiety is also a risk factor for later alcohol use and other substance use problems in clinical and nonclinical samples, including among MSM. Substance use tends to follow social anxiety disorder because alcohol and other substances are used to self-medicate anxiety symptoms in social situations. Alcohol and substance use in sexual situations is a relatively consistent risk factor for CAS among Canadian MSM, as well as higher HIV incidence. Due to consistent data linking substance use to HIV risks, it has been suggested that incorporating alcohol and substance use treatment into sexual risk reduction counselling may increase the efficacy of HIV prevention efforts for MSM. Social anxiety may have a specific relationship with substance use in creating risk for HIV and other STIs.

The present study will provide efficacy data for a novel and innovative STI/HIV prevention intervention for MSM. This intervention, called Sexual Confidence, builds upon empirically supported treatments for social anxiety, including social anxiety-related substance use, by adding risk reduction counselling to reduce STI/HIV sexual risk behaviour.

Over the past 10 years, numerous well-designed clinical trials have established the efficacy of behavioural interventions to reduce HIV risk behaviour among HIV-negative individuals. However, the success of these programs is limited, and evidence suggests these effects diminish over time. Improving the efficacy of these programs is a public health priority.

The current proposal seeks to reach individuals who are at higher risk for engaging in continued HIV risk behaviour due to the presence of social anxiety and substance use in sexual situations, and who may have difficulty responding to an intervention that targets sexual risk taking alone. Research from multiple countries suggests that all MSM are not equally at risk for HIV. A preponderance of data indicates the incremental presence of co-occurring mental health and substance use problems, called syndemic, predict both HIV prevalence and HIV sexual risk behaviour among MSM. One of the largest studies of MSM from 151 countries reported a strong dose-response relationship with increasing odds of HIV infection and greater number of syndemic problems. Compared with those without syndemic problems, the odds of HIV infection among those with 1, 2 and 3 or more syndemic problems were 1.67, 2.02 and 2.35 times greater respectively. The MSM in our study may account for almost half (22% of MSM who have double the risk for contracting HIV) of the HIV epidemic in Canada and beyond. Our team and others in sexual health promotion and HIV prevention have found that untreated mental health issues interfere with safer sexual practices. In order to adequately meet the HIV prevention needs of MSM, HIV prevention interventions need to address the mental health problems that contribute to increased risk in this group and that impede the efficacy of traditional HIV prevention efforts.

Meta-analyses suggest that behavioural interventions can be efficacious at reducing self-reported CAS among MSM. Other reviews found greater efficacy with active interventions rather than interventions where participants just receive information. In accordance with these findings, the Sexual Confidence program includes actively role-playing situations in which participants practice their safer sex negotiation skills. Given the lack of efficacious interventions designed in Canada, there is a need for rigorous empirical testing of the effects of HIV prevention interventions for Canadian MSM.

Rates of syphilis, chlamydia and gonorrhea have risen among MSM, especially among HIV+ MSM living in the Canadian urban centers including Toronto. Co-infection with HIV is particularly concerning as CAS with HIV+ MSM can increase likelihood of transmission. A cohort study from the Ontario HIV Treatment Network reported a 12-month cumulative incidence of 1.7% (95% CI 1.1% to 2.2%) for chlamydia/gonorrhea and having CAS with multiple HIV+ partners (HR=3.3, 95% CI 1.4 to 7.8) was a strong predictor of the infection.

CBT focuses on reducing psychological distress and the psychosocial interference of anxiety, substance use, and other psychological problems using cognitive restructuring techniques (i.e., teaching people to think more realistically about the situations they fear) and by repeated exposure to feared situations. CBT promotes learning adaptive responses to feared stimuli and is hypothesized to facilitate cognitive changes. Although behavioural therapies such as applied relaxation (AR) and exposure therapy are also brief and efficacious interventions, only CBT has evidence supporting its use when social anxiety disorder and substance use problems are comorbid.

CBT has proven to be acceptable among diverse MSM populations. A manualized format, the availability of training, and the presence of competence and adherence rating scales are additional characteristics that make CBT a feasible modality for HIV prevention programming. Lastly, the conceptual underpinnings for CBT (social cognitive theory) are shared by most other research-supported HIV risk reduction interventions.

The current state of the science identifies behavioural therapies, including CBT, as the psychosocial treatment of choice for social anxiety disorder. A review of 32 RCTs reported that the average CBT treated participant scored better than 80% of the waitlist and 66% of the placebo participants, and patients continue to increase the magnitude of their treatment gains over time. Pharmacotherapy treatments fare as well as CBT although pharmacotherapy often requires continuing treatment indefinitely to prevent relapse, whereas CBT has a finite number of sessions and is associated with strong maintenance of treatment gains.

Our main objective is to test the efficacy of this integrated program in reducing sexual risk behaviour using a 2-armed randomized controlled trial (RCT). Secondarily, we will compare the social anxiety, substance use in sexual situations, and the cumulative incidence of bacterial STIs between the two arms over a 9-month follow-up period. We hypothesize that, relative to the control condition, Sexual Confidence will reduce sexual risk behaviour, as defined by CAS with HIV-positive or unknown HIV status sexual partners without the use of PrEP. We further hypothesize that Sexual Confidence will result in greater reductions in social anxiety and substance use in sexual situations.

Our proposed research is a single blind, two arm randomized trial. Participants will be randomly assigned to the CBT or AR arm. Counsellors cannot be blinded as they are providing treatment, but assessors will be blinded to treatment condition. Participants will not be provided with details on the treatment condition to which they were not assigned to minimize the possibility that participant beliefs about which condition is better will impact outcomes.

Participants will be randomly assigned to CBT, the experimental arm, or the AR arm, which serves as the control condition. Both arms are active treatment conditions; however, we hypothesize that the Sexual Confidence CBT arm will have a significantly larger effect on sexual risk behaviour than the AR arm as AR does not discuss or address substance use, sexual risk reduction, or sexual behaviour.

We propose a conceptual model for the relationship between social anxiety, substance use, and sexual risk behaviour that purports the experience of social anxiety symptoms in sexual situations generates unwanted anxious affect (e.g., fear of sexual rejection, fear of negative evaluation), which individuals are motivated to avoid. Heavy alcohol and drug consumption before or during sexual activity is the behavioural mechanism in the model in this study for avoiding the anxiety associated with social interactions. In this way, using substances to avoid socially anxious thoughts leads to risky sexual behaviour both directly and indirectly through impaired safer sex negotiation.

These mechanisms are consistent with cognitive-behavioural theory. Behaviourally, substance use (and potentially risky sexual behaviour) is negatively reinforced through the avoidance of (or escape from) unwanted anxious affect. Cognitively, socially anxious predictions generate unrealistic risk appraisals of rejection that contribute to sexual risk. Sexual Confidence is designed to address these specified pathways: (a) planned exposures to learn adaptive strategies for reducing social anxiety, avoidance-motivated substance use, and avoidance-motivated sexual risk, (b) cognitive restructuring to train realistic cognitive appraisals of rejection for negotiating safer sex, and (c) risk reduction counselling to increase safer sex negotiation skills and to specify safer sex behaviour change targets.

During COVID 19, participants may not be able to attend in-person sessions for assessments and therapy. In order to reduce the risk of COVID-19 transmission among study participants, and to address potential participant reluctance to attend assessment and therapy sessions in person, we have created an online, or teletherapy, version of the study using Zoom for Healthcare. Teletherapy will be used for all assessment and therapy sessions. Participants who do not have access to a private location and/or the necessary technology to safely utilize Zoom will be offered a private room with a computer in order to utilize teletherapy from our research lab.

Inclusion Criteria:

• Participants must report engaging in CAS sex without PrEP with a man of HIV-positive or unknown HIV status within the last 3 months and report using substances within 2 hours before and/or during sexual activities. Participants must self-identify as a man (including trans* men who self-identify as men), and be aged 18 or older.

To minimize bias, participants will be asked to report to the research study clinician in case they decide to engage in other psychotherapies outside of the study protocol or decide to change their psychoactive medication or dose for the duration of their participation in the study. Participants who participate in outside psychotherapies or who change their medication regimens will still be permitted to conclude the therapy, but their data will not be included in the RCT analyses.

Exclusion Criteria:

• As mentioned previously, participants taking PrEP will be excluded from participating. This excludes individuals 1) on daily PrEP dosing, with no more then 3 doses missed per week at the time of exposure and 2) on-demand dosing at the time of the exposure, including at least one dose before and after exposure. Participants need to score within 1 SD of the clinical mean (M = 67.2) for social anxiety disorder on the Liebowitz Social Anxiety Scale.
• Persons will be excluded if our assessors/counsellors find that a participant's ability to respond to study measures is compromised by mental or physical disabilities or inability to speak and understand English.

Gay and bisexual men and other men who have sex with men (including trans men)

Counsellors cannot be blinded as they are providing treatment, however assessors will be blinded to treatment condition. Participants will not be provided with details on the treatment condition to which they were not assigned to minimize the possibility that participant beliefs about which condition is better will impact outcomes.