As planned, we identified whether or not interventions were solely focused on the prevention of alcohol or drug use; HIV, STIs and sexual risk behaviour; or mental ill health; and whether or not they had other aims such as access to HIV testing or adherence to HIV treatment. However, we ultimately developed a typology of interventions based on authors’ narrative descriptions of intervention methods because the strongest similarities and differences between interventions emerged in relation to the approaches used, and interventions addressing similar outcomes often took different approaches to doing so. We then further categorised each intervention in terms of its targeted health outcomes (Table 4). To determine targeted outcomes, we drew on authors’ descriptions of the nature, aims and outcome(s) of an intervention and the outcomes assessed in evaluations.
| Intervention category | Subcategory | Intervention name [report(s)] | Outcomes addressed | ||||
|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | Sexual health | Mental health | Substance use | ||
| Time-limited/modular | Interactive | Online modular (n = 9) | Cognitive therapy (n = 1) | Online mindfulness-based cognitive therapy (a,bAvellar96) | ✗ | ||
| Comprehensive sexual education for young people (n = 2) | myDEx (aBauermeister et al.97/a,b,cBauermeister et al.123)d | ✗ | ✗ | ✗ | |||
| Queer Sex Ed (a,bMustanski et al.111) | ✗ | ||||||
| HIV prevention/sexual health (n = 6) | China–Gate HIV Prevention Program Online Intervention (a,cCheng et al.124) | ✗ | |||||
| Hot and Safe M4M (a,cCarpenter et al.100) | ✗ | ||||||
| Keep it Up! (a,b,cMustanski et al.,122 a,bGreene et al.,103aMustanski et al.112/a,cMustanski et al.113/a,cMadkins et al.130)d | ✗ | ✗ | |||||
| MyPEEPS Mobile (aKuhns et al.129) | ✗ | ✗ | |||||
| Sexpulse (a,cRosser et al.116/aWilkerson et al.120)d | ✗ | ||||||
| WRAPP (a,bBowen et al.,98a,bBowen et al.,99a,bWilliams et al.,121a,cSchonnesson et al.117) | ✗ | ||||||
| Computer games (n = 4) | Gay Cruise (aKok et al.105) | ✗ | |||||
| Rainbow SPARX (a,bLucassen et al.108/aLucassen et al.107)d | ✗ | ||||||
| Role-playing game (aCoulter et al.126) | ✗ | ✗ | |||||
| SOLVE (a,cChristensen et al.101) | ✗ | ||||||
| Non-interactive(n = 4) | Online modular (n = 2) | Cognitive Vaccine Approach, non-tailored (a,cDavidovich et al.102)e | ✗ | ||||
| Cognitive Vaccine Approach, tailored (a,cDavidovich et al.102)e | ✗ | ||||||
| Video series (n = 2) | Sex Positive! (aHirshfield et al.104/a,cHirshfield et al.127)d | ✗ | ✗ | ✗ | |||
| People Like Us (aTan et al.131) | ✗ | ||||||
| Open-ended | Content organised by assessment (n = 2) | SMS (n = 1) | TXT-Auto (cReback et al.114/a,cReback et al.115/fReback et al.132)d | ✗ | ✗ | ||
| Static website (n = 1) | Internet-based safer sex intervention (cMilam et al.109/a,cMilam et al.110)d | ✗ | ✗ | ||||
| General content (n = 4) | Mobile multifeature app (n = 2) | HealthMindr (a,bSullivan et al.,118aJones et al.128) | ✗ | ||||
| Safe Behaviour and Screening (a,cChiou et al.125) | ✗ | ✗ | |||||
| Self-monitoring (n = 1) | Smartphone self-monitoring (a,bSwendeman et al.119) | ✗ | ✗ | ✗ | |||
| SMS (n = 1) | MOTIVES (aLinnemayr et al.106) | ✗ | ✗ | ✗ | |||
| Total (n) | 20 | 7 | 10 | ||||
Of the 23 interventions included in the review,96–132 11 addressed sexual health alone,98–102,105,111,116–118,120,121,124,128,131 two addressed mental health alone96,107,108 and none addressed substance use alone. The 10 remaining interventions addressed multiple outcomes of interest for this review: five addressed sexual health and substance use,103,109,110,112–115,122,125,129,130,132 one addressed mental health and substance use126 and four addressed all three outcomes – sexual health, mental health and substance use.97,104,106,119 In total, 20 interventions addressed sexual health,97–106,109–125,127–132 seven addressed mental health96,97,104,106–108,119,123,126,127 and 10 addressed substance use.97,103,104,106,109,110,112–115,119,122,123,125–127,129,130,132
In the case of the Keep it Up! intervention, which primarily targeted sexual health, five included study reports focused on two versions of the intervention.103,112,113,122,130 Of these, one, which reported on version 2.0, identified substance use as a secondary outcome.112 Because both this and the earlier version of Keep it Up! included content addressing substance use in the context of sexual health,122,130 we categorised Keep it Up! as addressing both sexual health and substance use.
Types of intervention methods
Interventions fell into two overarching types: time-limited/modular (n = 17)96–101,103,105,107,108,111–113,116,117,120–124,126,130 and open-ended (n = 6).102,104,106,109,110,114,115,118,119,125,127,128,131,132 Each contained intervention subtypes.
Time-limited or modular interventions
Seventeen interventions were designed as time-limited or modular interventions, guiding participants sequentially through intervention content from beginning to end.96–105,107,108,111–113,116,117,120–124,126,127,129–131 Of these, 14 addressed sexual health,97–105,111–113,116,117,120–124,127,129–131 five addressed mental health96,97,104,107,108,123,126,127 and five addressed substance use.97,103,104,112,113,122,123,126,127,129,130 These interventions could be subdivided into interactive96–101,103,105,107,108,111–113,116,117,120–124,126,130 and non-interactive interventions.102,104,127,131
Interactive interventions
The interactive time-limited/modular interventions presented opportunities for users to engage actively with the intervention via features such as activities, exercises, games, quizzes and selecting options within scripted dialogue. Interventions typically included an aspect of tailoring or personalisation, for example providing some activities tailored to the user’s readiness to change in Hot and Safe M4M,100 allowing the user to select a love- or sex-oriented movie to watch in Gay Cruise,105 providing printable feedback based on user responses in WRAPP99 and providing tailored goal-setting recommendations based on a user’s risks according to baseline assessments in Keep it Up!122 Based on author descriptions, three interventions appeared to include some tailoring of content based on individual reported needs, risks or behaviours,97,100,122 while two others referred to feedback that was tailored or personalised,99,105,121 but did not specify whether this was based on users’ needs, risks, behaviours or other factors such as participant characteristics or interaction with the intervention.
Interactive time-limited or modular interventions fell into two categories: online modular interventions and computer games.
Online modular interventions
The largest category in the typology, comprising nine of the review’s interventions, was interactive, modular programmes delivered online.96–100,103,111–113,116,117,120–124,129,130 These interventions delivered content in sequential modules via the internet, including, in one case, via e-mail.124 Among the online modular interventions, the online mindfulness-based cognitive therapy intervention delivered cognitive therapy via a modular approach for same-sex attracted men.96 Two were comprehensive sexual education interventions for young people: Queer Sex Ed targeted youth aged 16–20 years111 and myDEx targeted young adults aged 18–24 years.97,123 Six focused more narrowly on HIV prevention and sexual health.98–100,103,112,113,116,117,120–122,124,129,130
The timing of module delivery varied for interventions in this category. For example, the seven modules of Keep it Up! were delivered in three sessions with at least 24 hours between them,103,122 the online mindfulness-based cognitive therapy96 intervention was delivered as eight weekly sessions, and users of Queer Sex Ed111 completed the programme’s five modules at their own pace.
No online modular interventions targeted substance use alone, and one targeted mental health alone.96 The remaining eight targeted sexual health either alone (n = 5)98–100,111,116,117,120,121,124 or primarily but in conjunction with at least one other outcome (n = 3),97,103,112,113,122,123,129,130 making online modular interventions the most common type of sexual health intervention included in this review.
Online modular interventions took varied approaches to delivering intervention content. For example, WRAPP modules presented scripted discussions between peers, interspersed with interactive activities;99 multimedia Queer Sex Ed modules were guided by an avatar moderator and each module ended with a quiz;111 and the online mindfulness-based cognitive therapy intervention included lectures, meditation activities and homework activities.96
Computer games
Four interventions were designed as interactive computer games.101,105,107,108,126 Three (Gay Cruise,105 the role-playing game126 and SOLVE101) were sexual health101,105 or mental health/substance use126 interventions that immersed the user in a virtual environment where they interacted with one or more non-playable characters via scripted dialogue options. The fourth (Rainbow SPARX107,108) was a mental health intervention teaching CBT skills. Designed for sexual minority youth, Rainbow SPARX’s seven modules were set in a fantasy world where the user completed a mission in each level to collect a gem for their virtual shield and to progress to the next level.
Non-interactive interventions
Four time-limited or modular interventions did not appear to be interactive according to author descriptions.102,104,127,131 One subcategory of online modular interventions included both the tailored and non-tailored versions of the Cognitive Vaccine Approach,102 which targeted sexual health outcomes in the format of an eight-module online intervention. Although similar in structure to the interactive online modular interventions, the report described the delivery of programme messaging but did not refer to interactive components. The other subcategory was video series, which comprised two interventions.104,127,131 Targeting sexual health in conjunction with secondary outcomes of mental health and substance use, Sex Positive! comprised a weekly six-part video series following a gay man living with HIV.104,127 This core content was followed by four subsequent booster session videos. The People Like Us sexual health intervention was a web-based drama miniseries featuring stories of six ethnically and socioeconomically diverse gay, bisexual and queer men negotiating sexual health, mental health and relationship issues.131 Community-based organisation representatives appeared at the end of each video, summarising the episode and addressing key points about mental and sexual health.
Open-ended interventions
In contrast to the time-limited/modular interventions, six interventions appeared, according to author descriptions, to be open-ended.106,109,110,114,115,118,119,125,128,132 Rather than implying an infinite amount of content, this description merely indicates that these interventions were not designed as fixed and sequenced bodies of learning that all participants were intended to work through. All of these open-ended interventions addressed sexual health: one did so alone;118,128 three addressed sexual health and substance use;109,110,114,115,125,132 and two addressed sexual health, mental health and substance use.106,119
Two open-ended interventions were delivered via SMS.106,114,115,132 Although their content was limited (i.e. there was a finite set of SMS messages that could be delivered), we categorised these as open-ended because author descriptions suggested that they were not designed to guide participants from beginning to end through a sequential intervention. In TXT-Auto,114,115,132 users received five messages daily for 8 weeks and a weekly text-based assessment. The messages each user received were tailored to their risk profile at baseline. In the MOTIVES intervention, participants received a weekly SMS with HIV prevention information, followed 2 days later by a SMS asking a question about this information, as well as a HIV testing reminder message every 2.5 months.106
Among the open-ended interventions, two determined which core content to deliver based on a user assessment109,110,114,115,132 and four did not use this approach.106,118,119,125,132
Content organised by assessment
TXT-Auto114,115,132 and the internet-based safer sex intervention109,110 both addressed sexual health and substance use and tailored the delivery of core content based on user risk assessments. In the former, participants received a combination of general messages delivered to all users and messages tailored to their sexual health and substance use risk profile, assessed at baseline. In the latter, a user was directed to static websites with messaging tailored to their STI and HIV transmission risk, which was assessed monthly.
General content
Four open-ended interventions comprised general content delivered to all participants.106,118,119,125,128 Two addressed sexual health, mental health and substance use;106,119 one addressed sexual health and substance use;125 and one addressed sexual health alone.118,128 The format of these interventions varied. In the SMS intervention MOTIVES, based on behavioural economics, correctly recalling HIV prevention information increased users’ chances of winning a prize.106 The smartphone app HealthMindr included, among other features, resources to create a plan and schedule HIV testing, a feature for ordering condoms and HIV test kits and a quiz to self-assess pre-exposure prophylaxis (PrEP) and post-exposure prophylaxis eligibility.118,128 The features of the Safe Behaviour and Screening smartphone app for HIV prevention included a sexual behaviour and drug use log; sexual health, substance use and HIV testing information; and resources and a messaging platform to interact with other users.125 In the smartphone self-monitoring intervention, delivered via smartphone app, the user filled in daily and four-times daily surveys and used event-based reporting and text diary entries to self-monitor their behaviours.119 Although these interventions could include elements of tailoring, for example HealthMindr users received bespoke HIV prevention recommendations based on monthly risk assessments, tailoring did not determine delivery of core content.
Feedback from stakeholder consultation on typology of intervention approaches
The types of interventions described were felt, generally, to be relevant to meeting the challenges presented by the syndemic of poor sexual health, poor mental health and substance use among MSM. They were felt to be especially useful for the large group of MSM with less intense needs in these areas, for whom higher-intensity interventions, such as clinical psychology or counselling, might be unsuitable or inaccessible owing to service rationing. A minority of stakeholders felt that these interventions could be a useful supplement to face-to-face interventions for those with more intensive needs.
All stakeholders preferred interventions that provided some degree of personalisation and tailoring to the needs of individuals, ideally based on a risk assessment and providing content most relevant to their profile and/or needs. It was felt that this was critical in maintaining engagement with a diverse group of MSM who could benefit from these approaches. Interventions that were interactive and provided feedback mechanisms were favoured for increasing engagement and a sense of connection.
Open-ended interventions were favoured by some, over time-limited modular interventions, as approaches that would allow participants to select elements to work through that were most pertinent to their needs. These were perceived to increase engagement and, potentially, effectiveness. There was acknowledgement that, with these approaches, if the content were suitably engaging and the individual perceived benefits to themselves, they would probably engage with much of the content included in the intervention.
Computer game approaches were seen by most stakeholders as patronising and unsuitable, except if these were specifically targeting young MSM or were exceptionally well designed. There was significant concern that these interventions would alienate much of their target population. One stakeholder also raised a concern that computer game approaches could be unethical, as they were an incursion into spaces designed for entertainment and not usually occupied by health promotion initiatives.
Pathways from other existing services to online interventions, and to existing services from online interventions, were felt to be important; participants recommended that interventions should provide multiple referral pathways to face-to-face services, and vice versa. Without these, there was substantial concern that only the most highly motivated would access e-health interventions.