Chemsex is a growing public health concern associated with striking increases in HIV and STIs, particularly in communities of men who have sex with men (MSM).1-9 As in the Global North, this phenomenon is steeply rising across Southeast Asia, including Malaysia.10 Unfortunately, research on intervention options for chemsex has been severely limited. In the absence of evidence-based interventions for chemsex, a harm reduction approach (e.g., providing information and prevention resources, HIV prevention services, such as HIV testing and pre-exposure prophylaxis, and counseling) remains a cornerstone to reducing chemsex-associated harms.11,12 However, many unmet needs exist for HIV prevention and chemsex-related harm reduction services tailored for MSM in Malaysia and elsewhere.13-17 Ideally, chemsex support should be delivered when people need and want it,11,18 indicating the need for interventions to be adapted to users varying needs. Fortunately, advancement in technology has enabled the delivery of interventions in real time during the moments of need also known as just-in-time adaptive intervention (JITAI).19-25 Ecological momentary assessment (EMA), a gold standard for the assessment of in situ self-reports, has been an integral to the delivery of JITAI and remains the most accurate way to assess and predict risk behaviors in natural settings. Of relevance, JITAIs, coupled with EMA, have been successfully deployed for predicting and reducing mental health symptoms, sexual risk behaviors, and substance use, and thus, hold great promise to bridge the gap in HIV prevention efforts and mitigate chemsex-associated harms, particularly in low-resource settings, like Malaysia. Though effective, EMA heavily relies on user input and is subject to a degree of recall bias,26 thus, risking lack of user engagement with the intervention over time. Notably, the recent emergence of digital phenotyping27-30 – a moment-by-moment, in-situ quantification of individual-level data from personal digital devices – allows us to innovate JITAI further. Particularly, in the fields of chemsex and HIV, unobtrusively collected digital phenotype data (e.g., location via GPS, body motion via accelerometer, social dynamics), coupled with machine learning, could facilitate detection of HIV risk behaviors (e.g., chemsex). In turn, this can be used to trigger and deliver tailored antecedent chemsex intervention via the app. While DP has its merits (e.g., objective, continuous, low user burden), its application is still in its infancy. This indicates the need for additional research to develop and optimize machine learning algorithms capable of predicting stimulant use and seamlessly integrate with mHealth platform to deliver JITAI. In response to Notice of Special Interest (NOSI; NOT-DA-24-002), we therefore propose JomPrEP+, an app-based Mobile detection and intervention system (MDIS) that delivers JITAI based on identified moments of risk to prevent HIV transmission and mitigate chemsex-associated harms. The EMA-based JITAI incorporates several components, including: i) push interventions, in which reminders or prompts are delivered when the system detects states of risk or opportunity based on users response to EMA; and ii) pull interventions, which includes several on-demand features available for the users to access on the app (i.e., HIV self-testing, PrEP, eCounseling, safer chemsex package, tracking of medicine and drug use, multimedia educational materials). Our approach builds on extensive preliminary studies demonstrating the feasibility and promise of these strategies, including leveraging programming architecture from an earlier version of JomPrEP+,31 an eCounseling intervention for chemsex support,32 a safer chemsex package (PartyPack”),33 and the utility of Beiwe,34 a digital phenotyping (DP) tool to fully automate the delivery of JITAI with minimal user burden.35 To date, however, no studies have examined the impact of incorpor
Funding:
University of Connecticut
Funding Period:
07/15/2024 to 04/30/2029