# Smart Steps: A context-aware adherence intervention to improve PrEP adherence among men who have sex with men (MSM) with substance use disorder

> **NIH NIH DP2** · BRIGHAM AND WOMEN'S HOSPITAL · 2022 · $2,685,000

## Abstract

PROJECT SUMMARY
Adherence to once daily tenofovir disoproxil fumarate/emtricitabine as pre-exposure prophylaxis (PrEP) is
nearly 99% effective in preventing HIV infection yet adherence remains challenging especially among men who
have sex with men (MSM) with substance use disorder (SUD). MSM continue to experience disproportionate
risk of HIV; 10% of new HIV infections in the United States in 2018 were among MSM. Substance use is a
significant comorbidity and risk factor for HIV infection due to increased risky sexual activity, needle sharing,
disorganized routines and difficulty learning adherence skills. Among MSM with SUD, measuring adherence
remains critical to understanding methods in which to support PrEP adherence among at risk individuals. A
digital pill system (DPS) comprising an ingestible radiofrequency emitter integrated into a gelatin capsule which
overencapsulates PrEP can provide direct confirmation of PrEP ingestion events. The context in which these
events occur can provide important insights into the etiology of nonadherence and challenges that may be
addressed through personalized behavioral interventions. Our current work has demonstrated that MSM with
SUD will accept and operate the DPS to measure PrEP adherence and are willing to accept interventions that
account for patterns of PrEP ingestion from the digital pill. This Avenir Award takes a high risk, yet high reward
advance by developing SmartSteps, a context aware, antecedent intervention that digests PrEP adherence
data from the digital pill, combines it with digital phenotyping to anticipate potential PrEP nonadherence. In the
face of anticipated nonadherence SmartSteps will deliver empiric corrective feedback thereby mitigating PrEP
nonadherence before it occurs. This proposal will consist of two phases. In phase I, we will inform the
specification of an algorithm that utilizes smartphone digital phenotyping features (location, Bluetooth,
charging, accelerometry, app usage, phone and text usage) to anticipate PrEP nonadherence annotated by the
digital pill. We will utilize a digital phenotyping app, Beiwe, which securely and anonymously collects raw
smartphone use data. We will recruit N=40 MSM with substance use to utilize the digital pill for PrEP
adherence in concert with Beiwe over 60 days to generate a dataset that will be used to train the algorithm
behind SmartSteps. In phase II, we will conduct a pilot randomized controlled trial of SmartSteps compared to
treatment as usual among HIV negative MSM with substance use disorder with a primary outcome of
understanding the potential for an effect of SmartSteps in mitigating nonadherence and improving both overall
and weekly adherence. The overall impact of this proposal will be to develop a context aware, anticipatory
intervention that advances the science of adherence interventions in HIV prevention and SUD. SmartSteps will
also serve as an architectural framework to inform the development of similar interventions t...

## Key facts

- **NIH application ID:** 10468388
- **Project number:** 1DP2DA056107-01
- **Recipient organization:** BRIGHAM AND WOMEN'S HOSPITAL
- **Principal Investigator:** Peter R Chai
- **Activity code:** DP2 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2022
- **Award amount:** $2,685,000
- **Award type:** 1
- **Project period:** 2022-04-01 → 2026-03-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10468388

## Citation

> US National Institutes of Health, RePORTER application 10468388, Smart Steps: A context-aware adherence intervention to improve PrEP adherence among men who have sex with men (MSM) with substance use disorder (1DP2DA056107-01). Retrieved via AI Analytics 2026-05-26 from https://api.ai-analytics.org/grant/nih/10468388. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*