ARemind: A Personalized System to Remind for Adherence

The specific study aims are described below:

• Organize the Design, Evaluation, and Implementation Teams: The development of ARemind will be guided by design, evaluation and implementation teams. The design team will build the software architecture and execute the software development with inputs from the user studies. The evaluation team will prepare the institutional review board (IRB) submissions, develop the qualitative instruments, and analyze the qualitative and quantitative data. The implementation team will guide clinical and commercial design aspects including input on the remote adherence assessments, the intervention and clinic appointment modules, and the commercialization plan.
• Run first user study as a qualitative assessment of patient and provider preferences of ARemind: A web-based adherence report for providers will be developed. Software modules to assess adherence over the phone through 4-day SMS or IVR recalls will be created along with a protocol for performing phone-based unannounced pill counts. The interfaces will be tested as a part of in-depth qualitative semi-structured interviews with 15 providers and 15 patients to guide the system design.
• Study the content preferences through a second user study: A new version of ARemind (version 2.0) will be built based on the feedback from the first user study. It will include the three remote adherence assessment modules. It will be tested longitudinally for 3 weeks on 15 patients with HIV/AIDS. The patients will use Medication Event Monitoring System (MEMS) software, in the form of MEMS caps, and have weekly clinic visits with adherence counselors who will generate a composite adherence score at each visit. MEMS caps are electronic devices attached to a pill bottle which detect and record the time at which the bottles are opened and closed, using this as an estimate for the number of pills taken. These will be compared to the adherence scores from the remote adherence assessment modules. The patients will give qualitative feedback at each visit on the reminder messages to guide the design of content resistant to user fatigue.
• Explore the feasibility of ARemind through a third user study: An intervention module with an analytics engine will be built. It will personalize reminder messages based on adherence levels and facilitate patient phone calls with social workers/adherence counselors when appropriate. A clinic appointment reminder module will be created and interfaced with the health information exchange of Boston Medical Center (BMC). A new version of ARemind (version 3.0) will contain these new modules and content improved based upon the second study. ARemind 3.0 will be tested through a 24-week longitudinal study of 70 patients. Thirty five patients will be randomized to use cell phones with ARemind 3.0 (intervention) while 35 patients will use beepers (control). Patients will use MEMS caps for composite adherence scores that will be done at each of their 9 adherence visits during the study. Analysis of feasibility, efficacy, user-fatigue and cost-savings from improved adherence to medications and clinic appointments will be performed in this aim.

NOTE: OF THESE THREE USER STUDIES, THE FIRST TWO ARE CONSIDERED FEASIBILITY STUDIES WHILE THE LAST IS A MORE LONGITUDINAL CLINICAL STUDY. THIS REGISTRATION IN CLINICALTRIALS.GOV DEALS WITH THE THIRD USER STUDY, WHICH IS THE LONGEST, INVOLVES THE MOST PARTICIPANTS, AND DEMONSTRATES THE HIGHEST NUMBER OF INTERVENTIONS.

Inclusion criteria

• Stable ART (no change of ART for 3 months)
• Greater than 18 years of age
• Self-report adherence < 85%

Exclusion criteria

• HIV-infected patients not on ART
• Non-English speaking
• Dementia (via mini mental status exam)
• Incarceration
• Legally blind and./or deaf
• User of pillboxes