This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Formative Research, is properly cited. The complete bibliographic information, a link to the original publication on https://formative.jmir.org, as well as this copyright and license information must be included.
Digital pill systems (DPSs), which comprise ingestible radiofrequency sensors integrated into a gelatin capsule that overencapsulates a medication, can directly measure ingestion events.
Teaching users to operate a DPS is vital to ensure the collection of actionable ingestion and adherence data. In this study, we aim to develop and pilot a training program, grounded in the Technology Acceptance Model, to instruct individuals on DPS operation.
A two-part training program, comprising in-person and text message–based components, was used with HIV-negative men who have sex with men with nonalcohol substance use, who had enrolled in a 90-day pilot demonstration study using the DPS to measure adherence to pre-exposure prophylaxis. We assessed the number of responses to text check-ins, the number and types of episodes where technical support was requested, the resolutions of such issues, and engagement with the program over the study period. Participant feedback on the program was evaluated through qualitative user experience interviews.
A total of 15 participants were enrolled in and completed the program. Seven technical challenges related to DPS operations were reported across 5 participants. Most commonly, participants requested support connecting the wearable Reader device with their smartphone, charging the Reader, and operating the mobile app. A total of 6 issues were resolved asynchronously or in real time via phone; 1 required in-person evaluation and resolution. Preliminary qualitative findings indicate that both the in-person and remote follow-up components of the training program were perceived as acceptable. Suggested improvements included repeated DPS refresher sessions at in-person follow-up visits and enhanced written materials for the independent resolution of technological issues.
A brief two-part DPS training program, drawing from individuals’ experiences and from the Technology Acceptance Model, can provide valuable insights for users. The program also identifies and addresses several areas of actual or potential challenges related to operating a DPS and allows for the resolution of such issues within the first week of DPS use.
Digital pill systems (DPSs) use ingestible electronic sensors incorporated into gelatin capsules that overencapsulate medications. Ingestible sensors are activated upon entry into the stomach and exposure to gastrointestinal fluid, transmitting key data about medication ingestion patterns in real time. These data can be viewed by clinicians and patients via web-based platforms and smartphone apps. The provision of such feedback linked to medication ingestion patterns represents a novel, cybernetic closed-loop system that allows users to consider their medication adherence in light of digital pill–recorded data [
For men who have sex with men (MSM) with concomitant substance use, the use of once-daily tenofovir disoproxil fumarate/emtricitabine as pre-exposure prophylaxis (PrEP) for HIV prevention has been demonstrated to be highly efficacious; however, PrEP efficacy remains closely linked to adherence [
Ensuring consistent use of the DPS is vital to obtain actionable medication ingestion data that would enable real-time adherence interventions. Although several clinical trials have demonstrated the feasibility and acceptability of deploying a DPS to measure medication adherence, barriers to the adoption of this technology may exist [
Our team conducted qualitative exploration of these 3 TAM constructs with HIV-negative MSM who use substances, which found that such individuals are accepting the idea of using a DPS to measure adherence to once-daily PrEP and are willing to engage with the technology [
Accordingly, we developed a brief training program with features addressing the 3 pillars of the TAM to teach individuals how to operate and incorporate the DPS into their daily routine. Our goal is to design a program that could first be taught to individuals who were naïve to the DPS and subsequently be delivered to DPS users by these individuals once they obtained a basic understanding of the DPS, to maximize its scalability and use for individuals using a DPS. We piloted this training program as part of a clinical trial that deployed a DPS to measure PrEP adherence over a 90-day period in HIV-negative MSM with nonalcohol substance use. Finally, we aim to solicit qualitative feedback via user experience interviews to understand the experiences of individuals who participated in the DPS training program.
The parent study consisted of a 90-day, open-label demonstration trial to evaluate the feasibility and acceptability of the DPS (ID Cap System, etectRx Inc) to measure PrEP adherence, in which digital pills for PrEP were deployed among HIV-negative MSM (N=15) older than 18 years who self-reported nonalcohol substance use (NCT03842436). Participants ingested digital pills once daily for 90 days and attended 5 study visits over a period of 3-4 months. In addition to DPS technology training, study visits consisted of laboratory work (to confirm eligibility for PrEP), a brief quantitative assessment, pill counts, reviews of DPS adherence data, blood draws to evaluate PrEP adherence, and provision of three 30-day supplies of digital pills. As part of this parent study, we developed a training program grounded in our initial qualitative work to ensure participants’ successful operation of the DPS over the course of the study, which is the focus of this study, and all adherence metrics from participants’ use of the DPS during the parent study will be reported elsewhere.
A total of 15 participants enrolled in the parent study, a sample size that corresponded to the preliminary pilot nature of the study and our aim to demonstrate the feasibility and acceptability of the technology. All 15 participants were enrolled in the DPS training program. Participants met the following inclusion criteria: 18 years or older, cisgender MSM, self-reported use of nonalcohol substances in the past 6 months, currently taking PrEP, and qualifying laboratory tests for the use of PrEP (negative rapid HIV test, creatinine clearance ≥60 mL/min, evidence of hepatitis B immunization, and screening for sexually transmitted infections). Individuals were excluded if they did not speak English; self-reported living with HIV; were identified as transgender; had an estimated creatinine clearance <60 mL/min; were receiving active hepatitis B treatment; were taking proton pump inhibitors; had a history of Crohn disease or ulcerative colitis; had a history of bowel surgery, gastric bypass, or bowel stricture; had a history of gastrointestinal malignancy or radiation to the abdomen; did not own a smartphone; or were unable or unwilling to ingest the digital pill. This study was approved by the Fenway Health Institutional Review Board. A parent study was conducted between March 2019 and March 2020.
The DPS training program consisted of 2 components: an in-person training session during the study enrollment visit and a remote follow-up period of text message–based exchanges for 6 days after enrollment (
Overall structure of the digital pill system training program in-person session and remote follow-up. DPS: digital pill system.
Before the in-person session, the study team registered each participant on the web-based DPS interface, electronically assigned a Reader device to them, and prepared a prescription for their initial 30-day supply of digital pills. Each digital pill consisted of a capsule dosage form containing an ingestible sensor and an emtricitabine 200 mg/tenofovir disoproxil fumarate 300 mg tablet for PrEP, with instructions to take one capsule once daily (
Components of the digital pill system. The digital pill system comprises an ingestible electronic sensor integrated into a gelatin capsule overencapsulating tenofovir disoproxil fumarate/emtricitabine (Truvada) as a digital pill (A). Upon ingestion, the digital pill is activated and emits a radiofrequency signal that is transmitted to a wearable Reader device (B). Ingestion data then flow from the Reader device to a smartphone app (C).
Next, participants were instructed to download the smartphone app from either the Apple App Store or Google Play Store, depending on their device specifications. Following the installation of the mobile app, participants were instructed to activate the app by following in-app prompts and then pair the assigned Reader with the participant’s smartphone using a standard Bluetooth Low-Energy protocol (similar to how a smart speaker or smart watch pairs with a smartphone). Once the pairing process was complete, participants were asked to confirm that all DPS components were ready for use; they were then instructed to wear the Reader around their neck on a lanyard, open the mobile app on their smartphone, and ingest their first digital pill. A successful DPS operation was defined as the ability to activate and operate the digital pill, Reader, and smartphone app and successfully record digital pill ingestion. Finally, we discussed the potential technological challenges and user errors that participants may encounter and solutions to those barriers. The in-person portion of the training program lasted approximately 45 minutes.
Once the in-person portion of the DPS training program was completed, participants automatically entered the remote follow-up period of the program. In this portion, the study team programmed a series of 3 automated technology check-in text messages to be sent to each participant: the first at 24 hours following their in-person enrollment visit and then again on days 4 and 6. These messages asked participants to report via text message any technological barriers they were experiencing with the DPS (ie, by texting back 0 for “I need help with something” or 1 for “everything’s OK”). If technical assistance was requested in response to check-in messages, we attempted resolution via a standardized, hierarchical process, whereby resolution of all issues was first attempted via text message (through the web-based platform) and/or email with participants. If technical issues remained unresolved following attempts to rectify them via text message and/or email, a study team member escalated the issue to a phone call with participants for real-time evaluation and troubleshooting. Finally, participants were offered the opportunity to troubleshoot issues in person at the study site if they could not be resolved via any other means. Importantly, we did not offer advice through the training program regarding participants’ medication ingestion behaviors, device use, or feedback on their specific adherence (or nonadherence) trends. The DPS training program was considered complete at the end of the 6-day remote follow-up period. Study staff were available during the entire 90-day study period for additional technology-related concerns.
At the end of the 90-day study period, participants returned their DPS equipment and completed an individual, audio-recorded, semistructured qualitative interview. The aim of these interviews was to understand participants’ experiences operating the DPS and to obtain open-ended feedback on the DPS training program, including its perceived utility, adequacy for conveying the skills needed to operate the DPS, and suggested improvements to the training program. Interviewers were members of the study team who were trained in qualitative interviewing techniques (PRC, MJB, and YM).
We assessed the successful completion of the DPS training program among participants (attending the initial in-person training session and responding to text messages during the remote follow-up portion), reasons for failure to complete the DPS training program, and the portion (in-person or remote) that was not completed. In addition, we recorded reasons participants did not complete the training program. We also logged all instances where participants contacted the study team during the training program period and over the course of the 90-day parent study and calculated basic descriptive statistics regarding engagement with the DPS training program, defined as the number of times individuals responded to a text message check-in.
All interviews were professionally transcribed and checked for errors. Three study team members trained in methods of qualitative analysis (GG, YM, and JN) read all transcripts individually to generate a qualitative codebook. The codebook was developed iteratively using the semistructured interview guide as a framework and consisted of both inductive and deductive themes. Team members reviewed the codebook throughout the development process, adding new codes and resolving discrepancies at each stage. Data related to participants’ feedback on the DPS training program were extracted and discussed as a group to develop important themes around user experience.
We enrolled 15 individuals during the study period (
Demographics of study population (N=15).
Variables | Value | |
|
||
|
Median (IQR) | 32 (6.5) |
|
Range | 24-49 |
|
||
|
Yes | 14 (93) |
|
No | 1 (7) |
|
||
|
Yes | 3 (21) |
|
No | 11 (79) |
|
||
|
Android | 6 (40) |
|
Apple | 9 (60) |
aPrEP: pre-exposure prophylaxis.
bDefined as missing ≥2 doses over the past 2 weeks. The denominator for this variable is 14, as 14 participants had already been prescribed pre-exposure prophylaxis and 1 had not.
All participants enrolled in the training program were able to log an initial ingestion using the DPS during the in-person session and demonstrated the actual use of the DPS through recorded adherence events during the study period (
Participant enrollment and interaction with the digital pill system training program. DPS: digital pill system.
During the study period, we transmitted a total of 45 text messages to the 15 enrolled participants. Participants engaged with the DPS messages 91% (41/45 messages) of the time (ie, 41 messages were received from participants in response to DPS check-in messages). Of the 15 participants, 1 (7%) responded to just 1 of the 3 check-in messages, whereas 2 (13%) responded to 2 of the 3 messages, and 12 (80%) responded to all 3 messages. Of the 41 total responses received, 83% (n=34) of messages indicated that participants did not have difficulty operating the DPS, whereas 17% (n=7) of these messages included requests for technical assistance (a breakdown of message responses is given in
Characteristics of messaging and follow-up during the digital pill system training program.
Text message types | Value, n (%) | ||
Total check-in messages transmitted during study period | 45 (100) | ||
|
41 (91) | ||
|
Participants respondent to 1 of 3 check-in messages | 1 (7) | |
|
Participants respondent to 2 of 3 check-in messages | 2 (13) | |
|
Participants respondent to 3 of 3 check-in messages | 12 (80) | |
No difficulty operating the DPSa | 34 (83) | ||
|
7 (17) | ||
|
Asynchronous resolution | 3 (43) | |
|
Real-time telephone support from study team | 3 (43) | |
|
In-person assessment of technical issue | 1 (14) |
aDPS: digital pill system.
Of the messages requesting technical assistance, 43% (3/7) were successfully resolved asynchronously and 43% (3/7) were resolved in real time over the phone. One technical issue (1/7, 14%) required an in-person assessment, which was scheduled in addition to the existing study visits in the parent study. It was later determined that the participant had a malfunctioning mobile app and faulty Bluetooth permissions on their smartphone, which prevented successful pairing of the Reader with the phone. Once these issues were rectified, the participant was able to operate the DPS without any further problems. Although the study team remained available to troubleshoot further technological issues during the entire 90-day study period, no participants reported additional issues associated with the DPS after graduation from the training program.
All participants completed a qualitative user experience interview at the end of the 90-day study period. Overall, the participants accepted the DPS training program. They reported that the information provided during the in-person portion of the program was clear, helpful, and adequate for understanding the purpose of the DPS and for troubleshooting minor technological issues. Participants perceived the remote text message follow-up portion of the training program to be highly valuable, as it reminded them to operate the DPS during their first week of use and enabled rapid communication with the study team in the event of technical issues. Participants also discussed a number of potential refinements that could be made to the DPS training program, including repeated refresher sessions on the DPS technology at each subsequent in-person visit (during the 90-day study period), as well as additional written materials to be used for self-guided troubleshooting of technological issues. Selected participant quotes from the qualitative user experience interviews are presented in
Experiences with the in-person portion of the digital pill system (DPS) training program
“I thought that the first class and everything that happened that first day went really well, and you did a great job. [Study team member] made sure I understood everything… how to pair the phone, and how the technology worked, and how to keep track of it. There were clear instructions. There were lots of people standing by to answer questions if I ever had any questions after that” (age 29 years)
“I think it was fast, efficient. It’s very easy. I think the hardest part would be just to get it set up, but you do that in the meeting, so that was easy” (age 24 years)
“I think that was sufficient for me. I had no problem understanding the goals and how to interact with the device” (age 32 years)
Experiences with the remote portion of the DPS training program
“I think I had one issue with repairing my device when my device and my phone disconnected, so I had to ask for the pairing code, but someone was available to respond over text for that. [Study team member] was always available over email, but I didn’t really even need to ask many follow up questions, because that first session was very thorough… I thought it was very, very helpful” (age 29 years)
“The only thing was… I had the connectivity issues. I didn’t really know exactly what was going on. I think I had texted you, and you got back to me quickly, so I was able to figure it out” (age 40 years)
Suggested improvements to the DPS training program
“I think it's helpful, but I think every visit, maybe you should go over it, or review it, or actually have the person bring in the technology and show you. That way, it's drilled into their head… I learn by doing, so the more I do something, the easier it becomes to learn… I think that would've been helpful to have written instructions or be in the email… I just forgot some of the things you told me” (age 32 years)
“I think that someone else who might not be as technologically savvy might want more time, might want somebody to go through it more than once… but for me that was perfectly fine” (age 29 years)
“I think the only information [to add] is… tools to make sure that the Reader will pick up [the ingestion] and things to look for if it’s not picking up, so the user or the person has ways to troubleshoot it or figure it out when it’s not working” (age 28 years)
The adoption of a DPS to measure medication adherence has important implications for clinical trials, pharmaceutical drug evaluation, and real-world challenges to adherence in various disease states. Although DPSs are designed to be relatively unobtrusive, the requirements for using novel capsule-based dosage forms containing ingestible sensors and a wearable Reader device that is paired with a smartphone may present challenges to adherence to the DPS. Similar to other novel technologies, training participants to correctly operate the DPS is an important first step toward obtaining reliable and accurate adherence data. Although previous investigations have described the feasibility of deploying a DPS among individuals to measure medication adherence, few studies have described best practices around training study participants to use the DPS [
For MSM, a DPS is acceptable as a method to measure PrEP adherence, but a critical gap remains in understanding how best to impart tools to help MSM operationalize DPS. MSM account for most of the new HIV infections in the United States, and co-occurring psychosocial epidemics of mental illness and substance use in MSM are associated with further increased HIV incidence [
Operationalizing the deployment of the DPS will ultimately depend on users’ experiences and acceptance of the tool, as well as robust methods for ensuring consistent use of the technology to receive actionable adherence data. By using the TAM to inform the development of our DPS training program, we were able to identify and understand participants’ perceived uses of the DPS, integrate potential barriers related to the intended uses of the technology into the program, and provide ongoing support for actual real-world use. The preliminary findings from our qualitative user experience interviews demonstrate that the training program was acceptable for teaching participants important skills associated with operating the DPS. Participants reported that both the in-person and remote follow-up components of the program were valuable for providing and reinforcing operational skills, as well as for efficient communication with the study team for the purpose of troubleshooting technological issues. Continued DPS refresher sessions, as well as enhanced written materials for resolving technological issues related to DPS use, were among participants’ proposed improvements to the training program. These data suggest that a variety of individuals can effectively and efficiently train future DPS users using this training program.
In the future, we anticipate that a training program such as that described here can be integrated into a starter package for DPS enrollment. Users may access the training program in a number of settings, including independently as part of a package sent directly to their homes or within a pharmacy while picking up their medications. Alternatively, in models where a DPS is used as a booster package when other adherence monitoring methods such as self-report or pharmacological measures have already detected nonadherence, this training program can capitalize on TAM pillars of perceived use and actual use to rapidly orient the user to the DPS and produce actionable data to improve adherence behavior. Finally, we intentionally designed the training program in such a way that it could be deployed by individuals with minimal training, to maximize its potential scalability. The study team members underwent a brief orientation to the DPS from the manufacturer (etectRx Inc) and received instructions on how to operate the system. After demonstrating competence in the DPS operation internally, they were cleared by the principal investigator to deliver the structured training program directly to the study participants.
The operational skills taught in this DPS training program can serve as a toolbox to help individuals who use these systems to measure adherence. Although we used this training program to teach operational skills to individuals using a DPS for the first time, periodic booster training programs may also be used to reinforce the use of the digital pill over the long term. The ideal structure of a DPS should include detailed plans for responding to nonadherence detected through the technology, whether in the form of in-person adherence assessments at clinical appointments, phone calls from care teams monitoring adherence behavior, or the automated messaging architecture to respond directly to DPS users following missed doses or changes in patterns of medication use. Ideally, any nonadherence-related assessments should also include an exploration of adherence to the DPS technology itself. Our data suggest that using a simple text message–based system can help detect potential technical difficulties associated with a DPS operation and can mitigate these barriers before participants’ disengagement with the technology. Even when significant technical issues arise that impair the ability of the DPS to measure and report adherence, a training program can quickly diagnose these issues and allow a research or clinical team to determine the level of support needed to intervene.
This study had several limitations. First, we developed this DPS training program with a special focus on adherence to once-daily PrEP, and adherence to other medications may require specific and distinct instructions for integration with a DPS, to which PrEP-related guidelines may not be generalizable. However, we note that our training program can be easily modified to incorporate instructions specific to other xenobiotics. Second, we conducted the study at a single site with MSM who were young, well educated, and already engaged in health care services as well as research participation. The operation of a DPS may present specific barriers to other populations, especially those with less experience with novel technologies or who lack access to a smartphone. Third, the sample size of this preliminary pilot study was relatively small, and the findings may therefore not be representative of HIV-negative MSM who use substances. Finally, the parent study was conducted to understand the feasibility and acceptability of the DPS, and we therefore did not consider factors that would be required to integrate the training program of the DPS with clinical workflows.
Overall, this study demonstrates that a brief DPS training program consisting of in-person instruction and asynchronous text message–based follow-up can identify potential technical issues associated with the initial use of a DPS to measure medication adherence. The training program is simple to implement, requires only a single in-person visit, and can be adapted as a tool to integrate the DPS into adherence strategies in various disease states. Future studies to evaluate the efficacy of this educational intervention in the deployment of this novel technology for specific patient populations are required.
digital pill system
men who have sex with men
National Institutes of Health
pre-exposure prophylaxis
Technology Acceptance Model
PRC was funded by the National Institute on Drug Abuse (NIDA; K23DA044874 and R44DA051106). Research funding was provided by Gilead Sciences (ISR-17-1018), Hans and Mavis Psychosocial Foundation. KHM and CO were funded by National Institute of Allergy and Infectious Diseases NIAID P30AI060354 (Harvard Centers for AIDS Research). EWB and RKR were funded by National Institutes of Health (NIH) R01DA047236. JCM was funded by NIH R01AI145453.
PRC was the principal investigator of the study and was involved in all methodological decisions and data collection efforts. PRC, KHM, RKR, EWB, and CO were responsible for the conception and design of the study. MJB, GG, and YM contributed to the study design, recruited participants, and collected all the data. MJB, GG, SLB, and PEA conducted data management and analyses, with oversight from PRC. SLB and PEA assisted with the retrieval of data from the web-based platform and provided valuable insights regarding the interpretation of the results. JCM and RPC contributed to field expertise throughout the process of data collection and manuscript preparation. PRC, MJB, and GG drafted the manuscript, with contributions from YM, JN, MCS, JCM, RPC, KHM, RKR, SLB, PEA, EWB, and CO. All authors reviewed and approved the final manuscript.
SLB and PEA are employees of etectRx.