Digital pills allow data to be recorded automatically about patients’ medication adherence, but patients and providers have expressed significant ethical concerns.1,2 The same is true for IRBs reviewing study protocols that electronically track compliance.
“Ethical concerns relate to access, equity, and generalizability of results,” says Thomas J. George, Jr., MD, FACP, associate director for clinical research and director of the experimental therapeutics program at the University of Florida.
Digital pills represent a technology that is not readily available. Thus, inequality gaps could expand for those already underrepresented in medical research. George says IRBs should confirm any smart pill studies include provision of those pills for all study participants. How “compliance” is documented also is important.
Data suggesting a patient did not ingest a certain number of pills, for whatever reason, could end up documented as “noncompliance” in the patient’s medical record. This term carries negative implications, suggesting the patient actively chose not to adhere to the protocol, when there could be another explanation for the apparent noncompliance. “Regardless of the reasons, this could have significant implications for future participation in clinical research, insurance coverage, and decision-making,” George says.
To guard against this possibility, study investigators can determine the manner in which compliance of participants will be documented. “Particular attention should be paid to the impact the label of ‘non-compliance’ might have on subjects in the future, particularly those less likely to be in a position to advocate for themselves through review and editing of their medical records,” George advises.
Maria J. Silveira, MD, MPH, FAAHPM, has conducted multiple studies using the Medication Event Monitoring System (MEMS), medication bottles with an electronic cap, to track adherence. “The research is among the most difficult I’ve conducted, due to the combination of the IRB’s privacy and disclosure requirements and the technological limitations of the devices,” says Silveira, a clinician scientist at Ann Arbor VAMC’s Geriatric Research Education and Clinical Center.
IRB approval for those studies was particularly challenging. The central issue was privacy concerns on using the data, which are transferred from the medication bottle to the vendor, and reside on a secure server before a research team accesses the information. “We needed to assure the IRB that the data were stripped of personally identifiable information and as much protected health information as possible,” Silveira says.
To accomplish this, the medication type had to be coded so only the vendor and study team knew which medications were measured. The IRB also expressed informed consent concerns, and wanted researchers to tell study participants up front their adherence was going to be tracked. “They felt that not telling them was unethical,” Silveira notes.
Researchers explained it was necessary to keep information on how the bottles functioned from the subjects to prevent tampering with or gaming the device. On this point, the IRB agreed.
Then, the IRB raised the concern that if participants knew they were monitored, some might take more medication than they otherwise would, possibly leading to more side effects and adverse events. “We were measuring opioids, which can cause unintentional overdose and sedation when taken at doses higher than what the patient is accustomed,” Silveira notes.
Likewise, the IRB was concerned if some participants were aware they were monitored, they might take less medication and experience worsening of their original condition. To resolve these concerns, researchers agreed to monitor adherence in real time. Investigators also agreed to contact the patient immediately if they over- or underadhered by more than 20% on a given day.
Because of issues with the medication bottle, this requirement led to research staff spending a significant amount of time contacting study participants. “Research assistants felt this requirement significantly increased their workload and took the convenience out of the device,” Silveira says.
Ultimately, researchers resolved these tricky issues and obtained IRB approval to start the study. Even so, they still needed to contact the IRB regularly to report serious adverse events and protocol violations. “The devices are far from ideal,” Silveira laments. “Many vendors oversell their products and do not design them for research, but more for clinical use.”
It took months for researchers to find a device that included the right features and a vendor who understood the precision needed in clinical trials. Only after the first subject started to use the device did the researchers realize two medications were too light for the MEMS to detect. Thus, the MEMS device kept inaccurately reporting people were nonadherent. “Because we were measuring medications with a significant safety profile, it led to numerous contacts with the subject to ensure they were OK,” Silveira recalls.
That study participant became indignant at constant insinuations she was nonadherent, insisting she was taking the medication as instructed. At one point, the participant angrily dropped out. In the end, the researchers had to apply for modification of IRB approval to use a patient diary to confirm the data they were gathering from the MEMS.
Silveira says patient diaries result in faster IRB approval and “fewer headaches, overall” than the MEMS. IRBs are accustomed to seeing diaries as a “tried and true” approach to track medication adherence.
“MEMS devices are not at a level where the accuracy and convenience is sufficient to say that they are any better than patient diaries,” Silveira says.
REFERENCES
1. O’Sullivan D, Gibson JP, Jacob Y, et al. The future of medicine Is digital: Exploring the ethics of digital pills. Stud Health Technol Inform 2022;290:1120-1121.
2. Chevance A, Fortel A, Jouannin A, et al. Acceptability of and willingness to take digital pills by patients, the public, and health care professionals: Qualitative content analysis of a large online survey. J Med Internet Res 2022;24:e25597.
Digital pills allow data to be recorded automatically about patients’ medication adherence, but patients and providers have expressed significant ethical concerns.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content. TESTING
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content