Diabetes Mellitus, Type 2, Medication Adherence
Conditions
Brief summary
Reinforcement learning is an advanced analytic method that discovers each individual's pattern of responsiveness by observing their actions and then implements a personalized strategy to optimize individuals' behaviors using trial and error. The goal of this pilot study is to develop and test a novel reinforcement learning-enhanced text messaging program to support medication adherence in patients with type 2 diabetes. Type 2 diabetes is an optimal condition in which to test this program, as it is one of the most prevalent chronic conditions in the US adult population and requires most patients to be on daily or twice daily doses of medications. This pilot study will be a parallel randomized pragmatic trial comparing medication adherence and clinical outcomes for adults aged 18-84 with type 2 diabetes who are prescribed 1-3 daily oral medications for this disease. Participants will be randomized to one of two arms for the duration of the study period: (1) a reinforcement learning intervention arm with up to daily, tailored text messages based on time-varying treatment-response patterns; or (2) a control arm with up to daily, un-tailored text messages. Our outcomes of interest will be medication adherence, as measured by electronic pill bottles, and HbA1c levels.
Detailed description
The goal of this pilot study is to develop and test a novel reinforcement learning-enhanced text messaging program to support medication adherence in patients with type 2 diabetes. This pilot study will be a parallel randomized pragmatic trial comparing medication adherence and clinical outcomes for adults aged 18-84 with type 2 diabetes who are prescribed 1-3 daily oral medications for this disease. Participants will be randomized to one of two arms for the duration of the study period: 1) a reinforcement learning intervention arm with up to daily, tailored text messages based on time-varying treatment response patterns, or 2) a control arm with up to daily, untailored text messages. Our outcomes of interest will be medication adherence, as measured by electronic pill bottles, and HbA1c levels.
Interventions
BEHAVIORALReinforcement Learning
Participants in the intervention arm will receive up to daily, tailored text messages based on their electronic pill bottle-measured adherence. Given the participants' baseline characteristics and time-varying responses to the messages, a reinforcement learning algorithm will deliver different text messages and adapt over time to determine which type of messaging works best for each individual participant.
Sponsors
National Institute on Aging (NIA)
Brigham and Women's Hospital
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
DOUBLE (Investigator, Outcomes Assessor)
Eligibility
Sex/Gender
ALL
Age
18 Years to 84 Years
Healthy volunteers
No
Inclusion criteria
* Age between 18-84 years * Diagnosed with type 2 diabetes mellitus (T2DM) and are prescribed between 1-3 daily oral medications for this disease * Currently have a smartphone with a data plan or WiFi at home * HbA1c level ≥7.5% * Basic working knowledge of English * Willing and able to set up the platform and adhere to study procedures * Either not currently using a pillbox or willing to use electronic pill bottles (EDMs) for diabetes medications for the duration of the study
Exclusion criteria
* Patients with active enrollment in another diabetes trial within Mass General Brigham * Patients who receive daily assistance with taking their medications at home * Patients who are unable to receive text messages for more than 3 days in a row during the study period
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Medication Adherence | 6 months | Medication adherence to type 2 diabetes oral medications (averaged) as measured by the number of dates and times of pillbottle openings in the electronic pill bottles |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Glycemic Control | 6 months | Change in glycated hemoglobin A1c from baseline to end of the 6-month follow-up |
Countries
United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Reinforcement Learning Intervention Arm Up to daily, tailored text messages.
Reinforcement Learning: Participants in the intervention arm will receive up to daily, tailored text messages based on their electronic pill bottle-measured adherence. Given the participants' baseline characteristics and time-varying responses to the messages, a reinforcement learning algorithm will deliver different text messages and adapt over time to determine which type of messaging works best for each individual participant. | 29 |
| Control Arm Up to daily, untailored text messages. | 31 |
| Total | 60 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Overall Study | Death | 0 | 1 |
| Overall Study | Withdrawal by Subject | 1 | 2 |
Baseline characteristics
| Characteristic | Reinforcement Learning Intervention Arm | Control Arm | Total |
|---|---|---|---|
| Age, Continuous | 60.31 years STANDARD_DEVIATION 11.35 | 56.68 years STANDARD_DEVIATION 12.89 | 58.43 years STANDARD_DEVIATION 12.2 |
| Baseline HbA1c, mean (sd) | 8.99 percentage of glycated hemoglobin STANDARD_DEVIATION 1.39 | 9.10 percentage of glycated hemoglobin STANDARD_DEVIATION 1.47 | 9.05 percentage of glycated hemoglobin STANDARD_DEVIATION 1.42 |
| Education level, n(%) High school graduate or below | 7 Participants | 5 Participants | 12 Participants |
| Education level, n(%) Post-graduate | 7 Participants | 7 Participants | 14 Participants |
| Education level, n(%) Some college/college graduate | 15 Participants | 19 Participants | 34 Participants |
| Level of non-adherence, self-reported number of doses missed in past month, n(%) 0 | 12 Participants | 14 Participants | 26 Participants |
| Level of non-adherence, self-reported number of doses missed in past month, n(%) 1+ | 17 Participants | 17 Participants | 34 Participants |
| Number of Pillsy bottles, n(%) 1 | 20 Participants | 25 Participants | 45 Participants |
| Number of Pillsy bottles, n(%) 2+ | 9 Participants | 6 Participants | 15 Participants |
| Number of self-reported medications, n(%) 1 | 2 Participants | 2 Participants | 4 Participants |
| Number of self-reported medications, n(%) 2-4 | 6 Participants | 4 Participants | 10 Participants |
| Number of self-reported medications, n(%) 5+ | 21 Participants | 25 Participants | 46 Participants |
| Number of self-reported physicians, n(%) 1-3 | 10 Participants | 15 Participants | 25 Participants |
| Number of self-reported physicians, n(%) 4+ | 19 Participants | 16 Participants | 35 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Asian | 2 Participants | 1 Participants | 3 Participants |
| Race (NIH/OMB) Black or African American | 6 Participants | 7 Participants | 13 Participants |
| Race (NIH/OMB) More than one race | 1 Participants | 2 Participants | 3 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 3 Participants | 4 Participants | 7 Participants |
| Race (NIH/OMB) White | 17 Participants | 17 Participants | 34 Participants |
| Region of Enrollment United States | 29 participants | 31 participants | 60 participants |
| Self-reported automaticity of medication-taking, n(%) 0-1 (no or low automaticity) | 7 Participants | 12 Participants | 19 Participants |
| Self-reported automaticity of medication-taking, n(%) 2+ (high automaticity) | 22 Participants | 19 Participants | 41 Participants |
| Sex: Female, Male Female | 12 Participants | 14 Participants | 26 Participants |
| Sex: Female, Male Male | 17 Participants | 17 Participants | 34 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 0 / 29 | 1 / 31 |
| other Total, other adverse events | 0 / 29 | 0 / 31 |
| serious Total, serious adverse events | 0 / 29 | 0 / 31 |
Outcome results
Primary
Medication Adherence
Medication adherence to type 2 diabetes oral medications (averaged) as measured by the number of dates and times of pillbottle openings in the electronic pill bottles
Time frame: 6 months
Population: Participants were included in the analysis until they no longer participated in the study.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Reinforcement Learning Intervention Arm | Medication Adherence | 74 pill bottle openings | Standard Deviation 31 |
| Control Arm | Medication Adherence | 68 pill bottle openings | Standard Deviation 29 |
Secondary
Glycemic Control
Change in glycated hemoglobin A1c from baseline to end of the 6-month follow-up
Time frame: 6 months
Population: Participants were included in the analysis until they no longer participated in the study. Differences were calculated among those who had follow-up HbA1c values recorded.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Reinforcement Learning Intervention Arm | Glycemic Control | -0.82 percentage glycated hemoglobin | Standard Deviation 1.19 |
| Control Arm | Glycemic Control | -1.20 percentage glycated hemoglobin | Standard Deviation 2.28 |