Speakers
Michael Felker, MD, MHS
Professor of Medicine
Director, Cardiovascular and Metabolism Research, DCRI
Keywords
Pragmatic Clinical Trials, Heart Failure, Diabetes
Key Points
- Making lifestyle changes is difficult especially for people living with chronic diseases, where healthy behaviors could make a big difference. Medication adherence can be a barrier to effectiveness. This study asks how can we better facilitate health behaviors in patients with chronic diseases and can we leverage technology?
- The TARGET-HF-DM is a pragmatic multicenter randomized controlled trial held at 6 sites in the U.S. that focused on patients with heart failure and diabetes, two rapidly growing conditions that are highly morbid. The focus of the study was on physical activity and medication adherence. The primary endpoint was change in mean daily step counts from baseline to 3 months. Secondary endpoints were change in medication adherence, change in HRQOL, change in NT-proBNP, and change in hemoglobin A1C from baseline to 3 months. There were two exploratory endpoints: change in mean daily step counts from baseline to 6 months and change in metabolomic profiling from baseline to 3 months.
- The trial enrolled 187 patients, with 35% women, 47% African American, 10% Hispanic, age 59 years. Enrollment halted before planned sample size of 200 due to COVID-19. There was 1:1 randomization of participants to either the mHealth intervention or usual care.
- Both groups received a step counter and weekly text reminder to wear it. In the intervention group patients received 3 text messages weekly that were personalized encouraging them to be more active based on the past week’s activity. After 3 months of active intervention, there was an additional 3 months of data collection.
- In patients with both heart failure and diabetes, the 3-month mHealth intervention significantly improved daily volitional physical activity as measured by step counts; health related quality of life as measured by KCCQ, and metabolomic profiles of peripheral blood. Adherence to medical therapy was not measurably different between the arms.
- This trial turned out to be more challenging than anticipated. Lifestyle intervention trials have unique challenges, such as compliance/engagement challenges that lead to varying “dose” of intervention; unequal ascertainment of outcomes – patients in control group may be less engaged and more likely to discontinue follow up; intervention may interact with human behavior in complex ways; lack of blinding leads to strong “placebo” effects, and challenges of using a commercial step counter, which relied on people’s phones and thus software updates, etc.
- For the step-down RCT, patient recruitment started in August 2019 with 587 patients consenting and 552 included in the primary analysis. The median time to liberation in the HFNC group was 50 hours and the CPAP group was 43 hours. HFNC was not noninferior to CPAP following extubation.
Discussion Themes
– Did the mHealth group receive text messages on pill adherence? Can you talk about the compensation of participants? Patients did not get text messages about pill adherence. Compensation was modest and the same regardless of which group you were in.
– Were there specific approaches that you used to recruit under-represented populations? A huge part is where your sites are located. One of our leading sites was in Brooklyn, NY, and the patients there were largely African American and Hispanic. As we are thinking about trials in general we should be thinking about where the trial sites will be and who are the leaders and teams.
–Was it easier or harder than expected to recruit participants? It was harder than we expected. It’s almost always harder in my experience. It was a labor of love; the site payment was low; we asked friends to be sites in the trial. Sites have a lot of trials that are competing. One pays more than another. We did what we could to make the entry criteria as easy as possible.
Tags
#pctGR, @Collaboratory1