mHealth for diabetes patients: Lessons learned

November 12, 2013

Technology-based messaging support for at-risk and diabetic patients is being tested in 17 communities across the country. Three of the participating programs have released their evaluations and conclusions.

Mobile health (mHealth) technology has the potential to reach medically underserved populations and to positively influence behavior change among prediabetes and diabetes patients.

Knowing this is possible, the Beacon Community Cooperative Agreement Program, developed through funds from the Office of the National Coordinator for Health Information Technology, was created to launch community-based mHealth programs in 17 diverse communities throughout the country. Each Beacon community received $12-$15 million over a 3-year period for its health information technology systems and for implementing them to improve population health, healthcare quality, and reduce costs.

Target communities

Three Beacon communities released their insights on the implementation of mHealth for diabetes risk reduction and disease management. They were the Louisiana Public Health Institute, Crescent City Beacon Community in New Orleans, La.; HealthInsight, Southeast Michigan Beacon Community, Detroit, Mich.; and HealthInsight, Utah Beacon Community, Salt Lake City, Utah, said study author Alison L. Rein, MS, of AcademyHealth, Washington, DC. The study was recently published in the Journal of Medical Internet Research.

Two of these communities - the Crescent City and Southeast Michigan Beacon Communities - had at-risk populations of overweight or obese patients who had low incomes and/or no health insurance. The Utah Beacon Community targeted adults aged 18 and older who had type 2 diabetes.

In the first group of at-risk individuals, the program used txt4health, a text message service that delivered health information. After enrollment, patients completed a risk assessment for diabetes and received 4 to 7 messages a week that were tailored to their risk stratification level. Patients could also track the progress of their weight loss and exercise by responding to text messages.

In the second group, adults with type 2 diabetes, Care4Life, a two-way text messaging service, was used to help patients with self-management of their disease. It also provided coaching and an interactive support function, so patients could report monitoring levels for blood sugar and blood pressure, exercise, weight loss, and medication adherence.

“Beacon communities encountered a number of barriers at each stage [of implementing mHealth], including issues related to developing tailored, culturally competent messages; designing comprehensive outreach strategies; enrolling participants; engaging providers in mHealth programs; evaluating mHealth programs; and sustaining and scaling pilots,” Rein and her colleagues wrote in the study.

Factors that were critical to success included identification of community partners that could help to reach the targeted individuals, development of message content that could increase enrollment and engagement, and a budget for in-person engagement to increase enrollment. Other important elements are the need to engage providers in promoting mHealth among potential patients and a plan to evaluate the mHealth program strategy.

Enrollment

The Crescent City and Southeast Michigan Beacon communities reached out to their target populations through advertisements in settings where individuals congregate, including bus shelters and buses, laundromats, and hair salons. Other outreach efforts included health fairs, retail stores, and church meetings, in which community partners were helpful. The Utah Beacon Community directed its outreach effort toward 19 primary care clinics, which identified potential candidates with type 2 diabetes from electronic health records. Patients were invited by mail to enroll online, and staff later helped to assist with patient enrollment.

“These three Beacon Communities identified some key barriers to enrollment. First, although the txt4health and Care4Life programs were free, messaging rates applied for participants without unlimited texting plans, which proved cost-prohibitive for many,” Rein and colleagues wrote.

“Limited technology proficiency and/or access to computers or the Internet presented additional barriers to online and text-based enrollment. During a follow-up telephone survey of 104 patients invited to participate in Care4Life, 35% reported limited or no access to a computer and 38% reported having trouble using a Web browser,” the authors noted.

In-person engagement was helpful in driving enrollment numbers, but this is a more costly and labor-intensive option that needs to be budgeted, Rein said.

Provider engagement

While provider engagement is important for the success of mHealth initiatives with patients, the fee-for-service reimbursement model may be an impediment. With limited time for appointments and providers not being reimbursed for promotion of mHealth programs, this could have been a major problem for the Beacon communities.

To get around this challenge, practice managers in Southeast Michigan helped enroll patients during check-in and check-out times. At the Utah Beacon program, pay-for-performance incentives to clinics helped drive enrollment with medical assistants enrolling patients as a strategy to achieve their hemoglobin A1c levels.

“While new reimbursement structures may facilitate integration of mHealth into the primary care workflow, further advances in device interoperability and data integration will also be necessary to achieve this objective,” Rein noted.

Evaluation

The Beacon communities had to find resources to help with evaluation of their programs, which included community partners. The primary sources of data were the txt4health and Care4Life systems data, electronic health records, and surveys delivered online, by mail, and by phone. Limitations to data collection included nonresponse to surveys, incomplete electronic health records, sampling bias, and landline survey method that could introduce selection bias. Another challenge was determining the impact of mHealth on health behaviors and outcomes, they said.

“Those embarking on mHealth interventions should carefully consider their evaluation and research aims from the outset, as well as the resources they have at their disposal to achieve their desired outcomes,” Rein noted.

“In all three Beacon Communities, building relationships with community organizations and leveraging local resources was critical to the success of their mHealth initiatives,” she said.