Home / Resources / Articles / Personalized Text Messages Improve Physical Activity For Diabetes Patients

Personalized Text Messages Improve Physical Activity For Diabetes Patients

Mar 5, 2016

Study finds positive feedback can go a long way to reinforcing positive health habits.

Many patients with type 2 diabetes live sedentary lifestyles despite the clear benefit of regular physical activity, including better glucose control and improved quality of life. It is a known fact that an increase in physical activity can lead to a decreased risk of developing cardiovascular dysfunction, which is prevalent in increasing numbers of patients with diabetes. Given the prevalence of daily smartphone usage, it would not be unreasonable to think that they could also be used to improve patient care. It would provide continual communication with patients and sensors that quantify patient behavior. Nevertheless, the use of personalized messages that take into account the actual behaviors of patients and learn to reinforce the positive efforts has not yet been employed.

Previous studies have shown that patients receiving non-specific text messages have an insignificant amount of improved response to the self-management of long-term disease states, concluding that more research in this area is necessary. However, in this recent study, the investigators compared patients receiving non-changing generated text messages (control group) versus patients receiving positive feedback tailored to their physical activity (treatment group).

This study followed 27 patients with type 2 diabetes who did not engage in regular physical activity prior to recruitment for this study. Patients were randomized into a control group (n = 7) and a treatment group (n = 20), and received text messages via Short Message Services (SMS) one to seven times a week to encourage physical activity. Patients in the treatment group received messages comprised of positive feedback and negative feedback generated through a reinforcement learning algorithm, which functions by optimizing messages to improve each participant’s compliance with the activity regimen. Patients in the control group received constant weekly reminders to exercise, with follow-up HbA1c tests that were performed every 3 months.

Participants who received messages with positive and/or negative feedback increased the amount of activity and pace of walking over time, while the control group patients did not. In a questionnaire provided to participants at the conclusion of this study, patients in the treatment group reported that the messages helped them to increase (P = 0.01) and maintain (P = 0.07) their physical activity, while patients in the control group reported that the messages were ineffective. This study was able to predict which message would lead to better patient outcomes.

On average, the messages with the best outcomes were those that contained positive feedback with a social component. It was unclear as to what exactly that social component consisted of; however, on average the improvement in activity increased by 8.8% on the day following the message. Patients who had received a text message containing negative feedback, followed by consecutive positive feedback, had an improvement of 42.7%. The investigators of this study were also able to group patients by the response to messages in three subdivisions: (1) those who reacted negatively to all messages; (2) those who reacted positively to only positive feedback with a social component; and (3) those who reacted positively to all messages. These are important components to take into consideration. If contemplating implementing this type of patient interaction, the clinician should be aware of which category a patient may identify with in order to tailor interventions accordingly.

The study’s author concluded that the results suggest that a mobile phone application with a learning algorithm can improve adherence to exercise in patients with diabetes. Because a personalized learning algorithm is automated, it can be used in large populations to improve health and glycemic control.The use of technology has catapulted our healthcare light-years ahead of where it started. In terms of research and design, what is lacking now is the inclusion of these technologies in patient interventions. Text messaging could be a step in the right direction in regards to patient involvement in their own health. Continued research should be undergone to further solidify its use in the healthcare field today.

Practice Pearls:

  • Providing patients with long-term health conditions, such as diabetes, with textual reminders to increase physical activity along with positive feedback can lead to better adherence to lifestyle modifications.
  • With technology continuously growing as a major part of day-to-day life, the healthcare system must begin to create new initiatives incorporating these technologies for better patient outcomes.
  • The actual generation of the textual messages received by patients should be customized to patient performance to invoke a better response.

Researched and prepared by Adaisha C. Rutledge, Doctor of Pharmacy Candidate FAMU College of  Pharmacy, reviewed by Dave Joffe, BSPharm, CDE

Hochberg I, Feraru G, Kozdoba M, Manor S, Tennenholtz M, and Yom-To E. “Encouraging Physical Activity in Diabetes Patients Through Automatic Personalized Feedback Via Reinforcement Learning Improves Glycemic Control.” Diabetes Care. 2016 January 28.