Personalised text messages could improve adherence to exercise for type 2 diabetes patients

A smartphone application with a learning algorithm for sending personalised text messages could help type 2 diabetes patients stick to exercise regimens, Israeli researchers suggest.
Smartphones are being utilised more and more in treating diabetes, and the Rambam Health Care Campus, Haifa, Israel wanted to assess if personalised text messages could be used to reinforce adherence to exercise programs.
Exercise is an important treatment method for people with type 2 diabetes which can lead to improved blood glucose control and weight loss, but the researchers reported that “most patients with type 2 diabetes are sedentary despite the clear benefit of regular physical activity”.
27 patients with type 2 diabetes who did not regularly exercise had a pedometer installed on their smartphones. They were then randomised to either a treatment group or control group – all participants received text messages one to seven times a week to encourage exercise.
The treatment group personalised messages comprised of positive and negative feedback. These were generated by an automatic learning algorithm that assessed the effect of previous messages, tailoring messages to each individual based on the time they were given. Patients in the control group received messages simply encouraging exercise.
Treatment group participants increased their amount of physical activity and their walking pace compared to the control group. Patients in the treatment group, who had higher HbA1c levels at baseline, and lower activity targets, were also more likely to have reduced HbA1c.
Text messages were ineffective for increasing physical activity in the control group, but treatment group participants reported that the messages helped them increase and maintain physical activity.
The researchers wrote: “These results suggest that mobile phone application with a learning algorithm can improve adherence to exercise in patients with diabetes. Because a personalised learning algorithm is automated, it can be used in large populations to improve health and glycemic control.”
Due to the small study size, trials on a larger number of patients may be needed to confirm whether personalised text messages are clinically effective in encouraging exercise.
The findings were published in Diabetes Care.