A Randomized Controlled Trial of Sweet Talk, a Text-Messaging System to Support Young People With Diabetes

Franklin VL, Waller A, Pagliari C, Greene SA. Diabet Med. 2006;23:1332-1338.

Teenagers with type 1 diabetes in Scotland and elsewhere in Europe frequently have poor glycemic control, placing them at high risk for diabetic complications. Intensive insulin therapy has been established as the standard of care, but transitioning patients from conventional to intensive therapy is difficult, or even detrimental, without behavioral support interventions. Typically, such interventions are expensive and staff-intensive, so there is interest in a low-cost alternative. This study examined the effect of a low-cost automated text-messaging system called Sweet Talk on diabetes care behaviors and glycemic control in 92 preteens and teenagers (age range = 10.5 to 15.6 years).

Patients were eligible to participate if they had had type 1 diabetes for more than 1 year and were on conventional insulin therapy (CIT), defined as 2 to 3 daily injections of premixed insulin. Patients were excluded if they had serious social problems, severe learning difficulties, or needle phobia. Patients were randomized to 3 different treatment arms: CIT, CIT plus Sweet Talk, or multiple daily injection (MDI) plus Sweet Talk. For ethical reasons, this study did not include any group assigned to MDI without behavioral support intervention. Patients assigned to MDI were permitted to choose between basal-bolus injections or an insulin pump, and also received carbohydrate-counting education. Those assigned to Sweet Talk received goal-setting training. All participants were given a mobile telephone with text-messaging capability and a prepaid £10 phone card. There were no significant differences between treatment groups at the study baseline.

Sweet Talk messages were sent automatically from a database of over 400 messages in 4 categories of diabetes self-management tasks: insulin injection, blood glucose testing, diet, and exercise. Goal reminders were sent weekly; goal-reinforcing information or reminders tailored to the patient“s age, gender, and diabetes regimen were sent daily. Text newsletters on a variety of diabetes topics were sent occasionally. Each participant received 1 to 2 messages daily.

Initially, 22 patients in the MDI group chose pump therapy and 9 chose basal-bolus therapy, but over the course of the study, 5 patients discontinued pump therapy. Of these, 2 selected basal-bolus therapy and 3 selected CIT. Among the patients in the CIT groups, 8 patients switched to basal-bolus therapy for clinical reasons (3 from CIT and 5 from CIT plus Sweet Talk).

After 12 months, there was no change in glycemic control from baseline for the patients assigned to CIT or CIT plus Sweet Talk. Patients assigned to MDI plus Sweet Talk experienced a decrease in mean A1C from 10.0% to 9.2% (P<.001). Patients assigned to CIT plus Sweet Talk scored significantly higher on measures of self-efficacy (P = .003) and adherence (P = .042) than patients assigned to CIT only. Sweet Talk increased patients“ perception of diabetes team support but did not alter their perception of support from family and friends. The messaging system did not change scores on a test of diabetes knowledge. Patient satisfaction with Sweet Talk was high: 81% stated that it was helpful for their diabetes self-management, 90% wanted to continue receiving the messages, and 97% approved of the message frequency (ie, 1 to 2 per day). However, 20% complained that they received the same message repeatedly.

The authors of this study did not discuss whether the mobile phones were used by patients for any purposes other than communication with the diabetes clinic and receiving Sweet Talk messages. Furthermore, they did not estimate the degree to which having free mobile telephone service might have influenced patients“ attitudes toward study participation.

Although a clinically meaningful reduction in A1C in patients assigned to MDI plus Sweet Talk was seen, it was not possible to estimate how much of the reduction was attributable to each intervention within this group. This study shows that a low-cost behavioral support intervention based on mobile telephone text messaging is well-received by teenagers and improves psychological measures associated with treatment adherence.