Open AccessText message reminders for adolescents with poorly controlled type 1 diabetes: A randomized controlled trial
Author(s) -
Nour Ibrahim,
JeanMarc Tréluyer,
Nelly Briand,
Cécile Godot,
Michel Polak,
Jacques Beltrand
Publication year - 2021
Publication title -
plos one
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.99
H-Index - 332
ISSN - 1932-6203
DOI - 10.1371/journal.pone.0248549
Subject(s) - medicine , randomized controlled trial , glycemic , short message service , type 2 diabetes , diabetes mellitus , intention to treat analysis , text messaging , intervention (counseling) , pediatrics , physical therapy , insulin , endocrinology , world wide web , computer science , operating system , psychiatry
Background Among adolescents with type 1 diabetes, some experience great difficulties with treatment adherence, putting them at high risk of complications. We assessed the effect of text messaging (Short Messaging Service [SMS]) on glycemic control. Methods A two-arm open label randomized controlled trial enrolled adolescents with type 1 diabetes aged 12–21 years with baseline HbA1c ≥ 69 mmol/mol (8.5%). The intervention group received daily SMS reminders at self-selected times about insulin injections while the control group received standard of care. The patients allocated to the control group were not aware of the intervention. Results 92 patients were randomized, 45 in the SMS arm and 47 in the control arm. After 6 months, median HbA1c level was significantly lower in the intervention arm: 73 mmol/mol (8.8%) in the SMS arm and 83 mmol/mol (9.7%) in the control arm in the intent-to-treat analysis ( P = 0.03) but no longer in the per protocol analysis (P = 0.65). When we consider the proportions of patients whose HbA1c level decreased by at least 1% between baseline and 6 months, we find a significant difference among patients whose baseline HbA1c was ≥ 80 mmol/mol (9.5%) (n = 56): 60% in the SMS arm and 30.6% in the control arm had lowered their HbA1c level ( P = 0.03) in the intent-to-treat analysis but not in the per-protocol analysis (P = 0.50). Patients in the SMS arm reported high satisfaction with the intervention. Conclusions While there is a trend to lower HbA1c in the intervention group, no firm conclusions can yet be drawn. Further studies are needed to address methodological issues as we believe these interventions can support behavior change among adolescents with poorly controlled type 1 diabetes. ClinicalTrials.gov identifier: NCT02230137 .