Online Diabetes Education Matches In-Person Results in Type 1 Diabetes Pilot Trial

Adults with type 1 diabetes who took a structured self-management education course online ended up with essentially the same blood sugar control one year later as peers who completed the identical course face to face, according to a pilot randomized controlled trial conducted at Hospital Universitari Mútua Terrassa in Spain. The OPTIM@TE trial, published in Diabetes Therapy, randomized 51 adults with HbA1c of at least 7.5% to either an online or an in-person version of the same group-based diabetes self-management education and support program, and found no significant difference in glycated hemoglobin at 12 months between the two delivery formats.

The trial was designed to answer a question that has persisted as diabetes education migrates onto screens: when the curriculum, duration, educators, and learning objectives are held constant, does the delivery format itself change clinical outcomes? Previous studies of digital diabetes education have often compared online programs against usual care, or bundled online teaching with telemonitoring, extra professional support, or teleconsultation, making it hard to isolate the effect of the format. The researchers, led by Núria Alonso-Carril and Verónica Perea, instead built two versions of a single structured program and randomized participants between them in a 1:1 ratio.

The program itself delivered roughly 10 hours of structured group education covering glucose pattern interpretation, insulin dose optimization, carbohydrate counting and nutrition, physical activity, management of acute complications, diabetes technologies, and problem-solving strategies for everyday self-management decisions. Participants assigned to face-to-face delivery attended a single 10-hour educational day running from 08:00 to 18:00, with live teaching, practical exercises, case-based learning, discussion, and feedback woven through the sessions. Those assigned to online delivery covered the same curriculum over one week through a Microsoft Teams learning environment, working through eight prerecorded videos of 12 to 31 minutes each that could be replayed on demand, alongside complementary materials and self-directed activities, plus a one-hour introductory synchronous session and a final one-to-two-hour group videoconference.

Eligible participants were adults aged 18 or older who had lived with type 1 diabetes for at least three years, had HbA1c at or above 7.5%, and owned a personal digital device. The trial excluded people using continuous subcutaneous insulin infusion, pregnant women, those on systemic corticosteroids, people with severe comorbid conditions, and those unable to understand Spanish or Catalan. Recruitment ran from June 2021 to June 2024, but stopped short of the planned 60 participants because a higher-than-anticipated rate of nonattendance at the initial educational session limited the number of people who actually started their assigned program. Fifty-one participants were randomized, 26 to the online arm and 25 to the face-to-face arm, and the investigators explicitly note that the study was not powered to detect statistically significant between-group differences, so all effectiveness analyses are exploratory.

The primary outcome was HbA1c 12 months after randomization, analyzed by intention to treat with an analysis of covariance adjusting for baseline HbA1c. Twelve-month HbA1c data were available for 41 participants, 21 in the face-to-face group and 20 in the online group. Mean HbA1c moved from 8.69% to 8.56% in the face-to-face group and from 8.55% to 8.35% in the online group. The adjusted between-group difference was −0.07% (95% confidence interval −0.71 to 0.57; p = 0.831), a result indistinguishable from no difference. A sensitivity analysis additionally adjusted for university-level education, available for 29 participants, also found no significant difference (adjusted difference 0.23 percentage points; 95% CI −0.70 to 1.16; p = 0.621).

An exploratory per-protocol analysis restricted to participants who completed their assigned program told a more suggestive story. Among completers, mean HbA1c fell from 8.71% to 8.43% in the face-to-face group and from 7.99% to 7.32% in the online group, adjusted differences of roughly 0.3 and 0.7 percentage points respectively. The adjusted between-group difference of −0.56% narrowly missed statistical significance (95% CI −1.15 to 0.02; p = 0.056), and remained nonsignificant after adjusting for university-level education. The authors caution that this analysis is susceptible to selection bias because course completion occurred after randomization, and the small sample size makes the finding provisional at best. Still, they note that even modest HbA1c improvements can carry clinical benefit in people with suboptimal control, provided hypoglycemia risk does not rise.

Continuous glucose monitoring metrics, measured with 14-day FreeStyle Libre ambulatory glucose profiles, showed no significant between-group differences at 12 months. The analyzed metrics included time in range 70–180 mg/dL, time below range below 70 and below 54 mg/dL, time above range 180–250 and above 250 mg/dL, mean sensor glucose, glucose management indicator, and coefficient of variation, all reported according to international standardized glucose reporting recommendations. Anthropometric and metabolic variables, including body weight, blood pressure, lipid fractions, and urinary albumin-to-creatinine ratio, likewise showed no significant differences. Patient-reported outcomes, assessed with validated questionnaires covering diabetes knowledge, self-care, hypoglycemia awareness, fear of hypoglycemia, diabetes-related quality of life, and emotional wellbeing, were similarly flat between groups.

One secondary outcome did separate the formats: adherence to the Mediterranean diet, measured with the MEDAS questionnaire, improved significantly more in the online group, with an adjusted difference of 2.48 points (95% CI 0.71 to 4.24; p = 0.009) that persisted after adjustment for university-level education. The authors suggest, tentatively, that asynchronous access to prerecorded nutritional content may have let participants revisit dietary material at the moment of decision-making, while stressing that this hypothesis cannot be confirmed from the present data and that secondary analyses in a pilot trial are exploratory.

The most consequential finding may be about participation rather than physiology. Only 21 of 51 participants, 41.2%, completed the assigned educational program: 60.0% in the face-to-face arm versus 23.1% in the online arm. Engagement with the prerecorded videos was uneven, with the introductory module drawing 18 views and 1.2 cumulative viewing hours and the insulin dose adjustment module 16 views and 1.1 hours, while the diabetes technologies module drew only 6 views and 0.2 hours. Adherence did not differ by sex, and among the 19 completers who provided satisfaction data, overall satisfaction was high, with the most common suggestions being more practical exercises and refresher sessions. No adverse events related to the educational interventions were reported.

The authors argue that the lower online attendance should not be read as evidence that online education is intrinsically less effective, since engagement choices occurred after randomization and the study was not designed to diagnose the causes of nonattendance. Reasons for nonattendance were not systematically collected, though prior qualitative work points to competing priorities, emotional burden, perceived lack of benefit, and practical or organizational barriers. The trial’s strengths include the direct comparison of identical curricula delivered by the same multidisciplinary team and a pragmatic design embedded in routine clinical practice, with 12-month follow-up extending beyond the shorter horizons of many digital education studies. Its limitations are equally clear: a small pilot sample, no blinding of participants or educators, and modest overall glycemic change. The investigators conclude that strategies to improve engagement, such as shorter educational units, structured follow-up, personalized feedback, automated reminders, or gamification, are needed before larger pragmatic trials can establish whether online delivery of structured diabetes education can be scaled effectively.

Subject of Research: A pilot randomized controlled trial comparing online versus face-to-face delivery of an identical structured diabetes self-management education program in adults with type 1 diabetes and suboptimal glycemic control.

Article Title: Structured Online Diabetes Self-Management Education and Support for Adults with Type 1 Diabetes and Suboptimal Glycemic Control: The OPTIM@TE Pilot Randomized Controlled Trial

Article References: Alonso-Carril, N., Rodríguez-Rodríguez, S., Berrocal, A. B., Morató, C., Vidal, O., Barahona, M.-J., Martínez, D., Ferré, C., Amor, A. J., & Perea, V. (2026). Structured Online Diabetes Self-Management Education and Support for Adults with Type 1 Diabetes and Suboptimal Glycemic Control: The OPTIM@TE Pilot Randomized Controlled Trial. Diabetes Therapy. https://doi.org/10.1007/s13300-026-01933-w

Image Credits: AI Generated

DOI: 10.1007/s13300-026-01933-w

Keywords: type 1 diabetes, diabetes self-management education, HbA1c, continuous glucose monitoring, randomized controlled trial, online education, Mediterranean diet adherence, pilot trial, glycemic control, program completion, digital health, patient-reported outcomes