Effects of digital and remote exercise interventions on HbA1c in type 2 diabetes: a systematic review and network meta-analysis.

BACKGROUND

To compare the effects of different digitally and remotely deliveredexercise interventions on glycated hemoglobin (HbA1c) in adults with type 2diabetes mellitus (T2DM) using a network meta-analysis (NMA), and to clarifythe relative effectiveness of distinct delivery models.

METHODS

A systematic review and network meta-analysis of randomized controlledtrials was conducted following PRISMA-NMA guidelines and prospectivelyregistered in PROSPERO (CRD420261288210). Major databases were searched frominception to November 11, 2025. Exercise-centered digital and remoteinterventions were categorized as digital exercise interventions (DEX),structured exercise prescription (SEP), multi-behavior lifestyle interventions(MBLI), and tele-exercise/tele-rehabilitation (TEX). The primary outcome waschange in HbA1c.

RESULTS

Fifty-four RCTs were included. Compared with control, TEX demonstratedthe greatest reduction in HbA1c (MD -0.51%, 95% CrI -0.88 to -0.14), followedby MBLI (MD -0.32%, 95% CrI -0.51 to -0.13). SEP showed a non-significanttrend, and DEX did not significantly reduce HbA1c. SUCRA rankings favored TEX(0.87), followed by MBLI (0.66), SEP (0.62), and DEX (0.24). Subgroup analysessuggested that HbA1c reductions with TEX and MBLI were more evident among olderadults (≥60 years) and those with longer diabetes duration（>10 years）;however, network inconsistency was detected in these subgroups, and findingsshould be interpreted cautiously.

CONCLUSIONS

Digitally and remotelydelivered exercise interventions differ in their effectiveness for HbA1ccontrol in T2DM. Interventions incorporating ongoing professional supervisionare associated with the most consistent glycemic benefits, whereas largelyautomated approaches show limited effects, highlighting the importance ofdelivery structure and supervision intensity.