Home-based non-pharmacological interventions for cognition and diabetes self-management in adults with type 2 diabetes: a systematic review

Type 2 diabetes mellitus is a major public health challenge requiring sustained self-management. Cognitive impairment may compromise daily self-care behaviours, including medication adherence, dietary regulation, physical activity, and glucose monitoring. Home-based non-pharmacological interventions may offer scalable approaches to support cognition and self-management, but it remains unclear whether cognitive gains translate into meaningful behavioural or clinical benefits. This systematic review followed PRISMA 2020 guidelines and was prospectively registered in PROSPERO. Seven electronic databases were searched from inception to 11 March 2026. Randomised controlled trials and quasi-experimental studies evaluating home-based non-pharmacological interventions in adults with type 2 diabetes were eligible. Risk of bias was assessed using the Cochrane Risk of Bias 2 tool, and certainty of evidence was evaluated using the GRADE approach. Findings were synthesised narratively because of heterogeneity in interventions and outcomes. Ten randomised controlled trials were included, comprising exercise, cognitive training, multidomain, and sensory-based interventions. Exercise-based interventions showed the most consistent benefits for physical performance but limited and inconsistent effects on cognition, diabetes self-management, and glycaemic indicators. Cognitive training improved trained or near-transfer cognitive domains, but evidence on whether these gains translate into self-management behaviours, medication adherence, HbA1c, or other clinical outcomes remains insufficient, as few studies assessed both cognitive and self-management outcomes and none tested this pathway directly. Evidence for multidomain and sensory-based interventions was preliminary. Few studies recruited cognitively vulnerable participants, and none formally tested whether cognitive changes mediated diabetes-related functional outcomes. Current home-based non-pharmacological interventions may improve selected cognitive or physical outcomes in adults with type 2 diabetes, but current evidence is insufficient to determine whether these gains translate into diabetes self-management or glycaemic benefit, as few studies assessed both outcome domains and none formally tested mediation. Future public health and chronic disease self-management interventions should be scalable, behaviourally integrated, and designed to link cognitive processes with real-world self-management behaviours and clinical outcomes. PROSPERO CRD420261335132.