Effects of Mobile Health Interventions on Patients with Diabetes: A Systematic Review

Effects of Mobile Health Interventions on Patients with Diabetes: A Systematic Review ABSTRACT Rebecca Kartje, MD, MBA, MSHI A systematic review was carried out to critically appraise and consolidate evidence from multiple randomized controlled trials (RCTs) on the effectiveness of mHealth interventions on glycemic control (HbA1c) in diabetes self-management. Comprehensive searches conducted on four databases to identify relevant studies published between January 1996 and February 2016 yielded 2855 articles. 19 RCTs (2790 patients) that fulfilled inclusion criteria were included in the review. A systematic review with meta-analysis was performed on 18 studies using a random-effects model. Quality of the evidence was determined by the Cochrane Collaboration’s tool for assessing ROB. Subgroup analyses were conducted and funnel plots used to examine publication bias. The mean reduction in HbA1c in participants using an mHealth intervention involving clinical feedback compared with a control was 0.45% (95% CI -0.66, -0.24) with a moderate quality of evidence. Data were heterogeneous (I2 = 66%). Subgroup analysis indicated that mHealth interventions are more effective for type 2 diabetes than for type 1 diabetes, and studies of type 2 diabetes with shorter follow-up yielded larger effect sizes compared to studies with six months or more of follow-up. Reductions in HbA1c were found to be more pronounced in interventions that used automatic versus manual data entry. Studies with weekly provider feedback were found to have the largest effect on HbA1c but feedback on as needed basis also saw a significant effect. There was insufficient evidence to determine whether mHealth interventions based on behavioral change theories are more effective. Visual inspection of funnel plots suggested a publication bias. These findings are consistent with clinically relevant improvements and demonstrate that mHealth interventions may be an effective strategy to help control HbA1c, particularly with respect to type 2 diabetics. While the functionality and use of this technology needs to be standardized, mHealth represents a promising approach to the self-management of diabetes.