Task shifting to frontline community health workers for improved Diabetes care in low-resource settings in India: A phase II Non-randomized controlled clinical trial

Background: The epidemic of type 2 diabetes mellitus is growing at large globally, more so in Lowand Middle-Income Countries (LMIC) like India, who concurrently battles the burden of infectious diseases. In such a health setting, people with diabetes (PWD) often receive sub-optimal care due to lack of understanding, physician-patient time, social support, and financial resources. Thus, there is a need for innovative, feasible, targeted interventions that strengthen health system to reduce the impact of diabetes in the rural communities in India. We hypothesised that structured interventions delivered by trained community health workers (CHWs) would improve outcomes in PWD and sought to demonstrate non-inferiority compared to standard care.
Methods: A phase II, non-randomized, controlled clinical trial was conducted in a rural block in Tamil Nadu, India. Two sub-center areas (cluster) each were allocated to the control and intervention arms, with 50 participants in each cluster. The control group received standard care routinely provided and the intervention group received interventions namely – CHW delivered, Engage communities, Screen, Examine, Refer and Follow up with the aid of ‘education tools’ and a ‘tablet based SmartHealth Application for follow-up’. Non-inferiority of the intervention in terms of early case detection, reduction in number of unnecessary physician consultations and glycaemic control was demonstrated using generalized estimating equation (GEE) for repeated measures with exchangeable correlation structure, adjusting for age, duration of diabetes and family history of diabetes. Findings: Baseline characteristics of both groups were comparable. CHWs were successfully able to Engage, Screen, Examine, Refer and Follow-up patients in the community. The multipronged interventions delivered by CHWs were demonstrated to be non-inferior to standard care, with significant difference in case detection rate (P = 0.003), early recognition of complications, reduction in unnecessary physician consultations (P = 0.041) and better glycaemic control (P = 0.036).
Conclusion: The study has shown promising results in the limited sample size and a small geographic area and has been able to show that the task shifting to frontline community health workers for diabetes care works in this rural population. We recommend a scale up of the intervention by conducting a
multicentric randomized controlled trial including cost-effectiveness analysis and taking into account stakeholders‘opinions regarding this model of diabetes care.