A cluster-randomized trial to estimate the effect of mobile screening and treatment feedback on HbA1c and diabetes-related complications in Tshwane primary health care clinics, South Africa

Abstract

Listen

Translate article

Aim and objectivesOur study had two primary objectives. The first was to evaluate the effect of an expert review with an individualised patient management plan on change in HbA1c. The second was to determine whether a mobile screening unit would increase the proportion of patients screened for glycaemic and lipid control, as well as diabetic complications as compared to the standard guideline-driven clinic care. MethodsWe conducted a cluster-randomised trial and enrolled 599 diabetic patients from twelve primary health care clinics in the Tshwane district of South Africa. The study was carried out in three phases: Phase 1 was a historical twelve-month record review to determine baseline characteristics; Phase 2 was the provision of the intervention (active screening for metabolic control and diabetes-related complications and health education targeting health care workers and diabetic patients) compared to the control group (routine care with similar education given to health care workers and diabetic patients) with a follow-up which included a clinical assessment. Phase 3 was a repeat of the record review to assess post-intervention changes following the proposed individualised patient management plan in the intervention clinics, and self-initiated by the health care professionals in the control clinics, based on training and results-sharing. ResultsNo significant changes in the HbA1c between the intervention and control clinics after one year were observed. The mean Hba1c changed from 8.68 to 8.53% (intervention) and from 8.95 to 8.76% (control) (p=0.92). Significant differences were observed in screening for diabetic complications when active screening was done by a dedicated team. The increase observed was more than 60% in all indicators monitored. Significant changes were also seen in the control clinics for two measured outcome variables (HbA1c test ordered from 15% to 25% (p=0.02) and serum-creatinine test ordered increased from 11% to 19% (p<0.001)). There was no significant difference in the proportion of actions taken, or not taken, between the control and interventions arms (p=0.83). The proposed changed patient management plans by the expert review team were implemented in only 29% of patients in the intervention arm. ConclusionsA comprehensive programme that integrates clinical evaluations, active screening and an individualised patient management plan did not lead to significant improvement in the HbA1c levels. However, screening for glycaemic and lipid control, as well as diabetic complications, improved dramatically with a dedicated team. The lower than expected rate of uptake of the specialists-proposed patient management, with the resulting non-significant change in glycaemic control, needs further investigation.