Effectiveness of an integrated diabetes care package at primary healthcare facilities: a cluster randomised trial in Pakistan

Abstract

Background: There were an estimated 7 million people living with diabetes in Pakistan in 2014, and this is predicted to reach 11.4 million by 2030. Aim: To assess if an integrated care package can achieve better control of diabetes. Design & setting: The pragmatic cluster randomised controlled trial (cRCT) was conducted from December 2014–June 2016 at 14 primary healthcare facilities in Sargodha district. Opportunistic screening, diagnostic testing, and patient recording processes were introduced in both the control 'testing, treating, and recording' (TTR) arm, and the intervention 'additional case management' (ACM) arm, which also included a clinical care guide and pictorial flipbook for lifestyle education, associated clinician training, and mobile phone follow-up. Method: Clinics were randomised on a 1:1 basis (sealed envelope lottery method) and 250 patients recruited in the ACM arm and 245 in the TTR-only arm (age ≥25 years and HbA1c >7%). The primary outcome was mean change in HbA1c (%) from baseline to 9-month follow-up. Patients and staff were not blinded. Results: The primary outcome was available for n = 238/250 (95.2%) participants in the ACM arm and n = 219/245 (89.4%) participants in the TTR-only arm (all clusters). Cluster level mean outcome was -2.26 pp (95% confidence intervals [CI] = -2.99 to -1.53) for the ACM arm, and -1.44 pp (95% CI = -2.34 to -0.54) for the TTR-only arm. Cluster level mean ACM–TTR difference (covariate-unadjusted) was -0.82 pp (95% CI = -1.86 to 0.21; P = 0.11). Conclusion: The ACM intervention in public healthcare facilities did not show a statistically significant effect on HbA1c reduction compared to the control (TTR-only) arm. Future evaluation should assess changes after a longer follow-up period, and minimal care enhancement in the comparator (control) arm.