Improving diabetes care in the primary healthcare setting: a randomised cluster trial in remote Indigenous communities.

OBJECTIVE: To evaluate a system for improving diabetes care in remote Indigenous communities. DESIGN: Randomised, unblinded cluster trial over one year (1 March to 29 February 2000). PARTICIPANTS AND SETTING: Primary healthcare staff in 21 primary healthcare centres in the Torres Strait and Northern Peninsula Area (NPA) Health Service District, north Queensland, and 678 people with diabetes, mostly Torres Strait Islanders. INTERVENTION: Diabetes recall system established at eight of the 21 sites, as well as staff training in basic diabetes care, regular phone calls from the project officer, a two-monthly newsletter and a mid-project workshop. MAIN OUTCOME MEASURES: Regular checks of weight, blood pressure, eye and foot care, serum lipid levels and glucose monitoring and control, urinary albumin to creatinine ratio and serum creatinine levels, and administration of recommended vaccines; hospitalisation in the previous 12 months. RESULTS: There was improvement in most measures at most sites, except for blood pressure monitoring and control, and vaccination status. Intervention sites showed greater improvement in most indicators than control sites (combined relative risk RR], 1.21; 95% CI, 1.03-1.43). The intervention group showed a 32% reduction in hospital admissions for diabetes-related conditions over the study period (P=0.012). At follow-up, patients in intervention sites were 40% less likely to be hospitalised for a diabetes-related condition than those in control sites (RR, 0.60; 95% CI, 0.41-0.86; P=0.007). CONCLUSION: A simple recall system, managed by local healthcare workers and supported by a diabetes outreach service, achieved significant improvements in diabetes care and reduced hospitalisations in a high-risk population.