Clinical outcomes of an integrated primary–secondary model of care for individuals with complex type 2 diabetes: a non-inferiority randomised controlled trial
Published 3rd October 2018
Anthony W. Russell , Maria Donald, Samantha J. Borg, Jianzhen Zhang, Letitia H. Burridge, Robert S. Ware, Nelufa Begum, H. David McIntyre, Claire L. Jackson
Aims/hypothesis
The aim of the study was to determine if a Beacon model of integrated care utilising general practitioners (GPs) with special interests could achieve similar clinical outcomes to a hospital-based specialist diabetes outpatient clinic.
Methods
This pragmatic non-inferiority multisite randomised controlled trial assigned individuals with complex type 2 diabetes to care delivered by a Beacon clinic or to usual care delivered by a hospital outpatient department, in a 3:1 ratio. Owing to the nature of the study, researchers were only blinded during the allocation process. Eligible participants were aged 18 or over, had been referred by their usual GP to the hospital central referral hub with type 2 diabetes and had been triaged to be seen within 30 or 90 days. The intervention consisted of diabetes management in primary care by GPs with a special interest who had been upskilled in complex diabetes under the supervision of an endocrinologist. The primary outcome was HbA1c at 12 months post-recruitment. The non-inferiority margin was 4.4 mmol/mol (0.4%). Both per-protocol and intention-to-treat analyses are reported.
Results
Between 27 November 2012 and 14 July 2015, 352 individuals were recruited and 305 comprised the intention-to-treat sample (71 in usual care group and 234 in the Beacon model group). The Beacon model was non-inferior to usual care for both the per-protocol (difference −0.38 mmol/mol [95% CI −4.72, 3.96]; −0.03% [95% CI −0.43, 0.36]) and the intention-to-treat (difference −1.28 mmol/mol [95% CI −5.96, 3.40]; −0.12% [95% CI −0.55, 0.31]) analyses. Non-inferiority was sustained in a sensitivity analysis at 12 months. There were no statistically or clinically significant differences in the secondary outcomes of BP, lipids or quality of life as measured by the 12 item short-form health survey (SF-12v2) and the diabetes-related quality of life (DQoL-Brief) survey. Safety indicators did not differ between groups. Participant satisfaction on the eight-item client satisfaction questionnaire (CSQ-8) was good in both groups, but scores were significantly higher in the Beacon model group than the usual care group (mean [SD] 28.4 [4.9] vs 25.6 [4.9], respectively, p < 0.001).
Conclusions/interpretation
In individuals with type 2 diabetes, a model of integrated care delivered in the community by GPs with a special interest can safely achieve clinical outcomes that are not inferior to those achieved with gold-standard hospital-based specialist outpatient clinics. Individuals receiving care in the community had greater satisfaction. Further studies will determine the cost of delivering this model of care.
