Referral rates of patients with diabetes to secondary care are inversely related to the prevalence of diabetes in each primary care practice and confidence in treatment, not to HbA1c level.

<bold>Aims: </bold>To determine the factors affecting the referral rates of patients with diabetes from primary care to secondary care.<bold>Methods: </bold>A study based on 66 GP surgeries in the Cardiff and Vale University Health Board (population: 515,581) was conducted. We included patients who had an established clinical diagnosis of diabetes (type 1 and type 2) from September 2017 to September 2018. HbA1c outcome data of GP surgeries were obtained from the Quality and Outcomes Framework (QOF) database published for 2018. Referral rates were obtained from the electronic referral database of Cardiff and Vale University Health Board over the same period, and this was adjusted according to the number of patients with diabetes in each GP surgery. Confidence level on the treatment of diabetes among GPs was assessed as a sub-study conducted in nine GP surgeries in the same area, using a self-administered questionnaire. Linear regression was undertaken to assess the relationship between adjusted referral rate and key factors which might influence prescribing rate.<bold>Results: </bold>The average adjusted referral rate to secondary care in one year was 4.23% of patients with diabetes in each GP surgery, with a wide variation of 1.24% to 16.28%. The average percentage of patients with diabetes with HbA1c<59mmol/mol was 63.17% (range: 43.19-76.23%). The average confidence score of GPs in treating diabetes was 67% and ranged from 50-85% in the sub-study. Referral rates correlated inversely with the numbers of patients with diabetes in each practice β=-0.32; (95% CI -0.57, -0.08) p=0.01, but there was no significant correlation with the HbA1c outcome β=-0.13; (95% CI -0.39, 0.12); p=0.30. Borderline significant negative correlation was observed between referral rates and overall practice size β=-0.23; (95% CI -0.48, 0.02) p=0.07.<bold>Conclusions: </bold>Referral rates of patients with diabetes to secondary care are determined by the number of patients with diabetes in each practice and confidence level in treatment, not by the overall practice size or HbA1c level. Ensuring quality training in diabetes care for primary care teams as well as the development of integrated diabetes care may be the best way to optimise the volume and appropriateness of referrals to secondary care. [ABSTRACT FROM AUTHOR]