Treatment quality indicators predict short-term outcomes in patients with diabetes: a prospective cohort study using the GIANTT database

Objective To assess whether quality indicators for treatment of cardiovascular and renal risk factors are associated with short-term outcomes in patients with diabetes. Design A prospective cohort study using linear regression adjusting for confounders. Setting The GIANTT database (Groningen Initiative to Analyse Type 2 Diabetes Treatment) containing data from primary care medical records from The Netherlands. Participants 15 453 patients with type 2 diabetes mellitus diagnosed before 1 January 2008. Mean age 66.5 years, 47.5% men. Exposure Quality indicators assessing current treatment (CT) status or treatment intensification (TI) for patients with diabetes with elevated cardiovascular or renal risk factors. Main outcome measures Low-density lipoprotein cholesterol (LDL-C), systolic blood pressure (SBP), and albumin: creatinine ratio (ACR) before and after assessment of treatment quality. Results Use of lipid-lowering drugs was associated with better LDL-C levels (-0.41 mmol/litre; 95% CI -0.48 to -0.34). Use of blood pressure-lowering drugs and use of renin-angiotensin system inhibitors in patients with elevated risk factor levels was not associated with better SBP and ACR outcomes, respectively. TI was also associated with better LDL-C (-0.82 mmol/litre; CI -0.93 to -0.71) in patients with elevated LDL-C levels, and with better SBP (-1.26 mm Hg; CI -2.28 to -0.24) in patients with two elevated SBP levels. Intensification of albuminuria-lowering treatment showed a tendency towards better ACR (-2.47 mmol/mg; CI -5.32 to 0.39) in patients with elevated ACR levels. Conclusions Quality indicators of TI were predictive of better short-term cardiovascular and renal outcomes, whereas indicators assessing CT status showed association only with better LDL-C outcome.