Sodium-glucose co-transporter-2 inhibitor cardiovascular outcome trials and generalizability to English primary care

Aim To identify people in English primary care with equivalent cardiovascular risk to participants in the sodium-glucose co-transporter-2 inhibitor (SGLT-2i) cardiovascular outcome trials (CVOTs). A secondary objective was to report the usage of SGLT-2is. Methods Cross-sectional analysis of people registered with participating practices in the Royal College of General Practitioners (RCGP) Research and Surveillance Centre (RSC) network on the 31 December 2016. We derived: (1) proportions of the primary care population eligible for inclusion in each SGLT-2i CVOT (CANVAS, DECLARE, EMPA-REG and VERTIS); (2) characteristics of the eligible population compared with trial participants (demographics, disease duration and vascular risk); and (3) differences within the eligible population prescribed SGLT-2is. Results The proportions of people with type 2 diabetes (N = 84 394) meeting the inclusion criteria for each CVOT were: DECLARE 27% [95% confidence interval (CI) 26.5-27.1]; CANVAS 17% (16.6-17.1); VERTIS 7% (7.1-7.4); and EMPA-REG 7% (6.5-6.8). Primary care populations fulfilling inclusion criteria were 5-8 years older than trial cohorts, and <10% with inclusion criteria of each trial were prescribed an SGLT-2i; a greater proportion were men, and of white ethnicity. Conclusions There was variation in proportions of the primary care type 2 diabetes population fulfilling inclusion criteria of SGLT-2i CVOTs. The more stringent the inclusion criteria, the lower the proportion identified in a primary care setting. Prescription rates for SGLT-2is were low in this national database, and there were demographic disparities in prescribing.