Diabetes is a chronic metabolic disorder affecting over 30 million Americans, most of whom (up to 95%) have a diagnosis of type 2 diabetes. Diabetes confers substantial independent risk of atherosclerotic cardiovascular disease, heart failure, and chronic kidney disease; in turn, these comorbidities, which share common pathophysiologic risk with diabetes and are likely to be included in comprehensive diabetes management plans, amplify mortality risk in individuals living with diabetes. The complexity of interactions between type 2 diabetes, concordant comorbidities, and ensuing complications requires a clinical approach that manages risk while maintaining guideline-specified therapeutic targets. With the addition of new drug classes and an emphasis on self management along with shared decision-making, more patients are achieving individualized treatment goals. However, many patients struggle to meet targets for glycemic control or reduced cardiovascular risk. Inconsistencies in patient care quality suggest healthcare system level improvements may enable care teams, empower patients, and reduce therapeutic inertia (failure to intensify therapy when treatment targets are not met).
This paper summarizes the challenges associated with concordant comorbidities in individuals living with type 2 diabetes and further explores how real-world evidence and natural language processing may be used to offer insight regarding opportunities for management. Using de-identified data from an electronic health record platform Practice Fusion, a Veradigm® offering, a cohort analysis with case study was undertaken to 1) characterize ambulatory patients according to key demographics and comorbidities, 2) explore adoption of three of the latest glucose-lowering drug classes, and 3) evaluate the impact of concordant comorbidities on responsiveness to treatment intensification. The study identified one patient cohort as having greater incidence of microvascular and macrovascular complications, with more visits to healthcare providers. Forty-one percent (41%) of HbA1c values were supplemented through NLP enhancement. Across the cohorts, treatment intensification was associated with more patients achieving HbA1c values of less than 7%. Opportunities may exist for consideration of glucose-lowering drug classes with strong evidence of cardiovascular risk reduction and possibly nephro-protective effects to address unmet needs. Future studies that leverage real-world data from electronic health platforms may provide insight into drug research and development along with increased support for individualized diabetes management plans.