Overcoming therapeutic inertia in primary care: a multisite quality initiative to increase guideline-based prescribing for patients with diabetes.
Therapeutic inertia (TI), the failure to intensify or de-intensify treatment when appropriate, is a contributor to poor guideline adherence in diabetes treatment, including the suboptimal use of sodium-glucose cotransporter 2 inhibitors (SGLT-2is) and glucagon-like peptide-1 receptor agonists (GLP-1RAs).
We developed a multifaceted improvement initiative targeting TI at four academic primary care practices, aiming to increase rates of SGLT-2i and GLP-1RA use for patients with type 2 diabetes (T2DM). Prescribing trends for GLP-1RAs, SGLT-2is, sulfonylureas and insulin were compiled quarterly over a 12-month baseline period and 12-month intervention period and analysed using interrupted time series analyses. Providers completed a brief questionnaire assessing project feasibility and acceptability.
GLP-1RA prescribing showed an upward trend during both the baseline and intervention periods (+0.70% vs +0.87% per quarter; p=0.14 for difference) and increased significantly in the first intervention quarter (+1.73%; p=0.003). SGLT-2i prescribing was static during the baseline period, showed an upward trend during the intervention (0.0% vs +0.43% per quarter; p=0.05 for difference) and increased significantly in the first intervention quarter (+1.0%; p=0.03). In those prescribed a GLP-1RA or an SGLT-2i, sulfonylurea prescribing dropped significantly after intervention (-2.0% per quarter; p=0.02). Insulin prescribing rates did not change throughout the study period. There was no significant change in haemoglobin A1c among patients newly prescribed a GLP-1RA and/or an SGLT-2i during the intervention period (7.3% ±1.5% baseline vs 7.2% ±1.4% intervention period, p=0.23). Most providers indicated that they had prescribed (90%) and were more likely to prescribe (81%) GLP-1RAs and/or SGLT-2is in the future because of their participation.
A quality improvement initiative targeting drivers of TI was associated with increased rates of guideline-based medication prescribing for primary care patients with diabetes and may be applied to other conditions in which TI limits care optimisation or guideline adherence.
We developed a multifaceted improvement initiative targeting TI at four academic primary care practices, aiming to increase rates of SGLT-2i and GLP-1RA use for patients with type 2 diabetes (T2DM). Prescribing trends for GLP-1RAs, SGLT-2is, sulfonylureas and insulin were compiled quarterly over a 12-month baseline period and 12-month intervention period and analysed using interrupted time series analyses. Providers completed a brief questionnaire assessing project feasibility and acceptability.
GLP-1RA prescribing showed an upward trend during both the baseline and intervention periods (+0.70% vs +0.87% per quarter; p=0.14 for difference) and increased significantly in the first intervention quarter (+1.73%; p=0.003). SGLT-2i prescribing was static during the baseline period, showed an upward trend during the intervention (0.0% vs +0.43% per quarter; p=0.05 for difference) and increased significantly in the first intervention quarter (+1.0%; p=0.03). In those prescribed a GLP-1RA or an SGLT-2i, sulfonylurea prescribing dropped significantly after intervention (-2.0% per quarter; p=0.02). Insulin prescribing rates did not change throughout the study period. There was no significant change in haemoglobin A1c among patients newly prescribed a GLP-1RA and/or an SGLT-2i during the intervention period (7.3% ±1.5% baseline vs 7.2% ±1.4% intervention period, p=0.23). Most providers indicated that they had prescribed (90%) and were more likely to prescribe (81%) GLP-1RAs and/or SGLT-2is in the future because of their participation.
A quality improvement initiative targeting drivers of TI was associated with increased rates of guideline-based medication prescribing for primary care patients with diabetes and may be applied to other conditions in which TI limits care optimisation or guideline adherence.