De-implementation of low-value care practices in primary care: Results from the DE-imFAR study on abandonment of low-value pharmacological prescription for cardiovascular disease (CVD) primary prevention

Heather L Rogers, Alvaro Sanchez, The De-Imfar Phase 2 Study Team

Keywords: Implementation science, preventive medicine, cardiovascular diseases

Background:

Due to increasing recognition of low-value care provision, health systems are seeking ways to cease or reduce these practices.

Research questions:

This study aimed to compare the effectiveness of three additive de-implementation strategies to reduce potentially inappropriate prescribing (PIP) of statins in CVD primary prevention.

Method:

Design: Cluster randomized implementation trial with additional control group.
Sample: Family physicians (FPs) with non-zero incidence rates of PIP of statins in 2021 from 13 Integrated Healthcare Organizations (IHOs) of the Basque Health Service (n=621).
Intervention: All FPs were exposed to (1) a non-reflective decision assistance strategy based on reminders and decision support tools. FPs from two of the IHOs were randomized to additionally receive reflective strategies - either: (2) a decision information strategy based on knowledge dissemination (n=59), or (3) a decision information strategy plus an audit/feedback reflective decision structure strategy (n=59).
Target Population: 45- to 74-year-old patients with elevated cholesterol levels, but no diagnosed CVD and low cardiovascular risk, who attended at least one appointment between May 2022 and May 2023 (n=30,672).
Main Outcome: Change in the incidence rate of PIP of statins 12 months after FPs’ exposure to the strategies.
Clinicaltrials.gov identifier: NCT04022850.

Results:

All three strategies significantly reduced the pre-to-post incidence of PIP of statins in low-risk patients (p<0.001). There were no statistical differences when comparing all three strategies (p=0.07). Reduction was higher in the decision information strategy that adds a dissemination campaign to the decision support tools [adjOR=0.46(0.35-0.60)], while the audit/feedback strategy had no additional effect (p=0.32). A significant reduction was observed when comparing both reflective strategies with the non-reflective strategy (adjORs: 0.51 vs. 0.63; p=0.038).

Conclusions:

De-implementation strategies targeting clinical decision-making are effective in reducing PIP of statins in CVD primary prevention. An organizational culture promoting, prioritizing and increasing awareness to reduce low-value care is associated with better results.

Points for discussion:

What low-value care practices would you like to reduce/abandon in your practice?

What de-implementation strategies do you employ in your practice to achieve this reduction?

How effective are your de-implementation strategies? What do you think would make them more effective?

#157