Improving Guideline Concordant Primary Prevention Aspirin Use through a Multiclinic Antithrombotic Stewardship Intervention

Introduction: After the publication of several large, randomized clinical trials, guidelines have suggested a reduced role for primary prevention aspirin. Specifically, the 2019 American Heart Association and American College of Cardiology guidelines indicate that low dose aspirin should not be administered on a routine basis for primary prevention among adults >70 years of age and among adults who are at increased risk of bleeding. Furthermore, since 2022, the US Preventative Services Task Force recommends against initiating aspirin for primary prevention in adults ≥60 years, with discussion that it may be reasonable to consider stopping aspirin around 75 years of age.

There is not a well-established method to implement these guidelines in clinical practice. We sought to develop a multicomponent intervention to improve guideline concordant aspirin use and assess the impact of this intervention after 9 months.

Methods: We conducted a patient survey of over 1,400 Michigan Medicine (MM) patients to establish the baseline rate of primary prevention aspirin use and to determine the accuracy of the medication list for recording aspirin use. We conducted ten semi-structured interviews to get stakeholder perspectives on the problem. Additionally, we conducted a root cause analysis using a fishbone diagram and convened monthly meetings with a stakeholder group for implementation mapping. We developed an institutional primary prevention aspirin guideline that was approved for clinical use at MM.

Based on physician input, we implemented a best practice advisory alert (BPA) to support the uptake of the institutional aspirin guidelines. This was pilot tested in one family medicine and one internal medicine clinic starting in September 2023. The BPA programming was refined in December 2023 and then expanded to two additional family medicine and three internal medicine clinics in July 2024. Provider training was completed through electronic materials, grand rounds presentations, and faculty meetings. We evaluated guideline discordant aspirin use and BPA utilization by tracking patient medication lists monthly, monitoring BPA interactions, and through manual chart abstraction of charts that triggered the BPA.

Patients eligible for the BPA intervention were adults over 40 years of age who were seen for a health maintenance exam (HME) and had aspirin on their medication list. Patients were excluded if they had a past medical history or problem list with an indication for aspirin. Shared decision making on aspirin use was encouraged.

Results:

Based on our survey, primary prevention aspirin use was approximately 22.3% at baseline; about 50% of this was potentially inconsistent with guidelines. Post-intervention, aspirin was de-prescribed for 262 out of 1,379 eligible aspirin users (19.0%, 95% confidence interval 16.9-21.1%, p<0.001 assuming no increase in aspirin use) over 9 months. Monthly de-prescribing was stable overtime at 15.5-21.0% per month. Deprescribing was slightly more in family medicine clinics (23.8% average deprescribing/9 months) compared to internal medicine and medicine-pediatrics clinics (18.5% average deprescribing/9 months). The clinics with BPA alerts had higher rates of de-prescribing compared to other providers in their specialty without the BPA: 35.6% versus 23.8% for family medicine and 21.9% versus 18.5% for internal medicine during the study period.

For the two pilot test clinics, detailed chart review was done for 211 consecutive patients that triggered the BPA from September 2023 to June 2024. Providers directly interacted with the BPA directly for 66 patients (31.3%). Aspirin was confirmed to be for primary prevention for 193 out of 211 patients (91.5%). Aspirin was de-prescribed for 51 patients (24.2%), with all deprescribing being consistent with current guidelines. Aspirin was not de-prescribed for patients with an indication for aspirin. When documented, advanced age (over 75), falls, heavy alcohol use, gastrointestinal or peptic ulcer disease, liver disease, and bruising were the most common reasons for de-prescribing.

Conclusions:

Aspirin use for primary prevention is common but often inconsistent with guidelines. While shared decision making and individualized care remain critical, quality improvement interventions can improve guideline concordant aspirin use and are an important part of antithrombotic stewardship.