The goal of this study is to learn whether stopping beta-blockers can help older adults with heart failure with preserved ejection fraction (HFpEF) feel better and function better. This study will test whether "deprescribing" or stopping these medications in a careful, guided way can improve symptoms and quality of life.

Participants will be randomly assigned to one of two groups:

Deprescribing group: Beta-blockers are gradually reduced using capsules that contain decreasing doses.

Usual care group: Beta-blockers are continued at the usual dose in look-alike capsules.

All participants will:

• Take study medicine for about 4 months
• Have their blood pressure and heart rate monitored
• Complete regular phone calls and questionnaires about how they are feeling

This study does not involve any experimental medication. Participants active involvement in the study will last approximately 4 months. During these 4 months they will have 8 scheduled telephone visits.

Inclusion Criteria:

1. Ambulatory adults age ≥60 years with HFpEF (EF ≥50% based on most recent transthoracic echocardiogram)
2. Hospitalized with HF in the past 24 months (defined by diagnostic codes in any position)
3. Taking any of the following ß-blockers: atenolol, bisoprolol, carvedilol, metoprolol succinate, or metoprolol tartrate (i.e., pharmacy prescription data)
4. Kaiser Permanence Northern California Membership

Exclusion Criteria:

1. Alternative etiologies of HFpEF syndrome

2. Compelling guideline indication for ß-blocker:

   • Prior EF <50% (i.e., based on the structured data field on transthoracic echocardiogram)
   • Current anginal symptoms (i.e., based on the Rose Angina single-question screener for angina)
   • MI (i.e., diagnostic codes) and/or coronary revascularization (i.e., procedure codes) within 3 years
   • Hospitalization for a primary discharge diagnosis of atrial fibrillation or atrial flutter in the prior 12 months
   • Other documented reason in the opinion of the treating provider and/or enrolling clinician-investigator

3. Evidence of recent decompensated HF

   • Hospitalization for a primary discharge diagnosis of HF within the past 30 days
   • Change in loop diuretic pharmacy prescription in the past 30 days (i.e. pharmacy prescription data) (Note: This definition is based on documented prescription changes and does not include short-term dose adjustments that were verbally communicated to the patient by their provider.)

4. Estimated life expectancy <6 months (i.e., diagnostic codes)

5. Diagnosed dementia (i.e.., diagnostic codes)

6. Unable to provide informed consent

7. Loss of Kaiser Permanence Northern California Membership

Data collected for this study will be shared only after the publication of findings addressing the study's specific aims. Data generated under this project will be managed in accordance with the policies of Weill Cornell Medicine (WCM) and the NIH, including the NIH Data Sharing Policy and Implementation Guidance. After publication of our main findings, data will be made available for secondary analyses. External researchers may request access to the data by submitting a written proposal outlining the hypotheses to be tested. All data will be de-identified to protect participant confidentiality.

The data will be shared after we have published the primary results of this study. There is no end date to access the data from the research project.

Researchers granted access to the data will be required to use it solely for research purposes, ensuring that no individual participants are identifiable. They must also implement appropriate security measures, such as using password-protected servers and files, to safeguard the data. Upon completion of the analyses, the data must be returned or destroyed. Released datasets will be subject to stringent safeguards to ensure compliance with the Health Insurance Portability and Accountability Act (HIPAA). Additionally, a data-sharing agreement will be required, specifying that the data is used exclusively for research purposes and that no participants are identifiable. Any additional approvals required by participating institutions, including KPNC, will be obtained. A data management plan will also be necessary to ensure the continued security and confidentiality of the data, as well as to outline the process for returning or destroying the data once the analysis is complete.