The goal of this study is to increase shared decision making for patients considering treatment for severe aortic stenosis. The main questions it aims to answer are:

• Do patient decision aids and clinician skills training course improve the quality of decisions, and do they work well for different patient populations?
• Are heart clinics able to reach the majority of patients with decision aids before their specialist visit and do the majority of clinicians complete the training course?

All participating sites will start in the usual care group and then will be randomly assigned a time to switch to the intervention group. Participants will complete surveys before and after their specialist visit.

Researchers will compare data from patients seen during usual care with data from those seen after the interventions are implemented to see if there are improvements in the quality of decisions.

The goal of this study is to promote shared decision making for patients considering treatment for severe aortic stenosis. The main questions it aims to answer are:

• Do patient decision aids and clinician skills training course improve the quality of decisions, and do they work well for different patient populations?
• Are heart clinics able to reach the majority of patients with decision aids before their specialist visit and do the majority of clinicians complete the training course?

This study is a batched stepped wedge cluster randomized trial with 8 sites, about 56 surgeons and interventional cardiologists, and will enroll about 1300 patients. All participating sites will start in the usual care group and then will be randomly assigned a time to switch to the intervention group. The interventions include a patient decision aid that was created by the American College of Cardiology CardioSmart program and a online shared decision making skills training course for clinicians. The coordinating center will provide implementation support to sites when they transition to the intervention arm. Data collection is similar across usual care and intervention groups. Patient participants will complete surveys before and after their specialist visit. Clinician participants will be randomly selected to complete a survey after the patient visit. Limited clinical data will be collected from the medical record.

Researchers will compare data from patients seen during usual care with data from those seen after the interventions are implemented to see if there are improvements in the quality of decisions. Researchers will also calculate percentage of patients reached with decision aids and percentage of clinicians who complete training at each site to measure the success of the implementation.

Inclusion Criteria:

• Between 65-85 years of age
• Language is English or Spanish
• Diagnosed with severe aortic stenosis defined as (e.g. an aortic valve area < 1 cm2 or as determined by clinician)
• Attend a scheduled visit with an Interventional Cardiologist and/or Cardiac Surgeon from the Heart Valve Team at a participating site

Exclusion Criteria:

• Prior aortic valve replacement surgery
• High risk for either SAVR or TAVI (e.g., Society of Thoracic Surgery score >8% or clinician determined)
• Prior coronary artery bypass surgery (CABG)
• End stage renal disease on dialysis
• Severe lung disease (chronic obstructive pulmonary disease; COPD) requiring home oxygen
• Advanced Cirrhosis
• Unable to consent for self (proxy respondents are not allowed)

To promote research replicability, transparency and future use of the data, de-identified data sets will be created and will be available, by request, to outside researchers. The Data Package (comprised of the de-identified Analyzable Data Set, Full Protocol, metadata, data dictionary, full statistical analysis plan, and analytic code from the project) will be deposited in the Patient-Centered Outcomes Research Institute's Data Repository at the ICPSR. Participant information sheets for the study will indicate that de-identified information will be deposited in an open access service to promote use of the data by other researchers. No information that contains identifiers or that could be used to link an individual to the data will be included in the de-identified data set. No data will be provided from qualitative transcripts as this may lead to identifying information about participants. However, codes abstracted from qualitative transcripts may be made available.

The PCORI-designated repository will make the data available for third-party requests when PCORI makes the Final Research Report available, or at the time of publication the primary results, whichever comes first.

Researchers may access the data through the PCORI-designated repository (ICPSR's website) once it is made available.

There is limited blinding in the trial due to the pragmatic focus and batched stepped wedge design. Patients will be told that the study is testing different types of decision support and while they will not be blinded to the materials they receive, they will not know what materials other patients receive. Clinicians and staff will be involved in delivery of the decision aids and will not be blinded to the study arm. Research staff entering survey data and conducting chart review will not be given any explicit information about study assignment; however, it is likely that they will have a sense for the arm due to the timing. The statistician conducting analyses will be blinded to group assignment at each step. For the clinician survey, we will use computer generated random assignment to select one specialist (either interventional cardiologist or cardiac surgeon) to receive the post-visit survey for each patient participant.