Skip to main content

CardSort data for treatment features and goals for aortic stenosis

Cite this dataset

Col, Nananda (2022). CardSort data for treatment features and goals for aortic stenosis [Dataset]. Dryad.


Background: Guidelines recommend including the patient’s values and preferences when choosing treatment for severe aortic stenosis (sAS). However, little is known about what matters most to patients as they develop treatment preferences. Our objective was to identify, prioritize, and organize patient-reported goals and features of treatment for sAS.

Results: 51 adults with sAS and 3 caregivers with experience choosing treatment (age 36-92 years) were included. Participants were referred from multiple health centers across the U.S. and online. Eight nominal group meetings generated 32 unique treatment goals and 46 treatment features, which were grouped into 10 clusters of goals and 11 clusters of features. The most important clusters were: 1) trust in the healthcare team, 2) having good information about options, and 3) long-term outlook. Other clusters addressed the need for and urgency of treatment, being independent and active, overall health, quality of life, family and friends, recovery, homecare, and the process of decision-making.

Conclusions: These patient-reported items addressed the impact of the treatment decision on the lives of patients and their families from the time of decision-making through recovery, homecare, and beyond. Many attributes had not been previously reported for sAS. The goals and features that patients’ value, and the relative importance that they attach to them, differ from those reported in clinical trials and vary substantially from one individual to another. These findings are being used to design a shared decision-making tool to help patients and their clinicians choose a treatment that aligns with the patients’ priorities.


This multi-center mixed-methods study conducted structured focus groups using the nominal group technique to identify patients’ most important treatment goals and features. Patients separately rated and grouped those items using card sorting techniques. Multidimensional scaling and hierarchical cluster analyses generated a cognitive map and clusters. Data were collected via customized Qualtrics survey. Participants were presented with a list of either treatment features or treatment goals and ask to sort them into groups.

Usage notes

The data files are CSV and can be opened as XL files.


Edwards Lifesciences (United States), Award: HCP-8243002