Patient preferences for mitral valve regurgitation treatment: a discrete choice experiment

Figures & data

Table 1. Final DCE Attributes and Levels

Attribute	Attribute Type	Attribute Levels
		Level 1	Level 2	Level 3	Level 4
Procedure Invasiveness	Ordered Categorical	Less Invasive	Invasive
Procedure Recovery	Ordered Categorical	Moderate	Intensive
Risk that MR Symptoms Reappear (within 2 years)	Ordered Categorical or Continuous	1% (1 in 100 people)	8% (8 in 100 people)	15% (15 in 100 people)	40% (40 in 100 people)
Risk of Needing Reintervention (within 2 years)	Ordered Categorical or Continuous	1% (1 in 100 people)	8% (8 in 100 people)	15% (15 in 100 people)
Risk of Disabling Stroke (within 30 days)	Ordered Categorical or Continuous	1% (1 in 100 people)	2% (2 in 100 people)	5% (5 in 100 people)
Risk of New Onset Atrial Fibrillation (within 30 days)	Ordered Categorical or Continuous	1% (1 in 100 people)	10% (10 in 100 people)	15% (15 in 100 people)

Table 2. Treatment Profiles

Attribute	Treatment Profile (Attribute Levels)		‘beating heart’ (vs. ‘open-heart Surgery-like’ Treatment
	Beating heart (Base Case)	‘open-heart Surgery-Like’ (Comparator)	Attribute Level Change	Type
Procedure Invasiveness	Less Invasive	Invasive	Less Invasive (vs. Invasive)	Benefit
Procedure Recovery	Moderate Recovery	Intensive Recovery	Moderate (vs. Intensive)	Benefit
Risk that Mitral Valve Regurgitation Symptoms Reappear	10% Risk	1% Risk	9 pp Increase	Harm/Risk
Risk of Needing Reintervention	8% Risk	1% Risk	7pp Increase	Harm/Risk
Risk of Disabling Stroke	1% Risk	5% Risk	4 pp Decrease	Benefit
Risk of New Onset Atrial Fibrillation	10% Risk	20% Risk	10 pp Decrease	Benefit

pp = percentage point

Table 3. Error component logit model results (main effects, n = 118)

Variable/Parameter	Estimate	95% CI	P value
Treatment Attribute Levels (Fixed Coefficients)
Procedure invasiveness
Less invasive	0.56	(0.38, 0.73)	<0.001
Invasive (ref. level)	0.00
Procedure recovery
Moderate	0.21	(0.05, 0.38)	0.011
Intensive (ref. level)	0.00
Risk that symptoms reappear (per +1 pp)	−0.05	(−0.05, −0.04)	<0.001
Risk of needing reintervention (per +1 pp)	−0.07	(−0.09, −0.06)	<0.001
Risk of disabling stroke
1%	0.70	(0.46, 0.94)	<0.001
2%	0.28	(0.06, 0.51)	0.015
5% (ref. level)	0.00
Risk of new onset atrial fibrillation
1%	0.83	(0.58, 1.08)	<0.001
10%	0.61	(0.38, 0.84)	<0.001
20% (ref. level)	0.00
Error Component (Random Coefficient)
Alternative-specific intercept SD	0.61	(0.33, 0.89)	<0.001
Model Fit Statistics
Log Likelihood	−492.42
AIC	1002.84
BIC	1046.41

AIC = Akaike Information Criterion, BIC = Bayesian Information Criterion, CI = Confidence Interval, pp = percentage point, ref = reference, SD = Standard Deviation.

Table 4. Relative attribute importance (RAI) for differences in attribute levels based on ‘beating heart’ versus ‘open-heart surgery-like’ treatment profiles (n = 118)

Rank	Attribute	Difference between Treatments	Relative Importance Weight	95% CI
1	Risk of Disabling Stroke	4 pp (1% vs. 5%)	23.3%	(16.6%, 30.1%)
2	Risk of New Onset Atrial Fibrillation	10 pp (10% vs. 20%)	20.2%	(13.8%, 26.7%)
3	Procedure Invasiveness	Less Invasive vs. Invasive	18.5%	(13.6%, 23.5%)
4	Risk of Reintervention	7 pp (8% vs. 1%)	17.3%	(13.2%, 21.4%)
5	Risk that Symptoms Reappear	9 pp (10% vs. 1%)	13.5%	(11.2%, 15.8%)
6	Procedure Recovery	Moderate vs. Intensive	7.1%	(2.0%, 12.2%)

CI = Confidence Interval, pp = Percentage Point

Table 5. Maximum (or minimum) acceptable risk (MAR) thresholds for A) symptoms reappearing and B) reintervention

Maximum Acceptable Risk (MAR) Increase (or Minimum Decrease) in Exchange for Each Attribute
	Benefit Attributes				Harm Attributes		Total^a
	A Less Invasive Procedure	A Moderate Recovery	4 pp decreased risk of disabling stroke	10 pp decreased risk of new onset AFib	9 pp increased risk that symptoms reappear	7pp increased risk of reintervention
Versus	Invasive	Intensive	1% vs 5%	10% vs 20%	10% vs 1%	8% vs 1%
Risk That Symptoms Reappear (95% CI)	+12.4 pp (8.4, 16.3)	+4.7 pp (1.1, 8.4)	+15.5 pp (9.8, 21.3)	+13.5 pp (8.2, 18.8)		−11.5 pp (−14.3, −8.8)	+34.6 pp (23.8, 45.4)
Risk of Reintervention (95% CI)	+7.5 pp (4.6, 10.4)	+2.9 pp (0.6, 5.2)	+9.4 pp (5.4, 13.5)	+8.2 pp (4.7, 11.7)	−5.5 pp (−6.8, −4.2)		+22.6 pp (14.7, 30.4)

CI = Confidence Interval, MAR = Maximum (or Minimum) Acceptable Risk, pp = Percentage Point.^a ‘beating heart’ (vs. ‘open-heart surgery like’ treatment defined as the combination of all attribute level changes/risk differences specified in the other column headings for individual attribute trade-offs.

Figure 1. Probability of choosing a ‘beating heart’ treatment as each attribute shifts.

Table

	Janssen	Keuffel	Liden	Hanna	Rizzo
Conception and Design	Yes	No	Yes	Yes	No
Analysis and Interpretation of Data	Yes	Yes	Yes	Yes	Yes
Drafting of Paper and Critical Revision	Yes	Yes	Yes	Yes	Yes
Final Approval of Published Version	Yes	Yes	Yes	Yes	Yes
Accountable for All Aspects	Yes	Yes	Yes	Yes	Yes

Table A1. Initial (version 1) DCE attributes and levels

Attribute	Attribute Type	Attribute Levels
		Level 1	Level 2	Level 3	Level 4
Procedure Invasiveness	Ordered Categorical	Less Invasive	Invasive
Procedure Recovery	Ordered Categorical	Moderate	Intensive
Changes in Functioning and Symptoms	Ordered Categorical	Complete improvement	Partial Improvement	Worsening
Risk of having MR not eliminated (within 2 years)	Continuous or Ordered Categorical	1% (1 in 100 People)	8% (8 in 100 People)	15% (15 in 100 People)	40% (40 in 100 People)
Risk of needing a reintervention (withing 2 years)	Continuous or Ordered Categorical	1% (1 in 100 People)	8% (8 in 100 People)	15% (15 in 100 People)
Risk of disabling stroke (within 30 days)	Continuous or Ordered Categorical	1% (1 in 100 People)	2% (2 in 100 People)	5% (5 in 100 People)
Risk of New Onset Atrial Fibrillation (within 30 days)	Continuous or Ordered Categorical	1% (1 in 100 People)	10% (10 in 100 People)	20% (20 in 100 People)

Table A2. Full description of final attributes and levels

Attribute	Attribute Description
Procedure invasiveness	To treat and alleviate the symptoms of mitral valve regurgitation, there are the options to receive a medical procedure that is either less invasive or invasive. Less invasive: A cut 2–3 inches long is made between your ribs just below the nipple to access the heart. While the heart is beating, the surgeon uses an x-ray camera to insert a small device into the heart to repair the leaking valve. The procedure takes less than two hours. You will likely go to the intensive care unit (ICU) for less than one day. Invasive: A cut 8–10 inches long is made in your chest and your chest is opened to access your heart. Then, your heart is stopped while a machine takes over your heart and lung function. Stopping the heart allows the surgeon to see the valve more easily. An incision is made in the heart, so the surgeon can repair the leaking valve. Your heart is started again, and your chest is stitched closed. The procedure normally takes 3–5 hours. You will likely go to an intensive care unit (ICU) for up to two days.
Procedure recovery	If you undergo a medical procedure for mitral valve regurgitation, your recovery time and experience may be different depending on the type of procedure. Moderate recovery: You should be able to leave the hospital 2–4 days after the procedure. Upon leaving, you will be able to go home, possibly with some restrictions on physical activity and return to work as soon as one week after the procedure. Intensive recovery: You should be able to leave the hospital 4–8 days after the procedure. Depending on your age, most patients will go home, but some patients will go to a rehabilitation facility before going home. You will need at least one month of rest and recovery and may require additional rehabilitation before returning to work.
Risk that MR symptoms reappear (within 2 years)	Mitral valve regurgitation can cause symptoms including shortness of breath, fatigue, heart palpitations, and swollen feet or ankles. These symptoms might have an impact on your daily functioning (ability to do ordinary physical activities such as walking, climbing the stairs, or dressing yourself). If treatment for mitral valve regurgitation is effective, your symptoms will likely improve or disappear so you can do all of your daily activities. If you undergo a medical procedure to treat your mitral valve regurgitation, there is a chance that your symptoms will not be eliminated. You might continue to experience mitral valve regurgitation symptoms, or symptoms might reappear (or develop, if you previously had none) within two years of having the procedure. Because of these symptoms, you may have difficulty doing some or all physical activities. Depending on the treatment option you choose in the choice tasks later, there will be different risks that your mitral valve regurgitation symptoms reappear within two years:
Risk of needing a reintervention (within 2 years)	If you undergo a medical procedure to treat your mitral valve regurgitation, there is a risk you will require a reintervention within two years to further correct your mitral valve regurgitation. This medical reintervention is another medical procedure that can be either minimally invasive or invasive. Depending on the treatment option you choose in the choice tasks later, there will be different risks of needing a reintervention within two years:
Risk of disabling stroke (within 30 days)	You might have different risks of experiencing a disabling stroke within 30 days of the medical procedure. If you experience a disabling stroke, you will be hospitalized for medical care. Depending on the severity of the stroke and the timing of treatment, a stroke may be fatal, or could lead to temporary or permanent disability, such as paralysis, reduced mobility, and problems with thinking, memory, and speech. Within 30 days of having the procedure, the risk of disabling stroke might range from:
Risk of new onset atrial fibrillation (within 30 days)	You might develop new onset atrial fibrillation after the procedure. Atrial fibrillation means that you sometimes experience an irregular, rapid heart rate that causes poor blood flow. Atrial fibrillation may require treatment with blood thinners until resolved. Within 30 days of having the procedure, the risk of new onset atrial fibrillation might range from:

Table A3. Final DCE design

Block	Task^1	Option	Attribute Levels
			Procedure invasiveness	Procedure recovery	Risk that symptoms reappear	Risk of re-intervention	Risk of disabling stroke	Risk of new onset atrial fibrillation
1	1	A	2	2	2	3	3	1
1	1	B	1	1	4	2	2	2
1	2	A	2	1	4	1	2	1
1	2	B	1	2	3	3	3	2
1	3	A	2	1	1	3	2	2
1	3	B	1	2	4	1	3	3
1	4	A	2	2	2	2	1	2
1	4	B	1	1	1	1	3	3
1	5	A	2	1	1	3	2	3
1	5	B	1	2	3	2	1	2
1	6	A	1	1	4	1	1	2
1	6	B	2	2	1	2	3	1
1	7	A	1	2	1	1	1	1
1	7	B	2	1	2	2	2	3
1	8	A	1	1	2	2	3	3
1	8	B	2	2	3	1	2	2
2	1	A	1	1	3	2	2	1
2	1	B	2	2	1	1	1	2
2	2	A	1	2	4	1	3	2
2	2	B	2	1	3	2	1	3
2	3	A	2	1	1	1	3	2
2	3	B	1	2	4	3	1	3
2	4	A	1	2	2	3	2	1
2	4	B	2	1	1	2	1	2
2	5	A	1	1	1	3	1	1
2	5	B	2	2	2	1	3	3
2	6	A	1	1	3	3	3	2
2	6	B	2	2	2	1	1	1
2	7	A	2	1	4	3	3	2
2	7	B	1	2	1	2	2	1
2	8	A	1	2	1	2	1	3
2	8	B	2	1	3	1	2	1
3	1	A	2	2	4	2	3	1
3	1	B	1	1	2	1	2	3
3	2	A	2	2	3	2	3	3
3	2	B	1	1	2	3	1	1
3	3	A	1	2	2	2	2	2
3	3	B	2	1	4	3	1	3
3	4	A	2	2	3	3	1	3
3	4	B	1	1	4	2	3	1
3	5	A	1	2	3	1	2	3
3	5	B	2	1	2	3	3	2
3	6	A	2	1	3	1	1	1
3	6	B	1	2	1	3	2	2
3	7	A	2	2	4	2	2	3
3	7	B	1	1	3	3	3	1
3	8	A	1	1	2	1	1	3
3	8	B	2	2	4	3	2	1
All	9 (Dominance)^2	A	1	1	1	1	1	1
All	9 (Dominance)^2	B	2	2	4	3	3	3
All	10 (Holdout)^2	A	1	1	3	2	1	2
All	10 (Holdout)^2	B	2	2	1	1	3	3

¹Choice tasks were presented in a randomized order for each participant. ² The dominance test and holdout tasks were not part of the experimental design and were not included in the analytic models.

Table A4. Dominance test task

Attribute	Treatment Option A (More Preferable/Dominant)	Treatment Option B (Less Preferable/Dominated)
Procedure invasiveness	Less invasive	Invasive
Procedure recovery	Moderate recovery	Intensive recovery
Risk that MR symptoms reappear	1% risk	40% risk
Risk of needing a reintervention	1% risk	15% risk
Risk of disabling stroke	1% risk	5% risk
Risk of new onset atrial fibrillation	1% risk	20% risk

Table A5. Holdout task for comparison of choice probabilities

Attribute	Treatment Option A (“Minimally Invasive Procedure-like’ Profile)	Treatment Option B (‘open-heart Surgery-like’ Profile)
Procedure invasiveness	Less invasive	Invasive
Procedure recovery	Moderate recovery	Intensive recovery
Risk that symptoms reappear	15% risk	1% risk
Risk of needing reintervention	8% risk	1% risk
Risk of disabling stroke	1% risk	5% risk
Risk of new onset atrial fibrillation	10% risk	20% risk

Table A6. Inclusion and exclusion criteria

Attribute	Cohort A: Clinically Confirmed dMR Diagnosis Cohort	Cohort B: Self-Reported MR Diagnosis
Inclusion Criteria	Age > 18 Years Old
	Willing To Participate
	Speaks, Reads and Understands English
	Clinically confirmed diagnosis of current or previously surgically treated severe MR due to primary/degenerative MV disease. and MR is severe (or was previously, if treated surgically), meaning the patient meets the clinical guidelines for surgical intervention, but may not yet have been treated surgically.	Self-Reported Diagnosis of MR
Optional Inclusion Criteria	Patients were not required to meet the following criteria but remained eligible to participate if they had been: 1) Treated surgically or procedurally with MV repair or replacement for severe MR due to primary/degenerative MV disease. 2) Treated for aortic stenosis surgically or with transcatheter devices
Exclusion Criteria	Does not have access to a web interface on a computer or tablet for the purposes of completing the survey
	Diagnosis of secondary/functional MR (as clinically confirmed or self-reported, depending on screening method for the relevant cohort)
	Has cognitive or visual impairment that would interfere with the ability to participate	MR is self-reported as currently mild and not previously treated surgically or procedurally with MV repair or replacement
	Has acute psychopathology that would interfere with ability to participate

Table A7. Participant sociodemographic characteristics (n = 118)

Sociodemographic Characteristic, n (%)^a	Diagnosis Cohort		Total (n = 118)	P-value^b
	Clinically Confirmed (n = 30)	Self-reported (n = 88)
Gender^1				0.073
Male	20 (66.7%)	42 (47.7%)	62 (52.5%)
Female	10 (33.3%)	46 (52.3%)	56 (47.5%)
Age				0.034
Mean (SD)	68.8 (12.0)	62.4 (14.7)	64.0 (14.3)
Median (IQR)	70 (9)	66 (16.5)	67 (13)
Min – Max	22–92	23–88	22–92
Age Group^1				0.042
64 or younger	7 (23.3%)	39 (44.3%)	46 (39.0%)
65 or older	23 (76.7%)	49 (55.7%)	72 (61.0%)
Ethnicity^2				0.571
Hispanic, Latino or Spanish origin	0 (0.0%)	4 (4.5%)	4 (3.4%)
Not Hispanic, Latino or Spanish origin	30 (100%)	84 (95.5%)	114 (96.6%)
Race
American Indian or Alaska Native^2	0 (0.0%)	2 (2.3%)	2 (1.7%)	1.000
Asian^2	0 (0.0%)	2 (2.3%)	2 (1.7%)	1.000
^2Black or African American^2	0 (0.0%)	3 (3.4%)	3 (2.5%)	0.570
Native Hawaiian or Other Pacific Islander	0 (0.0%)	0 (0.0%)	0 (0.0%)
White^2	30 (100%)	81 (92.0%)	111 (94.1%)	0.189
Other: Mexican, Puerto Rican^2	0 (0.0%)	1 (1.1%)	1 (0.8%)	1.000
Marital status^2				0.731
Single, never married	2 (6.7%)	9 (10.2%)	11 (9.3%)
Married or domestic partnership	23 (76.7%)	70 (79.5%)	93 (78.8%)
Widowed	2 (6.7%)	4 (4.5%)	6 (5.1%)
Divorced or separated	3 (10.0%)	5 (5.7%)	8 (6.8%)
Parental status^1				0.223
Has children	26 (86.7%)	67 (76.1%)	93 (78.8%)
No children	4 (13.3%)	21 (23.9%)	25 (21.2%)
Highest education level^2				<.001
No high school	0 (0.0%)	0 (0.0%)	0 (0.0%)
High school or GED	9 (30.0%)	2 (2.3%)	11 (9.3%)
Some college or technical school	14 (46.7%)	27 (30.7%)	41 (34.7%)
College (Bachelor’s degree)	4 (13.3%)	30 (34.1%)	34 (28.8%)
Graduate (Master’s degree)	1 (3.3%)	21 (23.9%)	22 (18.6%)
Post-Graduate (Doctoral degree)	2 (6.7%)	8 (9.1%)	10 (8.5%)
Employment status^1				0.140
Employed full-time	2 (6.7%)	23 (26.1%)	25 (21.2%)
Employed part-time	4 (13.3%)	13 (14.8%)	17 (14.4%)
Unemployed	2 (6.7%)	3 (3.4%)	5 (4.2%)
Student	0 (0.0%)	0 (0.0%)	0 (0.0%)
Stay-at-home	0 (0.0%)	0 (0.0%)	0 (0.0%)
On disability	2 (6.7%)	6 (6.8%)	8 (6.8%)
Retired	20 (66.7%)	43 (48.9%)	63 (53.4%)
Annual Household Income^2				0.011
$10,000 or less	3 (10.0%)	0 (0.0%)	3 (2.5%)
$10,001-$25,000	2 (6.7%)	9 (10.2%)	11 (9.3%)
$25,001-$50,000	4 (13.3%)	14 (15.9%)	18 (15.3%)
$50,001-$75,000	8 (26.7%)	12 (13.6%)	20 (16.9%)
$75,001-$150,000	8 (26.7%)	19 (21.6%)	27 (22.9%)
$150,001-$250,000	0 (0.0%)	16 (18.2%)	16 (13.6%)
$250,001 or more	1 (3.3%)	5 (5.7%)	6 (5.1%)
Prefer not to answer	3 (10.0%)	13 (14.8%)	16 (13.6%)
Missing	1 (3.3%)	0 (0.0%)	1 (0.8%)
Area of Residence Population Size^2				<0.001
50,000 individuals or more (urban)	9 (30.0%)	53 (60.2%)	62 (52.5%)
2,500 to 50,000 individuals (suburban)	10 (33.3%)	31 (35.2%)	41 (34.7%)
Less than 2,500 individuals (rural)	11 (36.7%)	4 (4.5%)	15 (12.7%)
US Region of Residence^2,c				<0.001
Northeast	0 (0.0%)	10 (11.4%)	10 (8.5%)
Midwest region	19 (63.3%)	9 (10.2%)	28 (23.7%)
South region	0 (0.0%)	44 (50.0%)	44 (37.3%)
West region	10 (33.3%)	24 (27.3%)	34 (28.8%)
Missing	1 (3.3%)	1 (1.1%)	2 (1.7%)
Terminal Illness^2				0.065
Yes	0 (0.0%)	3 (3.4%)	3 (2.5%)
No	27 (90.0%)	84 (95.5%)	111 (94.1%)
Prefer not to answer	3 (10.0%)	1 (1.1%)	4 (3.4%)

IQR = Interquartile Range, SD = Standard Deviation, ^aMissing data included in calculation of percentages where applicable, ^bParametric P value for between-group comparisons. T-test for continuous variables, ^c P value corresponds to group comparison of US regions (not states), ¹Chi-squared, ²Fisher’s Exact Test for categorical variables

Table A8. Participant clinical characteristics (n = 118)

Clinical Characteristic, n (%)^a	Diagnosis Cohort		Total (n = 118)	P-value^b
	Clinically Confirmed (n = 30)	Self-reported (n = 88)
Years since diagnosis				<0.001
Missing	30 (0)	87 (1)	117 (1)
Mean (SD)	5.50 (6.93)	19.47 (18.70)	15.89 (17.57)
Median (IQR)	3.22 (5.26)	12.56 (25.86)	9.49 (20.80)
Min – Max	0.16–27.64	0.02–79.44	0.02–79.44
Median (IQR)	70 (9)	66 (16.5)	67 (13)
Min – Max	22–92	23–88	22–92
NYHA Classification^2,c				<0.001
Class I	13 (43.3%)	2 (2.3%)	15 (12.7%)
Class II	9 (30.0%)	7 (8.0%)	16 (13.6%)
Class III	6 (20.0%)	6 (6.8%)	12 (10.2%)
Class IV	0 (0.0%)	1 (1.1%)	1 (0.8%)
Not specified/Not sure	2 (6.7%)	72 (81.8%)	74 (62.7%)
Type of Mitral Prolapse^2,c				0.044
P2	5 (16.7%)	3 (3.4%)	8 (6.8%)
Bi leaflet	9 (30.0%)	19 (21.6%)	28 (23.7%)
Anterior	1 (3.3%)	5 (5.7%)	6 (5.1%)
Not specified/Not sure	15 (50.0%)	61 (69.3%)	76 (64.4%)
Comorbidities and Complications
Atrial fibrillation^1	15 (50.0%)	45 (51.1%)	60 (50.8%)	0.914
Chronic obstructive pulmonary disease (COPD)^2	2 (6.7%)	4 (4.5%)	6 (5.1%)	0.643
Diabetes^1	4 (13.3%)	17 (19.3%)	21 (17.8%)	0.459
Kidney failure^2	2 (6.7%)	6 (6.8%)	8 (6.8%)	1.000
Angina^2	0 (0.0%)	7 (8.0%)	7 (5.9%)	0.189
Coronary artery disease^2	5 (16.7%)	13 (14.8%)	18 (15.3%)	0.775
Other: Aortic valve disease (regurgitation/stenosis/unspecified)^2	1 (3.3%)	3 (3.4%)	4 (3.4%)	1.000
Other: Tricuspid valve disease (regurgitation/unspecified)^2	0 (0.0%)	3 (3.4%)	3 (2.5%)	0.570
Other: Regurgitation in all heart valves^2	0 (0.0%)	1 (1.1%)	1 (0.8%)	1.000
Other: Other arrhythmia(s) (PVCs/PACs/SVT/AV block)^2	0 (0.0%)	3 (3.4%)	3 (2.5%)	0.570
Other: Atrioventricular septal defect^2	0 (0.0%)	2 (2.3%)	2 (1.7%)	1.000
Other: Atypical chest pain^2	0 (0.0%)	1 (1.1%)	1 (0.8%)	1.000
Other: Cardiac myxoma, genetic^2	0 (0.0%)	1 (1.1%)	1 (0.8%)	1.000
Other: Congenital heart disease^2	0 (0.0%)	1 (1.1%)	1 (0.8%)	1.000
Other: Congestive heart failure^2	0 (0.0%)	2 (2.3%)	2 (1.7%)	1.000
Other: Endocarditis^2	0 (0.0%)	2 (2.3%)	2 (1.7%)	1.000
Other: Heart attack(s)^2	0 (0.0%)	1 (1.1%)	1 (0.8%)	1.000
Other: Interstitial lung disease^2	0 (0.0%)	1 (1.1%)	1 (0.8%)	1.000
Other: Peripheral artery disease^2	1 (3.3%)	0 (0.0%)	1 (0.8%)	0.254
Other: Pulmonary hypertension^2	0 (0.0%)	3 (3.4%)	3 (2.5%)	0.570
Other: Stroke^2	0 (0.0%)	1 (1.1%)	1 (0.8%)	1.000
None^1	9 (30.0%)	26 (29.5%)	35 (29.7%)	0.962
Missing^2	1 (3.3%)	0 (0.0%)	1 (0.8%)	0.254
Mitral regurgitation symptoms experienced
Anxiety^1	4 (13.3%)	20 (22.7%)	24 (20.3%)	0.270
Blood clots^2	1 (3.3%)	5 (5.7%)	6 (5.1%)	1.000
Depression^2	3 (10.0%)	11 (12.5%)	14 (11.9%)	1.000
Fatigue^1	19 (63.3%)	42 (47.7%)	61 (51.7%)	0.140
Heart murmur^1	22 (73.3%)	73 (83.0%)	95 (80.5%)	0.251
Heart palpitations^1	11 (36.7%)	42 (47.7%)	53 (44.9%)	0.293
High blood pressure (hypertension)^1	10 (33.3%)	12 (13.6%)	22 (18.6%)	0.017
Loss of appetite^2	5 (16.7%)	6 (6.8%)	11 (9.3%)	0.144
Shortness of breath (dyspnea)^1	22 (73.3%)	51 (58.0%)	73 (61.9%)	0.134
Swelling of the abdomen (ascites)^2	3 (10.0%)	9 (10.2%)	12 (10.2%)	1.000
Swelling of the hands, legs, or feet (edema)^1	11 (36.7%)	31 (35.2%)	42 (35.6%)	0.887
Weight change (loss or gain)^1	9 (30.0%)	11 (12.5%)	20 (16.9%)	0.027
Other: Chest pressure/tightness^2	0 (0.0%)	1 (1.1%)	1 (0.8%)	1.000
Other: Fainting^2	0 (0.0%)	1 (1.1%)	1 (0.8%)	1.000
Other: Fluid retention in lungs^2]	0 (0.0%)	1 (1.1%)	1 (0.8%)	1.000
No symptoms experienced^2	1 (3.3%)	1 (1.1%)	2 (1.7%)	0.445
Global physical health rating^2				0.202
Excellent	1 (3.3%)	10 (11.4%)	11 (9.3%)
Very good	12 (40.0%)	22 (25.0%)	34 (28.8%)
Good	12 (40.0%)	37 (42.0%)	49 (41.5%)
Fair	2 (6.7%)	16 (18.2%)	18 (15.3%)
Poor	2 (6.7%)	3 (3.4%)	5 (4.2%)
Missing	1 (3.3%)	0 (0.0%)	1 (0.8%)
Ability to carry out everyday activities (physical functioning)^2				0.778
Completely	14 (46.7%)	39 (44.3%)	53 (44.9%)
Mostly	8 (26.7%)	18 (20.5%)	26 (22.0%)
Moderately	5 (16.7%)	16 (18.2%)	21 (17.8%)
A little	2 (6.7%)	12 (13.6%)	14 (11.9%)
Not at all	0 (0.0%)	3 (3.4%)	3 (2.5%)
Missing	1 (3.3%)	0 (0.0%)	1 (0.8%)
Activities (ever) limited by mitral regurgitation
recreational activities1	18 (60.0%)	48 (54.5%)	66 (55.9%)	0.603
Working or doing household chores1	16 (53.3%)	44 (50.0%)	60 (50.8%)	0.752
Visiting family or friends out of home1	8 (26.7%)	27 (30.7%)	35 (29.7%)	0.678
None of the above1	11 (36.7%)	33 (37.5%)	44 (37.3%)	0.935
Admitted to hospital for heart failure in the last year^1				0.715
Yes	6 (20.0%)	15 (17.0%)	21 (17.8%)
No	24 (80.0%)	73 (83.0%)	97 (82.2%)
Any previous procedure requiring overnight hospital stay^1				0.359
Yes	23 (76.7%)	74 (84.1%)	97 (82.2%)
No	7 (23.3%)	14 (15.9%)	21 (17.8%)

IQR = Interquartile Range, SD = Standard Deviation, ^aMissing data included in calculation of percentages where applicable, ^bParametric P value for between-group comparisons. T-test for continuous variables, ^c Differences between diagnosis cohorts are not statistically significant when ‘Not Specified/Not sure’ responses are excluded, ¹Chi-squared, ²Fisher’s Exact Test for categorical variables

Table A9. Participant mitral regurgitation treatment history characteristics (n = 118)

Treatment Characteristic, n (%)^a	Diagnosis Cohort		Total (n = 118)	P-value^b
	Clinically Confirmed (n = 30)	Self-reported (n = 88)
Current mitral regurgitation symptom management
Taking medicines^1	13 (43.3%)	44 (50.0%)	57 (48.3%)	0.528
Watching and waiting^1	3 (10.0%)	34 (38.6%)	37 (31.4%)	0.004
Waiting to undergo a medical procedure^2	1 (3.3%)	7 (8.0%)	8 (6.8%)	0.678
Received a medical procedure^1	24 (80.0%)	48 (54.5%)	72 (61.0%)	0.014
Other ^2,c	0 (0.0%)	4 (4.5%)	4 (3.4%)	0.571
Treated with previous mitral procedure^1,d
Yes	24 (80.0%)	57 (64.8%)	81 (68.6%)	0.121
No	6 (20.0%)	31 (35.2%)	37 (31.4%)
Length of hospital stay after mitral procedure (nights)	(N = 24)^e	(N = 47)^e	(N = 71)^e	0.759
SD	8.9 (17.4)	8.0 (7.1)	8.3 (11.5)
Median (IQR)	5 (3.5)	6 (5)	6 (4)
Min – Max	1–90	2–45	1–90
Transportation to hospital for mitral procedure2	(N = 24)^e	(N = 47)^e	(N = 71)^e	0.471
Walked	0 (0.0%)	3 (6.4%)	3 (4.2%)
Public transportation	0 (0.0%)	2 (4.3%)	2 (2.8%)
Car/taxi	22 (91.7%)	35 (74.5%)	57 (80.3%)
Ambulance	2 (8.3%)	4 (8.5%)	6 (8.5%)
Airplane	0 (0.0%)	3 (6.4%)	3 (4.2%)
Travel time to hospital for mitral procedure (minutes)	(N = 24)^e	(N = 47)^e	(N = 71)^e	0.584
n (missing)	24 (0)	47 (0)	71 (0)
Mean (SD)	72.1 (59.1)	93.1 (181.5)	86.0 (151.3)
Median (IQR)	45 (72.5)	40 (40)	40 (50)
Min – Max	15–210	5–1080	5–1080
Two main reasons for choosing hospital/surgeon for previous/hypothetical mitral procedure
Location/easy to get to^2	8 (26.7%)	8 (9.1%)	16 (13.6%)	0.027
Length of wait for an appointment^2	2 (6.7%)	4 (4.5%)	6 (5.1%)	0.643
Convenient appointment times/days^2	1 (3.3%)	2 (2.3%)	3 (2.5%)	1.000
Reputation of hospital/surgeon^1	15 (50.0%)	55 (62.5%)	70 (59.3%)	0.229
Hospital’s/surgeon’s level of experience with the procedure^1	13 (43.3%)	58 (65.9%)	71 (60.2%)	0.029
Personal experience^2	3 (10.0%)	4 (4.5%)	7 (5.9%)	0.368
Recommended by friends/family^2	1 (3.3%)	3 (3.4%)	4 (3.4%)	1.000
Recommended by other healthcare provider^1	10 (33.3%)	26 (29.5%)	36 (30.5%)	0.697
Cost of the procedure at the hospital^2	1 (3.3%)	2 (2.3%)	3 (2.5%)	1.000
Previous procedure to repair or replace aortic valve^2				0.746
Yes	5 (16.7%)	16 (18.2%)	21 (17.8%)
No	25 (83.3%)	68 (77.3%)	93 (78.8%)
Not sure	0 (0.0%)	4 (4.5%)	4 (3.4%)
Any other previous heart procedure^2				0.058
Yes	9 (30.0%)	44 (50.0%)	53 (44.9%)
No	18 (60.0%)	43 (48.9%)	61 (51.7%)
Not sure	2 (6.7%)	1 (1.1%)	3 (2.5%)
Missing	1 (3.3%)	0 (0.0%)	1 (0.8%)

IQR = Interquartile Range, SD = Standard Deviation, ^aMissing data included in calculation of percentages where applicable, ^bParametric P value for between-group comparisons. T-test for continuous variables, ^c ‘Other’ symptom management responses included ‘my cardiologist recommends ablation,’ ‘clinical trial candidate,’ ‘diet,’ and ‘maintain awareness of limitations.,’ ^dBased on response to either of 2 survey questions, ^ePercentage denominator represents only participants who saw the survey question ¹Chi-squared, ²Fisher’s Exact Test for categorical variables

Table A10. DCE data quality metrics for the full sample including all complete survey responses (n = 145)

Data Quality Metric, n (%)^a	Diagnosis Cohort		Total (n = 145)	P-value^b
	Clinically Confirmed (n = 34)	Self-reported (n = 111)
Failed dominance test^1
Yes	3 (8.8%)	15 (13.5%)	18 (12.4%)	0.765
No	30 (88.2%)	96 (86.5%)	126 (86.9%)
Missing	1 (2.9%)	0 (0.0%)	1 (0.7%)
Straight-lined^1
Yes	3 (8.8%)	14 (12.6%)	17 (11.7%)	0.763
No	31 (91.2%)	97 (87.4%)	128 (88.3%)
Failed dominance test^1 or straight-lined^2,c
Either	4 (11.8%)	22 (19.8%)	26 (17.9%)
Both	2 (5.9%)	7 (6.3%)	9 (6.2%)
Failed dominance test only	1 (2.9%)	8 (7.2%)	9 (6.2%)
Straight-lined only	1 (2.9%)	7 (6.3%)	8 (5.5%)
Neither	30 (88.2%)	89 (80.2%)	119 (82.1%)
Completed survey in < 7 minutes^1
Yes	0 (0.0%)	1 (0.9%)	1 (0.7%)	1.000
No	34 (100%)	110 (99.1%)	144 (99.3%)
Excluded from final analytic sample^2
Yes	4 (11.8%)	23 (20.7%)	27 (18.6%)	0.240
No	30 (88.2%)	88 (79.3%)	118 (81.4%)

^aMissing data included in calculation of percentages where applicable.

^bParametric P value for between-group comparisons: ¹Fisher’s exact for categorical variables or ²Chi-squared ^c P value corresponds to group comparison of ‘Either’ vs ‘Neither’ categories.

Table A11. Participant evaluation of DCE (n = 118)

DCE Task Evaluation, n (%)^a	Diagnosis Cohort		Total (n = 118)	P-value^b
	Clinically Confirmed (n = 30)	Self-reported (n = 88)
Questions easy to understand^2				0.120
Strongly agree/agree	25 (83.3%)	83 (94.3%)	108 (91.5%)
Strongly disagree/disagree/neutral	5 (16.7%)	5 (5.7%)	10 (8.5%)
Questions easy to answer^1				0.282
Strongly agree/agree	21 (70.0%)	70 (79.5%)	91 (77.1%)
Strongly disagree/disagree/neutral	9 (30.0%)	18 (20.5%)	27 (22.9%)
Answered all questions consistent with preferences				1.000
Strongly agree/agree	29 (96.7%)	86 (97.7%)	115 (97.5%)
Strongly disagree/disagree/neutral	1 (3.3%)	2 (2.3%)	3 (2.5%)
Questions were relevant^2				0.091
Strongly agree/agree	24 (80.0%)	81 (92.0%)	105 (89.0%)
Strongly disagree/disagree/neutral	6 (20.0%)	7 (8.0%)	13 (11.0%)

^aMissing data included in calculation of percentages where applicable, ^bParametric P value for between-group comparisons. T-test for continuous variables, ¹Chi-squared, ²Fisher’s Exact Test for categorical variables

Table A12. Swait-Louviere scale test procedure results

Conditional Logit Model		Log Likelihood	1^st LR test: P Value (9 df)	2^nd LR test: P Value (1 df)
Clinically confirmed diagnosis cohort (N = 30)		−124.52
Self-reported diagnosis cohort (N = 88)		−367.65	0.477
Both cohorts pooled (N = 118)	Varying Scale Factors (Scale Adjusted/Heteroscedastic)	−496.46		0.756
	Equal scale factors	−496.51

Table A13. Conditional/multinomial logit model results (main effects, n = 118)

Variable/Parameter	Estimate	95% CI	P value
Treatment Attribute Levels
Procedure invasiveness
Less invasive	0.51	(0.35, 0.67)	<0.001
Invasive (ref. level)	0.00
Procedure recovery
Moderate	0.19	(0.04, 0.35)	0.015
Intensive (ref. level)	0.00
Risk that symptoms reappear (per +1 pp)	−0.04	(−0.05, −0.03)	<0.001
Risk of needing reintervention (per +1 pp)	−0.07	(−0.09, −0.05)	<0.001
Risk of disabling stroke
1%	0.65	(0.43, 0.88)	<0.001
2%	0.26	(0.05, 0.48)	0.015
5% (ref. level)	0.00
Risk of new onset atrial fibrillation
1%	0.77	(0.53, 1.00)	<0.001
10%	0.56	(0.35, 0.78)	<0.001
20% (ref. level)	0.00
Model Fit Statistics
Log Likelihood	−496.51
AIC	1009.01
BIC	1047.75

AIC = Akaike Information Criterion, BIC = Bayesian Information Criterion, CI = Confidence Interval, pp = percentage point, ref = reference.

Table A14. Error component logit model results (main effects, n = 118) – categorical specification

Variable/Parameter	Estimate	95% CI	P value
Treatment Attribute Levels (Fixed Coefficients)
Procedure invasiveness
Less invasive	0.56	(0.38, 0.73)	<0.001
Invasive (ref. level)	0.00
Procedure recovery
Moderate	0.20	(0.03, 0.36)	0.019
Intensive (ref. level)	0.00
Risk that symptoms reappear (per +1 pp)
1% (ref. level)	0.00
8%	−0.55	(−0.85, −.025)	<0.001
15%	−0.63	(−0.91, −0.35)	<0.001
40%	−1.81	(−2.14, −1.47)	<0.001
Risk of needing reintervention (per +1 pp)
1% (ref. level)	0.00
8%	−0.56	(−0.79, −0.34)	<0.001
15%	−1.02	(−1.28, −0.75)	<0.001
Risk of disabling stroke
1%	0.70	(0.46, 0.94)	<0.001
2%	0.26	(0.03, 0.49)	0.027
5% (ref. level)	0.00
Risk of new onset atrial fibrillation
1%	0.85	(0.59, 1.10)	<0.001
10%	0.59	(0.36, 0.82)	<0.001
20% (ref. level)	0.00
Error Component (Random Coefficient)
Alternative-specific intercept SD	0.62	(0.34, 0.90)	<0.001
Model Fit Statistics
Log Likelihood	−490.64
AIC	1005.28
BIC	1063.38

AIC = Akaike Information Criterion, BIC = Bayesian Information Criterion, CI = Confidence Interval, pp = percentage point, ref = reference, SD = Standard Deviation

Table A15. Conditional/multinomial logit model results (main effects, n = 118) – categorical specification

Variable/Parameter	Estimate	95% CI	P value
Treatment Attribute Levels (Fixed Coefficients)
Procedure invasiveness
Less invasive	0.51	(0.35, 0.67)	<0.001
Invasive (ref. level)	0.00
Procedure recovery
Moderate	0.18	(0.03, 0.34)	0.021
Intensive (ref. level)	0.00
Risk that symptoms reappear (per +1 pp)
1% (ref. level)	0.00
8%	−0.50	(−0.78, −.022)	<0.001
15%	−0.59	(−0.84, −0.33)	<0.001
40%	−1.68	(−1.99, −1.37)	<0.001
Risk of needing reintervention (per +1 pp)
1% (ref. level)	0.00
8%	−0.53	(−0.74, −0.32)	<0.001
15%	−0.95	(−1.20, −0.71)	<0.001
Risk of disabling stroke
1%	0.66	(0.43, 0.88)	<0.001
2%	0.24	(0.02, 0.46)	0.030
5% (ref. level)	0.00
Risk of new onset atrial fibrillation
1%	0.79	(0.55, 1.03)	<0.001
10%	0.55	(0.33, 0.77)	<0.001
20% (ref. level)	0.00
Model Fit Statistics
Log Likelihood	−494.90
AIC	1011.80
BIC	1065.06

AIC = Akaike Information Criterion, BIC = Bayesian Information Criterion, CI = Confidence Interval, pp = percentage point, ref = reference, SD = Standard Deviation.