Estimating Utility Values Using the DMD-QoL: A Disease-Specific Preference-Based Measure for Duchenne Muscular Dystrophy (DMD)

Abstract

Introduction

The DMD-QoL is a disease-specific, preference-based measure (PBM). Its design was informed by the experiences of individuals with Duchenne muscular dystrophy (DMD), and it was developed to reflect a broader conceptualization of wellness and quality of life (QoL) compared with other PBMs. This study estimated and compared DMD-QoL and Health Utilities Index (HUI) utility in DMD.

Methods

Individuals with DMD or their caregivers completed the DMD-QoL and HUI-2 at baseline and 12 months. Participants were classified into health states based on upper and lower limb function. Mean utility and change over 12 months were calculated by baseline health state and respondent type; change in utility was also estimated among those in whom declining ambulatory function led to a health state transition.

Results

Among 155 participants (caregiver-respondents, 78.7%), mean (standard deviation [SD]) patient age at baseline was 14.1 (6.9) years. Across health states, mean (SD) baseline DMD-QoL utility ranged from 0.62 (0.35) to 0.78 (0.10) (patient-assessed) and 0.63 (0.17) to 0.84 (0.12) (caregiver-assessed). Mean HUI-2 values ranged from 0.49 (0.07) to 0.96 (0.06) (patient-assessed) and 0.44 (0.09) to 0.85 (0.11) (caregiver-assessed). While DMD-QoL and HUI-2 utilities were relatively stable over 12 months, a clinically important and significant loss in HUI-2 utility (mean, − 0.16 [95% confidence interval, − 0.22, − 0.10]) was observed among 17 (11.0%) participants with worsening ambulatory function. No such change was observed with the DMD-QoL (0.00 [− 0.06, 0.06]).

Conclusion

DMD-QoL utilities did not vary meaningfully across health states compared with HUI-2 scores, which tended to decline with deteriorating mobility. The relative stability of DMD-QoL scores may reflect its broader conceptualization of QoL, grounded in the experiences of individuals with DMD.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s12325-026-03700-6.

Key Summary Points

Why carry out this study?

A number of generic (i.e., disease agnostic) instruments (e.g., the Health Utilities Index Mark 2 [HUI-2]) exist for estimating health state utilities, which are required inputs for cost effectiveness analysis of health interventions. However, generic measures may not always be appropriate for measuring utilities for a particular condition or population.

For estimating utilities in Duchenne muscular dystrophy (DMD), the relatively new DMD-QoL measure has been designed to incorporate attributes relevant to the DMD community within its descriptive system and to be used across the DMD life course.

At present, how utility values estimated using the DMD-QoL compare with those of established generic measures is unclear. The objective of this study was to quantify and compare utility values for a range of health states in DMD using both the DMD-QoL and the HUI-2.

What was learned from the study?

In this study, distinctions were observed between DMD-QoL and HUI-2 for estimating utilities in DMD. While the HUI-2 appears more sensitive to functional decline, the DMD-QoL reflects a broader conceptualization of wellness and quality of life based on the experiences of individuals with DMD.

These findings suggest measure choice should match goals, capturing functional change or broader wellbeing.

Introduction

Studies of health-related quality of life (HRQoL) have documented the devastating effects of Duchenne muscular dystrophy (DMD) from the perspective of patients, their caregivers, and their families [2, 3]. The clinical progression of DMD is characterized by loss of ambulation and upper limb function, respiratory insufficiency, cardiomyopathy, and premature mortality [4]. The dramatic impact of clinical progression on HRQoL [2, 3] has been elucidated by studies examining DMD-specific health states; health state utility declines with declining ambulatory, upper limb, and respiratory function in DMD [5, 6].

Health state utility values are required inputs for cost-utility models, and many health technology assessment agency decision-makers recommend using standardized generic preference-based measures (PBMs) such as the EQ-5D and the Health Utilities Index (HUI) to estimate these values [7, 8]. However, generic PBMs may not always be available, appropriate, or sufficient for measuring utilities for a particular condition or population [9]. In DMD, for example, evidence on the ability of existing generic PBMs to adequately capture HRQoL is equivocal [10, 11]. A study by Crossnohere et al. (2021) “found support for the appropriateness of EQ-5D to assess health status in Duchenne” [10]. However, Powell et al. countered that the study’s assessment of the content validity of the EQ-5D was limited [11].

The DMD-QoL, which was developed in collaboration with the DMD community, is the first condition-specific HRQoL measure that can also be used to estimate utility values in DMD [9, 12]. The DMD-QoL is a 14-item instrument developed in the UK to measure HRQoL based on patient responses. It was designed to incorporate attributes relevant to the DMD community within its descriptive system and to be used across the DMD life course. In general, the DMD-QoL reflects a broader conceptualization of mobility, wellbeing, and quality of life than other non-disease-specific PBMs. While initially developed specifically to assess HRQoL, it was subsequently adapted to estimate utility values based on responses to 8 of the 14 items and incorporates weightings provided by members of the UK general public [9, 12]. However, how utility values estimated using the DMD-QoL compared with those of other PBMs is unclear. The objective of this study was to quantify and compare utility values for a range of health states in DMD using both the DMD-QoL and the HUI.

Methods

Participants and Recruitment

Individuals with DMD, and/or family caregivers reporting on their behalf, were recruited through Parent Project Muscular Dystrophy (PPMD), a USA-based patient advocacy organization. Members of the PPMD mailing list (n = 2550) received email notification of the study; among families that expressed interest in participating, a screening phone call was conducted to determine which household members were eligible. Parents often asked their child whether they would like to participate, and parents of younger patients were present during screening. Eligible participants had a confirmed diagnosis of DMD, lived in the US, and could read and respond to survey questions in English and provide informed consent. Self-reporting patients were required to be 12–40 years old; caregivers could report on behalf of someone aged 4–40 years. Enrollment and baseline data collection occurred between August 2021 and February 2022, and participants received an honorarium approved by the ethics board.

Survey Materials and Measures

Participants completed an online survey at baseline and 12 months that included the DMD-QoL and HUI [5]. Utilities are derived from responses to the HUI questionnaire according to two complementary systems: the HUI mark 2 (HUI-2) and the HUI mark 3 (HUI-3) [13]. Presented here are HUI-2 data that were also selected for use in a recent economic evaluation in DMD [14]. Prior research suggests that the HUI-2’s descriptive system may be more relevant to the DMD community than other generic PBMs [15], hence its selection in the present study. The HUI-2 considers levels of impairment in sensation, mobility, emotion, cognition, self-care, pain, and fertility (optional) on a scale of 1 (no impairment) to 4 or 5 (most severe impairment). DMD-QoL attributes score mobility, hand function, difficulty breathing, pain, tiredness, worry, participation, and feeling good about oneself on a scale of 1 to 4, with higher scores indicating greater impairment [9].

Self-report or proxy versions of each measure were administered as appropriate. The self-reported version of the DMD-QoL is designed for patients aged ≥ 10 years, and the proxy version for those aged ≥ 7 years [12]. For the HUI, the recommended age for the self-administered version used here is ≥ 12 years, and ≥ 5 years for proxy report [13]. Given the lack of PBMs available for measuring HRQoL among very young children, caregivers were allowed to report on behalf of children below the recommended age threshold.

Demographic and clinical data, including data on treatment with phosphorodiamidate morpholino oligomers (PMOs), cardioprotective medications, and steroids, were collected to characterize the sample and patient functional health status. Responses to the clinical questions were used to classify patients into one of four health states based on lower and upper limb function: early ambulatory, late ambulatory, early non-ambulatory, and late non-ambulatory [5]. These questions were based on existing functional classification systems used in DMD [16, 17], developed in collaboration with a clinical expert in DMD [18], and adapted to enable patients and caregivers to self-classify without clinical input.

Before implementation, the online platform was pilot tested for functionality; the relevance and acceptability of the demographic and clinical questions had previously been examined in a qualitative interview study in DMD [18]. A license was obtained to use the HUI for this study, and permission to use the DMD-QoL was granted by the developers. The study was conducted in accordance with the Declaration of Helsinki. Study conduct was approved by Advarra Institutional Review Board (IRB; study number: Pro00055581) and WCG IRB (IRB tracking number: 20224611). All participants provided informed consent.

Analysis

Participant demographic and clinical characteristics were summarized using measures of central tendency for continuous data and frequency distributions for categorical data. Participant responses to HUI questions were converted into an index score using the developers’ algorithm and the original tariffs (derived from a Canadian population) [19, 20]. The range of possible utility values derived from HUI-2 attribute levels is 1.0 to − 0.03 [13]. For the DMD-QoL, utility values were estimated by adding the sum of attribute-level-specific utility decrements to 1 [21]. Utility decrements based on UK general public preferences were applied; country-specific tariffs for the US are not presently available. Possible DMD-QoL utility values calculated may range from 1 to − 0.559.

Mean (standard deviation [SD]) DMD-QoL and HUI-2 utility values at baseline and 12 months were stratified by patient baseline functional health state and respondent type (patient vs. caregiver-proxy). Pearson’s correlation coefficient was calculated for DMD-QoL vs. HUI-2 utility values at baseline, stratified by health state.

Linear regression was used to explore factors associated with baseline utility. In these analyses, responses from caregivers and patients were considered together, with respondent type included as a model covariate, along with baseline health state (i.e., lower and upper limb function), use of ventilation, presence of cardiomyopathy, patient age, and treatment status. Covariates that were significant predictors of baseline utility on either the DMD-QoL or HUI-2 in univariate analyses were retained in the multivariable model for both measures. R² values were calculated for each multivariable model to understand the amount of variability in DMD-QoL vs. HUI-2 utility estimates, explained by the predictors.

Baseline attribute levels that informed utility values were descriptively compared across health states and respondent type for each measure. Median (first quartile, third quartile [Q1, Q3]) attribute levels were calculated and visualized using heat maps to qualitatively evaluate the impact of different attributes on DMD-QoL or HUI-2 utility values.

Change in utility over 12 months was calculated, stratified by patient baseline health state and respondent type. The percentage of patients experiencing a decline in ambulatory function resulting in a health state transition over 12 months was reported. Change in utility was estimated for this subgroup using linear mixed regression to adjust for the impact of age, respondent type, and baseline health state. Changes in utility over 12 months for the HUI-2 were considered in the context of the documented minimal clinically important difference (0.03) [13]; to the best of our knowledge, similar data on the DMD-QoL are not available.

All analyses were conducted in R 4.4.1 (R Core Team, 2024).

Patient and Public Involvement

Patients and caregivers were first involved in this research at a prior qualitative stage, which sought to understand, through patient and caregiver insights, the clinical progression of DMD and the impacts of progressive symptoms on HRQoL. The findings from this work informed the health states used to classify patients in the present study. As well, parts of the survey materials were pilot tested to ensure their relevance and acceptability.

During enrollment for the current study, a phone call discussion was conducted with each family considering participation. Patients and their caregivers were given the opportunity to ask questions and request to be notified of publications or other dissemination of the study findings. In addition, the study results will be shared with patient advocacy groups to ensure that results are accessible to the broader community of families affected by DMD.

Results

Participants

The baseline and follow-up surveys were completed by 155 participants: 33 (21.3%) patients with DMD and 122 (78.7%) caregivers (Table 1). The sample included 15 patient-caregiver dyads (i.e., cases in which both the patient and their caregiver provided responses); therefore, 140 unique patients are represented in the data. Mean (SD) patient age at baseline was 14.1 (6.9) years; self-reporting patients were older (20.0 [6.3] years) than those included via caregiver report (12.8 [6.0] years). Health status at baseline was more progressed among self-reporting patients; 24 (72.7%) were non-ambulatory (vs. 50 [41.0%] among caregiver reports), and 22 (66.7%) had some degree of impairment in upper limb function (vs. 51 [41.8%] among caregiver reports). Four (12.1%) patient respondents used both nighttime and daytime ventilation for respiratory support compared with six (4.9%) patients included via caregiver report. Symptomatic cardiomyopathy was present among four (12.1%) self-reporting patients and eight (6.6%) caregiver reports (Table 1).

Table 1.

Baseline characteristics of the study sample by respondent type

	Patient reported (n = 33)	Caregiver reported (n = 122)
Mean (SD) age, years	20.0 (6.3)	12.8 (6.0)
Ambulatory status, n (%)
Ambulatory	7 (21.2%)	55 (45.1%)
Transitional	2 (6.1%)	17 (13.9%)
Non-ambulatory	24 (72.7%)	50 (41.0%)
Upper limb function, n (%)
Preserved	11 (33.3%)	71 (58.2%)
Mildly impaired	8 (24.2%)	26 (21.3%)
Moderately impaired	11 (33.3%)	20 (16.4%)
Loss of function	3 (9.1%)	5 (4.1%)
Ventilation use, n (%)
No ventilation	16 (48.5%)	97 (79.5%)
Nighttime ventilation only	13 (39.4%)	19 (15.6%)
Nighttime and daytime ventilation	4 (12.1%)	6 (4.9%)
Cardiomyopathy, n (%)
None identified	15 (45.5%)	82 (67.2%)
Asymptomatic	14 (42.4%)	32 (26.2%)
Symptomatic	4 (12.1%)	8 (6.6%)
Treatment status, n (%)
PMOs (with or without steroids)	0 (0.0%)	14 (11.5%)
Steroids without PMOs	26 (78.8%)	95 (77.9%)
Neither PMOs or steroids	7 (21.2%)	13 (10.7%)
Cardiac medication, n (%)
Yes	30 (90.9%)	88 (72.1%)
No	0 (0.0%)	1 (0.8%)
Missing	3 (9.1%)	33 (27.0%)

PMO phosphorodiamidate morpholino oligomer, SD standard deviation

Utility Values

Across health states at baseline, mean (SD) patient-assessed DMD-QoL utility ranged from 0.78 (0.10) to 0.62 (0.35), and caregiver-assessed from 0.84 (0.12) to 0.63 (0.17) (Table 2). Mean (SD) patient-assessed HUI-2 utility ranged from 0.96 (0.06) to 0.49 (0.07), and caregiver-assessed from 0.85 (0.11) to 0.44 (0.09).

Table 2.

Mean (SD) DMD-QoL/HUI-2 utility and change in utility over 12 months by health state and respondent type

		Mean (SD) DMD-QoL utility scores			Mean (SD) HUI-2 utility scores
Patient health state at baseline	n	At baseline	At follow-up	Change in utility	At baseline	At follow-up	Change in utility
Patient assessed
Early ambulatory	7	0.78 (0.10)	0.81 (0.10)	0.03 (0.11)	0.96 (0.06)	0.87 (0.11)	-0.09 (0.08)
Late ambulatory	2	0.62 (0.35)	0.63 (0.20)	0.01 (0.55)	0.59 (0.24)	0.83 (0.03)	0.24 (0.21)
Early non-ambulatory	10	0.69 (0.16)	0.68 (0.14)	0.00 (0.08)	0.52 (0.14)	0.51 (0.11)	0.00 (0.11)
Late non-ambulatory	14	0.68 (0.19)	0.72 (0.17)	0.04 (0.15)	0.49 (0.07)	0.47 (0.08)	-0.01 (0.07)
Caregiver proxy assessed
Early ambulatory	55	0.84 (0.12)	0.82 (0.12)	-0.02 (0.11)	0.85 (0.11)	0.80 (0.16)	-0.04 (0.12)
Late ambulatory	17	0.70 (0.18)	0.75 (0.13)	0.05 (0.12)	0.69 (0.19)	0.66 (0.14)	-0.03 (0.20)
Early non-ambulatory	25	0.72 (0.17)	0.70 (0.17)	-0.02 (0.16)	0.53 (0.11)	0.50 (0.11)	-0.03 (0.07)
Late non-ambulatory	25	0.63 (0.17)	0.64 (0.21)	0.01 (0.15)	0.44 (0.09)	0.42 (0.09)	-0.02 (0.10)

Early ambulatory: able to walk

Late ambulatory: able to walk with difficulty; transitioning to using a wheelchair or scooter

Early non-ambulatory: unable to walk with no upper limb impairment or mild impairment

Late non-ambulatory: unable to walk with moderate or severe upper limb impairment

Across health states and for both respondent types, the range of scores observed on the DMD-QoL was more restricted than on the HUI-2. For those in the early ambulatory health state (n = 7), patient-assessed mean (SD) utility values on the DMD-QoL (0.78 [0.10]) were lower than on the HUI-2 (0.96 [0.06; p-value = 0.016). The opposite was true for those in more progressed health states. For example, mean patient-assessed baseline DMD-QoL utility values for the late non-ambulatory health state (n = 14), were 0.68 (0.19) versus 0.49 (0.07) on the HUI-2 (p < 0.001). The same trend was observed for mean caregiver-assessed utility values (Table 2).

The correlation between DMD-QoL and HUI-2 utilities for the overall sample at baseline was r = 0.63 (Supplementary Table 1). Correlations by health state ranged from r = 0.44 (early ambulatory; p < 0.001) to r = 0.86 (late ambulatory; p < 0.001).

Factors Associated with Baseline Utility

Regression models analyzing the impact of baseline clinical and demographic factors on baseline utility values demonstrated that model covariates explained 70.4% of variance for baseline HUI-2 utility but only 26.0% of variance for DMD-QoL utility scores. Initial health state was significantly associated with both DMD-QoL and HUI-2 utility (all p < 0.001), as was treatment status (p = 0.044) for the HUI-2. Respondent type, ventilation use, cardiomyopathy, and patient age were not significant predictors.

Attribute Levels

Scores for most DMD-QoL attributes—including mobility, pain, being tired, being worried, and feeling good—showed little difference across the progressive health states describing declining patient function (Table 3). The exception was caregiver responses to the DMD-QoL hand function attribute and, to a lesser extent, participation, both of which tended to worsen with disease progression. For the HUI-2, mobility and self-care were the primary contributors to utility decrement, with impairments in these attributes generally increasing alongside the severity of health states (Table 4).

Table 3.

Median (Q1, Q3) DMD-QoL attribute levels at baseline by health state and respondent type

Patient health state at baseline	n	Median (Q1-Q3) DMD-QoL attribute levels at baseline
Patient assessed		Mobility	Hands	Breathe	Pain	Tired	Worried	Take part	Feel good
Early ambulatory	7	2.0 (1.0, 2.0)	1.0 (1.0, 1.0)	1.0 (1.0, 1.0)	2.0 (1.0, 2.0)	2.0 (1.5, 2.5)	2.0 (1.5, 2.0)	2.0 (1.5, 2.0)	3.0 (1.0, 3.5)
Late ambulatory	2	1.5 (1.2, 1.8)	1.0 (1.0, 1.0)	2.0 (1.5, 2.5)	3.0 (2.5, 3.5)	3.0 (2.5, 3.5)	2.5 (2.2, 2.8)	2.0 (1.5, 2.5)	2.0 (1.5, 2.5)
Early non-ambulatory	10	2.0 (2.0, 2.0)	1.5 (1.0, 2.0)	1.0 (1.0, 1.8)	2.0 (2.0, 2.0)	2.0 (2.0, 3.0)	2.0 (2.0, 3.0)	2.5 (1.2, 3.0)	2.0 (1.2, 3.0)
Late non-ambulatory	14	2.0 (1.2, 3.0)	2.0 (2.0, 3.0)	1.0 (1.0, 2.0)	2.0 (2.0, 2.0)	2.0 (2.0, 2.0)	2.0 (1.2, 2.0)	2.0 (1.0, 2.0)	2.0 (1.0, 2.0)
Caregiver proxy assessed
Early ambulatory	55	1.0 (1.0, 2.0)	1.0 (1.0, 2.0)	1.0 (1.0, 1.0)	2.0 (1.0, 2.0)	2.0 (2.0, 2.0)	2.0 (1.0, 2.0)	2.0 (1.0, 3.0)	2.0 (1.0, 2.0)
Late ambulatory	17	2.0 (2.0, 3.0)	1.0 (1.0, 2.0)	1.0 (1.0, 1.0)	2.0 (2.0, 2.0)	2.0 (2.0, 3.0)	2.0 (2.0, 2.0)	2.0 (2.0, 3.0)	2.0 (2.0, 3.0)
Early non-ambulatory	25	2.0 (2.0, 3.0)	2.0 (1.0, 2.0)	1.0 (1.0, 1.0)	2.0 (1.0, 2.0)	2.0 (2.0, 2.0)	2.0 (1.0, 2.0)	3.0 (2.0, 3.0)	2.0 (1.0, 3.0)
Late non-ambulatory	25	2.0 (2.0, 3.0)	3.0 (2.0, 4.0)	1.0 (1.0, 2.0)	2.0 (2.0, 2.0)	2.0 (2.0, 2.0)	2.0 (2.0, 2.0)	3.0 (2.0, 3.0)	2.0 (2.0, 3.0)

Attributes levels range from 1 (best possible response) to 4 (worst possible response)

Table 4.

Median (Q1, Q3) HUI-2 attribute levels at baseline by health state and respondent type

Patient health state at baseline	n	Median (Q1, Q3) HUI-2 attribute levels at baseline
		Sensation	Mobility	Cognition	Self-care	Emotion	Pain
Patient assessed
Early ambulatory	7	1.0 (1.0, 1.0)	1.0 (1.0, 1.0)	1.0 (1.0, 1.0)	1.0 (1.0, 1.0)	1.0 (1.0, 1.0)	1.0 (1.0, 1.0)
Late ambulatory	2	2.5 (2.2, 2.8)	1.5 (1.2, 1.8)	1.5 (1.2, 1.8)	1.0 (1.0, 1.0)	2.5 (2.2, 2.8)	3.0 (2.5, 3.5)
Early non-ambulatory	10	1.5 (1.0, 2.0)	4.0 (4.0, 4.0)	1.0 (1.0, 2.0)	3.5 (1.0, 4.0)	1.5 (1.0, 2.8)	2.0 (1.2, 2.0)
Late non-ambulatory	14	2.0 (1.0, 2.0)	4.0 (4.0, 4.0)	1.0 (1.0, 1.8)	4.0 (4.0, 4.0)	1.0 (1.0, 2.0)	2.0 (1.0, 2.0)
Caregiver proxy assessed
Early ambulatory	55	1.0 (1.0, 2.0)	1.0 (1.0, 2.0)	1.0 (1.0, 2.0)	1.0 (1.0, 1.0)	2.0 (1.0, 2.0)	2.0 (1.0, 2.0)
Late ambulatory	17	1.0 (1.0, 3.0)	3.0 (2.0, 3.0)	1.0 (1.0, 2.0)	2.0 (1.0, 2.0)	2.0 (1.0, 2.0)	2.0 (2.0, 2.0)
Early non-ambulatory	25	1.0 (1.0, 2.0)	4.0 (4.0, 4.0)	1.0 (1.0, 2.0)	4.0 (2.0, 4.0)	2.0 (1.0, 2.0)		2.0 (1.0, 2.0)
Late non-ambulatory	25	1.0 (1.0, 2.0)	4.0 (4.0, 5.0)	1.0 (1.0, 2.0)	4.0 (4.0, 4.0)	2.0 (1.0, 2.0)		2.0 (2.0, 2.0)

Sensation, cognition, and self-care levels range from 1 to 4, with 4 representing the highest level of impairment; all other attributes range from 1 to 5

Change in Utility over Time

In the overall sample, utility values derived from the DMD-QoL and HUI-2 showed limited changes over 12 months (Table 2). However, 17 individuals (11.0%) had a decline in ambulatory function resulting in a transition of health states over 12 months. Among these 17 individuals, a significant and clinically important [13] loss in HUI-2 utility (− 0.16 [− 0.22, − 0.10] 95% CI) was observed, but not in DMD-QoL utility (0.00 [− 0.06, 0.06] 95% CI).

Discussion

Both disease-specific and generic instruments have potential roles for documenting HRQoL impact and providing data for healthcare decision-making. Understanding the similarities and differences in the measures and their outcomes can inform their use. In this study, health state-specific DMD-QoL utility values differed considerably from those derived using the generic HUI-2 instrument. DMD-QoL utility values were relatively stable and exhibited a narrow range across ambulatory and non-ambulatory health states. In contrast, HUI-2 utility values declined with increasing health state severity and displayed a wider range across health states than did DMD-QoL utility values. The HUI-2 utility for the least progressed health state (early ambulatory) was higher, and for the most progressed health state (late non-ambulatory) was lower, than DMD-QoL utilities for those states.

There was considerable variability in utility among individuals in the same health state for both instruments; however, regression modeling indicated that variability in baseline HUI-2 utility could be better explained by available clinical or treatment-related covariates than for the DMD-QoL. DMD-QoL attribute scores did not appear to differentiate by health state, unlike the HUI, in which key attribute scores (i.e., mobility and self-care) declined alongside the progressive loss of function in the health states. Finally, it was interesting to note that limited change was observed in HUI-2 and DMD-QoL scores for the whole cohort over 12 months. However, among the subset who experienced a decline in ambulatory status according to their health state, a significant decline in HUI-2, but not DMD-QoL, utility was observed. The longitudinal findings from the present study are important, as there are limited other published data on how utility changes over time in DMD [22]. Findings from this study provide insight into the relative value of the two instruments for measuring outcomes in DMD and suggest that the HUI appears to be more sensitive for distinguishing levels of disease severity, and potentially more useful for charting progression, compared to the DMD-QoL.

To our knowledge, this is the first study to estimate utility values for a range of health states in DMD based on the DMD-QoL. Previously published estimates were based on generic PMBs and tended to show lesser utility for health states describing more highly progressed disease, as was the case for the HUI-2 values in the present study. For example, a study of the EQ-5D-3L among adult patients with DMD and caregivers as proxies reported utility values ranging from 0.65 (early ambulatory) to 0.26 (late non-ambulatory) [10]. These values are notably lower than the HUI-2 values estimated here, but comparability is limited by the combination of patient and caregiver values. Other available estimates show higher values for ambulatory boys with DMD based on the HUI-2 and HUI-3, with declines in utility as ambulation is lost [23], consistent with our findings. However, these prior estimates are generally limited by health state definitions that stratify patients based on single dimensions of functional status rather than capturing a more nuanced model of disease progression [6, 23].

Generic and condition-specific measures are complementary, and each has strengths and limitations. The DMD-QoL was developed to more broadly reflect the impact of DMD on quality of life compared with generic PBMs assessing the HRQoL impact of impaired function [9, 12]. While not tracking directly with the current understanding of clinical progression in DMD, the DMD-QoL documents aspects of HRQoL that are important to the DMD community and that may transcend loss of physical function. Observed differences in utility values calculated from the DMD-QoL vs. the HUI may be driven by differences in the descriptive systems underlying individuals’ scores, as well as how these impacts are interpreted by the members of the general public who provide the tariffs for each instrument. For instance, the general public might perceive the DMD-QoL’s description of mobility issues in DMD (‘I find it hard to get around’) as less severe compared with the EQ-5D (‘I am unable to walk’), which would lead to a lesser weighting for mobility impairments on the DMD-QoL versus the EQ-5D.

Considering non-ambulatory individuals, relatively high HRQoL scores and utility values were observed on the DMD-QoL compared with the HUI-2. This result seems consistent with findings from analyses of visual analog scale data in DMD, a non-PBM that documents a respondent’s self-rated health on a scale of 0–100 [24]. The differences between self-reports of HRQoL and estimates of utility available from generic instruments raise questions about which aspects of function are important in driving the HRQoL of individuals with DMD who are non-ambulatory.

At present, data on the DMD-QoL are relatively limited. The original work was conducted in the UK, and data from other jurisdictions are not yet available. HUI utilities are based on Canadian value sets; DMD-QoL data from other countries will help contextualize comparisons to PBMs like the HUI. The relative contribution of tariffs versus the scoring algorithms in producing observed differences between scale scores is not clear, nor are data available to understand what represents a meaningful change in DMD-QoL utility over time. Given this paucity of data, our study provides initial insights into the potential role of the DMD-QoL in measuring HRQoL and utility in DMD.

Strengths of this study include the large sample size of individuals representing varied levels of lower and upper limb function, both important determinants of HRQoL in DMD [23]. Outcomes from the relatively new DMD-QoL were compared with those from the HUI, an instrument more widely used to evaluate outcomes in DMD [6, 23].

Limitations include the omission of clinical manifestations that reflect important aspects of disease progression in DMD, such as cardiac and respiratory function, from the health states due to sample size constraints. The natural history model developed by Project HERCULES recommends the use of eight health states to describe the decline in ambulatory, upper limb, cardiac, and respiratory functions in DMD [25]. Health states in this study are not directly comparable to those from HERCULES, which are more granular, but also based on clinician-administered functional assessments. Importantly, health states in the present study were designed to enable study participants to self-report health status. Other potential limitations include that participants were required to be capable of completing a survey in English and providing informed consent, which may have limited generalizability. Participants were allowed to receive assistance with the survey, which may have helped to address this limitation; however, the presence of another individual recording responses may have influenced the responses provided. Finally, many factors that could impact utility within a given health state are difficult to capture, for example, modifications to the built environment, individual characteristics, or access to financial resources [23, 26], which may also contribute to within-health-state variability.

Conclusions

The study findings highlight distinctions between the DMD-QoL and HUI-2 in the context of estimating utility values in DMD. While the HUI-2 appears more sensitive to functional decline, the DMD-QoL reflects a broader conceptualization of wellness and quality of life based on the experiences of individuals with DMD. Utility values estimated using the DMD-QoL appear to be relatively consistent over a range of DMD-relevant health states and tracked less with clinical status than utility estimated using the HUI. These findings underscore the importance of selecting outcome measures that align with the specific goals of assessment, whether to track functional change or to understand lived experience. While based on our findings the DMD-QoL may not offer advantages over generic PBMs for assessing meaningful clinical benefit in health technology assessment, further research is warranted to understand the DMD-QoL’s potential value as a complementary measure to assess patient wellbeing over time.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Shelagh M. Szabo and Andrea Bever contributed to the collection of data and were responsible for drafting the manuscript. Andrea Bever, Shelagh M Szabo, Pramoda Jayasinghe, David Feeny, Peter J Neumann, Daniel C Malone, Susan T Iannaccone, Katherine L Gooch, and Ivana F Audhya contributed to the study design, analysis and interpretation of the data, and the critical review of the manuscript. Andrea Bever, Shelagh M Szabo, Pramoda Jayasinghe, David Feeny, Peter J Neumann, Daniel C Malone, Susan T Iannaccone, Katherine L Gooch, and Ivana F Audhya read and approved the final version of the manuscript. All authors comply with ICMJE guidelines for authorship.

Funding

This work was supported by Sarepta Therapeutics (grant number: n/a). The journal’s Rapid Service Fee and Open Access Fee was funded by Sarepta Therapeutics.

Data Availability

The datasets generated and/or analysed during the current study are not publicly available due the need to maintain participant privacy but may be available from the corresponding author on reasonable request.

Declarations

Conflict of Interest

Ivana F Audhya and Katherine L Gooch are employees of Sarepta and may own stock/options. Shelagh M Szabo, Andrea Bever and Pramoda Jayasinghe are employees of Broadstreet HEOR, which received funds from Sarepta to conduct this study. David Feeny, Peter J Neumann, and Daniel C Malone received consulting fees related to this work. Susan T Iannaccone has received research funding or consulting fees from Avexis, Biogen, Fibrogen, Mallinckrodt, Regeneron, Sarepta, Scholar Rock, PTC Therapeutics, Pfizer, MDA, CureSMA, NIH, Genentech-Roche, and BCBS. It should be noted that David Feeny has a proprietary interest in Health Utilities Incorporated, Dundas, Ontario, Canada. HUInc. distributes copyrighted Health Utilities Index (HUI) materials and provides methodological advice on the use of HUI.

Ethical Approval

The study was conducted in accordance with the Declaration of Helsinki. Study conduct was approved by Advarra Institutional Review Board (IRB; study number: Pro00055581) and WCG IRB (IRB tracking number: 20224611). All participants provided informed consent.