Validation and Reliability of the Thai Pediatric Charcot–Marie–Tooth Quality of Life Outcome Measure

TL;DR

Researchers validated a Thai version of a quality of life questionnaire for children with Charcot–Marie–Tooth disease, finding it reliable and feasible for use.

Contribution

The study provides the first validated Thai adaptation of the pCMT-QoL questionnaire for children with CMT.

Findings

Thai pCMT-QoL showed high test-retest reliability (ICC > 0.85) and good internal consistency (Cronbach's alpha > 0.7).

Parents reported higher quality of life scores than children, consistent with findings in other neuromuscular diseases.

The questionnaire was completed in under 15 minutes, indicating good feasibility for clinical use.

Abstract

Charcot–Marie–Tooth disease (CMT) is a hereditary neuropathy that causes progressive muscle weakness, sensory deficits, and impaired mobility, significantly affecting quality of life (QoL). The Pediatric Charcot–Marie–Tooth Quality of Life (pCMT‐QoL) instrument was developed specifically for children with CMT. However, a validated Thai version is not yet available.

We conducted a cross‐sectional study at the Pediatric Neuromuscular Clinic from July 2023 to December 2024. Using a forward–backward translation method, we adapted the pCMT‐QoL into Thai. Twenty‐three children with CMT and their caregivers completed the Thai questionnaire. We evaluated internal consistency using Cronbach's alpha and test–retest reliability using intraclass correlation coefficients (ICCs). Convergent validity was examined via Pearson correlation between child self‐reports and parent‐proxy reports across functional, mental, and physical domains.

The Thai pCMT‐QoL demonstrated high test–retest reliability (ICC > 0.85) and satisfactory internal consistency (Cronbach's alpha > 0.7) across most domains. Convergent validity was strong for the total and mental domains but weaker for the physical domain, reflecting differences in perception between children and parents. Parents generally reported higher QoL scores than children did, a finding consistent with studies in other neuromuscular diseases. Most participants completed the questionnaire within 15 min, suggesting good feasibility.

The Thai pCMT‐QoL is a reliable, culturally adapted tool for assessing QoL in children with CMT. It is suitable for both remote and in‐clinic administration. Future studies with larger cohorts are needed to confirm its responsiveness to clinical changes and to broaden its application in diverse settings.

Tables

TABLE 1. Demographic characteristics of pediatric patients with Charcot–Marie–Tooth disease.

Characteristics	N = 23
Sex, male	13 (56.5%)
Age at evaluation (years)	12.1 ± 4.7
Age at onset (years)	4.6 ± 4.9
Previous foot surgery	4 (17.4%)
Non‐ambulatory patients	3 (13.0%)
Education level
Kindergarten	4 (17.4%)
Primary school	7 (30.4%)
Secondary school	11 (47.8%)
Bachelor's degree	1 (4.3%)
Family history	4 (17.4%)
Electrophysiologic subtype
Axonal	14 (60.9%)
Demyelinating	7 (30.4%)
Intermediate	1 (4.3%)
Undetermined	1 (4.3%)

TABLE 2. Demographic characteristics of parents and caregivers of pediatric patients with Charcot–Marie–Tooth disease.

Characteristics	N = 23
Sex, male	4 (17.4%)
Age (years)	47.5 ± 9.0
Marital status
Single	3 (13.0)
Married	16 (69.6)
Divorced	4 (17.4)
Education level
Primary school	3 (13.0)
Secondary school	8 (34.8)
Bachelor's degree	10 (43.5)
Master's degree	1 (4.3)
Doctoral degree	1 (4.3)
Employment status
Employed	20 (87.0)
Unemployed	3 (13.0)
Household income per month in Thai baht
< 10 000	10 (43.5)
10 000–30 000	9 (39.1)
30 000–50 000	3 (13.0)
> 50 000	1 (4.3)

TABLE 3. Test–retest reliability of the Thai version of the pCMT‐QoL.

Domain	Age 0–7 (parent)		Age 8–18 (child)		Age 8–18 (parent)
	(N = 5)		(N = 16)		(N = 16)
	ICC	95% CI	ICC	95% CI	ICC	95% CI
Symptoms	0.84	(0.07, 0.98)	0.60	(0.17, 0.84)	0.71	(0.37, 0.88)
Function	0.90	(0.34, 0.99)	0.72	(0.36, 0.89)	0.90	(0.76, 0.96)
Social activities	0.99	(0.88, 1.00)	0.83	(0.59, 0.94)	0.79	(0.51, 0.91)
Feelings	0.89	(0.30, 0.99)	0.88	(0.70, 0.96)	0.88	(0.71, 0.95)
Cognition	0.65	(−0.34, 0.96)	0.82	(0.56, 0.93)	0.83	(0.60, 0.93)
Social skills	0.83	(0.05, 0.98)	0.36	(−0.15, 0.72)	0.73	(0.41, 0.89)
Physical domain score	0.96	(0.70, 0.99)	0.67	(0.28, 0.87)	0.74	(0.42, 0.89)
Mental domain score	0.84	(0.10, 0.98)	0.93	(0.81, 0.98)	0.88	(0.70, 0.95)
Total score	0.90	(0.34, 0.99)	0.86	(0.65, 0.95)	0.91	(0.79, 0.97)

TABLE 4. Internal consistency of the Thai pCMT‐QoL across domains and age groups.

Domain	Standardized Cronbach's alpha
	Ages 0–7 (parent) (N = 5)	Ages 8–18 (child) (N = 16)	Ages 8–18 (parent) (N = 18)
Physical: symptoms	0.85	0.79	0.57
Physical: function	0.79	0.66	0.89
Physical: social activities	0.83	0.61	0.88
Mental: feelings	0.75	0.89	0.95
Mental: cognition	0.83	0.78	0.92
Mental: social skills	0.78	0.73	0.83
Physical domain score	0.82	0.66	0.75
Mental domain score	0.79	0.89	0.96
Total score	0.83	0.88	0.91

TABLE 5. Comparison of parent‐proxy and child self‐reported pCMT‐QoL scores for ages 8–18 (N = 16).

Domain	Paired t‐test			Convergent validity
	Child version score mean (SD)	Parent version score mean (SD)	p	Correlation (r)	p
Symptoms	34.6 (16.3)	61.8 (16.6)	< 0.001
Function	43.9 (17.8)	50.7 (21.8)	0.175
Social activities	53.4 (17.5)	48.7 (25.7)	0.432
Feelings	41.9 (22.1)	38.8 (30.3)	0.640
Cognition	25.8 (15.5)	31.4 (25.4)	0.256
Social skills	33.6 (18.1)	26.1 (18.1)	0.102
Physical domain score	44.0 (10.0)	53.8 (13.1)	0.006	0.47	0.063
Mental domain score	33.8 (14.4)	32.1 (23.1)	0.673	0.73	0.001
Total score	38.9 (10.6)	42.9 (14.5)	0.128	0.72	0.002

TABLE 6. Correlation between pCMT‐QoL and PedsQL 4.0 across age groups.

pCMT‐QOL	PedsQL 4.0	Age 0–7 (parent) (N = 5)		Age 8–18 (child) (N = 16)		Age 8–18 (parent) (N = 18)
		r	p	r	p	r	p
Symptoms	Physical functioning	0.93	0.023	0.10	0.715	−0.48	0.057
Function
Social activities
Feelings	Emotional functioning	0.49	0.398	0.70	0.004	0.76	< 0.001
Cognition	School functioning	0.29	0.638	0.56	0.031	0.52	0.040
Social skills	Social functioning	0.35	0.558	−0.15	0.591	0.47	0.063
Total score		0.59	0.292	0.55	0.036	0.78	< 0.001

Full text

Introduction

1

Charcot–Marie–Tooth disease (CMT), also known as hereditary motor and sensory polyneuropathy, is a hereditary peripheral neuropathy that affects about 1 in 2500 individuals [1]. It is characterized by progressive muscle weakness, sensory loss, and foot deformities. More than 100 implicated genes confer substantial genetic and phenotypic heterogeneity [2]. This complexity can profoundly affect patients' quality of life (QoL).

Recent investigations into CMT therapies highlight the importance of early‐stage interventions, especially in pediatric populations. To support clinical trial readiness, researchers have developed CMT‐specific outcome measures and categorized them into clinical outcomes, functional outcomes, patient‐reported outcomes, biomarkers of disease burden, and treatment‐specific biomarkers of target engagement [3].

Clinical and functional outcome measures for pediatric CMT, such as the CMT Pediatric Scale and the CMT Infant Scale, are available worldwide [4, 5]. However, patient‐reported and self‐reported outcome measures remain limited. The Pediatric Charcot–Marie–Tooth Quality of Life (pCMT‐QoL) instrument was developed as a disease‐specific tool, with versions in English and Italian [6, 7, 8, 9]. The original English version of the pCMT‐QoL comprises three instruments [6, 8, 10]. These instruments include a parent‐report version for children aged 0–7 years, a parent‐report version for children aged 8–18 years, and a child self‐report version for children aged 8–18 years.

In Thailand, several studies have described the clinical spectrum, gene distribution, and genotype–phenotype correlation of pediatric CMT [11, 12]. However, no validated Thai QoL assessment tool exists specifically for this population. This deficiency hinders accurate evaluation of disease burden and the effectiveness of therapeutic interventions.

This study aimed to establish the validity and reliability of a Thai version of the pCMT‐QoL as a clinically useful measurement for Thai physicians. Our findings should enhance understanding of QoL in pediatric CMT and provide valuable patient‐reported data for international drug trials targeting this rare disease.

Materials and Methods

2

Study Design and Patient Population

2.1

This cross‐sectional study was conducted at the Pediatric Neuromuscular Clinic at Siriraj Hospital, Thailand, from July 2023 to December 2024. We included children aged 0 to 18 years with a confirmed diagnosis of CMT by electrophysiological testing or nerve conduction studies. All patients underwent genetic testing. The Siriraj Institutional Review Board approved the study protocol (354/2566 [IRB3] and COA no. Si 446/2023). The research was conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all participating children and their parents or caregivers.

Each patient and parent or caregiver independently completed the Thai pCMT‐QoL (Supporting Information [Link], [Link], [Link]). For participants with physical or visual limitations, a research assistant read the questions aloud and recorded the responses. The working version of the pCMT‐QoL was administered twice, separated by a 2–4‐week interval. The first administration took place in the clinic, and the second was mailed to participants' homes.

We collected patients' demographic information, education level, ambulatory status, surgical history, electrophysiologic subtype, and genetic findings. For parents, we recorded demographic data, education level, marital status, socioeconomic status, employment status, and household income.

Measures and Procedures

2.2

We obtained formal permission from the original developers (S.R. and M.S.) to use and translate the pCMT‐QoL instrument. The translation process followed established linguistic guidelines and used a forward–backward method [13, 14].

Two certified Thai translators, both fluent in English, performed the initial forward translation. Three pediatric neurologists with extensive experience in CMT then reviewed and refined this draft version.

Next, two additional certified translators, who were blinded to the original version, back‐translated the revised text into English (Appendices A, B, C). We compared the back‐translated copy with the original pCMT‐QoL to confirm conceptual accuracy and resolve any inconsistencies. The finalized Thai version of the pCMT‐QoL (Supporting Information [Link], [Link], [Link]) was then implemented in the study.

In addition to the pCMT‐QoL, we administered the Thai version of the Pediatric Quality of Life Inventory (PedsQL 4.0) Generic Core Scales on the same day that patients and caregivers completed the first pCMT‐QoL assessment. The 23‐item PedsQL 4.0 is a well‐established, validated instrument designed to measure health‐related quality of life in children and adolescents [15, 16]. It includes four multidimensional scales: Physical Functioning (8 items), Emotional Functioning (5 items), Social Functioning (5 items), and School Functioning (5 items). Items are rated on a 5‐point Likert scale (0 = never a problem to 4 = almost always a problem), with scores transformed on a scale from 0 to 100, where higher scores indicate better functioning. The tool is available in both child self‐report and parent‐proxy versions and was used to evaluate convergent validity by correlating domain scores with those of the Thai pCMT‐QoL.

pCMT‐QoL Instrument

2.3

The pCMT‐QoL instrument contains 57 items across six domains: symptoms (12 items), function (10), social activities (7), feelings (10), cognition (10), and social skills (8). The first three domains reflect physical aspects, and the last three address mental aspects.

Child self‐reports and parent‐proxy versions are available for children aged 8–18 years, while only parent‐proxy versions exist for children aged 0–7 years. Both versions feature identical core items, adjusted for age‐appropriate language and first‐ or third‐person tense. Each item is scored on a 5‐point scale (0 = never a problem, 1 = almost never a problem, 2 = sometimes a problem, 3 = almost always a problem, and 4 = always a problem), with reverse scoring for positively framed items.

Scoring proceeds in two steps. First, a weighted sum of the domain items is calculated based on dataset‐derived weights. Then, the summed scores are converted to a 0‐to‐100 scale, where 100 indicates the worst QoL and 0 indicates the best. Study data were collected and managed through REDCap (Research Electronic Data Capture), a secure, web‐based application designed to support validated data entry, audit trails, streamlined data export, and data import from external sources [17]. The system was hosted at the Research Data Management Unit, Faculty of Medicine Siriraj Hospital, Mahidol University.

Statistical Analysis

2.4

All analyses were performed using IBM SPSS Statistics, version 29 (IBM Corp., Armonk, NY, USA). Statistical significance was set at p < 0.05. Demographic data are reported as frequencies, percentages, means, and standard deviations. We used a t test to assess mean differences between groups.

Test–Retest Reliability

2.4.1

Parent–child dyads completed the pCMT‐QoL instrument on two separate occasions, spaced 2–4 weeks apart. We evaluated test–retest reliability using intraclass correlation coefficients (ICCs). According to established guidelines, ICC < 0.5 indicates poor reliability, 0.5–0.75 indicates moderate reliability, 0.75–0.9 indicates good reliability, and > 0.90 indicates excellent reliability [18, 19].

Internal Consistency

2.4.2

We assessed the internal consistency of each pCMT‐QoL domain using Cronbach's alpha. Values above 0.7 were considered acceptable, reflecting satisfactory internal consistency [20].

Convergent Validity

2.4.3

We examined convergent validity using Pearson correlations in two ways. First, we assessed the correlations between parent‐proxy and child self‐reported pCMT‐QoL scores (ages 8–18), including total scores, physical composite domain scores, and mental composite domain scores. Second, to further support convergent validity using an independent measure, we assessed correlations between corresponding domain scores of the Thai pCMT‐QoL and the Thai version of PedsQL 4.0 Generic Core Scales (Physical, Emotional, Social, and School Functioning). Correlations were classified as negligible (0.00–0.10), weak (0.10–0.39), moderate (0.40–0.69), strong (0.70–0.89), or very strong (0.90–1.00) [21].

Results

3

Demographic Characteristics

3.1

We enrolled 23 pediatric patients with CMT and their parents. Among these participants, five had children aged 0–7 years, and 18 had children aged 8–18 years, corresponding to the respective versions of the Thai CMT‐QoL. The mean age of the children was 12.1 years, with an average symptom onset at 4.6 years. Most children (60.9%) had the axonal subtype of CMT. A majority of caregivers were married, had completed secondary school or held a bachelor's degree, and were employed (87.0%). Nearly half reported a household income of less than 10 000 baht per month (Table 1 and Table 2).

Feasibility

3.2

On average, parents took 14.8 min (SD = 4.7, range 6.2–28.4) to complete the pCMT‐QoL, while children took 11.5 min (SD = 3.6, range 4.2–16.2). Most parents and all children (82%) finished within 15 min. The remaining 18% of respondents, all of whom were parents with visual impairments, required assistance from a reader. Among the 18 children aged 8–18 years, child self‐reports were unavailable for 2 (8.7%) because of intellectual disabilities and behavioral problems, but their parents completed the parent‐proxy version instead. No data were missing from the parent reports.

Reliability

3.3

Test–Retest Reliability

3.3.1

A subset of children (n = 21) and parents (n = 16) completed the pCMT‐QoL a second time by mail after 2–4 weeks. The parent‐proxy versions for ages 0–7 years and 8–18 years yielded ICCs of 0.90 and 0.91, respectively. The child version for ages 8–18 years had an ICC of 0.86. Domain‐specific ICCs ranged from 0.36 (social skills) to 0.88 (feelings). Notably, the social skills domain in the child version had lower reliability with an ICC of 0.36 (Table 3).

Internal Consistency

3.3.2

Internal consistency was evaluated using Cronbach's alpha at the first assessment (Table 4).

The pCMT‐QoL exhibited strong internal consistency in most domains and age groups, with Cronbach's alpha values typically exceeding 0.70. Total scores for all groups showed high reliability, ranging from 0.83 to 0.91. However, three domains demonstrated lower reliability, including symptoms (α = 0.57) in the parent‐proxy version for ages 0–7 years, and social activities (α = 0.61) and function (α = 0.66) in the child self‐report for ages 8–18 years.

Parent–Child Agreement

3.4

Among children aged 8–18 years, parent‐proxy versus child self‐reported pCMT‐QoL scores differed across several domains (Table 5). Overall, the parent‐proxy total score (42.9 ± 14.5) was higher than the child self‐report total score (38.9 ± 10.6; p = 0.128), with a strong correlation (r = 0.72). In the physical domain, the parent score (53.8 ± 13.1) significantly exceeded the child score (44.0 ± 10.0; p = 0.006), showing a moderate correlation (r = 0.47). Conversely, in the mental domain, the child score (33.8 ± 14.4) was higher than the parent score (32.1 ± 23.1; p = 0.673).

Convergent Validity

3.5

Pearson correlation analysis revealed strong convergence between parent‐proxy and child total scores as well as between mental composite domain scores. In contrast, the physical composite domain showed moderate correlation (Table 5). Additional analysis using the PedsQL 4.0 showed that total scores consistently demonstrated moderate to strong correlations across groups (Table 6).

Discussion

4

We translated three original versions of the pCMT‐QoL instrument into Thai: parent‐proxy versions for ages 0–7 and 8–18 years, and a child version for ages 8–18 years [6, 8, 10]. The Thai pCMT‐QoL was then administered to 21 pediatric patients and 23 parents to assess its reliability and validity. No participants reported difficulties completing the questionnaire. The validated Thai version demonstrated high validity and reliability, suggesting strong potential for clinical and research use. It can effectively gauge health‐related QoL among pediatric CMT patients and identify areas for therapeutic improvement.

Test–retest reliability was high for all versions of the Thai pCMT‐QoL, with ICCs surpassing 0.85, which is consistent with findings from the English and Spanish versions of the pCMT‐QoL [7, 8, 9]. However, the lower ICC in the cognitive domain for the parent‐proxy version (ages 0–7) suggests variability in parental interpretations. This discrepancy may stem from caregivers' differing perceptions rather than actual changes in a child's condition. In future applications, providing brief guidance or instructions to caregivers may help standardize interpretations of symptoms and behaviors, reduce subjective differences, and improve the reliability of proxy assessments.

In addition, it was observed that scores in some domains, particularly Social Skills, fluctuated at the 2‐week follow‐up compared with baseline, resulting in lower ICC values. These changes likely reflect real‐life variability in children and adolescents with CMT—shaped by school attendance, peer interactions, family support, independence, and daily symptom fluctuations—rather than measurement error. Prior studies have similarly highlighted greater variability in psychosocial domains compared with physical functioning [22, 23].

Cronbach's alpha values showed high internal consistency across most domains, especially in the mental composite domain, which displayed robust reliability. However, the function and social activities domains for children aged 8–18 years exhibited notably low reliability, and similar concerns arose in the symptoms domain for the parent‐proxy version aged 0–7. These findings underscore the complexities of measuring QoL in pediatric populations, where rapid developmental changes, subjective interpretations, and varying parental observations can lead to inconsistent responses. Targeted modifications to these domains, along with including multiple informants, may help improve reliability.

Among children aged 8–18 years, parent‐proxy and child self‐reported pCMT‐QoL scores showed strong correlations for total and mental composite domains, confirming convergent validity. However, the physical composite domain correlated only moderately, reflecting divergent perceptions of physical limitations between children and parents. Parents also reported significantly higher total scores than children, a pattern consistent with the English version of the pCMT‐QoL [8] and other neuromuscular disease studies [24]. These discrepancies illustrate the importance of integrating both parent‐proxy and child self‐reports to capture a comprehensive view of the child's QoL. Indeed, research suggests that discordant responses between child and parent reports are common in health‐related QoL questionnaires, even outside chronic illness contexts [25, 26]. Thus, incorporating both perspectives remains crucial for accurately assessing pediatric well‐being.

This study confirms that the Thai version of the pCMT‐QoL instrument is culturally adapted, validated, and reliable, with performance comparable to the original English version. It also offers practical advantages, requiring about 15 min to complete and permitting remote administration. However, the small sample size—particularly in the 0–7‐year‐old parent‐proxy group—limits the generalizability of the findings, and the instrument's responsiveness to clinical changes over time remains untested. Given the rarity of pediatric CMT and the constrained sample, future research should involve larger and more diverse cohorts, along with its systematic use in clinical evaluations and follow‐up.

Conclusions

5

In summary, the Thai pCMT‐QoL is a reliable and valid measure for assessing the QoL of pediatric CMT patients in Thailand. Its comprehensive coverage of physical and mental health domains makes it a valuable resource for both patient‐centered care and research. Further validation in broader settings will maximize its clinical utility and ensure its adaptability across diverse populations.

Author Contributions

Pimchanok Kulsirichawaroj: conceptualization, methodology, project administration, writing – original draft, visualization, funding acquisition. Arisa Phochaisarn: investigation. Sivaporn Limpaninlachat: methodology, formal analysis, validation, writing – review and editing. Nutchavadee Vorasan: formal analysis. Surachai Likasitwattanakul: resources, writing – review and editing. Oranee Sanmaneechai: conceptualization, methodology, resources, formal analysis, validation, writing – review and editing, supervision. Michael E. Shy: instrument development, writing – review and editing. Sindhu Ramchandren: instrument development, writing – review and editing.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: Supporting Information.

Data S2: Supporting Information.

Data S3: Supporting Information.