The modified Care and Comfort Questionnaire score in children with severe cerebral palsy

Abstract

Aims

The modified Care and Comfort Hypertonicity Questionnaire (mCCHQ) is a validated caregiver-reported outcome measure for children with Gross Motor Function Classification System (GMFCS) IV/V cerebral palsy (CP). This study aimed to establish reference values for the mCCHQ and determine if a floor or ceiling effect exists.

Methods

This was a population-based cross-sectional study of GMFCS IV/V children enrolled in the CP Integrated Pathway Scotland (CPIPS) national surveillance programme. At each CPIPS assessment, age, sex, GMFCS and CP subtype were recorded, and caregivers were asked to complete the mCCHQ. The Scottish Index of Multiple Deprivation (SIMD) was calculated according to the child’s home address.

Results

The mCCHQ scores were available from 685 clinical assessments of 405 GMFCS IV/V children across four years (2017 to 2020). The mean age at assessment was 8.3 years (2 to 16). Overall, 40% (n = 163) were GMFCS IV and 60% (n = 242) GMFCS V. The mean total mCCHQ score was 52.1 (SD 12.7; 15 to 92). The mean scores for the personal care, positioning/transferring, comfort, and interaction/communication subdomains were 17.2 (SD 5.9; 6 to 36), 11.4 (SD 4.0; 1 to 24), 8.7 (SD 3.2; 0 to 20), and 14.9 (SD 4.5; 0 to 25), respectively. Although no notable floor or ceiling effect was observed for the total mCCHQ score, 13% and 15% of respondents gave the minimum possible score for the positioning/transferring and comfort subdomains, respectively, demonstrating a moderate/significant floor effect. There was no discernible trend in total mCCHQ scores across different ages. Demographic variables had limited influence on the mCCHQ score.

Conclusion

The mCCHQ provides a valuable outcome score for quantifying the experience of caring for a child with GMFCS level IV/V CP. While it demonstrates good sensitivity and reliability, the observed floor effects in the positioning/transferring and comfort subdomains highlight areas for potential refinement.

Cite this article: Bone Jt Open 2026;7(2):241–248.

Introduction

Cerebral palsy (CP) has a wide range of neuromotor and musculoskeletal manifestations in children, notably hip displacement and knee flexion contractures.^1-3 This may be associated with pain, reduced range of motion (ROM), gait disturbance, difficulties in seating, providing personal care, and difficulties with activities of daily living.^4-9

The subjective experience of living with CP, and caring for a child with CP is complex, and it is essential that clinicians attempt to understand and quantify this to provide the best standards of care.^8,10 Many Gross Motor Function Classification System (GMFCS) level IV/V children may struggle to communicate in a way that others can understand, creating reliance on caregivers’ reporting by proxy. The modified Care and Comfort Hypertonicity Questionnaire (mCCHQ) is a validated, modified version of the Care and Comfort Hypertonicity Questionnaire (CCHQ), designed specifically for caregivers of children with CP at GMFCS levels IV and V.^11-13 Caregivers rate the ease of 27 specific tasks on a modified Likert scale from 1 (very easy) to 4 (impossible) across four domains: 1) personal care; 2) positioning/transferring; 3) comfort; and 4) interaction/communication. Higher scores denote greater difficulty with tasks, with a maximum total score of 108. The mCCHQ is free to access, quick and simple for caregivers to complete. It also shows responsiveness to surgical intervention and good test-retest reliability.¹⁴

Cerebral Palsy Integrated Pathway Scotland (CPIPS) is the national CP hip surveillance programme in Scotland, established in 2013.^15-17 Around 95% of the Scottish paediatric CP population are enrolled, currently over 2,700 children. All children undergo standardized serial clinical and radiological examinations, the frequency of which is determined by their age and GMFCS level. Additionally, there has been routine collection of caregivers’ mCCHQ scores for GMFCS IV/V children since 2016.¹⁴

There is limited knowledge of the reference values and sensitivity of the mCCHQ. If a high proportion of caregivers report the lowest or highest scores (i.e. a ‘floor’ or ‘ceiling’ effect), it may indicate that the mCCHQ is not sensitive enough to capture variation in scores and therefore does not accurately represent the experience of caregivers of children with CP. This study aimed to present mCCHQ data for an undifferentiated population of children with GMFCS level IV/V CP to determine reference values and examine if a floor or ceiling effects exist for the mCCHQ.

Methods

Patient demographic characteristics

This was a population-based cohort study in children with CP from the CPIPS national surveillance programme. At each assessment, the child’s age, sex, GMFCS, CP subtype (spastic, dystonic, ataxic, mixed, dyskinetic, or other) and distribution of CP (monoplegia, hemiplegia, diplegia, triplegia, quadriplegia, or other) were recorded. The Scottish Index of Multiple Deprivation (SIMD)¹⁸ was calculated according to the child’s home address postal code. The SIMD ranks 6,976 geographical areas derived from patient postcodes across seven standardized domains (income, employment, education, health, access to services, crime, and housing), creating a relative, rather than absolute, indicator of deprivation reflecting in-country geographical variation. SIMD values were presented in quintiles, with 1 being the most and 5 the least deprived.¹⁸

Data were available from 405 GMFCS IV and V children across a four-year period (2017 to 2020). Demographic details are shown in Table I .

Table I.

Demographic details of the study cohort.

Variable	Data (n = 405)
Mean age, yrs (range)	8.3 (2 to 16)
Sex, n (%)
Male	222 (55)
Female	183 (45)
GMFCS, n (%)
IV	163 (40)
V	242 (60)
Distribution of CP, n (%)
Hemiplegia	8 (2)
Diplegia	36 (9)
Triplegia	6 (1)
Quadriplegia/total body	347 (87)
Other	8 (2)
Type of CP, n (%)
Spastic	235 (58)
Dystonic	61 (15)
Ataxic	2 (0.5)
Mixed	72 (18)
Dyskinetic	23 (6)
Other	12 (3)
SIMD quintile, n (%)
1 (most deprived)	91 (23)
2	83 (21)
3	68 (17)
4	82 (20)
5 (least deprived)	79 (20)
Missing data	2 (1)

CP, cerebral palsy; GMFCS, Gross Motor Function Classification System; SIMD, Scottish Index of Multiple Deprivation.

The mCCHQ

At each CPIPS assessment, caregivers are asked to complete the mCCHQ. They rate the difficulty of 27 specific tasks across four domains: 1) personal care (ie. putting on pants/trousers?); 2) positioning (ie. ease of positioning in a wheelchair?); 3) comfort (ie. is there pain or discomfort during position change?), and 4) interaction/communication (i.e. how easy is it for your child to extend arms to reach communication devices?). These are scored on a Likert scale from 1 (very easy) to 4 (impossible). Higher scores denote greater difficulty with tasks. If every question were answered, the maximum scores for the personal care, positioning/transferring, comfort, and interaction/communication subdomains are 36, 24, 20, and 28, respectively, giving a total maximum potential score of 108. If every question were answered, the minimum scores are 9, 6, 5, and 7, respectively, giving a total minimal potential score of 27 (Supplementary Material). All questions can also be answered ‘n/a’; for example, if the question ‘Ease of getting in or out of a car?’ is not relevant to that child, denoting a score of 0. Aggregate scores for each of the four subdomains and a total score were collected.

Statistical analysis

Data analysis and visualization were performed using SPSS v. 29.0.1.0 (IBM, USA). Descriptive statistics were used as appropriate for baseline characteristics. A p-value < 0.05 defined statistical significance. Normality was assessed using histograms and Shapiro-Wilk testing. The following statistical analyses were used:

• Independent-samples t-test with equal variances assumed was used to compare the means of two independent groups for continuous variables with normally distributed data.

• Analysis of variance (ANOVA) with Bonferroni correction was used for comparisons involving more than two groups to determine whether significant differences existed among group means. The assumption of sphericity was assessed by Mauchly’s test of sphericity. ε (ε) was calculated according to Greenhouse and Geisser and was used to correct the one-way repeated measures ANOVA.

• Pearson correlation coefficient was used to assess the strength and direction of relationships between continuous variables with normally distributed data.

• Spearman’s rank correlation coefficient was used for non-parametric correlation analysis to evaluate monotonic relationships between variables that were ordinal or not normally distributed.

• Floor and ceiling effects were defined as the proportion of caregivers who reported the lowest (‘best’) or highest (‘worst’) possible score for the mCCHQ. The reported thresholds of floor and ceiling effects vary throughout the literature, and there is limited equivalent paediatric CP literature available for comparison.¹⁹ A significant ceiling effect is considered to be present if ≥ 15% of respondents score the lowest or highest possible score, a moderate ceiling is 10% to 15%, a minor if 5% to 10%, and negligible if < 5%.^19-22

Results

Distribution of mCCHQ scores

The mCCHQ scores were available from 685 clinical assessments, of which 684 had complete data (a personal care score was missing in one assessment of a GMFCS V child). Of these, 259 of the assessments were in GMFCS level IV children and 426 in GMFCS level V. The mean total mCCHQ score was 52.1 (SD 12.7; 15 to 92). Mean scores for the personal care, positioning/transferring, comfort and interaction/communication subdomains were 17.2 (SD 5.9; 6 to 36), 11.4 (SD 4.0; 1 to 24), 8.7 (SD 3.2; 0 to 20), and 14.9 (SD 4.5; 0 to 25), respectively. The distribution of scores by GMFCS level are shown in Figures 1 to 5 and Table II. There was no statistically significant difference between mean mCCHQ scores between GMFCS levels IV and V, apart from in the comfort score (p = 0.038), although the absolute difference in means was small (9.0 in GMFCS IV and 8.6 in GMFCS level V, respectively). There was no discernible trend in total mCCHQ scores across different ages (Figure 6).

Fig. 1.

Frequency of total modified Care and Comfort Hypertonicity Questionnaire score by Gross Motor Function Classification System (GMFCS) level.

Fig. 5.

Frequency of interaction/communication modified Care and Comfort Hypertonicity Questionnaire score by Gross Motor Function Classification System (GMFCS) level.

Table II.

The modified Care and Comfort Hypertonicity Questionnaire score (mCCHQ) presented by Gross Motor Function Classification System (GMFCS) level.

mCCHQ	GMFCS	N	Minimum score (n, %)	Maximum score (n, %)	Mean (SD)	p-value*
Personal care	IV	259	6	36	17.5 (6.0)	0.306
	V	425	6	33	17.0 (5.9)
Positioning/transferring	IV	259	1	24	11.6 (3.9)	0.417
	V	426	1	23	11.3 (4.1)
Comfort	IV	259	2	20	9.0 (3.1)	0.038
	V	426	0	18	8.6 (3.2)
Interaction/communication	IV	259	3	25	15.1 (4.6)	0.306
	V	426	0	25	14.7 (4.6)
Total score	IV	259	23	91	53.1 (12.4)	0.096
	V	426	15	92	51.6 (12.9)

^*Analysis of variance (ANOVA) with Bonferroni correction.

Fig. 6.

The mean total modified Care and Comfort Hypertonicity Questionnaire score (mCCHQ) by age in years and Gross Motor Function Classification System (GMFCS) level.

Fig. 2.

Frequency of personal care modified Care and Comfort Hypertonicity Questionnaire score by Gross Motor Function Classification System (GMFCS) level.

Fig. 3.

Frequency of positioning/transferring modified Care and Comfort Hypertonicity Questionnaire score by Gross Motor Function Classification System (GMFCS) level.

Fig. 4.

Frequency of the comfort modified Care and Comfort Hypertonicity Questionnaire score by Gross Motor Function Classification System (GMFCS) level.

Within-subjects trends in mCCHQ scores

A total of 66 children had at least three clinical assessments across the four-year period. A one-way repeated measures ANOVA was conducted to determine whether there was a statistically significant difference in total mCCHQ scores over time. The assumption of sphericity was not met, χ2(2) = 10.10, p = 0.006. ε was 0.873 and was used to correct the one-way repeated measures ANOVA. There was no significant difference in mean total mCCHQ scores when they were compared across these time points (p = 0.554) (Table III).

Table III.

Mean total modified Care and Comfort Hypertonicity Questionnaire (mCCHQ) across three timepoints.

mCCHQ	Mean (SD)
Total Score t1	54.26 (11.964)
Total Score t2	51.38 (13.931)
Total Score t3	53.58 (12.188)

Floor and ceiling effects

The proportion of maximum and minimum responses for the four subdomains and the total score of the mCCHQ is shown in Table IV. A floor effect was observed for the positioning/transferring and comfort subdomains, with minimum scores reported by 13% and 15% of respondents, respectively. However, there was no significant floor effect observed for the total mCCHQ score. No ceiling effects were observed.

Table IV.

Proportion of floor and ceiling responses to the modified Care and Comfort Hypertonicity Questionnaire score (mCCHQ).

mCCHQ subdomain	Minimum score (floor), n (%)	Maximum score (ceiling), n (%)
Personal care	55/684 (8)	2/684 (0.3)
Positioning/transferring	87/685 (13)	1/685 (0.1)
Comfort	104/685 (15)	1/685 (0.1)
Interaction/communication	41/685 (6)	0/685 (0.0)
Total score	17/684 (2.5)	0/685 (0.0)

mCCHQ scores and demographic variables

The mCCHQ scores were compared with the demographic variables shown in Table V. There was no statistically significant association between any demographic variable or mCCHQ score except for a weak, but statistically significant, association between GMFCS level and the comfort subdomain (p = 0.038) and SIMD rank and the positioning subdomain (p = 0.039). despite this, the absolute difference in mean positioning subdomains scores between SIMD quintiles (Table VI) and the absolute difference in mean Comfort subdomains scores between GMFCS IV and V was small (9.0 vs 8.6, respectively, Table II).

Table V.

The statistical relationship between the modified Care and Comfort Hypertonicity Questionnaire score (mCCHQ) and demographic variables.

	mCCHQ score
Variable	Personal care	Positioning	Comfort	Communication	Total score
Age*	0.63	0.20	0.44	0.35	0.19
Sex†	0.77	0.33	0.87	0.85	0.08
GMFCS level‡	0.31	0.42	0.04§	0.31	0.10
Type of CP‡	0.34	0.56	0.59	0.61	0.86
Distribution of CP‡	0.49	0.94	0.38	0.65	0.80
SIMD rank¶	0.43	0.04§	0.75	0.33	0.13

^*Two-tailed Spearman test.

^†Independent-samples t-test.

^‡Analysis of variance.

^§Statistically significant values.

^¶Pearson correlation coefficient.

CP, cerebral palsy; GMFCS, Gross Motor Function Classification System; SIMD, Scottish Index of Multiple Deprivation.

Table VI.

Distribution of mean Positioning subdomain scores by Scottish Index of Multiple Deprivation (SIMD) quintile.

SIMD quintile	Mean positioning score (SD)
1	10.93 (4.078)
2	11.24 (4.049)
3	11.65 (3.532)
4	11.33 (4.118)
5	12.11 (4.222)

Discussion

This is the first study to show the distribution of mCCHQ scores in an undifferentiated population of children with GMFCS level IV/V CP enrolled on a national surveillance programme. The findings provide valuable insight into caregivers' experiences, the sensitivity of the mCCHQ, and further support its utility as a clinical tool.¹⁴

Distribution of scores, floor, and ceiling effects

The mCCHQ scores showed considerable variation across all four subdomains, with mean scores suggesting moderate difficulty in most caregiving tasks. While no significant floor or ceiling effects were observed for the total mCCHQ score, moderate to significant floor effects were noted in the positioning/transferring and comfort subdomains (13% and 15%, respectively). This is likely because it is possible to have a low score in one subdomain, but higher scores in other subdomains bring the overall total score above the minimum floor score. These findings suggest that while the mCCHQ can generally capture a broad range of caregiver challenges, the sensitivity may vary depending on the domain being assessed, and it may inadequately differentiate among tasks with lower levels of difficulty. Another potential explanation is that certain tasks are perceived as less challenging due to caregivers becoming accustomed to the needs of their child, such as rating the difficulty of ‘ease of positioning in a wheelchair?’. A floor effect has also been observed in the ‘pain and feelings about disability’ subdomain of the caregiver-reported Cerebral Palsy Quality of Life Questionnaire for Children (CP QoL-Child) and for all subdomains of the Child Health Questionnaire, except ‘general health’.¹⁹

The lack of a significant ceiling effect across all subdomains and the total score is encouraging, suggesting that the mCCHQ can capture the upper range of challenges caregivers may face. This robustness against ceiling effects implies that the tool is effective at identifying the most challenging caregiving scenarios, an essential feature when monitoring treatment interventions and tailoring support strategies. Comparing the mCCHQ to other patient-reported outcome measures (PROMs), ceiling effects have been observed in four of the seven CP QOL-Child subdomains and in all 12 of the Child Health Questionnaire subdomains.

A variety of PROMs have been developed for children with CP (Table VII), many of which use caregiver-proxy reporting. This is particularly important for children with severe CP, whose ability to self-report may be compromised due to intellectual and/or communicative difficulties.¹⁰ However, there are several limitations to caregiver reporting by proxy. Systemic bias can arise due to caregivers’ reliance on indirect cues, personal knowledge, and the limitations of capturing subjective non-observable phenomena such as emotions, thoughts, and feelings.¹⁰ This highlights the importance of qualitative data in understanding the experience of caring for a child with CP,⁸ although this would be challenging to longitudinally collect within a national surveillance programme. Quantitative PROMs are also routinely collected by the Netherlands CP register,²³ and tools such as the Gait Outcomes Assessment List (GOAL)²⁴ are being used in prospective physiotherapy trials in CP such as SPELL²⁵ and ROBUST.²⁶

Table VII.

An overview of patient-reported outcome measures specifically developed for children with cerebral palsy and/or their caregivers.

Name	Completed by	GMFCS	Domains assessed	Scoring	Validation
Caregiver Priorities and Child Health Index of Life with Disabilities (CPCHILD)	Caregiver only	IV to V	Personal care Positioning, transferring, and mobility Communication and social interaction Comfort and emotions Health Overall quality of life Importance of items to your child’s quality of life	Seven-point Likert	(23, 24)
The Cerebral Palsy Quality of Life Questionnaire (CP QOL-Child/Teen)	Either caregiver (for children aged 4 to 12 years) or child (aged 9 to 12 years)	I to V	Family and friends Participation Communication Health Special equipment Pain and bother Additional caregiver-reported questions on access to services and caregiver health	Nine-point Likert	(25 to 27)
paediatric Quality of Life Inventory (PEDSQL 3.0) Cerebral Palsy Modules	Either caregiver (for children aged 2 to 4 years) or child (aged 5 to 7, 8 to 12, or 13 to 18)	I to V	Daily activities School activities Movement and balance Pain and hurt Fatigue Eating activities Speech and communication	Five-point Likert and three-point Likert	(28)
Gait Outcomes Assessment List Questionnaire (GOAL)	Either caregiver or child	I to V	Activities of daily living and independence Gait function and mobility Pain, discomfort, and fatigue Physical activities, sports, and recreation Gait pattern and appearance Use of braces and mobility aids Body image and self-esteem	Seven-point Likert	(29, 30)
The Care and Comfort Score (and subsequent modifications) (mCCHQ)	Caregiver only	III to IV	Personal care Positioning Comfort Communication	Original: 7-point Likert CareQ: 5-point Likert Modified: Four-point Likert	(12 to 14)
DISABKIDS-CP Module (CPM)	Either caregiver or child	I to V	Activities of daily living and independence Emotion Independence Social inclusion and exclusion	Five-point Likert	(31, 32)
KIDSCREEN	Either caregiver or child (10-, 27-, and 52-item versions)	I to V	Physical activities and health Feelings and general mood About yourself/your child Free time, family and home life Money matters Friends, school, and learning Bullying	Five-point Likert	(18, 33)

GMFCS, Gross Motor Function Classification System.

Relationship between mCCHQ scores and demographics

There was no statistically significant association between mCCHQ scores and most demographic variables, including age, sex, type of CP, distribution of CP, and socioeconomic status (as measured by SIMD rank). The absence of an association may reflect the pervasive challenges faced by caregivers of children with GMFCS level IV/V CP, regardless of demographic differences. The exception was a weak, but statistically significant, association between GMFCS level and comfort subdomain scores, and SIMD rank and the positioning subdomain scores. While the absolute difference in score was small, this finding may warrant further exploration to better understand potential nuances in caregiving challenges across different functional levels.

Strengths of this study are using data from a large national surveillance cohort of GMFCS level IV/V children, a group that is historically under-represented in the CP literature.¹⁰ Children were represented relatively evenly across all ages, sexes, and SIMD quintiles, enhancing the generalizability of the findings.

Although all CPIPS assessments are performed by clinical experts according to standardized protocol and therefore should be highly reliable, there remains potential for database transcription errors. Data are only presented up to 2020 due to concerns of disruptions to the completeness of data collection during the COVID-19 pandemic. Due to the way data were exported from the CPIPS database, a carer response of ‘N/A’ was scored as 0 (i.e. best possible). This had the potential to conflate ‘not applicable’ with ‘no difficulty’, potentially depressing scores and inflating floor effects. However, across all 685 assessments, only two subdomain scores were recorded as ‘0’ (for one comfort score and interaction/communication score in two different patients). Therefore, this was unlikely to significantly adversely affect the results.

This was a cross-sectional study across four years with limited ability to monitor mCCHQ trends over time within individuals. It was also not known what treatment or surgical intervention(s) children may have had during this period (such as hip reconstructive surgery, soft-tissue surgery, Botulinum toxin, physiotherapy, orthoses, or baclofen) which could have affected their mCCHQ score. Therefore, this data cannot contribute to calculating the minimal clinically important difference (the smallest change that would be considered meaningful by the patient or clinician) for the mCCHQ. However, previous work has shown that the mCCHQ is responsive to surgical intervention for hip displacement, showing improvement in the subdomains of personal care, positioning/transferring, and total score from preoperative assessment to six-month postoperative follow-up.¹⁴ It was also not possible to assess the internal consistency of the mCCHQ using Cronbach’s α as only aggregate scores for each subdomain were available.

In conclusion, the mCCHQ provides a valuable framework for understanding and quantifying the complex and multifaceted caregiver experience of children with GMFCS level IV/V CP. While it demonstrates good sensitivity and utility, the observed floor effects in the positioning/transferring and comfort subdomains highlight areas for potential refinement. By continuing to integrate mCCHQ data into routine practice and research, clinicians and policymakers can better support caregivers and improve the quality of care for children with severe CP. Additionally, incorporating caregiver feedback into the refinement process may enhance the tool’s sensitivity and relevance. Future efforts should focus on enhancing the tool’s sensitivity, exploring longitudinal trends, and leveraging qualitative insights to ensure it remains a robust and relevant measure of caregiver burden.

Take home message

- The modified Care and Comfort Hypertonicity Questionnaire provides a valuable outcome score for quantifying the experience of caring for a child with severe cerebral palsy.

- While the scores demonstrate good sensitivity and reliability, the observed floor effects in some subdomains highlight areas for potential refinement.

Data Availability

The data that support the findings for this study are available to other researchers from the corresponding author upon reasonable request.