Sage Journals: Discover world-class research

Abstract

PURPOSE:

The study sought to examine the content validity, the intra- and inter-rater reliability, and concurrent validity of the Japanese versions of the Manual Ability Classification System, the Communication Function Classification System, and the Eating and Drinking Ability Classification System, and representation of the distribution of the levels of each classification systems in comparison to the Gross Motor Function Classification System levels for children with cerebral palsy.

METHODS:

The Japanese versions were developed using the back-translation method. For content validity, professionals were asked to complete a questionnaire including items on the appropriateness of the translation, its validity, and distinctions among levels. For reliability, professionals used the three classification systems twice in children with cerebral palsy. For concurrent validity, relationships among the four classification systems were examined by correlation analyses.

RESULTS:

Participants included twenty-one professionals and 290 children with cerebral palsy (mean age: 12 years two months, female: 132, male: 158). The content validity was generally good. For reliability, the lower limit of the 95% confidence interval for the intraclass correlation coefficients was greater than 0.89, and the correlation coefficients were high.

CONCLUSION:

The results of this study showed good reliability and validity of the functional classification systems in Japan.

Keywords

Cerebral palsy manual ability classification system communication function classification system eating and drinking ability classification system

1 Introduction

Cerebral palsy (CP), the most common cause of physical disability affecting children, is a heterogeneous medical condition [1, 2]. It is well known that even children with similar health conditions have very diverse functional abilities. Thus, this heterogeneity of functional abilities remains a major challenge in implementing effective rehabilitation for children and adolescents with CP [3].

Using classification systems is useful to categorize children with CP into more homogeneous groups based on their functional abilities. The Gross Motor Function Classification System (GMFCS) [4], Manual Ability Classification System (MACS) [5], Communication Function Classification System (CFCS) [6], and Eating and Drinking Ability Classification System (EDACS) [7] were developed to objectively classify children and adolescents with CP for the purpose of effective communication, setting goals, informing decisions on services and intervention, and applying research findings to practice [8, 9]. The GMFCS was the first tool developed and is the gold standard among classification systems for children and adolescents with CP [10]. Each system has five levels, ranging from level I (the least affected) to level V (the most affected), and are intended to represent differences in function that are meaningful in daily life. These four functional classification systems have been translated into a variety of languages [11 –14] and are used in many countries around the world.

It is well known that there are differences in the prevalence of CP between regions and/or populations [15]. Classification systems serve to categorize individuals into groups with the purpose of comparing different cohorts [2]. Therefore, examining the distribution of functional classification systems in various cultures is useful for international comparisons. Furthermore, understanding the inter-relationship between these classification systems may be useful in terms of intervention for children and adolescents with this heterogeneous health condition.

Although Japanese versions of the GMFCS and MACS exist in Japan, there are no versions of the CFCS or EDACS. In Japan, functional classification of communication, eating, and drinking have so far been noted empirically in terms like mild, moderate, and severe to describe the severity. The lack of a functional classification system remains a clinical and research challenge. Although the Japanese version of the GMFCS has been examined for content validity using the Delphi survey and the inter-rater reliability [16], these factors have not yet been tested for MACS, CFCS, or EDACS. The MACS has twenty-seven different language versions [12], and several validation studies have been conducted [17 –25]. Similarly, there are 26 different language versions of the CFCS and EDACS [13, 14], and several validation studies have been conducted [26 –34]. Clarifying the relationship between the four classification systems will help to improve the effectiveness of rehabilitation interventions for children and adolescents with CP, and also allow international comparisons. Therefore, the aim of this study was to develop the CFCS and EDACS, examine the content validity, intra- and inter-rater reliability, and concurrent validity of the Japanese versions of the MACS, CFCS and EDACS, and represent the distribution of levels in each classification systems in comparison with the GMFCS levels.

2 Methods

The present study consisted of three phases. First, translation of the original English CFCS and EDACS into Japanese was carried out. Second, the content validity of the Japanese MACS, CFCS, and EDACS was examined by a multidisciplinary team of professionals involved in therapy for children and adolescents with CP. For the last phase, the intra- and inter-rater reliability of the Japanese MACS, the CFCS, and the EDACS were tested on children and adolescents with CP. The distribution of levels for each classification system in comparison with the GMFCS levels as representing concurrent validity was described. The study protocol was approved by the Ethics Committee of Hokkaido Medical Center for Child Health and Rehabilitation. With regard to the team of professionals, informed consent was obtained from all participants. Children and adolescents with CP along with parents, were provided with the opportunity to opt out. Informed consent was obtained in the form of opt-out placed on the center’s website and notice board. No written informed consent was obtained. The study design was publicized via the website and notice board in the center in accordance with the Ethics Guidelines for Medical and Health Research in Japan.

2.1 Participants

2.1.1 Children with cerebral palsy

Participants were recruited from the Hokkaido Medical Center for Child Health and Rehabilitation, Hokkaido, Japan. This center has fulfilled central roles in medical services for people with CP in Hokkaido. Consequently, most children and adolescents with CP in Hokkaido consult the Center. Participants consisted of children aged over four years with a diagnosis of CP. All children and adolescents with CP who visited the outpatient center from November 2016 to May 2017 were included. However, because of language, cognitive, or other difficulties, children and adolescents were excluded if their parents were unable to understand the study information or instructions.

2.1.2 Professionals for the content validity

Professionals who have practical experience in and specific knowledge of therapy for children and adolescents with CP were recruited through the Hokkaido Medical Center for Child Health and Rehabilitation.

2.2 Procedures

2.2.1 Phase 1: Translation of the CFCS and EDACS

For the development of the Japanese versions of the CFCS and EDACS, a forward-backward translation method was used. The Japanese translation was made with the permission of the developers of the CFCS and EDACS. The translation was executed according to the international standard for translation of self-report measures through forward translation by two or more persons and back-translation of the combined forward ones [35]. The original English CFCS and EDACS were translated into Japanese separately by three individuals (physical therapists). In the Japanese version of the EDACS, some words were changed in consideration of cultural differences as most Japanese people are not familiar with “Halve,” “Tahini,” and “Toffee.’ Instead, the Japanese version used “Mochi,” “Dango,” and “Caramel.” The consensus version, which was discussed until all three individuals agreed, was then translated back into English by a professional translator. After some minor adjustments, such as the reconciliation of word meanings, the developers of the CFCS and EDACS approved the Japanese versions of the CFCS and EDACS and authorized their usage in Japan.

2.2.2 Phase 2: Content validity of the Japanese versions of the MACS, CFCS and EDACS

The content validity of the Japanese version in terms of Japanese values was examined by professionals. They read the Japanese versions of the CFCS and EDACS carefully and provided feedback on the content (particularly its utility in Japan), wording, and readability of the items orally or in writing. Their feedback led to some changes (i.e., from Kanji to Hiragana and from technical terms to general terms), although it should be noted that the construct validity was unchanged [36]. Changes were implemented by discussion among all the authors.

In addition, with regard to the content validity of the MACS, CFCS, and EDACS, the professionals were asked to complete a questionnaire.

2.2.3 Phase 3: Intra- and inter-rater reliability, the distribution of the MACS, CFCS, and EDACS levels compared with the GMFCS levels, and relationships among the four classification systems in terms of concurrent validity

To increase the understanding of three classification systems among the professionals who participated in the study, Japanese versions of three classification systems were distributed one week before the study. The evaluators were asked to read them carefully. Each professional was instructed to perform the tests in a non-specific order. Assessments were conducted independently by interviewing children and/or parents and, where possible, by clinical observation. For intra- and inter-rater reliability, each child with CP was observed by two or more professionals, and the MACS, CFCS, and EDACS functional levels were determined according to the instructions in the respective manual. Each professional performed the three classification systems rating twice with a two-month interval between assessments. A two-month interval was employed as the time interval as it did not allow significant changes in the participants’ functioning while being sufficient to prevent evaluators from remembering their previous assessments [37]. The GMFCS level data was extracted from medical records, which were made known to the raters prior to the rating process.

2.3 Evaluation tools

2.3.1 Questionnaire for content validity

A questionnaire for content validity was developed in accordance with the previous validation study for the Japanese version of the GMFCS [16]. This questionnaire had four items, including overall appropriateness of the translation (“Do you feel that excessively formal expressions are present due to the influence of translation?” and “Do you feel that difficulties in understanding are present due to the influence of translation?”), its validity as a means of classification of function in children with CP (“Is the description of each level appropriate as representative of function in real children with CP?”), and distinctions among levels (“It is difficult to make clear distinctions among levels?”). The response for each item was given as “Yes,” “No,” or “Neither.”

2.3.2 Gross Motor Function Classification System –Expanded and Revised (GMFCS-E&R)

The GMFCS-E&R is a system for the classification of gross motor function in children and youth with CP [4]. Function is classified on the basis of the child’s self-initiated movements, with a focus on gross motor function, and is scored across five levels of function from Level I = independent movement to Level 5 = complete assistance. Each level of the GMFCS provides functional descriptions for five age groups: 1–2, 2–4, 4–6, 6–12, and 12–18 years. In the current study, the Japanese translation of the GMFCS was used, which has been found to have good inter-rater reliability and evidence of content validity [16, 38].

2.3.3 Manual Ability Classification System (MACS)

The MACS is a system for the classification of manual ability in children and youths with CP [5]. The MACS was developed for children aged 4 to 18 years. Function is classified based on the child’s self-initiated ability to handle objects during daily activities. Like the GMFCS, the MACS consists of five levels, which are intended to be clinically meaningful. The Japanese version of the MACS can be downloaded from the MACS website [12].

2.3.4 Communication Function Classification System (CFCS)

The CFCS is a system for the classification of communication functions in individuals with any disability, although it was originally developed for use with individuals with CP [6]. The CFCS was developed for use with individuals aged two years and older. Function is classified based on the child’s everyday performance of all methods of communication, including speech, gestures, eye contact, facial expressions, and augmentative and alternative communication. This classification, too, is distributed across five levels. The Japanese version of the CFCS can be downloaded from the CFCS website [13].

2.3.5 Eating and Drinking Ability Classification System (EDACS)

The EDACS describes the functional eating and drinking abilities of children with CP aged three years and older, using five distinct levels [7]. It refers to the key features of ‘safety’ (aspiration and choking) and ‘efficiency’ (amount of food lost and time taken to eat). The EDACS also provides a three-level ordinal rating scale to describe the degree of assistance required (independent; requires assistance; totally dependent). The Japanese version of the EDACS manual and algorithm can be downloaded from the EDACS website [14].

2.4 Statistical analysis

Statistical analyses were performed by a blinded analyst independent of the evaluators. For the content validity questionnaire, positive responses to each question were considered “positive,” negative responses were considered “negative,” and neither responses were considered “neutral.” To evaluate the content validity, the desired level of the sum of positive and neutral responses for each item was set at≥80%.

For descriptive analyses, the results from each occupational therapist’s rating of the MACS were used. If the occupational therapist did not provide a rating, the physical therapist’s rating as a second option or the physician’s rating as a third option was used. Similarly, for the CFCS, the speech-language-hearing therapist’s rating was the first option, that of the occupational therapist, the second, and the physician’s rating, the third. Therapist rating options for the EDACS were the same as those for the CFCS.

As not all subjects could be evaluated twice during the study period, only those who could be evaluated twice were included in the intra- and inter-rater reliability analysis. Intra- and inter-rater agreements were assessed using percentage agreement and the weighted kappa (κ). Intra- and inter-rater reliability also were assessed using intraclass correlation coefficients (ICCs). The desired level of percentage agreement was set at≥90%. Values of κ between 0.41 and 0.60 indicated moderate agreement, from 0.61 to 0.80 substantial agreement, and from 0.81 to 1.00 almost perfect agreement [37]. ICCs exceeding 0.7 were considered acceptable for measures in groups and ICCs exceeding 0.9 reliable for clinical use with individual patients [37].

For concurrent validity, relationships among the four classification systems were examined by correlation analyses using the Spearman rank correlation coefficient.

Errors in the data were treated as missing values [e.g., if two or more responses were checked for the degree of assistance required (independent; requires assistance; totally dependent) on the EDACS]. Data were analyzed using IBM SPSS 27.0 for all analyses. A p-value < 5% was taken to indicate statistical significance for all the analyses performed.

3 Results

3.1 Participants

Participants included 21 professionals, comprised of three physicians, ten physical therapists, five occupational therapists, and three speech-language-hearing therapists (Table 1), along with 290 children with CP, aged 4 to 21 years (mean 12 years two months [SD 4 years nine months]) (Table 2). One participant with cerebral palsy for whom there was an error in the data was removed from the analysis of the degree of assistance required in the EDACS.

Table 1
Characteristics of professionals employed in the examination of content validity

Total Physician PT OT ST

(N = 21) (n = 3) (n = 10) (n = 5) (n = 3)

Experience (in years)

<5 3 0 3 0 0

5 –10 4 0 2 1 1

10 –15 1 1 0 0 1

15 –20 4 0 2 0 1

20 –25 4 0 2 2 0

>25 5 2 1 2 0

Experience in CP (in years)

<5 7 0 4 1 2

5 –10 2 0 2 0 0

10 –15 1 1 0 0 0

15 –20 3 0 1 2 0

20 –25 5 0 2 2 1

>25 3 2 1 0 0

Age

<29 4 0 4 0 0

30 –39 4 0 2 0 2

40 –49 9 1 4 3 1

50 –59 3 1 0 2 0

>60 1 1 0 0 0

Female/Male 10/11 1/2 4/6/ 3/2 2/1

	Total	Physician	PT	OT	ST
Experience (in years)
<5	3	0	3	0	0
5 –10	4	0	2	1	1
10 –15	1	1	0	0	1
15 –20	4	0	2	0	1
20 –25	4	0	2	2	0
>25	5	2	1	2	0
Experience in CP (in years)
<5	7	0	4	1	2
5 –10	2	0	2	0	0
10 –15	1	1	0	0	0
15 –20	3	0	1	2	0
20 –25	5	0	2	2	1
>25	3	2	1	0	0
Age
<29	4	0	4	0	0
30 –39	4	0	2	0	2
40 –49	9	1	4	3	1
50 –59	3	1	0	2	0
>60	1	1	0	0	0
Female/Male	10/11	1/2	4/6/	3/2	2/1

PT: Physical Therapist, OT: Occupational Therapist, ST: Speech Language Hearing Therapist, CP: Cerebral palsy.

Table 2

Characteristics of participants with cerebral palsy

		Gross Motor Function Classification System levels
	Total	Level I	Level II	Level III	Level IV	Level V
	(N = 290)	(n = 57)	(n = 55)	(n = 35)	(n = 61)	(n = 82)
	n (%)	n (%)	n (%)	n (%)	n (%)	n (%)
Age, mean:SD (in years:months)	12:2 (4:9)	12:10 (4:9)	11:4 (4:7)	12:3 (4:6)	12:11 (4:10)	11:6 (4:9)
Sex
Male	158 (54)	34 (60)	36 (65)	16 (46)	30 (49)	42 (51)
Female	132 (46)	23 (40)	19 (35)	19 (54)	31 (51)	40 (49)
Motor subtype
Unilateral spastic	33 (11)	20 (35)	12 (22)	1 (3)	0 (0)	0 (0)
Bilateral spastic	239 (82)	35 (61)	41 (75)	32 (91)	55 (90)	76 (93)
Dyskinetic	5 (2)	0 (0)	0 (0)	0 (0)	2 (3)	3 (4)
Ataxia	8 (3)	2 (4)	2 (4)	1 (3)	3 (5)	0 (0)
Unknown/Other	5 (2)	0 (0)	0 (0)	1 (3)	1 (2)	3 (4)
Intellectual disability	163 (56)	8 (14)	21 (38)	14 (40)	40 (66)	80 (98)
Epilepsy	137 (47)	10 (18)	11 (20)	12 (34)	33 (54)	71 (87)
Autism Spectrum Disorders	22 (8)	7 (12)	6 (11)	2 (6)	6 (10)	1 (1)
Pervasive Developmental Disorders	17 (6)	6 (11)	5 (9)	2 (6)	4 (7)	0 (0)
Total blindness	10 (3)	0 (0)	2 (4)	0 (0)	5 (8)	3 (4)
Severe deafness	6 (2)	1 (0)	2 (4)	0 (0)	2 (3)	1 (1)

SD: Standard Deviation, CP: Cerebral palsy.

3.2 Content validity of the Japanese versions of the MACS, CFCS and EDACS

Table 3 shows the results of the content validity questionnaire. The sum of positive and neutral responses for each item were≥80%, except for the item “It is difficult to make a clear distinction among levels?” on the CFCS and EDACS.

Table 3
Content validity questionnaires

Item Positive Negative Neutral

n (%) n (%) n (%)

MACS CFCS EDACS MACS CFCS EDACS MACS CFCS EDACS

Do you feel that excessively formal expressions are present due to the influence of translation? 16 19 18 3 2 3 2 0 0

(Yes = Negative, No = Positive) (76.2) (90.5) (85.7) (14.3) (9.5) (14.3) (9.5) (0) (0)

Do you feel that difficulties in understanding are present due to the influence of translation? 15 16 17 4 4 3 2 1 1

(Yes = Negative, No = Positive) (71.5) (76.2) (81.0) (19.0) (19.0) (14.3) (9.5) (4.8) (4.8)

Is the description of each level appropriate as representative of function in real children with CP? 12 11 15 1 1 1 8 9 5

(Yes = Positive, No = Negative) (57.1) (52.4) (71.4) (4.8) (4.8) (4.8) (38.1) (42.9) (23.8)

It is difficult to make clear distinctions among levels? 6 6 9 3 7 5 12 8 7

(Yes = Negative, No = Positive) (28.6) (28.6) (42.9) (14.3) (33.3) (23.8) (57.1) (38.1) (33.3)

Item	Positive	Negative	Neutral
Do you feel that excessively formal expressions are present due to the influence of translation?	16	19	18	3	2	3	2	0	0
(Yes = Negative, No = Positive)	(76.2)	(90.5)	(85.7)	(14.3)	(9.5)	(14.3)	(9.5)	(0)	(0)
Do you feel that difficulties in understanding are present due to the influence of translation?	15	16	17	4	4	3	2	1	1
(Yes = Negative, No = Positive)	(71.5)	(76.2)	(81.0)	(19.0)	(19.0)	(14.3)	(9.5)	(4.8)	(4.8)
Is the description of each level appropriate as representative of function in real children with CP?	12	11	15	1	1	1	8	9	5
(Yes = Positive, No = Negative)	(57.1)	(52.4)	(71.4)	(4.8)	(4.8)	(4.8)	(38.1)	(42.9)	(23.8)
It is difficult to make clear distinctions among levels?	6	6	9	3	7	5	12	8	7
(Yes = Negative, No = Positive)	(28.6)	(28.6)	(42.9)	(14.3)	(33.3)	(23.8)	(57.1)	(38.1)	(33.3)

MACS: Manual Ability Classification System, CFCS: Communication Function Classification System, EDACS: Eating and Drinking Ability Classification System, CP: Cerebral palsy.

3.3 Intra- and inter-rater reliability of the Japanese versions of the MACS, CFCS and EDACS

Table 4 shows the results for intra-rater reliability. The percentage of perfect agreement for each level was calculated based on the rater 1:82% for the MACS level I, 50% for II, 31% for III, 57% for IV, and 92% for V; 91% for the CFCS level I, 54% for II, 36% for III, 65% for IV and 67% for V; 90% for the EDACS level I, 58% for II, 36% for III, 17% for IV and 88% for V; 93% for ‘Independent’ on the EDACS degree of assistance required scale, 72% for ‘Required assistance’ and 94% for ‘Totally dependent.’ Table 5 shows the results for inter-rater reliability. The percentage of perfect agreement for each level was calculated based on the time 1:73% for the MACS level I, 82% for II, 64% for III, 82% for IV, and 97% for V; 81% for the CFCS level I, 100% for II, 46% for III, 76% for IV and 100% for V; 96% for the EDACS level I, 60% for II, 67% for III, 83% for IV and 54% for V; 83% for ‘Independent’ on the EDACS degree of assistance required scale, 65% for ‘Required assistance’ and 100% for ‘Totally dependent.’ Table 6 shows the results for agreement percentage and reliability coefficients. All ICC values were greater than or equal to 0.91, and the lower limit of the 95% confidence interval was greater than or equal to 0.89.

Table 4
Agreement between raters for the Japanese version of the Manual Ability Classification System, the Communication Function Classification System, and the Eating and Drinking Ability Classification System

MACS rater 1 Total

I II III IV V

MACS rater 2 I 50 13 1 0 0 64

II 10 20 21 1 0 52

III 1 6 12 10 0 29

IV 0 1 5 17 5 28

V 0 0 0 2 55 57

Total 61 40 39 30 60 230

CFCS rater 1 Total

I II III IV V

CFCS rater 2 I 73 8 4 0 0 85

II 6 15 11 3 0 35

III 1 3 13 4 0 21

IV 0 2 8 22 17 49

V 0 0 0 5 35 40

Total 80 28 36 34 52 230

EDACS rater 1 Total

I II III IV V

EDACS rater 2 I 98 10 3 0 0 111

II 10 23 12 1 0 46

III 1 7 10 6 1 25

IV 0 0 2 2 4 8

V 0 0 1 3 36 40

Total 109 40 28 12 41 230

EDACS degree of assistance required scale rater 1 total

Ind RA TD

EDACS degree of assistance required scale rater 2 Ind 109 8 1 118

RA 8 33 3 44

TD 0 5 63 68

Total 117 46 67 230

		MACS rater 1	Total
MACS rater 2	I	50	13	1	0	0	64
	II	10	20	21	1	0	52
	III	1	6	12	10	0	29
	IV	0	1	5	17	5	28
	V	0	0	0	2	55	57
Total		61	40	39	30	60	230
		CFCS rater 1	Total
		I	II	III	IV	V
CFCS rater 2	I	73	8	4	0	0	85
	II	6	15	11	3	0	35
	III	1	3	13	4	0	21
	IV	0	2	8	22	17	49
	V	0	0	0	5	35	40
Total		80	28	36	34	52	230
		EDACS rater 1	Total
		I	II	III	IV	V
EDACS rater 2	I	98	10	3	0	0	111
	II	10	23	12	1	0	46
	III	1	7	10	6	1	25
	IV	0	0	2	2	4	8
	V	0	0	1	3	36	40
Total		109	40	28	12	41	230

Table 5

Agreement between the first and second rating by the same rater for the Japanese version of the Manual Ability Classification System, the Communication Function Classification System, and the Eating and Drinking Ability Classification System

		MACS time 1					Total
		I	II	III	IV	V
MACS time 2	I	8	1	0	0	0	9
	II	3	14	3	0	0	20
	III	0	2	9	1	0	12
	IV	0	0	2	9	1	12
	V	0	0	0	1	29	30
Total		11	17	14	11	30	83
		CFCS time 1					Total
		I	II	III	IV	V
CFCS time 2	I	13	0	0	0	0	13
	II	2	11	4	0	0	17
	III	1	0	6	4	0	11
	IV	0	0	3	16	0	19
	V	0	0	0	1	22	23
Total		16	11	13	21	22	83
		EDACS time 1					Total
		I	II	III	IV	V
EDACS time 2	I	24	4	0	0	0	28
	II	1	9	3	1	0	14
	III	0	2	10	0	1	12
	IV	0	0	2	5	21	8
	V	0	0	0	0	22	21
Total		25	15	15	6	41	83

		EDACS degree of assistance required scale time 1			total
Ind	RA	TD
EDACS degree of assistance required scale time 2	Ind	25	1	0	26
	RA	5	13	0	18
	TD	0	6	32	38
Total		30	20	32	82

MACS: Manual Ability Classification System, CFCS: Communication Function Classification System, EDACS: Eating and Drinking Ability Classification System, Ind: Independent, RA: Requires assistance, TD: Totally dependent. One participant with cerebral palsy for whom there was an error in the data was removed from the analysis of the degree of assistance required for the EDACS.

Table 6

Agreement percentages and reliability coefficients of the Japanese version of the Manual Ability Classification system, Communication Function Classification System, and Eating and Drinking Ability Classification System

		Perfect agreement (%)	Close agreement (%)	Kappa (95% CI) p < 0.001	ICC (95% CI) p < 0.001
MACS	Inter-rater	67.0	31.3	0.80 (0.76 –0.84)	0.92 (0.90 –0.94)
	Intra-rater	83.1	16.9	0.90 (0.85 –0.95)	0.96 (0.94 –0.97)
CFCS	Inter-rater	68.7	27.0	0.80 (0.75 –0.84)	0.91 (0.89 –0.93)
	Intra-rater	81.9	16.9	0.88 (0.82 –0.94)	0.95 (0.92 –0.97)
EDACS	Inter-rater	73.5	23.5	0.81 (0.77 –0.86)	0.92 (0.90 –0.94)
	Intra-rater	83.1	15.7	0.90 (0.85 –0.95)	0.96 (0.95 –0.97)
EDACS degree of assistance required scale	Inter-rater	89.1	10.4	0.88 (0.83 –0.92)
	Intra-rater	85.4	14.6	0.85 (0.76 –0.93)

Close agreement indicates a difference of one level either side. MACS: Manual Ability Classification System, CFCS: Communication Function Classification System, EDACS: Eating and Drinking Ability Classification System, ICC: intraclass correlation coefficient, CI: Confidence Interval.

3.4 Distribution of the MACS, CFCS, and EDACS levels compared with the GMFCS levels and relationships among the four classification systems in terms of concurrent validity

Distribution between classification systems based on the GMFCS is shown in Table 7. The Spearman rank correlation coefficients between the GMFCS and the MACS, CFCS and EDACS were 0.87, 0.76 and 0.79, respectively (p < 0.001). Those between the MACS and the CFCS and EDACS were 0.87 and 0.86, respectively (p < 0.001). Those between the CFCS and EDACS were 0.86 (p < 0.001).

Table 7
Distribution between classification systems based on the Gross Motor Function Classification System

GMFCS levels n (%)

I II III IV V Total n (%)

MACS n (%) I 39 (13.4) 19 (6.6) 4 (1.4) 3 (1.0) 0 (0) 65 (22.4)

II 15 (5.2) 31 (10.7) 19 (6.6) 12 (4.1) 0 (0) 77 (26.6)

III 2 (0.7) 4 (1.4) 11 (3.8) 24 (8.3) 0 (0) 41 (14.1)

IV 0 (0) 1 (0.3) 1 (0.3) 19 (6.6) 13 (4.5) 34 (11.7)

V 1 (0.3) 0 (0) 0 (0) 3 (1.0) 69 (23.8) 73 (25.20

CFCS n (%) I 41 (14.1) 31 (10.7) 12 (4.1) 10 (3.4) 1 (0.3) 95 (32.8)

II 10 (3.4) 12 (4.1) 11 (3.8) 10 (3.4) 2 (0.7) 45 (15.5)

III 4 (1.4) 4 (1.4) 4 (1.4) 20 (6.9) 2 (0.7) 34 (11.7)

IV 1 (0.3) 7 (2.4) 7 (2.4) 17 (5.9) 16 (5.5) 48 (16.6)

V 1 (0.3) 1 (0.3) 1 (0.3) 4 (1.4) 61 (21.0) 68 (23.4)

EDACS n (%) I 53 (18.3) 41 (14.1) 21 (7.2) 15 (5.2) 2 (0.7) 132 (45.5)

II 3 (1.0) 12 (4.1) 13 (4.5) 24 (8.3) 4 (1.4) 56 (19.3)

III 0 (0) 2 (0.7) 1 (0.3) 19 (6.6) 17 (5.9) 39 (13.4)

IV 0 (0) 0 (0) 0 (0) 1 (0.3) 11 (3.8) 12 (4.1)

V 1 (0.3) 0 (0) 0 (0) 2 (0.7) 48 (16.6) 51 (17.6)

Total n (%) 57 (19.7) 55 (19.0) 35 (12.1) 61 (21.0) 82 (28.3) 290 (100)

		GMFCS levels n (%)
MACS n (%)	I	39 (13.4)	19 (6.6)	4 (1.4)	3 (1.0)	0 (0)	65 (22.4)
	II	15 (5.2)	31 (10.7)	19 (6.6)	12 (4.1)	0 (0)	77 (26.6)
	III	2 (0.7)	4 (1.4)	11 (3.8)	24 (8.3)	0 (0)	41 (14.1)
	IV	0 (0)	1 (0.3)	1 (0.3)	19 (6.6)	13 (4.5)	34 (11.7)
	V	1 (0.3)	0 (0)	0 (0)	3 (1.0)	69 (23.8)	73 (25.20
CFCS n (%)	I	41 (14.1)	31 (10.7)	12 (4.1)	10 (3.4)	1 (0.3)	95 (32.8)
	II	10 (3.4)	12 (4.1)	11 (3.8)	10 (3.4)	2 (0.7)	45 (15.5)
	III	4 (1.4)	4 (1.4)	4 (1.4)	20 (6.9)	2 (0.7)	34 (11.7)
	IV	1 (0.3)	7 (2.4)	7 (2.4)	17 (5.9)	16 (5.5)	48 (16.6)
	V	1 (0.3)	1 (0.3)	1 (0.3)	4 (1.4)	61 (21.0)	68 (23.4)
EDACS n (%)	I	53 (18.3)	41 (14.1)	21 (7.2)	15 (5.2)	2 (0.7)	132 (45.5)
	II	3 (1.0)	12 (4.1)	13 (4.5)	24 (8.3)	4 (1.4)	56 (19.3)
	III	0 (0)	2 (0.7)	1 (0.3)	19 (6.6)	17 (5.9)	39 (13.4)
	IV	0 (0)	0 (0)	0 (0)	1 (0.3)	11 (3.8)	12 (4.1)
	V	1 (0.3)	0 (0)	0 (0)	2 (0.7)	48 (16.6)	51 (17.6)
Total n (%)		57 (19.7)	55 (19.0)	35 (12.1)	61 (21.0)	82 (28.3)	290 (100)

GMFCS: Gross Motor Function Classification System, MACS: Manual Ability Classification System, CFCS: Communication Function Classification System, EDACS: Eating and Drinking Ability Classification System.

4 Discussion

This study provides evidence of the content validity, inter- and intra-rater reliability, and concurrent validity of the Japanese versions of the MACS, CFCS and EDACS and represents the distribution of the levels of these classification systems in comparison to the GMFCS levels in Japan. The distribution of functional profiles using classification systems for established psychometric properties helps professionals organize interventions according to individual function.

The content validity of the Japanese classification systems was generally good, but some difficulty in differentiating between levels may be present in the CFCS and EDACS. The results of the inter-rater reliability suggest that it was particularly difficult to distinguish between levels III and IV. The terminology used in the CFCS involves the subjective judgment of the rater. For example, whether or not the pace of the conversation is comfortable may depend on the subjectivity of the rater. It may be difficult to determine because of the frequently asked questions on the CFCS website [13] also explain how to define comfortable, conversational pace. In addition, the familiarity between the rater and the child may affect the classification [6, 30]. The difference between levels III and IV is the effectiveness of communication with “familiar partners”. However, whether or not the rater is a “familiar partner” may affect the classification. If the rater is a “familiar partner,” classification may be easy; however, if not, it may be difficult to accurately assess the effectiveness of communication with a “familiar partner.” The same is noted in the Dutch version [30]. Therefore, it may be necessary to clearly define the effectiveness of communication. For the EDACS, in terms of the difference between levels I and II, the rater may be confused in determining eating efficiency as in a previous study [33]. For example, if only clinical physical findings are available, it may be difficult to determine aspiration risk. In particular, the difference between levels III and IV is expected to make it difficult to determine silent aspiration. As noted in the EDACS [7], silent aspiration is clinically difficult to identify and may affect the EDACS classification. It is important to check for aspiration to address the difficulty in distinguishing between levels III and IV, which has also been shown in the English version [7, 31].

For reliability, all percentage agreement values were at a desirable level. The reliability coefficients were also all adequate, including the 95% confidence interval. In a previous study, the reliability coefficients for the intra-rater reliability of the MACS, CFCS, and EDACS were greater than 0.8 for professionals [6 , 29]. In terms of inter-rater reliability, the reliability coefficients among professionals were greater than 0.6 [5–7 , 39]. Thus, the results of this study were similar to those of previous studies.

Correlations among the four classification systems show that the GMFCS was highly correlated with the MACS but slightly less correlated with the CFCS and EDACS. These results are similar to those of previous studies and indicate that gross motor function differs from communication and feeding functions [26 –34]. The correlation coefficients were higher than those obtained in previous studies. The reason for this was that the participants in this study were classified at higher levels on the GMFCS than the participants in the previous studies. According to the study by Almasri et al., the correlation coefficient between the GMFCS and MACS was higher in children with spastic quadriplegia with higher GMFCS levels and lower in children with hemiplegia with lower GMFCS levels [3]. In Japan, there tends to be a greater proportion of individuals with cerebral palsy at GMFCS levels IV and V [40], as well as the participants of this study. Furthermore, the distribution of the MACS levels in comparison with the GMFCS levels illustrated that the MACS was correlated with the GMFCS, as indicated by the correlation coefficient. The CFCS was also correlated with the GMFCS as well as the MACS, but they scattered throughout the levels of the GMFCS. In the EDACS, eating and drinking functions were preserved in levels I to III of the GMFCS but were variably distributed in levels IV and V of the GMFCS. These results reflect the results of the correlation coefficients above and may also show the concurrent validity of these classification systems.

The current study has several limitations. As the data collection was performed at a central facility in Hokkaido, there may be limitations in generalizing the results to other parts of Japan, taking into account the regional nature of health care systems. The specific order in which the rating is performed may affect the results of the rating. It is also questionable whether the intra-rater assessment at a two-month interval is appropriate. Furthermore, using only GMFCS to examine the concurrent validity may be regarded as a limitation.

5 Conclusions

The content validity of the Japanese version of the MACS, CFCS and EDACS were generally good. For reliability, all percentage agreement values were at a desirable level, and the reliability coefficients were all adequate. The correlation coefficients among the GMFCS, MACS, CFCS, and EDACS were high.

Footnotes

Acknowledgments

We are very grateful to Dr. Hidecker and Dr. Sellers for sharing their knowledge and providing support. We also would like to thank Dr. Garry Heterick for the back-translation and English language editing. This work was supported by JSPS KAKENHI Grant Number JP 17K01460.

Conflict of interest

The authors have no conflicts of interest to report.

References

Surveillance of cerebral palsy in Europe: a collaboration of cerebral palsy surveys and registers. Surveillance of Cerebral Palsy in Europe (SCPE). Dev Med Child Neurol. 2000;42(12):816–24. doi: 10.1017/s0012162200001511

Rosenbaum

, Paneth

, Leviton

, et al. A report: the definition and classification of cerebral palsy April 2006. Dev Med Child Neurol Suppl. 2007;109:8–14.

Almasri

, Saleh

, Abu-Dahab

, Malkawi

, Nordmark

. Functional profiles of children with cerebral palsy in Jordan based on the association between gross motor function and manual ability. BMC Pediatr. 2018;18(1):276. doi: 10.1186/s12887-018-1257-x

Palisano

, Rosenbaum

, Bartlett

, Livingston

. Content validity of the expanded and revised Gross Motor Function Classification System. Dev Med Child Neurol. 2008;50(10):744–50. doi: 10.1111/j.1469-8749.2008.03089.x

Eliasson

, Krumlinde-Sundholm

, Rösblad

, et al. The ManualAbility Classification System (MACS) for children with cerebralpalsy: scale development and evidence of validity and reliability. Dev Med Child Neurol. 2006;48(7):549–54. doi: 10.1017/s0012162206001162

Hidecker

, Paneth

, Rosenbaum

, et al. Developing and validating the Communication Function Classification System for individuals with cerebral palsy. Dev Med Child Neurol. 2011;53(8):704–10. doi: 10.1111/j.1469-8749.2011.03996.x

Sellers

, Mandy

, Pennington

, Hankins

, Morris

. Development and reliability of a system to classify the eating and drinking ability of people with cerebral palsy. Dev Med Child Neurol. 2014;56(3):245–51. doi: 10.1111/dmcn.12352

Rosenbaum

, Eliasson

, Hidecker

, Palisano

. Classification in childhood disability: focusing on function in the 21st century. J Child Neurol. 2014;29(8):1036–45. doi: 10.1177/0883073814533008

Palisano

, Avery

, Gorter

, Galuppi

, McCoy

. Stability of the Gross Motor Function Classification System, Manual Ability Classification System, and Communication Function Classification System. Dev Med Child Neurol. 2018;60(10):1026–32. doi: 10.1111/dmcn.13903

10.

Graham

, Rosenbaum

, Paneth

, et al. Cerebral palsy. Nat Rev Dis Primers. 2016;2:15082. doi: 10.1038/nrd2015.82

11.

GMFCS – E&R. CanChild [cited 9 September 2021]. Available from: https://canchild.ca/en/resources/42-gross-motor-function-classification-system-expanded-revised-gmfcs-e-r

12.

Manual Ability Classification System [homepage on the Internet]. [cited 9 September 2021]. Available from: https://www.macs.nu

13.

Communication Function Classification System [homepage on the Internet]. [cited 9 September 2021]. Available from: http://cfcs.us

14.

Eating and Drinking Ability Classification System. Brighton: Sussex Community NHS Foundation Trust; 2022 [cited 9 September 2021]. Available from: https://www.sussexcommunity.nhs.uk/get-involved/research/chailey-research/eating-drinking-classification.htm

15.

Rosenbaum

, Rosenbloom

. Cerebral palsy: from diagnosis to adult life. London: Mac Keith Press; 2012.

16.

Kondo

, Teranishi

, Iwata

, Sonoda

, Saitoh

. Reliability Study of Gross Motor Function Classification System and Delphi Survey of Expert Opinion for Clinical Use of this System in Japan. The Japanese Journal of Rehabilitation Medicine. 2009;46(8):519–26. doi: 10.2490/jjrmc.46.519

17.

Morris

, Kurinczuk

, Fitzpatrick

, Rosenbaum

. Reliability of the manual ability classification system for children with cerebral palsy. Dev Med Child Neurol. 2006;48(12):950–3. doi: 10.1017/s001216220600209x

18.

Öhrvall

, Eliasson

. Parents’ and therapists’ perceptions of the content of the Manual Ability Classification System, MACS. Scand J Occup Ther. 2010;17(3):209–16. doi: 10.1080/11038120903125101

19.

Kuijper

, van der Wilden

, Ketelaar

, Gorter

. Manual ability classification system for children with cerebral palsy in a school setting and its relationship to home self-care activities. Am J Occup Ther. 2010;64(4):614–20. doi: 10.5014/ajot.2010.08087

20.

Akpinar

, Tezel

, Eliasson

, Icagasioglu

. Reliability and cross-cultural validation of the Turkish version of Manual Ability Classification System (MACS) for children with cerebral palsy. Disabil Rehabil. 2010;32(23):1910–6. doi: 10.3109/09638281003763796

21.

Mutlu

, Kara

, Gunel

, Karahan

, Livanelioglu

. Agreement between parents and clinicians for the motor functional classification systems of children with cerebral palsy. Disabil Rehabil. 2011;33(11):927–32. doi: 10.3109/09638288.2010.514645

22.

Jang

, Sung

, Kang

, et al. Reliability and validity of the Korean version of the manual ability classification system for children with cerebral palsy. Child Care Health Dev. 2013;39(1):90–3. doi: 10.1111/j.1365-2214.2012.01408.x

23.

Silva

, Funayama

, Pfeifer

. Manual Ability Classification System (MACS): reliability between therapists and parents in Brazil. Braz J Phys Ther. 2015;19(1):26–33. doi: 10.1590/bjpt-rbf.2014.0065

24.

Burgess

, Boyd

, Ziviani

, Chatfield

, Ware

, Sakzewski

. Stability of the Manual Ability Classification System in young children with cerebral palsy. Dev Med Child Neurol. 2019;61(7):798–804. doi: 10.1111/dmcn.14143

25.

Piscitelli

, Vercelli

, Meroni

, Zagnoni

, Pellicciari

. Reliability of the gross motor function classification system and the manual ability classification system in children with cerebral palsy in Tanzania. Dev Neurorehabil. 2019;22(2):80–6. doi: 10.1080/17518423.2017.1342710

26.

Choi

, Hwang

, Rha

, Park

. Reliability and validity of the Korean-language version of the Communication Function Classification System in children with cerebral palsy. Child Care Health Dev. 2018;44(1):140–6. doi: 10.1111/cch.12507

27.

Cunningham

, Rosenbaum

, Hidecker

. Promoting consistent use of the communication function classification system (CFCS). Disabil Rehabil. 2016;38(2):195–204. doi: 10.3109/09638288.2015.1027009

28.

Randall

, Harvey

, Imms

, Reid

, Lee

, Reddihough

. Reliable classification of functional profiles and movement disorders of children with cerebral palsy. Phys Occup Ther Pediatr. 2013;33(3):342–52. doi: 10.3109/01942638.2012.747584

29.

Soleymani

, Joveini

, Baghestani

. The Communication Function Classification System: cultural adaptation, validity, and reliability of the Farsi version for patients with cerebral palsy. Pediatr Neurol. 2015;52(3):333–7. doi: 10.1016/j.pediatrneurol.2014.10.026

30.

Vander

Zwart KE

, Geytenbeek

, de Kleijn

, et al. Reliability of the Dutch-language version of the Communication Function Classification System and its association with language comprehension and method of communication. Dev Med Child Neurol. 2016;58(2):180–8. doi: 10.1111/dmcn.12839

31.

Benfer

, Weir

, Bell

, Ware

, Davies

PSW

, Boyd

. The Eating and Drinking Ability Classification System in a population-based sample of preschool children with cerebral palsy. Dev Med Child Neurol. 2017;59(6):647–54. doi: 10.1111/dmcn.13403

32.

Hyun

, Yi

, Shin

H-I

. Reliability and Validity of the Eating and Drinking Ability Classification System in Adults with Cerebral Palsy. Dysphagia. 2021;36(3):351–61. doi: 10.1007/s00455-020-10141-y

33.

Tschirren

, Bauer

, Hanser

, Marsico

, Sellers

, van Hedel

HJA

. The Eating and Drinking Ability Classification System: concurrent validity and reliability in children with cerebral palsy. Dev Med Child Neurol. 2018;60(6):611–7. doi: 10.1111/dmcn.13751

34.

van Hulst

, Snik

DAC

, Jongerius

, Sellers

, Erasmus

, Geurts

ACH

. Reliability, construct validity and usability of the Eating and Drinking Ability Classification System (EDACS) among Dutch children with Cerebral Palsy. J Pediatr Rehabil Med. 2018;11(2):115–24. doi: 10.3233/prm-170515

35.

Beaton

, Bombardier

, Guillemin

, Ferraz

. Guidelines for the process of cross-cultural adaptation of self-report measures. Spine (Phila Pa 1976). 2000;25(24):3186–91. doi: 10.1097/00007632-200012150-00014

36.

Streiner

, Norman

, Cairney

. Health measurement scales: a practical guide to their development and use. USA: Oxford University Press; 2015.

37.

Portney

. Foundations of clinical research: applications to evidence-based practice. 4^th ed. Philadelphia: FA Davis; 2020.

38.

Kondo

, Hosokawa

, Soma

, et al. Gross motor function classification system: preliminary study for Japanese children. Am J Phys Med Rehabil. 2003;82(2):116–21. doi: 10.1097/00002060-200302000-00006

39.

Plasschaert

, Ketelaar

, Nijnuis

, Enkelaar

, Gorter

. Classification of manual abilities in children with cerebral palsy under 5 years of age: how reliable is the Manual Ability Classification System?Clin Rehabil. 2009;23(2):164–70. doi: 10.1177/0269215508098892

40.

Touyama

, Touyama

, Ochiai

, Toyokawa

, Kobayashi

. Long-term survival of children with cerebral palsy in Okinawa, Japan. Dev Med Child Neurol. 2013;55(5):459–63. doi: 10.1111/j.1469-8749.2012.04429.x

	Total	Physician	PT	OT	ST
	(N = 21)	(n = 3)	(n = 10)	(n = 5)	(n = 3)
Experience (in years)
<5	3	0	3	0	0
5 –10	4	0	2	1	1
10 –15	1	1	0	0	1
15 –20	4	0	2	0	1
20 –25	4	0	2	2	0
>25	5	2	1	2	0
Experience in CP (in years)
<5	7	0	4	1	2
5 –10	2	0	2	0	0
10 –15	1	1	0	0	0
15 –20	3	0	1	2	0
20 –25	5	0	2	2	1
>25	3	2	1	0	0
Age
<29	4	0	4	0	0
30 –39	4	0	2	0	2
40 –49	9	1	4	3	1
50 –59	3	1	0	2	0
>60	1	1	0	0	0
Female/Male	10/11	1/2	4/6/	3/2	2/1

Psychometric evaluation and distribution of classification systems in children with cerebral palsy in Japan

Abstract

PURPOSE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1 Introduction

2 Methods

2.1 Participants

2.1.1 Children with cerebral palsy

2.1.2 Professionals for the content validity

2.2 Procedures

2.2.1 Phase 1: Translation of the CFCS and EDACS

2.2.2 Phase 2: Content validity of the Japanese versions of the MACS, CFCS and EDACS

2.2.3 Phase 3: Intra- and inter-rater reliability, the distribution of the MACS, CFCS, and EDACS levels compared with the GMFCS levels, and relationships among the four classification systems in terms of concurrent validity

2.3 Evaluation tools

2.3.1 Questionnaire for content validity

2.3.2 Gross Motor Function Classification System –Expanded and Revised (GMFCS-E&R)

2.3.3 Manual Ability Classification System (MACS)

2.3.4 Communication Function Classification System (CFCS)

2.3.5 Eating and Drinking Ability Classification System (EDACS)

2.4 Statistical analysis

3 Results

3.1 Participants

5 Conclusions

Footnotes

Acknowledgments

Conflict of interest

References