• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Dev Med Child Neurol. Author manuscript; available in PMC Aug 1, 2012.
Published in final edited form as:
PMCID: PMC3130799

Developing and validating the Communication Function Classification System for individuals with cerebral palsy



The purpose of this study was to create and validate a Communication Function Classification System (CFCS) for children with cerebral palsy (CP) that can be used by a wide variety of individuals who are interested in CP. This paper reports the content validity, interrater reliability, and test–retest reliability of the CFCS for children with CP.


An 11-member development team created comprehensive descriptions of the CFCS levels, and four nominal groups comprising 27 participants critiqued these levels. Within a Delphi survey, 112 participants commented on the clarity and usefulness of the CFCS. Interrater reliability was completed by 61 professionals and 68 parents/relatives who classified 69 children with CP aged 2 to 18 years. Test–retest reliability was completed by 48 professionals who allowed at least 2 weeks between classifications. The participants who assessed the CFCS were all relevant stakeholders: adults with CP, parents of children with CP, educators, occupational therapists, physical therapists, physicians, and speech–language pathologists.


The interrater reliability of the CFCS was 0.66 between two professionals and 0.49 between a parent and a professional. Professional interrater reliability improved to 0.77 for classification of children older than 4 years. The test–retest reliability was 0.82.


The CFCS demonstrates content validity and shows very good test–retest reliability, good professional interrater reliability, and moderate parent–professional interrater reliability. Combining the CFCS with the Gross Motor Function Classification System and the Manual Ability Classification System contributes to a functional performance view of daily life for individuals with CP, in accordance with the World Health Organization’s International Classification of Functioning, Disability and Health.

Individuals with cerebral palsy (CP) have sensorimotor and developmental issues that affect their daily lives by restricting their mobility, manipulation of objects, and/or communication.1 Within the framework of the World Health Organization’s International Classification of Functioning, Disability and Health (ICF),2,3 the Gross Motor Function Classification System (GMFCS)4 and the Manual Ability Classification System (MACS) for children with CP5 make it possible to classify mobility and handling objects respectively, at the ICF activity/participation level.6 However, no analogous classification of functional communication has been available for use in CP practice and research. The lack of a communication classification tool that is quick, reliable, valid, and easy to use limits the comparison of descriptive CP epidemiology studies as well as the interpretation and generalizability of CP treatment studies.

Communication disorders can be described from several perspectives: body structure and function level, activity level, and participation level, as well as environmental and personal levels.2,3,713 Estimates of communication disorders in CP have varied from 31%14 to 88%.15 This wide range is partly a result of the lack of a consensus definition of communication disorders within CP research and practice. A recent study from a Norwegian CP registry reported that 51% of children with CP had speech problems as classified by ratings of ‘slightly indistinct’, ‘obviously indistinct’, ‘severely indistinct’, or ‘no speech’, including 19% who had ‘no speech’.16 This population-based estimate reporting indistinct or no speech may underestimate CP communication disorders as it may not capture other types of communication problems resulting from hearing or language impairments. However, reporting speech, language, and hearing difficulties simply suggests the range of associated impairments in CP, not the more pertinent daily-life issues of how well a child with CP communicates with family, friends, acquaintances, and strangers.13

The purpose of this study was to create and validate a communication function classification system (CFCS) for children with CP, for use by a wide variety of individuals interested in CP. This required a shift from the traditional focus on body structure and function (i.e. assessing components of speech, language, and hearing problems), to a focus on activity/participation, specifically the way in which to classify a person’s communication capacity within real-life situations.


The development and validation of the CFCS involved four phases that were methodologically similar to the development of the GMFCS and the MACS.6 In the first phase, the CFCS was drafted by a development team. The second and third phases focused on revision and validation and included a series of four nominal group studies and two Delphi survey consensus rounds. The fourth phase measured interrater reliability among professionals and parents, and test–retest reliability. Ethical approval of this research was granted by the institutional review board at Michigan State University. Institutional partners in the reliability phase either accepted the institutional review board approval from Michigan State University or sought and received their own institutional review board approval.

Phase 1: Initial development


The 11 participants in this phase were from eight groups of individuals who had experience with CP and communication: one adult with CP, one parent of a child with CP, one educator, one neurologist, two occupational therapists, two paediatricians, one physical therapist, and three speech–language pathologists/researchers (note: one participant was an educator who has CP so she was counted in both groups).


The participants discussed and created the first draft of the CFCS, the development of which was based on reviews of the literature on functional communication7,8,10,11,1721 and the participants’ experiences of communication in real-life situations. Participants shared their ideas through a series of five conference calls and subsequent e-mails. The number of conference-call participants varied from two to three per call to accommodate participants’ schedules. Discussions and suggested revisions were then shared by e-mails and subsequent phone calls. Each element of the CFCS was evaluated for usefulness and clarity.


The initial draft of the CFCS included the following variables that the development team viewed as important components of functional communication: (1) sender skills: conveying a message to a communication partner, which may (but not necessarily) include intelligible speech and expressive language skills; (2) receiver skills: understanding a message sent by a communication partner, which may (but not necessarily) include functional hearing and receptive language skills; (3) pace of communication: conversational turn taking that occurs at an expected speed without long pauses between sending and receiving; (4) degree of familiarity with the communication partners: how well one knows the partner can influence the success of the communicative event as conversations are constructed between two or more people; (5) age-appropriateness: communication skills are expected to be commensurate with the child’s developmental level; and (6) use of augmentative and alternative communication (AAC): including communication methods such as manual signs, pictures, communication boards, communication books, and talking devices (sometimes called voice output communication aids or speech-generating devices).

The initial development process concluded when the participants were satisfied with the CFCS draft. The draft, which included the six components discussed by the phase 1 development team, was then evaluated by the nominal groups in the next phase.

Phase 2: Nominal groups


The nominal group included 27 participants from Canada and the USA. See Table I for participant characteristics.

Table I
Participant Characteristics of the participants in the development of the Communication Function Classification System


The nominal group process fosters discussions among individuals in order to create a consensus opinion.22,23 Four nominal group rounds were held to achieve content validity. Participants were placed into nominal groups by schedule preferences. When scheduling permitted, at least one member from each of the eight stakeholder groups was included in each nominal group. Two weeks before the nominal group process began, participants were sent the open-ended questions and the latest version of the CFCS (questions are available upon request from the corresponding author).

Each group was given the opportunity to discuss the CFCS during a 4-hour time block. An experienced group facilitator (PLR for one and JL for the other three nominal groups) was present to ask questions, move between topics, restate participants’ thoughts, suggestions, and ideas, provide structure during the meeting, and ask the group to vote on different suggestions to determine consensus. Individuals’ first names were used to encourage equality among members, despite differing experiences and educational backgrounds. The first author (MJCH) was also present during the nominal group process to answer questions about the CFCS, but she refrained from adding comments or suggestions.


This nominal group process included progressive feedback whereby each group suggested changes and raised issues, building upon the previous group’s revisions. The changes and issues that received a majority vote from that nominal group were incorporated into the next revision. This serial consideration continued until the last nominal group mostly suggested wording changes and a consensus emerged about the concepts.

Discussion from the early nominal group clarified the point that the CFCS levels encompassed the range of communication effectiveness and that the CFCS wording was understandable to stakeholders’ differing perspectives. The creation of a CFCS level identification chart was suggested and, subsequently an algorithm was incorporated into the CFCS.

Phase 3: Delphi surveys


The Delphi phase consisted of 112 participants who were recruited internationally. All participants from round 1 were invited to participate in round 2, and 69 of the original 112 completed Round 2. See Table I for participant characteristics.


Delphi surveys provide a structured method for stakeholders to provide anonymous feedback.22,23 The purpose of the surveys was to reach a large number of individuals with expertise in CP in order to evaluate the clarity, conciseness, and usefulness of the CFCS draft. Delphi surveys were conducted in rounds until a preset 80% target agreement was reached on all closed-ended questions. The round 1 Delphi survey, with 36 questions similar to those used in the nominal groups, was completed in a Web- or paper-based format (questions are listed in Table SI, supporting information published online). A second Delphi survey round was held for the one question that did not reach the targeted 80% agreement. Open-ended responses were analysed for themes and possible changes for the CFCS.


All but one closed-ended question received greater than 80% agreement in round 1. That closed-ended question (on being able to differentiate between the CFCS levels) received more than 80% agreement in round 2 after wording changes were made to the draft. Responses to some of the open-ended questions had the common themes of describing communication pace, making global judgements of communication performance, and including all communication methods such as AAC. Delphi survey respondents felt that the CFCS might be useful in classifying functional communication, providing a common terminology, assisting with clinical decision making, being used with the GMFCS and MACS to provide a functional profile of an individual’s performance, and recognizing all the communication methods including AAC. A ‘frequently asked questions’ section was created from these themes and is available on the CFCS website (http://cfcs.us).

Phase 4: Interrater and test–retest reliability


Participants in the reliability phase comprised 61 professionals and 68 family members (primarily parents) from the USA and Canada who classified 69 children with CP (age range 2y–18y; median age 5y 0mo; mean age 6y 6mo, SD 3y 6mo). The children’s GMFCS and MACS classifications varied (see Table I for participant characteristics; see Table II for characteristics unique to the children).The test–retest reliability was completed by 48 of the professionals.

Table II
Characteristics of children with cerebral palsy who participated in the reliability phases of development of the CFCS (also see Table I)


To calculate the interrater reliability, at least two professionals and one parent were asked to classify the communication performance of a child with CP. As stated in the CFCS instructions, the classification should be made by someone who is familiar with the child’s everyday communication. Interrater reliability (weighted kappa) was calculated between each pair of professionals and between each parent/relative and each professional involved with the same child. Professionals classified a child’s communication performance twice with at least a 2-week interval between classifications. In 25% of cases, the child was not present at the reclassification, and the professional used case notes and/or his previous interactions with the child to make the second classification. In addition, the raters were asked to provide information about their educational level, CP experiences, and team role (i.e. parent or professional), as well as the child’s characteristics including birth date, comorbidities, communication methods, and GMFCS and MACS levels.


Table III presents the five CFCS levels as well as a comparison with the five levels of the GMFCS and MACS. The CFCS, with instructions, definitions, and additional details, can be downloaded at www.cfcs.us for clinical and research use. The weighted kappa interrater reliability among professionals for the 68 children classified was 0.66 (95% confidence interval [CI] 0.55–0.78). Table IV shows the distribution of the resulting 73 CFCS classification comparisons. The interrater reliability may be related to the child’s age at the time of the CFCS classification: on the sample of children who were aged 5 years and older, weighted kappa improved to 0.77 (95% CI 0.66–0.89). The professionals’ familiarity with the child varied from only a first visit to the provision of ongoing services. However, sub-analyses by the professionals’ familiarity or by the settings of clinic, school, or home did not change CFCS agreement.

Table III
The five levels of gross motor function (GMFCS) manual ability, (MACS), and communication function (CFCS) classification systems
Table IV
Reliability of the Communication Function Classification System (CFCS)

A subset of 48 professionals reclassified the children’s communication performance at least 2 weeks after the initial classification, after noting that the child had not had any intervening health change. The weighted kappa test–retest reliability was 0.82 (95% CI 0.74–0.90). Table IV shows the distribution of the resulting 89 CFCS classification test–retest comparisons.

A parent and a professional classified the communication performance of the same child. The weighted kappa interrater reliability between professionals and parents for 69 children was 0.49 (95% CI 0.40–0.59). Table IV shows the distribution of the resulting 138 CFCS classification comparisons. Professionals tended to classify the child’s communication as less effective than the parent’s classification. When the parent–professional interrater reliability was calculated on the sample of children aged 5 years and older, the weighted kappa decreased to 0.42 (95% CI 0.29–0.55).


The CFCS was empirically derived from the communication disorders literature and expert experience to classify patterns of an individual’s communication performance in one of five levels of everyday communication effectiveness with a partner (Table III). Communication occurs when conversation partners establish a shared understanding.24,25 This shared understanding (i.e. communication) results from an interaction among components of ICF activity, participation, and body structure and function (e.g. speech, language, and hearing skills) with contextual factors (e.g. setting, conversational partner skills, AAC methods, and cultural backgrounds).2,3,12

Systematically including multiple stakeholders in the development and validation of the CFCS provided different viewpoints that led to important insights and helped to create a classification that is more likely to be stakeholder-friendly for professionals and lay people. It provides a global judgement of functional communication performance at the ICF activity/participation level, regardless of the specific body structure and functions relating to a communication disorder or another specific diagnosis. The CFCS classification considers all communication methods including AAC use.

The test–retest reliability of the CFCS was very good, and professional interrater reliability was good.26 The interrater reliability between parents and professionals was moderate, with parents tending to classify their children’s communication as being more effective than professionals perceived it to be. Parents are likely to see the children in more environments and with more communication partners, but at the same time parents may underestimate the difficulty that unfamiliar communication partners may have in communicating with the children. Research is needed to explore and understand the differences underlying parents’ and professionals’ classifications.

Communication is quite different from walking and object manipulation, which depend largely or entirely on the capacity of the individual doing those activities, although environmental factors do influence whether and how these activities are performed. Communication is the exchange of information between people, and the responsibility for communication, therefore, is typically shared between conversational partners. The CFCS classification requires a rater to know how the individual with CP communicates with both unfamiliar and familiar communication partners in different environments in order to establish shared understandings.

A distinction between classification and assessment is critical in order to understand the CFCS and similar instruments. The GMFCS, MACS, and CFCS are not assessments or tests, although they may be complementary to detailed assessments.6 Classification systems are designed to recognize and distinguish varied ‘levels’ of functional patterns. In contrast, assessments or tests are systematic, usually standardized, methods of collecting or measuring aspects of the person. As with the GMFCS and the MACS, the CFCS does not explain the reasons why a person’s performance falls within a particular classification level. Building on the conceptual foundation of the ICF, functional patterns may be due to differing aspects of the person and contextual factors including speech intelligibility, hearing sensitivity, language skills, AAC competencies, and familiarity of conversational partners and their communication skills. The CFCS classification task of making a global judgement of effective communication may be novel to some individuals, especially those who are not speech–language pathologists or otherwise trained in communication. Familiarity and practice with the CFCS concepts may be needed to improve interrater reliability.

Possible age effects were considered throughout the development of the CFCS. As mobility,27 hand function, and communication performances typically improve as children develop, the effect of age on performance is not surprising. For example, in one study using the GMFCS, younger children were better described by three levels that differentiated into five more consistent GMFCS levels as the child grew older.25 The improvement in interrater reliability for the CFCS in children older than 4 years may also suggest an age effect. To address the effect of age, the GMFCS uses age bands to describe mobility performance at different ages. The MACS is not recommended for use with children under 4 years of age, whereas the CFCS has been used with children as young as 2 years old. An age-cohort study about the stability of the CFCS in CP is needed and could provide an insight into CFCS levels and age effects across the lifespan of an individual.

Several research projects are currently using the CFCS, including some that are validating the CFCS in languages other than English. These CFCS translations will be posted on the CFCS website (www.cfcs.us) as they become available. Similar CFCS validation and reliability studies are needed to determine the applicability to other populations, including individuals with autism, Down syndrome, hearing loss, and those who have suffered a stroke. If the CFCS classifies communication performance at an activity/participation level then, in theory, the CFCS should be applicable to communication regardless of the underlying body structure and function issues. What would vary by diagnostic group would be any potential prognostic implications associated with a particular classification.

The development of the CFCS addressed some of the limitations of classification by (1) proposing a valid and reliable tool for clinical and research applications that seek to classify communication at an ICF activity/participation level;2,3 (2) including previously underrepresented groups such as adults with CP and parents of children with CP in its development, and (3) providing a common language when talking with an individual with CP, family members, and other professionals. Combining the CFCS with GMFCS and MACS levels contributes to a functional performance view of daily life for individuals with CP.

What this paper adds

  • Describes the development and validation of a communication classification system for children with CP (CFCS), including content validity and reliability.
  • The CFCS categorizes communication at the activity/participation level of the ICF.
  • The system was developed as an analogue of the GMFCS and the MACS, potentially to create functional profiles of people with CP.

Supplementary Material

Supp Material

Supp Table S1


This research was supported in part by a National Institutes of Health postdoctoral fellowship (NIDCD F32 DC008265–02) to the first author, as well as grants from the Cerebral Palsy International Research Foundation and The Hearst Foundation to the first three authors.

We thank those who chose to participate in the research anonymously. We also thank all individuals and institutions that contributed to this study. (A list of all those who contributed to this study can be found online).


Augmentative and alternative communication
Communication Function Classification System
Manual Ability Classification System


ONLINE MATERIAL/SUPPORTING INFORMATION Additional material and supporting information may be found in the online version of this article. Please note: Wiley–Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author of the article.


1. Rosenbaum P, Paneth N, Leviton A, et al. A report: the definition and classification of cerebral palsy April 2006. Dev Med Child Neurol. 2007;49(Suppl. 109):8–14. [PubMed]
2. World Health Organization . International Classification of Functioning, Disability and Health: ICF. World Health Organization; Geneva: 2001.
3. World Health Organization . International Classification of Functioning, Disability, and Health: Children & Youth Version: ICF-CY. World Health Organization; Geneva: 2007.
4. Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol. 1997;39:214–23. [PubMed]
5. Eliasson A, Krumlinde-Sundholm L, Rösblad B, et al. The Manual Ability Classification System (MACS) for children with cerebral palsy: scale development and evidence of validity and reliability. Dev Med Child Neurol. 2006;48:549–54. [PubMed]
6. Rosenbaum PL, Palisano RJ, Bartlett DJ, Galuppi BE, Russell DJ. Development of the Gross Motor Function Classification System for cerebral palsy. Dev Med Child Neurol. 2008;50:249–53. [PubMed]
7. Frattali C. Measuring Outcomes in Speech-Language Pathology. Thieme; New York: 1998.
8. Yorkston KM, Strand EA, Kennedy MRT. Comprehensibility of dysarthric speech: mplications for assessment and treatment planning. Am J Speech Lang Pathol. 1996;5:55–66.
9. World Health Organization . Towards a Common Language for Functioning, Disability and Health: ICF. World Health Organization; Geneva: 2002.
10. Threats TT, Worrall L. Classifying communication disability using the ICF. Adv Speech Lang Pathol. 2004;6:53–62.
11. Worrall L, McCooey R, Davidson B, Larkins B, Hickson L. The validity of functional assessments of communication and the activity/participation components of the ICIDH-2: Do they reflect what really happens in real-life? J Commun Disord. 2002;35:107–37. [PubMed]
12. Hidecker MJ. Building the evidence for communication interventions. Dev Med Child Neurol. 2010;52:316–17. [PubMed]
13. Hidecker MJC. Communication activity and participation research. Dev Med Child Neurol. 2010;52:408–9. [PubMed]
14. Wolfe WG. A comprehensive evaluation of fifty cases of cerebral palsy. J Speech Hear Disord. 1950;15:234–51. [PubMed]
15. Achilles RF. Communicative anomalies of individuals with cerebral palsy. Cereb Palsy Rev. 1955;16:15–24.
16. Andersen G, Mjoen TR, Vik T. Prevalence of speech problems and the use of augmentative and alternative communication in children with cerebral palsy: a registry-based study in Norway. Perspectives on Augmentative and Alternative Communication. 2010;19:12–20.
17. Kent RD, Miolo G, Bloedel S. The intelligibility of children’s speech: a review of evaluation procedures. Am J Speech Lang Pathol. 1994;3:81–95.
18. Duffy JR. Motor Speech Disorders: Substrates, Differential Diagnosis, and Management. Elsevier Mosby; St. Louis, Mo.: 2005.
19. Yorkston KM, Beukelman DR. Communication efficiency of dysarthric speakers as measured by sentence intelligibility and speaking rate. J Speech Hear Disord. 1981;46:296–301. [PubMed]
20. Blackstone S, Berg M Hunt. Social Networks: A Communication Inventory for Individuals with Complex Communication Needs and their Communication Partners. Augmentative Communication, Inc; Monterey, CA: 2003.
21. Msall ME, Tremont MR. Functional outcomes in self-care, mobility, communication, and learning in extremely low-birthweight infants. Clin Perinatol. 2000;27:381–401. [PubMed]
22. Delbecq AL, Van de Ven AH, Gustafson DH. Group Techniques for Program Planning: A Guide to Nominal Group and Delphi Processes. Scott Foresman; Glenview, IL: 1975.
23. Fink A, Kosecoff J, Chassin M, Brook RH. Consensus methods: characteristics and guidelines for use. Am J Public Health. 1984;74:979–83. [PMC free article] [PubMed]
24. Clark HH. Using Language. Cambridge University Press; New York: 1996.
25. The Joint Commission . Advancing Effective Communication, Cultural Competence, and Patient- and Family-centred Care: A Roadmap for Hospitals. The Joint Commission; Oakbrook Terrace, IL: 2010.
26. Byrt T. How good is that agreement? Epidemiology. 1996;7:561. [PubMed]
27. Gorter JW, Ketelaar M, Rosenbaum P, Helders PJ, Palisano R. Use of the GMFCS in infants with CP: the need for reclassification at age 2 years or older. Dev Med Child Neurol. 2009;51:46–52. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...