Abstract
Classification systems in health care are usually based on current understanding of the condition. They are often derived empirically and adopted applying sound principles of measurement science to assess whether they are reliable (consistent) and valid (true) for the purposes to which they are applied. In the past 15 years, the authors have developed and validated classification systems for specific aspects of everyday function in people with cerebral palsy—gross motor function, manual abilities, and communicative function. This article describes the approaches used to conceptualize each aspect of function, develop the tools, and assess their reliability and validity. We report on the utility of each system with respect to clinical applicability, use of these tools for research, and the uptake and impact that they have had around the world. We hope that readers will find these accounts interesting, relevant, and applicable to their daily work with children and youth with disabilities.
What Are Classification Systems, and Why Do We Need Them?
Classification systems used in health care are created systematically, for particular purposes, to address specific clinical or research questions. They are developed with a structure and “properties,” like reliability and validity, that indicate how well they “work.” Their primary purpose is to discriminate meaningfully among variations in the expression or stages of progression of a health condition. There are innumerable examples of classifications in biomedicine, notably in the field of cancer, where staging systems like the TNM (tumor, nodes, metastases) categorizations 1 or the classification of childhood lymphocytic leukemia into risk and management categories based primarily on clinical features 2 provide considerable information to guide management and prognostication. Newer systems that use basic biological markers (eg, Yeoh et al 3 ) may now replace some of these earlier classifications.
In the field of developmental disability, it is well known that children with the same health condition or medical diagnosis vary considerably in their functional abilities. We have traditionally applied unhelpful terms like mild, moderate, and severe to describe severity of cerebral palsy or intellectual disability, or high-functioning and low-functioning to categorize children with autism spectrum disorders. Among the big challenges with these terms are that (1) they are value-laden: mild can easily be taken to imply “it could be worse”—something parents may find insulting, upsetting, and unhelpful; (2) they are imprecise, insofar as they are rarely defined carefully and used systematically; and (3) there is no evidence that these words are used reliably (ie, consistently from time to time and person to person) or that they are valid (ie, meaningfully distinct from one another or associated with different outcomes).
This paper provides a broad background and overview of 3 purpose-built functional classification systems that describe variations that are thought to be meaningful in daily life regarding gross motor, manual, and communicative function in young people with cerebral palsy. All 3 systems move us well below the surface of diagnostic labels, which in themselves tell us virtually nothing about what a person can do, and at best indicate what aspects of function are probably affected. Each system was developed by a team led by experts in that aspect of function—and all have been created by colleagues—the authors of this paper and our respective research groups—who have shared ideas, methods, and experiences as we worked together. We intend to illustrate how these classification systems promote consistency in the language we use to talk about things, how they provide a means for communication between families and professionals and among professionals, and how the distinctions between classification levels allow clinicians and researchers in our field to assess whether something they are hearing or reading, for example, about a treatment or prognosis, is relevant to a specific child or their work. We will also show that these systems allow us to learn about aspects of people’s lives that are not related to “severity” as described by the functional level of that system. Finally, there is no assumption that this is a systematic or comprehensive review of the literature; rather, our goal is to report how these systems have been developed and to provide examples of published studies that illustrate their usefulness in the clinical and research worlds in which we work.
Common Aspects of These Cerebral Palsy Classification Systems
The functional classification systems described in this report share a number of common features. 4 The first important element of these systems is that each was built upon on a clear articulation of the specific area of function that was the focus of the system. For example, the original system (the Gross Motor Function Classification System [GMFCS] 5,6 ) explicitly looked only at “gross motor” function, this being by definition the hallmark impairment of cerebral palsy. 7 The Manual Ability Classification System (MACS) 8 was created by Swedish colleagues whose research interests and clinic work concern “manual function.” They had initially challenged the developers of the Gross Motor Function Classification System regarding its apparent disregard of “manual abilities,” whereas the Gross Motor Function Classification System had explicitly focused only on gross motor function in order not to confound one aspect of function with another. In creating the Communication Function Classification System (CFCS), 9 Hidecker et al wanted to categorize levels of communicative functioning beyond traditional approaches used by speech-language pathologists. Like the earlier classifications, the intent of the Communication Function Classification System was to describe people’s ability to communicate without regard to the way the function was achieved, and without the traditional emphasis on “normal” function.
A second common element among these systems, and others now under development, is their embrace of the concepts in the World Health Organization’s International Classification of Functioning, Disability and Health (the ICF), 10,11 whose approach includes a framework for considering any health issue across several interlinked components, of which the concepts of “activity” and “participation” (execution of a task or action by an individual in day-to-day life) were the guiding focus for these systems. The emphasis in all these classification systems was to capture descriptions of what children and youth do rather than to describe their difficulties or focus on how “normally” they do what they do. In other words, all these systems look at achievement rather than deficits.
Third, because the focus was on daily function, each classification describes “performance” (people’s usual activity) rather than “capacity” (what people can do at their best). All 3 systems share these basic philosophical and structural underpinnings.
Fourth, the 5 levels in each system are meant to be mutually exclusive and collectively exhaustive. They were derived empirically rather than statistically; there was never any assumption that the levels in any of the systems were equally distributed in the population of people with cerebral palsy (as quintiles), or that the “distance” between the levels was equal. Observant readers will note that the levels in all 3 systems, though “ordered,” are labeled with Roman numerals rather than Arabic numbers. This was designed to prevent people from assuming that levels 1 to 5 had some quantitative value associated with them, or that they could be averaged (which of course they cannot).
Finally, none of these systems is a test, an outcome measure, or a diagnostic tool—nor should they replace appropriate assessment instruments. Each simply describes “levels” of functioning in their respective domains.
What Tools Have Been Developed?
The Gross Motor Function Classification System 5,6
The Gross Motor Function Classification System 5,6 was the original system in this family of classifications. It was developed to address the researchers’ recognition that while terms like mild, moderate, and severe had some crude heuristic and discriminative value (see Scrutton and Rosenbaum 12 ) the uncertainty of what they meant and the absence of any evidence that they were used reliably meant that they could not be trusted.
There is no accepted objective biomedical way to measure day-to-day function in childhood disabilities. Furthermore, the use of developmental assessments with norms for children with typical development is inappropriate for special populations. 13 The impetus to create a gross motor function classification system arose in part from early research to validate the Gross Motor Function Measure (GMFM). 14 Plots of Gross Motor Function Measure scores against age produced crude curves that appeared to differ when children were grouped based on therapists’ judgment of severity (mild, moderate, severe). 12 Our impression was that a standardized and reliable classification system based on gross motor function would have wide applicability including creation of motor development curves. Therefore, the Gross Motor Function Classification System was developed empirically. We started with a review of literature, examination of scores of children on Gross Motor Function Measure items that represent common motor milestones, and discussion among our group. By analogy with how geneticists often work with families in exploring a possible syndrome, the idea was to use a pattern-recognition approach with “word pictures” that would make it possible for users to identify a descriptive level that fit that child’s current function.
The draft of the Gross Motor Function Classification System was further revised following nominal group process and Delphi survey consensus methods. 5 The participants in the Delphi survey were 20 physical therapists, occupational therapists, and developmental pediatricians from North America, Europe, and Australia. Participants were asked to use the Gross Motor Function Classification System, rate level of agreement with statements on content and construct validity, and provide open-ended comments. The consensus was that 5 levels best captured the differences in gross motor function among children with cerebral palsy. One obvious “source of variation”—children’s age—was recognized as a strong influence on gross motor functioning; we know that most children with cerebral palsy develop over time, regardless of interventions. The original Gross Motor Function Classification System had 4 age strata (under 2 years, 2 to <4 years, 4 to <6 years, and 6 to 12 years). Following validation, interrater reliability was established in a study involving 51 therapists and 75 children. 5
At the time the Gross Motor Function Classification System was first developed in the late 1990s, there was little systematic understanding of the functional aspects of the lives of adolescents with cerebral palsy, so the developers of the Gross Motor Function Classification System deliberately limited the age-band descriptions to age 12 years. When further clinical and research evidence became available, we used a Delphi consensus method similar to the one used to validate the Gross Motor Function Classification System to add descriptions of adolescent gross motor function and publish an “expanded and revised” version of the Gross Motor Function Classification System (GMFCS E&R) that includes a >12- to 18-year age band. 6
The following example illustrates how the Gross Motor Function Classification System is structured. These specific “word pictures” capture the range of functioning of children with cerebral palsy between the ages of 2 and 4: LEVEL I: Children floor sit with both hands free to manipulate objects. Movements in and out of floor sitting and standing are performed without adult assistance. Children walk as the preferred method of mobility without the need for any assistive mobility device. LEVEL II: Children floor sit but may have difficulty with balance when both hands are free to manipulate objects. Movements in and out of sitting are performed without adult assistance. Children pull to stand on a stable surface. Children crawl on hands and knees with a reciprocal pattern, cruise holding onto furniture and walk using an assistive mobility device as preferred methods of mobility. LEVEL III: Children maintain floor sitting often by “W-sitting” (sitting between flexed and internally rotated hips and knees) and may require adult assistance to assume sitting. Children creep on their stomach or crawl on hands and knees (often without reciprocal leg movements) as their primary methods of self-mobility. Children may pull to stand on a stable surface and cruise short distances. Children may walk short distances indoors using a hand-held mobility device (walker) and adult assistance for steering and turning. LEVEL IV: Children floor sit when placed, but are unable to maintain alignment and balance without use of their hands for support. Children frequently require adaptive equipment for sitting and standing. Self-mobility for short distances (within a room) is achieved through rolling, creeping on stomach, or crawling on hands and knees without reciprocal leg movement. LEVEL V: Physical impairments restrict voluntary control of movement and the ability to maintain antigravity head and trunk postures. All areas of motor function are limited. Functional limitations in sitting and standing are not fully compensated for through the use of adaptive equipment and assistive technology. At Level V, children have no means of independent movement and are transported. Some children achieve self-mobility using a powered wheelchair with extensive adaptations.
Once the system was shown to be both acceptable to clinicians and feasible to use in actual daily practice, its utility and validity were explored in a new way. Inter-rater reliability involving classification of written clinical reports was established to be very high. 15 The predictive validity of Gross Motor Function Classification System classification at younger ages on later functioning, based on reviews of clinical notes made at the younger ages, was reported in that same paper, with Wood and Rosenbaum 15 showing that after the age of 2 years a child’s Gross Motor Function Classification System level predicted walking function at age 12 with a positive predictive validity of 0.74 and a negative predictive validity of 0.90. In a subsequent prospective longitudinal study of “motor growth” in a randomly selected population of 657 children with cerebral palsy, the very strong discriminative and predictive validity of the Gross Motor Function Classification System was firmly established. 16 Children in the Rosenbaum et al study 16 were classified an average of 4.3 times over a 2- to 3-year period. Seventy-three percent of children remained in the same level for all ratings. 17 The weighted kappa coefficient between the first and last ratings was 0.84 for children younger than age 6 years and 0.89 for children at least 6 years old, indicating excellent chance-corrected agreement. Children initially classified in Levels I and V were least likely to be reclassified. These results provided evidence of the stability of the Gross Motor Function Classification System.
In summary, the Gross Motor Function Classification System was developed with front-line service providers. It was shown to be reliable in real-time application and as a means of categorizing gross motor function using chart-based descriptive accounts of previous motor function; to discriminate gross motor function meaningfully; and to predict later motor function with a high rate of accuracy. Furthermore, stability of classification over time has been well established. 15,17,18
The Manual Abilities Classification System 8
There is an important link between what can be measured and what is important to identify and highlight. The impact and uptake of the Gross Motor Function Classification System provided the stimulus for creating the Manual Ability Classification System. Starting with a vague idea in the shadow of gross motor function, hand (manual) function has gained great traction in the past few decades. It was in congruence with this recognition that the wish to create a classification similar to Gross Motor Function Classification System for hand function occurred. The assumption was that although there might be correlations between the 2 aspects of functioning, gross and fine motor ability would not necessarily follow each other directly, nor would one be a surrogate for the other.
The first step in developing the Manual Ability Classification System was to identify and articulate the functional concept to be classified; to clarify the conceptual background of these ideas; and to define and formulate the basic underlying construct. To address these questions the researchers assembled an expert group of 6 people—2 occupational therapists, 2 physical therapists, a hand surgeon, and a developmental pediatrician—each with over 20 years of clinical experience in pediatric rehabilitation and research in the field. The construct development of the classification was the most challenging process, and the research group worked together over several years, using workshops, telephone conferences, and email. We did an inventory of available tests and classifications of hand function but wanted to describe something beyond the available construct of “fine motor skills.”
An import milestone was reached when the Manual Ability Classification System construct—“children’s ability to handle objects in daily life”—was agreed upon. Thereafter, it became important to describe the content of the levels of the classification. For that purpose the researchers video-recorded children with different subtypes of cerebral palsy and areas of functioning performing various manual activities in their natural environments, mainly in their homes or schools. Using those videos, a discussion and definition of the classification levels progressed, based on the assumption that 5 levels would be appropriate, if possible, to correspond to the structure of Gross Motor Function Classification System. As an example, Manual Ability Classification System Level I is described as follows: “
After the first version of the classification was outlined an external validation process was initiated, involving both professionals in pediatric rehabilitation and parents of children with cerebral palsy. To collect opinions from professionals, the researchers presented preliminary versions of the classification at national and international conferences (Sweden 2002, Canada 2003, Norway 2003). Interactive poster presentations were used at which participants were encouraged to classify children from video clips and to comment on the classification. Comments and suggestions collected on these occasions were brought back to the expert group to be considered, and refinement of the wording and the distinctions continued over a long period. An additional step for validity testing was to investigate whether the concept of “manual ability” was meaningful for parents of children with cerebral palsy. 19 They were asked if this concept was useful for describing their child’s hand function, how they were thinking when deciding a particular level instead of others, and if they felt that there was an overall usefulness of the classification. The positive responses from families helped the research group to continue their work. An important comment from parents was that in using Manual Ability Classification System people were asking about the children’s ability, not their disability.
Since the initial development of the Manual Ability Classification System, further studies have investigated its validity, reliability, and stability. Further validation was performed by linking the Manual Ability Classification System to the International Classification of Functioning, Disability and Health–Child and Youth version, by which it was shown that the former system was related to 7 chapters in the latter. 20 Thus, the Manual Ability Classification System covers a broad perspective of handling objects in different contexts of daily life, supporting the evidence that it describes an ecological perspective of hand function.
The interrater reliability has been tested and found to be good to excellent between professionals and between professionals and parents for children between 4 and 18 years of age. 21 –24 Likewise, it seems that the Manual Ability Classification System is stable over time and that the classification has predictive value, 21,25 such that children rarely change Manual Ability Classification System levels after age 4 years.
It might seem surprising that children do not reach more functional Manual Ability Classification System levels as they get older. This does not mean that children fail to learn new manual skills with development or by intervention. Rather it means that the improvement happens mainly within the levels, but not to the extent that they typically change ability levels. In developing and using the Manual Ability Classification System, we expected that children’s abilities would be judged from an age-related perspective, meaning that they are expected to use their hands in different activities dependent on age. It is important to highlight that, as with the Gross Motor Function Classification System, the Manual Ability Classification System is a classification system with stability over time and cannot be used as an outcome measure.
Several studies have explored the relationship between the Manual Ability Classification System and the Gross Motor Function Classification System, showing that only about half of the group are classified at the same levels on the two systems. 25 –28 Work is needed for a version for younger children, and a field version of Mini-Manual Ability Classification System is available for which field trials are ongoing.
The Communication Function Classification System (CFCS) 9
The reported prevalence of communication problems in cerebral palsy has varied widely (eg, Odding et al 29 ). Part of this variance is due to a lack of consensus on the definitions of functional communication problems. 30 For example, is it a problem in using speech, using other modes of communication, having a hearing loss, understanding language, and/or expressing language? Another challenge has been how to compare the severity of communication problems when these problems have different underlying features. For example, is “severe dysarthria” similar to or different from “severe expressive language disorder”? Obtaining clinical agreement on “mild,” “moderate,” and “severe” communication disorder ratings was difficult.
On learning about the development of the Gross Motor Function Classification System and the Manual Ability Classification System, the question arose as to whether a similar development process could lead to a useful classification of communication performance. As reported by Hidecker et al, 9 multiple stakeholders participated in each development phase of what became the Communication Function Classification System. In addition to speech-language pathologists, adults with cerebral palsy, parents of children with cerebral palsy, occupational therapists, physical therapists, physicians, and educators provided important perspectives to the development of the Communication Function Classification System. These stakeholders used this multidisciplinary critique to increase its understandability. This resulted in a classification system that can be made by the person with cerebral palsy, the parent, and/or a professional who knows the person well.
Part of the key to the Communication Function Classification System development was to consider how to stage communication at an International Classification of Functioning, Disability and Health activity level, regardless of the underlying speech, language, and/or hearing anatomy and physiology. An important difference between communication and gross motor/hand function is that effective communication occurs between 2 or more individuals. These individuals work together to construct a shared understanding. This led to many hours of consideration on what the role of the communication partner should be in the development of the Communication Function Classification System level descriptions.
People with significant communication disabilities have noted that the types of people with whom they wish to communicate, and the speed of the communication, are important to participation, saying “full participation is unobtainable whenever communication with unfamiliar partners is restricted” (p. 200), and “a slow rate of participation can have a damaging effect on the interaction.” 31 Consequently, the Communication Function Classification System levels include a distinction between communicating with unfamiliar partners (those who are strangers or acquaintances) and familiar partners (those who know the person and can use their shared experiences when communicating).
Much discussion also occurred about what we meant by a “slow rate.” After many drafts, we settled on “a comfortable pace of communication.” By this, we mean that the conversation occurs with few communication breakdowns (which require time to recognize and repair) and little wait time between partners’ conversational turns. The person is perceived to be an effective sender and receiver when they can quickly and easily shift between transmitting and understanding messages. One must be able to be both a sender and a receiver, which is why the Communication Function Classification System requires a global judgment of both. Thus, Communication Function Classification System Level 1 is not “error-free” or “perfect” communication; rather, communication performance at Communication Function Classification System Level 1 (“Effective sender and receiver with unfamiliar and familiar partners”) involves being able to communicate with any partner in a reasonable amount of time.
Another important Communication Function Classification System development was an agreement that, given the focus on function, all effective communication methods should be considered in making the classification. These methods may include gestures, sign language, speech, and augmentative and alternative communication (AAC). (Augmentative and alternative communication refers to the use of unaided and aided systems to augment communication understanding and expression. See Beukelman and Mirenda 32 for more details.) To highlight the possible use of multiple methods of communication, we added a checklist of common methods of communication to the second page of the Communication Function Classification System.
Additional studies are currently underway to understand the reliability and stability of the Communication Function Classification System. The test–retest reliability has been excellent 9 ; interrater reliability among professionals has been good to excellent, 9,24,33 and reliability between professionals and parents has been moderate to excellent 9 (Geytenbeek, personal communication; Mutlu, personal communication). Some aspects of communication are known to develop well into adolescence, which may be important to the question of the stability of the Communication Function Classification System. We are just beginning to look at the Communication Function Classification System levels by age ranges and over time. Research is needed to assess whether the Communication Function Classification System will be useful as a predictor of later communication performance.
Research is also underway using the Communication Function Classification System to classify communication performance with persons who do not have cerebral palsy 34 (also Lenski, personal communication). Again, the purpose of this global judgment is to describe the person’s usual communication regardless of any underlying speech, language, and hearing difficulties. The preliminary results are promising, in that professionals are able to assign a Communication Function Classification System level regardless of the underlying reason for the communication disorders.
Systems Recently Published or Under Development
Work has recently been published by colleagues in the United Kingdom to develop an analogue system to describe feeding and eating function. 35 In Canada, an Autism Classification System of Functioning: Social Communication (ACSF:SC) is being developed and field tested. 36,37
How Are These Tools Used?
Clinical Uses
There are a number of clinical applications for all 3 classification systems. Each enables clinicians to communicate with families, and especially to respond to parents’ almost universal question “How bad is it?” Using the “word pictures,” parents can recognize their own child’s functioning. There is good evidence that parents find this approach acceptable and feasible, and that they can be reliable in their classification of their child’s functional status. 27,38,39 Experience suggests that the presentation of the materials to families should start with the description of the least-functional level at the top of the page with successive levels of better functioning presented as the reader looks down the list. 40 This enables parents of children with limited function to avoid having to read the several levels of function that their child cannot do, while parents of children with more abilities can see what their child has accomplished relative to other children within that age band (where these exist).
Clinically, each classification system describes functional status, which, among other advantages, may allow clinicians and service program managers to begin to explore management options by “level” (see, eg, Heinen et al 41 ). The Gross Motor Function Classification System has also been validated to predict future gross motor status, 16,42 again enabling clinicians to provide an evidence-based response to parents’ inevitable question “Will our child walk?” For Manual Ability Classification System, 78% of the children between ages 4 and 18 years stay in the same ability level, indicating that they do not lose function over time. Work is currently underway to assess the extent to which Communication Function Classification System predict later functional abilities.
The combined use of the Gross Motor Function Classification System and the motor development curves for children and youth with cerebral palsy 16,43 has raised questions regarding interpretation. Most important is the evidence that children with cerebral palsy are most likely to change within their Gross Motor Function Classification System level. In our opinion, it is encouraging to be able to see that a child is performing as expected or better than expected on the Gross Motor Function Measure 44 compared to children of the same age and Gross Motor Function Classification System level, and this kind of information can and should be communicated to families. In contrast, a goal of intervention should be to improve a child’s function rather than their Gross Motor Function Classification System level. As previously stated, the Gross Motor Function Classification System is not an outcome measure.
Use of the classification systems as part of the process of anticipatory guidance has not been examined. Anticipatory guidance is a collaborative process in which children, families, and professionals discuss future aspirations and plan ahead. For example, for a 14-year-old with cerebral palsy who excels academically, what are the considerations for mobility, manual ability, and communication function for attending university?
Administrative Purposes
The Gross Motor Function Classification System has been used to describe populations of people with cerebral palsy, as illustrated by a report from Northern Ireland. 45 The Manual Ability Classification System is included in several registers, for example, the Victorian Cerebral Palsy Register in Australia and the Cerebral Palsy Follow-Up Programme, a population-based National Health Care Quality Register in the Nordic countries. 46 The Communication Function Classification System is also beginning to be included in several registers 33 (Andersen, personal communication). Systematic data from such overviews of the (relatively stable) functional status of the population of people with a lifelong condition like cerebral palsy can enable policy makers and planners to anticipate future community needs, for example for accessible assisted living facilities, and to plan for the human, fiscal, and technical resources needed to make such services available.
Clinical and Epidemiological Research
Researchers have used the Gross Motor Function Classification System and other systems to study the relationships among markers of functional status, as described by these systems, and other dimensions of the lives of people with cerebral palsy. Thus, for example, we have reported the wide range of correlations (from substantial to trivial) between Gross Motor Function Classification System levels and several other aspects of the health status of preschool children 47 and of older children and adolescents with cerebral palsy. 48 A study with 657 children and youth with cerebral palsy looked at the correlations between Gross Motor Function Classification System levels and traditional classifications of cerebral palsy by motor type and by topographical distribution. 49 These relationships were statistically significant (the power of the analyses being very substantial, given the sample size) but the explanatory value of these relationships was weak to modest at best. This study illustrated, among other points, that while the classical clinical descriptions of cerebral palsy may have clinical usefulness, they relate poorly to the functional realities of people with this condition. 49 Thus, to describe the functional capacity of someone with cerebral palsy, the Gross Motor Function Classification System should be used as it provides a picture that the topographic and motor impairment descriptions cannot capture.
Rosenbaum et al 50 used the Gross Motor Function Classification System to illustrate how different approaches to describing “quality of life” and “health-related quality of life” are, and are not, associated with functional abilities and limitations, and how these approaches to “quality of life” are only very weakly related to one another. Using these 3 reliable and valid classification systems together, Hidecker et al 51 have begun to explore how gross and fine motor function, and communication function, are and are not related statistically to each other. Harvey et al 52 used the 3 classifications to consider whether gestational age had a role in later functional performance in 109 children with cerebral palsy and periventricular white matter injury. Raghavendra et al 53 used the 3 classifications along with other measures to understand intervention outcomes in social participation.
The Manual Ability Classification System has been used to describe manual ability in different groups and populations of children with cerebral palsy. It has been shown to be an important complement to the diagnosis. For example, it is now known that most children with unilateral cerebral palsy function in Levels I and II whereas some have more limitations and function at Level III. Children with dyskinetic cerebral palsy more commonly function in the lower levels whereas children with bilateral cerebral palsy may function across all 5 levels, 54 sometimes with similar Gross Motor Function Classification System levels but at other times with rather different levels in the 2 systems.
The Manual Ability Classification System also seems somewhat predictive for the development of self-care. Only children at Levels I and II seem to come close to independence in self-care, though interestingly even children at Manual Ability Classification System Level 1 do not necessarily become independent until later ages. Knowledge of a child’s Manual Ability Classification System and Gross Motor Function Classification System levels can be useful when discussing expectations of, and goals for, the development of functional skills.
In clinical research, these classification systems are being used to stratify people prospectively for clinical trials and prospective population-based studies (eg, Boyd et al, 55 Law et al 56 ) and have also been used after the fact to report findings by levels (eg, McLaughlin et al 57 ). Soo et al 58 and Hagglund et al. 59 have reported the very significant inverse relationship between Gross Motor Function Classification System levels and a person’s risk for hip subluxation or dislocation, making our approach to hip surveillance far more evidence based than was traditionally possible.
What Has Been the Impact of These Systems?
In the academic world, one measure of impact is the frequency with which others cite peer-reviewed papers. To date, the papers reporting the Gross Motor Function Classification System 5,6 have been cited in the literature more than 1650 times; the Manual Ability Classification System 8 285 times; and the Communication Function Classification System 9 15 times in less than 2 years. We can anticipate that journal reviewers and editors will increasingly expect authors of clinical and research papers about people with cerebral palsy to use these systems to describe their populations.
Are They Being Used?
Below are some highlights of the evidence of utility: Citations and downloads provide some perspectives (see Gray et al
60
). Uptake by parents and families has been exciting. See, for example, the Cerebral Palsy Alliance webpage apparently aimed at parents and individuals with cerebral palsy https://www.cerebralpalsy.org.au/what-is-cerebral-palsy/types-of-cerebral-palsy/thatsummarizesthe3classifications; the website “My Child Without Limits” describes the Gross Motor Function Classification System, http://www.mychildwithoutlimits.org/understand/cerebral-palsy/cerebral-palsy-diagnosis-and-classification/functional-classification/; and the Gross Motor Function Classification System is listed on DisabilityMeasures.org http://www.disabilitymeasures.org/gmfcs/. The “Google-ability” of these systems means that people can become aware of and familiar with them. Completed research and work-in-progress, such as the On Track Study (http://canchild.ca/en/ourresearch/on_track_study.asp), is exploring, among other factors, the relationships among gross motor function, manual ability, and communication function. At the preservice level of education, students in professions working with cerebral palsy are being taught the International Classification of Functioning, Disability and Health model as well as its classifications at different levels. Learning to use the Gross Motor Function Classification System, Manual Ability Classification System, and Communication Function Classification System will support their interprofessional education as multiple professionals, individuals with cerebral palsy, and their families describe everyday performance.
What Do These Classification Systems Not Do?
It has been said that any good idea contains within it the seeds of perversity. The developers of the Gross Motor Function Classification System have at times been challenged—mostly by fellow professionals rather than by parents—about several issues. One concern is about the potential for parents who learn about the relatively limited functional prognosis for their child to simply “give up hope.” These sceptics believe that using these systems to prognosticate may be harmful. People have raised the same concern about the possibility that insurance providers will discontinue coverage for a child who is predicted to make slow progress, especially when people learn about the relative stability of these classification systems over the course of the developing years.
One answer, based on sound ethical and clinical principles, is to wait to prognosticate until parents ask! It is almost always the case that parents know that these “issues” loom in the background; some choose to anticipate the future and ask early in their journey into parenting a child with cerebral palsy, whereas others take a longer time to come to this point, or indeed never ask directly. Another aspect of the answer is that when parents know the probable direction of their child’s development (at least on that element of functioning), they can make informed decisions about their child’s therapies, for example, for accessible housing and vehicles, recreational programs, and so on. In addition, “therapies” can and should address more than the acquisition of basic skills. There is a host of opportunities to explore access to improved equipment and strategies to enable people to be as functional as possible in the face of “natural” limitations. Taking our cue from families and children, we can work with them on skill development, fitness, social function, and “participation.” The list of relevant directions for intervention is long and exciting, 11 and in our view should never be based solely on the current level of functioning in any of these classification systems.
Occasionally one reads about the naïve effort to report a “mean” Gross Motor Function Classification System level in a population being described in a research report. This is of course entirely meaningless. As noted earlier, the levels in all 3 systems, though “ordered,” are labeled with Roman numerals rather than Arabic numbers. There is no intent either that the levels are equally distributed in the population of people with cerebral palsy, or that the “distance” between the levels is equal. It may make sense to report median numbers of people in a cohort (as one would for other discrete variables in a population), but numerical averages are meaningless and in fact distracting.
These classifications are not meant to replace standard clinical assessments or be used to assess change. The challenge is to help professionals understand that these systems are based on a global judgment of the usual, daily performance of an individual, and that they describe overall categories of functioning.
Summary
We hope that readers will recognize that in the development and validation of these and other classification systems we have applied sound research methods to create systems that address clinical challenges, and that their acceptability to families as well as clinical and research colleagues suggests that they are meeting a need. Given the 21st-century expansion of our purview beyond a traditional biomedical focus to include descriptions of everyday functioning, we believe that these systems—and hopefully other well-constructed classifications—will enrich the field of developmental disability and perhaps other areas of clinical practice as well.
These systems can be found and downloaded for free at:
GMFCS: http://motorgrowth.canchild.ca/en/GMFCS/resources/GMFCS-ER.pdf
MACS: www.macs.nu
CFCS: Communication Function Classification System: http://cfcs.us
Key Take Home Messages
Simple clinical classification systems are available to describe gross motor, manual ability, and communication function of young people with cerebral palsy.
These systems are complementary to traditional biomedical descriptions of disorders and disabilities.
The systems are free, easily accessible, usable by, and acceptable to parents.
Footnotes
Author Contributions
PR took the lead in drafting this paper, to which A-CE, MJCH, and RJP all contributed actively. The authors share the responsibility for any errors or misinterpretations herein.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.