Measuring participation in childhood disability: how does the capability approach improve our understanding?

In recent years, regular readers of this journal have become familiar with the International Classification of Functioning, Disability and Health (ICF).1,2 There has been a consensual acceptance amongst many childhood disability researchers that maximizing children’s effective ‘participation’ is the overarching goal in providing services for disabled children once their survival is secured and the they are protected from harm. I am grateful to an anonymous peer reviewer of a grant application for encouraging me to consider the concept of ‘capability’. First devised by Nobel Prize winner Amartya Sen for development studies, and further developed by Martha Nussbaum for assessing sex inequalities, the capability approach provides an alternative framework to the ICF for examining inequalities in well-being and social arrangements.3,4 In this paper I suggest that, when measuring participation, incorporating a ‘capability’ perspective may be more enlightening than measuring solely ‘capacity’ and ‘performance’ as proposed in the ICF. Participation is defined in the ICF as ‘involvement in life situations’ and, for children, includes domains of learning and applying knowledge, communication, home life, school life, social life, relationships, and leisure and recreation. The ICF offers two qualifiers for activities and participation: these are ‘capacity’ which indicates the highest probable level of functioning (what a child can do in an ideal environment) and ‘performance’ (what a child actually does in the environment in which they live). Inconsistencies between children’s capacity and performance are presumed to be due to environmental or personal ‘contextual’ factors. Hence, differences in aspects of children’s participation have been attributed to movement, manual and intellectual abilities,5 and also to the environment in which they live.6 Environmental factors are classified under the following categories (1) products and technology, (2) natural environment and human-made changes to environment (3) support and relationships, (4) attitudes, and (5) services, systems, and policies. Personal factors are not classified in the ICF, and, as such, have received less attention, but include sex, race, age, lifestyle, habits, coping styles, and other such factors. It has been argued that ‘choice’ is a personal factor in the ICF;7 however, this is an implicit assumption. The ICF definition of pertinent personal factors describes traits that affect decision-making, but ‘choice’ is not stated explicitly. The capability approach considers the range of experiences and life situations as ‘possible functionings’, described by Sen as ‘beings’ and ‘doings’.3 Participation, as defined in the ICF, is consistent with the ‘doings’, whilst ‘beings’, such as being happy, are more consistent with subjective ‘quality of life’. In the capability approach, ‘achieved functioning’ (which is analogous to performance in the ICF) is moderated by a person’s capacity, opportunity, and choice. Capability in this context describes a person’s capacity and the opportunities typically afforded them by environmental factors; Sen describes ‘conversion factors’ for available resources as personal factors, and social and physical environmental factors. The range of ‘possible functionings’ for disabled children, their ‘capability set’, may be restricted due to their capacity (by virtue of impairments), or be limited by their social and physical environment. According to Sen’s paradigm, what is important is that children ‘have the capability to participate, or the real, effective opportunities and freedom to participate’.8 Then, whether functioning is achieved is a matter of personal choice to participate in specific life situations. Lists of key ‘capabilities’ have been proposed; however, the need for a definitive list is debated. Sen proposes that a democratic process should determine which capabilities should be included in specific circumstances.9 My thoughts on this topic have been informed by the small body of literature that has considered the issues of applying the capability approach to disability and health research. Readers interested in learning more about this approach and its relationship to other models of disability will find the papers by Robeyns, Mitra, and Terzi instructive.8–10 The distinctions among ‘capacity’, ‘capability’, and ‘performance’ (achieved functioning) may not be immediately apparent and Figure 1 seeks to make the distinction clearer. Assessing a child’s capacity to perform in an ideal environment is consistent with taking a clinical perspective. Think about how we evaluate walking in the gait laboratory: in effect we are asking the child ‘Can you, and how do you, walk in ideal circumstances?’ In contrast, the construct of capability represents the child’s predicament in the life they lead, taking into account their capacity and available resources, social and physical environment. In order to evaluate capability, one asks the question ‘Could you walk at home, school and around your neighbourhood if you want to?’ If assessing performance, one asks the question ‘Do you walk at home, school and around your neighbourhood?’ As is evident, a child with a movement disorder may be able to walk in ideal circumstances (have the capacity), and perhaps could walk in certain environments if they wanted (have the capability), but might choose not to walk, even though they can, and for instance favour aided or wheeled mobility at school. The relationship between capacity (in a standardized environment), and capability (range of possible functionings) and performance/achieved functioning in the lived environment, modified after Robeyns.9 To consider the issue further, let us take an example from an item in the Assessment of Life Habits (LIFE-H) questionnaire: using a computer.11 Under the ICF, the distinction is usually between capacity and performance. Does the child have the cognitive and physical capacity? If so, how often do they perform the activity? If the child has the capacity, environmental factors such as the presence of a computer adapted for the child to use (products and technology) and their parents’ permission to use the computer (attitudes) may make it possible for them to perform the activity (be capable). However, let us suppose that they prefer not to use the computer and thus consequently rarely achieve performance in this life situation. In this case the distinction between capacity and performance may be assumed to be due to some environmental factor when in fact it is simply a matter of the child’s choice, unless the child’s preferences are specifically recorded as a personal mediating factor. Considering the same example using the capability approach, we would first ask whether this is within their ‘capability set’ of ‘possible functionings’. If the child does not have the cognitive ability to use the computer then their capacity and capability are the same. However, a child might have the capacity but not the capability. For instance, they may have the cognitive capacity but there is no computer at home, or the computer is not adapted to be made useable by the child, or perhaps a parent does not permit the child to use the computer. In these circumstances the child’s ‘freedom to choose’ whether to make this an achieved functioning is curtailed by the lack of opportunity imposed by inadequate resources or the social context of their family, hence they do not have the capability (despite having the capacity). If resources are available and the family context is enabling then the activity is within their capability set and they have the choice to make this an achieved functioning. Take two other examples, this time from the Children’s Assessment of Participation and Enjoyment (CAPE).12 First, doing water sports; a child may have the capacity to take part in water sports but if the swimming pool is not accessible for them, then they are not capable. Similarly, a child may have the capacity for playing a musical instrument but to be capable they require an appropriate instrument and permission and a place to play. To be clear, the distinction is between capacity as a ‘person at best’ construct and capability as a concept that combines the person, resources and context. Capability in this context is not captured directly in existing instruments measuring children’s participation. If we accept that this is an important concept to measure then methods need be developed to assess capability alongside assessments of capacity and performance. The distinction between the ICF and capability approach is subtle, and in terms of the constructs of capacity and capability is somewhat semantic. However, in many respects the ICF and capability approach are complementary. The capability approach does not provide a definitive list of important ‘capabilities’. Sen recommends that situation or application dependent lists can be produced in a democratic process with input from those whose capabilities are being assessed. The range of life situations classified in the ICF provides a useful starting list for assessing children’s participation; however, the ICF is a classification and not a scaled measure. The capability approach provides a framework for scaling the measurement of participation by asking whether a child can achieve in a life situation if they want to do so. As the answer will vary by circumstances the response categories are likely to be the frequency that the child has the opportunity ranging incrementally from ‘I am never able to do this’ to ‘I can always do this if I want’. Maximizing children’s effective participation may be the overarching goal of health and social services but this has to be consistent with the child’s choices about how they want to live their life. It follows that children should also be consulted on their own treatment goals, as can be done, for example, with the Canadian Occupational Performance Measure (COPM).13 There is preliminary evidence that paediatric rehabilitation based on goals set by the child and family are more effective than conventional intervention;14 and a trial in Canada is further evaluating the approach.15 Therefore, what services should aim to do is equalize children’s capability sets by removing restrictions to enable participation, thereby giving children a degree of choice and autonomy. If we are evaluating services, assessing increases in children’s capability becomes a key outcome even when performance is not achieved. In certain circumstances, intervention in the psychosocial processes by which children make their decisions might be the most likely way to increase children’s participation. Measuring children’s capacity and performance without assessing capability misses the opportunity to include the fundamental rights of individuals, in our case children, to choose the life they lead. It may also lead to misleading conclusions about the effect of environmental factors. I am grateful to the grant reviewer for making me think about these issues, to Lorella Terzi and Proochista Ariana for helping me grasp the principles of capability, and to Peter Rosenbaum for commenting on a previous version of the manuscript.