Abstract
Introduction
Evaluation of hand function has traditionally focused on objective measurements involving strength and range of movement measurements. The World Health Organisation International Classification of Disability, Functioning and Health (ICF) framework has introduced a more patient-orientated approach to assessment, incorporating four distinct areas of body structures, body functions, environment and activity limitation. These can be evaluated with both questionnaires and functional tests. Whereas there seems to be agreement over adult hand assessment, children’s hand assessment leaves involved therapists and clinician often wondering, which test to perform. The objective of this review was to evaluate the currently available childrens' hand and upper limb assessment tools within the framework of the ICF.
Method
A review of 19 of the most commonly used paediatric hand and upper limb assessment tools in the context of the components of the ICF. The International Classification of Disability, Functioning and Health linking rules composed by Cieza et al. were used in order to map each of the meaningful concepts in these assessment tools to the domains of the ICF.
Results
The activities domain of the ICF was found to be well represented in all assessment tools. Four out of 19 assessment tools had items that mapped to the environment domain. The highest number of different concepts mapped was found in musculoskeletal questionnaires. The lowest number of concepts was found in the pegboard tests.
Conclusion
Tests are often combined to attempt to cover as many aspects of the ICF yet our findings show that this may not always be achieved. Clinicians and therapists must be aware that in order to get a more complete overview of a child’s hand and upper limb functioning sometimes time- and resource-consuming tests have to be combined.
Introduction
A valid and reliable assessment of hand and upper limb function is essential to identify a baseline level of hand function. This enables clinicians to evaluate the effectiveness of medical or surgical interventions by comparisons with pre-intervention data and to monitor the progression or regression of ability over time. 1 The need for evidence-based intervention has been driven by the introduction of the internal market in health care provision, which has pushed health care providers to evaluate outcomes and establish cost-effective and efficient treatments. 2 Outcome measures or assessment tools also provide a common research language in order to be able to compare outcomes between patients and between centers. 3
Studies have reported that children spend more than one-third of a typical school day on hand skill–related activities.4,5 Achieving optimal hand function in the paediatric population is a priority. 6 In children hand and upper limb outcome assessment is made even more difficult by the need to incorporate normal development into the evaluation. Over the last few years there has been a move away from only measuring range of movement and strength towards evaluating in a developmental, patient-centred manner in order to capture the quality of the observed task. 7 There are multiple assessment tools available for hand function assessment, all either being disease-specific, region-specific or both.
The World Health Organisation International Classification of Disability, Functioning and Health (WHO ICF) framework was developed and endorsed in May 2001 and provides an international standard to describe and measure health and disability.8,9 It provides a holistic approach to patient health assessment and gives equal importance to each of its components: body structures and functions, participation, activities and contextual factors – environment (Figure 1). The body structures (s) are anatomical parts of the body such as organs, limb and their components. The body functions (f) refers to the physiological function of body systems. Activity (d) is defined as the execution of a task or action by an individual and participation (d) is the involvement in a life situation. Although both are denoted by the same letter, they are easily divided into distinct categories within the ICF.
50
The environmental and contextual factors (e) make up the physical social and attitudinal environment in which people live and conduct their lives. The International Classification of Disability, Functioning and Health, Children and Youth version (ICF-CY) was launched in 2007 and identifies that the nature and form that function takes in children differs from those in adults.
10
Domains of the World Health Organisation International Classification of Disability, Functioning and Health (WHO ICF)
11
.
The aim of this study was to evaluate all validated paediatric hand and upper limb assessment tools within the framework of the WHO ICF, in order to enable clinicians and allied health professionals to make informed decisions about the composition of outcome measures and to highlight areas that may not be represented when completing the assessments. We also compared cost of the outcome measures and the time taken for completion for both the clinician and the patient or parent. Outcome measures or assessment tools were divided into two main groups: questionnaires and performance based tests.
Methods
A comprehensive literature search of available databases was conducted using Medline/Pubmed (1950 to January 2014), CINAHL (1982 to January 2014), Embase (1980 to January 2014). The keywords used were as follows: child, children, paediatric and pediatric combined with upper limb, upper extremity, hand, arm. This was then also combined with test, questionnaire and assessment. All results were screened by the first author and by screening the text of the abstract. Articles that were not in English, did not have an abstract or did not have published validation data for children were excluded. Validation data were defined as psychometric data that included construct validity and a measure of intra- or inter-rater reliability. If validation data were not at first apparent in the abstract, the full text was then screened. The abstracts and references were analysed to produce a list of hand and upper limb assessment tools validated for use in children. Assessment tools that involved the use of a prosthesis were included if they could be used for non-prosthesis users also. The final subset of outcome measures for evaluation included seven questionnaires and 12 performance-based tests.
Brief description of each of the selected questionnaires and performance based tests, time taken to administer associated cost, and number of derived meaningful concepts.
PODCI: POSNA paediatric musculoskeletal functional health questionnaire; DCDQ: Developmental Co-ordination Disorder Questionnaire; CHAQ: Childhood Health Assessment Questionnaire; PUFI: Prosthetic Upper Extremity Functional Index; CHEQ: Children’s Hand use Experience Questionnaire ABILI-K: Abilihand kids; CAPP-FSI: Child Amputee Prosthetics- Functional Status Instruments; AHA: Assisting Hand Assessment; AMPS: Assessment of Motor and Process Skills; M-ABC: Movement ABC; MA: Melbourne assessment of Unilateral Upper Limb; QUEST: Quality of Upper Extremity Skills Test; JTHT: Jebsen-Taylor Hand Test; FDT: Functional Dexterity Test; UBET: Unilateral Below Elbow Test; NH-PT: Nine Hole Peg Test; SHUEE: Shriners Hospital for Children Upper Extremity Evaluation; ACHS: Assessment of Children’s Hand Skills Test CHSQ: Childrens Hand Skills ability Questionnaire; BOT-2:Bruininks-Oseretsky Test of Motor Proficiency 2; DCD: Development Co-ordination Disorder; JIA: Juvenile Idiopathic Arthritis; CP: Cerebral Palsy; OBPP: Obstetric Brachial Plexus Palsy; General: general development.
Each assessment tool was analysed by the first and second author both of whom had studied the ICF framework to improve familiarity. Consensus between the two assessors was used to decide which meaningful concept was identified in all items within the questionnaire or performance-based test, and then which ICF categories should be linked to each concept. Where a consensus could not be achieved between the two assessors a third person (senior author) also familiar with the ICF was consulted who made a final decision on the linking.
ICF linking
Each question or task that contributed to the overall composite score was evaluated for a meaningful concept which was then mapped to a domain within the ICF, as per the linking rules generated by Cieza et al.19,20 described below. The term “meaningful concept” describes the purpose of either the question or the task that is put forward to the patient. Within each of the four domains of the ICF there are a series of steps that lead on to adding further letters and numbers, which leads to a final code for a particular item. An example of this is asking if a child can throw a ball, would come under d for activities and participation leading to level d445 hand and arm use and from there the final level of d4454 throwing.
Linking to the ICF requires a careful consideration of the question in terms of the concept that it represents on the ICF. There is an important difference in observed activity from tests and asking a similar question in a patient/parent report questionnaire format. For example picking up beads under direct observation would be more likely to assess grip function or a specific movement of the hand resulting in a bodily function code b7101 mobility of several joints, but asked in a questionnaire would represent picking up objects d4400. Equally, it is possible for one question or task to represent two or more concepts on the ICF, and conversely there are items that may not be represented on the ICF at all. Where an item or task did not have a specific 4-numbered code on the ICF we used the “other” function e.g. D4458 hand and arm use, other, specified.
Results
The search of databases generated 239 articles in PUBMED/Medline, 88 in EMBASE and 25 in CINAHL. Following screening for eligibility a total of 19 assessment tools were identified.
Evaluated questionnaires
The POSNA paediatric musculoskeletal functional health questionnaire (PODCI) 21 is an outcomes assessment with questions orientated around the musculoskeletal health of the child. Questions reflect upper and lower limb functions, with subdivisions of activities of daily living, satisfaction and participation. There are 73 individual questions and two age groups (2–10 years and 11–18 years). The questionnaires for children aged 2–10 years were parent reports whereas the older children (>10 years) had a parent report as well as a self-report to complete. The average time to administer was 10–18 minutes and it is freely available to download.
The Childhood Health Assessment Questionnaire (CHAQ) 22 has been developed to assess health status in children with juvenile idiopathic arthritis. It forms a series of questions reflecting activities of daily living within 8 subscales; dressing, grooming, arising, eating, walking, reach, grip and various activities. It is valid for use in children aged 1–19 years and is completed by parents. Average time to complete is 10 minutes with an added 10 minutes for scoring. It has no cost associated with its use.
The Prosthetic Upper Extremity Functional Index (PUFI) 23 looks at the extent to which a prosthesis is used in daily activities. It is valid for use in children aged 3–18 years and can be used for congenital or acquired conditions that result in using a prosthesis. Each of the 50 activity questions is duplicated to identify improvements with and without prosthesis. The average time to complete is 20–30 minutes and it is free at point of use.
The Children’s Hand use Experience Questionnaire (CHEQ) 24 is an online 25 questionnaire developed for participants with decreased function in one hand, e.g. obstetric brachial plexus, cerebral palsy etc. It can be completed by a parent or self-reported and is for age range 6–18 years. There are 31 bimanual activity-based questions and the results are scored and emailed to participants at the end. The CHEQ does not provide an average completion time and there is no cost associated with its use.
The Developmental Co-ordination Disorder Questionnaire (DCDQ) 26 is a parent-report questionnaire validated for screening of coordination disorders in children, aged 5 to 15 years. Half of the 15 items are worded negatively, the other half positively. It takes 10–15 minutes to complete and there is no cost associated with its use.
The Abilihand kids 27 is a questionnaire validated for children with cerebral palsy aged 6–15 years with 22 activity-related questions that represent bimanual activities only. It is a parent report that takes and average of 10 minutes to complete and is free to use.
The Child Amputee Prosthetics – Functional Status Instruments (CAPP-FSI) 28 was developed for children with congenital or acquired upper and lower limb deficiencies. It can be used in children aged 1–17 years and is a parent-reported questionnaire. There are three age groups and 34 questions on upper limb per questionnaire. There was no described average time for completion and it was unclear as to whether there was a cost associated with use.
Evaluated performance-based tests
Assisting Hand Assessment (AHA) 29 is a measure of bimanual performance validated for use in children aged 18 months to 5 years, with cerebral palsy or obstetric brachial plexus palsy. Patients are observed undertaking a series of bimanual tasks in a structured play session that is video-recorded. The test takes an average of 10–15 minutes to complete with a further 15–30 minutes to score. The combined cost training and buying the approved kit was >€1000.
The AMPS (Assessment of Motor and Process Skills) 30 is an observational, general health assessment that is used to evaluate performance of activities of daily living. There are more than 100 standardised ADL tasks for children aged 2–18 years to choose 16 motor and 20 process skill items from. It takes the form of observation of play over a 30- to 40-minute period. The combined cost of the training course and the manual was €827.
The Movement – ABC 31 is validated for the developmental co-ordination disorder group and was developed to identify mild to moderate motor impairment in children. It measures manual dexterity, ball skills, static balance and dynamic balance. There are four different age groups within 4--12 years with a combined total of 32 observed tasks. The total time taken for completion was on average 20–40 minutes with an additional 10 minutes for scoring. The cost to acquire the test was >€1000.
The Melbourne Assessment of Unilateral Upper Limb (MA) 32 measures quality of unilateral upper limb movement in children with neurological impairment, e.g. cerebral palsy. It is validated for ages 5–15 years and assesses one arm at a time. There are 16 items involving reach, grasp, release and manipulation that are video-taped and subsequently scored. The average time taken to complete was 20–30 minutes with a 30-minute scoring period after. The cost of the test was >€500.
The Quality of Upper Extremity Skills Test (QUEST) 33 is an outcome measure designed to assess movement patterns and hand function in children with cerebral palsy. It can be used in children aged from 18 months to 8 years; 36 items assess domains that include dissociated movement, grasp, weight-bearing and protective extension. It takes the form of observed non-standardised play activity within a natural environment for the child. The average time to complete was 15 minutes with 15–30 minutes for scoring afterwards. The manual was free and kit was described for self-assembly.
The Jebsen-Taylor (JT HT) 34 Hand Function Test is not disease-specific and is validated for use in children aged 5–19 years. It takes the form of seven manual everyday timed activities and is scored based on the time taken to complete the tasks. The test takes on average 10–15 minutes to complete and costs €270 to acquire.
The Functional Dexterity Test (FDT) 35 is an observed and timed pegboard test assessing manual dexterity. It is validated for use in children aged 3–17 years and is not disease-specific. It is composed of 16 pegs hence 16 observed tasks. As a timed test the length of time taken to complete is user-dependent. The cost of the test is €93.
The Unilateral Below Elbow Test (UBET) 36 is an observational tool for children with upper limb prostheses. It is validated for use in children with transverse reduction deficiency and can be used by prosthesis and non-prosthesis wearers. It is designed for a population aged 2 to 21 years (in four age groups) and involves nine different tasks, which differ between the four age groups, hence a combined total of 36 tasks. The test takes an average of 20 minutes to complete and is again for self assembly, hence no direct cost associated with acquiring it.
The Nine Hole Peg Test (NH-PT) 37 is an observed, timed test of fine motor dexterity completed by non-disease specific patients. It is validated for children aged 4–19 years. It is a timed pegboard test with nine pegs, hence nine tasks. Scores can be taken for how long it takes to move all pegs or how many pegs can be moved within a 50- to 100-second time slot. The test costs €125 to acquire with no ongoing cost.
The Shriners Hospital for Children Upper Extremity Evaluation (SHUEE) 38 is a video-based assessment validated for use in children with hemiplegic cerebral palsy. It assesses spontaneous functional movement as well as dynamic positional assessments, grasp and release analyses. It is validated for children aged 3–18 years with a total of 22 questions/observed tasks. It takes 15 minutes to complete with an additional 15–30 minutes to score afterwards. The test can be assembled from a department's own items and hence there is no apparent cost associated with acquisition.
The Assessment of Children’s Hand Skills Test (ACHS) 39 is a test designed to measure a comprehensive range of hand skills in children aged 2–12 years. It is an observational assessment with parent report questionnaire (CHSQ) 40 on a range of 42 tasks, where only 2–3 tasks are chosen for the observational evaluation. We have included mapping of all of the tasks/areas described in the test even though only 2–3 are used as it is not possible to say which tasks would be selected at each sitting. The validation process involved assessment of participants with and without generic upper limb disability. The performance-based test component takes 10 minutes to complete and there is no cost associated with use or acquisition.
The Bruininks-Oseretsky Test of Motor Proficiency 2 (BOT2) 41 measures gross and fine motor skills through a series of 51 tasks in children aged 4–14 years. It is not specific to the upper limb and is developed for use in development co-ordination disorder covering the domains of fine manual control, manual co-ordination, body co-ordination, strength and agility and total motor composite. This test takes 45–60 minutes to complete. The cost of the test is €618 with a requirement to watch a training video.
Linking to the ICF
Each of the questions/tasks was mapped to a domain and concept within the WHO ICF. The number of questions and corresponding number of meaningful concepts are shown in Figure 2 for the questionnaires and Figure 3 for the performance-based tests.
Bar chart showing the number of questions and the number of meaningful concepts generated from each of the questionnaires. Number of tasks and number of meaningful concepts generated from each performance based test.

The PODCI questionnaire showed the lowest proportion (60%) of meaningful concepts per questions asked. This was due to a comparatively larger proportion of questions not linking to the ICF. The PUFI questionnaire showed the highest proportion of meaningful concepts generated as the result of questions representing more than one concept. In the case of the CAPP-FSI and the PUFI this was due to the follow-up questions regarding the use of prosthesis following each standard question.
On analysis of the functional tests we found that the pegboard tests mapped to the highest proportion on concepts per task. Each peg manoeuvre itself represented a separate task and each task had a number of components. For example: in the nine-hole peg test we concluded that three concepts were being requested: b7602 co-ordination of voluntary movement, d4400 picking up and d4403 releasing. Conversely, if it was felt that the test itself represented one task the proportion of meaningful concepts would still remain unchanged, i.e. one task, three meaningful concepts generated.
The AHA test formed the lowest number of generated meaningful concepts when linked to the ICF with 22 questions deriving only 17 meaningful concepts. Follow-up scores for each item in the UBET questionnaire was based on stability and co-ordination, hence generating a second concept of b760 control of voluntary movement function for each question asked.
The analysis involved the actual number of concepts mapped to the four domains, as it is less useful to have a single concept repeatedly mapped to the same domain. Figure 4 shows the percentage of concepts mapped to the various domains of the ICF. The activities domain is well represented since most questionnaires are developed to ascertain activities that a child can or cannot complete in daily life. It is therefore to be expected that not only is this area well-represented but it also duplicates many of the concepts in order to ascertain whether a number of essential tasks can be successfully completed. The CHEQ, for example, predominantly asks questions regarding activities as is its primary objective.
The only questionnaire to show representation of all four domains of the ICF-CY was the PODCI, although only 2% of responses were representative of the environments domain (Figure 4). The participation domain is an area of the ICF, along with environment, that is most often underrepresented in these tests. It has been possible to make a clear distinction between activities and participation within their “d” category within the ICF. 50 Participation is considered to be the involvement in a real-life situation and therefore is not simply task fulfillment. The analysis of all included questionnaires revealed that only four out of the eight of the questionnaires included questions that reflected the participation domain. The PODCI showed the highest representation of this domain with 18% of questions reflecting participation.
The functional performance-based tests were found to be lacking in both the participation and environment domain (Figure 5). This is perhaps to be expected since their main aim is to observe a task being completed, which takes the patients out of their natural environment and outside of real-life situations. If a test involved the use of technology (e.g. computers, keyboards, games consoles) we determined that this represented an area of the environments domain. Tests that did not rely on observed task completion but also objective measurements such as joint movement and stability were determined to reflect the body functions (f) domain.
The percentage distribution of each of the questionnaires mapped to four of the ICF domains. Percentage distribution of each of the performance-based tests mapped to four of the ICF domains.

Discussion
The ICF can be seen as a unifying model for physical and rehabilitation medicine. This is a specialty that has been among the first to recognise the potential of the ICF to develop rehabilitation care, to inform health policy and management. 42 It also has the potential to stimulate research with the common goal of optimising functioning and minimising disability from both the individual and the public health perspective.43,44 There are multiple assessment tools available for hand and upper limb function assessment in children, all either validated to be disease-specific, region-specific or both. None have been found to be validated for use in a wide range of disease processes or age groups. There are advantages and disadvantages to using disease-specific or general health assessment tools. General health measures can provide a wider picture of health assessment whereas disease-specific are more useful for comparison purposes and hence research.
The contribution that adopting the ICF can make to paediatric occupational therapy is well recognised but requires integration of the ICF into everyday practice. 43 Hand therapists spend 20% of their time on taking measurements from patients which is deemed to be the most valuable part of their daily regime.44,45 Therefore, the time taken for each assessment is important in a system, which is constantly demanding increased efficiency and decreased manpower availability. Over the last few years there has been a move away from simply testing hand function with objective measurements and towards evaluating in a patient-centred approach in order to capture the quality of function. 46 The time taken for each assessment thus becomes an important issue as assessments that take longer to complete and become a less efficient use of time for the therapist and clinicians. Predictably, pegboard tests are the quickest (Nine hole peg test – 1 minute). It does, however, give the least variability in the number of meaningful concepts mapping onto the ICF and hence limited information on functionality. Conversely, tests that create a more detailed impression of function are also those that are not only time-consuming but also carry a cost burden. The AHA and Movement-ABC were tests that required a more significant financial investment and also a time investment for training the assessor. Thus in order to standardise the outcome assessment of hand function in children across national/international units we should bear in mind that not only time but also cost can be a limiting factor.
With children there is also the expectation of the parents to be incorporated into the decision-making process, which often does not differentiate between function and cosmesis. 47 This is particularly important since many assessment tools for children are completed by parents whose expectations for cosmesis may leave them feeling procedures or management have not necessarily given the benefit they actually have. It therefore becomes more important to develop an evidence-based approach to discussions with parents in order to manage expectations. This can more easily be supported by standardised assessment.
Our review aimed to identify the WHO ICF content of 19 validated and ready-to-use paediatric outcome measures that are used in the assessment of upper limb function in the paediatric population. None of the selected measures was conceptually developed from the ICF and it is possible this is the reason that none actually represented all of the domains of the ICF. Developing an ICF-based comprehensive tool is a challenging task since the ICF is vast and, although it has been derived to give us a holistic approach to patient evaluation, it is an ambitious task to expect a single assessment tool to represent all of the domains. We must therefore look at the over-arching domains of the ICF and see how well these can be represented in order to achieve a general overview of wellbeing, or conversely, disability. The activities and participation domain (d) of the ICF is unanimously the most well-represented in all performance-based tests and questionnaires. This is clearly from the activities’ perspective as it is easier to assess activity-based functioning or have it demonstrated in a series of observed tasks. Participation on the other hand is comparatively poorly represented. The difficulty being, perhaps, the subjective nature of assessing participation in everyday life and the perceived usefulness of this in clinical practice.
The recognition of the central role played by environmental factors has changed the focus of rehabilitative intervention from the individual to the environment in which the individual lives. Disability is no longer understood only as a feature of the individual but rather as the outcome of an interaction of the person with a health condition and the environmental factors. 48 Environment is the least represented of all domains in the tests and questionnaires. In the PUFI and CAPP-FSI questionnaires, the repeated questions regarding the use of prosthesis led us to classify it as an environment concept on each occasion it appeared. Although this was not the only concept generated per question it did lead to a higher proportion of responses within the environments domain than any of the other tests. In determining which outcome measures represent the environments domain the most we should take into consideration how many differing meaningful concepts (and hence domains) are actually generated. If a test or questionnaire is composed of 50 activity-based tasks or questions and each of these has a follow-up question of “is this easier with a prosthesis?” then along with 50 links to the activities domain there is also an additional 50 links to the environments domain (e1151 Assistive products and technology for personal use in daily living). This can give the impression of representing the environments domain well but it is only a single item within the ICF.
Body structures is an area of the ICF that is rarely mentioned in the literature and previous linking reviews have found no assessment tool that links to this area of the ICF. 49 It is unusual to find tests that include any mapping to this domain and none of the tests in this study did so. Most tests are designed for single-disease entities, yet their questions do not always relate to the disease specifically. Questionnaires on hand and upper limb function that ask specifically about presence or absence of digits or levels of amputations would capture this domain. That information would need to be deemed to confer an advantage or disadvantage in order to be influential in the final evaluation of a composite score.
Our review aims to highlight the content that each assessment tool has in common and highlights how other areas such as participation, e.g. social interaction, and body functions, e.g. pain, are distributed amongst the various assessment tools.
One of the limitations of our review is that although there is clear guidance from Cieza et al. 20 regarding the linking rules, the decision as to the nature of the meaningful concept is to a very small degree subjective in a few cases. Our approach has attempted to be objective and the use of two reviewers does aim to reduce this potential bias.
Conclusions
Our study has shown that although there are a wide variety of questionnaires and procedure-based tests available assessing function and functioning in a seemingly holistic manner, when mapped to the WHO ICF they are unlikely to represent all of the domains. Many clinicians often combine these two groups in an attempt to obtain a better evaluation of a patient's functional ability, yet our study has shown that this may still not be the case. We feel that new generation assessment tools will need to attempt to satisfy all domains of the ICF in order to truly provide a well-rounded approach to long-term disability evaluation and care. This may only be achieved by deriving these new assessments directly from the ICF itself and using a combination of modalities i.e. questionnaires with performance-based tests.
Footnotes
Acknowledgment
We would like to acknowledge Angela Harth (Formerly of the BG Trauma Centre Ludwigshafen) for her contribution to this article.
Conflict of Interest
None declared.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
