Appendix EOutcome measures

Publication Details

Reduction of spasticity

Data assessing muscle tone were preferentially extracted if measured using Ashworth scores. The Guideline Development Group (GDG) acknowledged that although there was no consistent method of presenting or summarising outcomes using this categorical scoring method, it was more readily and commonly performed in clinical practice than estimates derived using Tardieu scores. The NCC-WCH technical team recorded in the evidence tables how Ashworth scores were estimated in the included studies. Where Ashworth scores were not available, Tardieu scores (the estimate R2 – R1) were included.

Optimisation of movement and function

Movement incorporates joint movement and walking ability. Functional ability was assessed using validated tools that estimate the child or young person's skills and ability to reach developmental milestones, predefined tasks, general mobility, or in combined activities reflecting domains of participation (for example, self-care) in the World Health Organization's International Classification of Functioning, Disability and Health (ICF Framework; http://www.who.int/classifications/icf/en/). Functional ability could be reported by the child or young person, their parent or other carers.

The GDG recognised the importance of individualised goal setting in determining functional optimisation. The GDG considered the limitations of each tool (for example, its application to children and young people of different ages and with different disabilities and levels of comprehension, its sensitivity to detect change, and floor and ceiling effects).

The GDG prioritised the following assessments of movement and function.

Movement

Range of movement

Estimates of active and passive range of movement were included where possible. Range of movement is the distance and direction a joint can move between its limits. Active range of movement is ideally measured with a goniometer and estimates the range of movement through which a child or young person can move a joint actively (without assistance) using the adjacent muscles (that is, active movements use contractile muscle tissues as well as inert tissues). Further testing with passive motion and manual resistance helps to clarify the tissues at fault and the source of any associated pain. Passive range of movement is assessed while the joint is moved with assistance and with no effort from the child or young person (that is, the contractile muscle tissues are not engaged and movement, or pain, is a function of inert tissues only).

The GDG's view was that improvements in active and passive range of movement would become clinically worthwhile when they resulted in clinically important improvement in function, For example, passive range of movement might translate to an improvement in ease of care, reduction of pain, or improved Gross Motor Function Measure (GMFM; see below). In the absence of long-term studies, the GDG's view was that passive range of movement was a reasonable early indicator (proxy) for development of contractures.

Function

Walking

Gait assessment varies in complexity. It can range through observation of walking, examination of footprints in paint on a strip of paper, to a full gait analysis using specialised computerised equipment to observe the effort required to walk and specific parameters of motion, positioning and forces generated by the leg and foot during walking. The GDG prioritised estimations of walking speed and distance because clinically these would be universally estimable and because they believed that ultimately they are the most important outcomes for children and young people. The GDG noted that inability to walk would not be a relevant outcome for girls aged under 15 months or boys aged under 18 months, although for early walkers a pattern of walking typical of cerebral palsy (for example, walking on the toes or dragging a leg) would warrant further investigation.

Goal Attainment Scaling

The GDG agreed with emerging evidence that goals are more likely to be achieved if children and young people are involved in setting them. The GDG therefore gave a high prioritisation to Goal Attainment Scaling (GAS). This is a mathematical technique for quantifying achievement (or non achievement) of set goals. Usually three to five specific, measurable, achievable, realistic, and timebound (SMART) goals are identified individually to suit the child or young person, and levels are set around current and expected levels of performance. Goals may be weighted in order of priority for an individual child or young person and anticipated difficulty. Each goal is rated on a five-point scale indicating improvement or deterioration. The composite goal score can be transformed mathematically into a standardised T-score, with a mean of 50 and standard deviation (SD) of 10.

The GDG believed that this technique would avoid some potential problems of other standardised measures, such as a ceiling effect, lack of sensitivity and disjunction between the child or young person's main concerns and domains of the measure.

Canadian Occupational Performance Measure

The Canadian Occupational Performance Measure (COPM) records change in performance over time as perceived by the child or young person. It is completed by the child or young person with assistance from an occupational therapist. The GDG noted that this tool may be difficult to use with children or young people who have communication problems or who cannot understand the scoring system. The developers of the COPM reported difficulty using it with children aged under 8 years. The COPM is individualised for the child or young person, restricting generalised interpretation of its results.

There are two components to the COPM: performance (COPM-P) and satisfaction (COPM-S). The GDG considered COPM-P as a measure of function and COPM-S as a measure of acceptability and tolerability. Goals are set with the child or young person for daily activities in three domains: self-care, productivity, and leisure. The child or young person then chooses their five most important activities and grades each of them from 1 to 10 for performance and satisfaction. The mean values of the five scores become the baseline performance and satisfaction scores (minimum of 1, maximum of 10). Scoring is repeated for the same activities after a period of time. Evidence suggests that a change of two or more points at reassessment is clinically meaningful.

Paediatric Evaluation of Disability Inventory

The Paediatric Evaluation of Disability Inventory (PEDI) is a paediatric clinical assessment tool that supports a programme of intervention priorities and goal setting by describing functional skill attainment (rather than component skills) and detecting delays in attainment in children aged 6 months to 7.5 years with a range of disabling conditions and mixed impairments (Haley 1992). It can also be used to evaluate skills of older children whose abilities are below those expected of a child aged 7.5 years. It can be used to evaluate progress over time in individuals or groups or changes following an intervention.

The tool is administered by healthcare professionals and educators through observation or structured interviews with parents or carers. It evaluates three separate domains (self-care, mobility and social function), which can be examined separately or in combination. Scores for capability and performance are calculated for each domain and can be further analysed as a normative standard or scaled score.

Clinically meaningful differences in scores are those that exceed two standard errors (SEs) above or below the standard normative score for a particular domain (self-care, mobility, or social function) or two SEs above or below the scaled score for a particular domain. This has been expressed as a change score of 11.5, or approximately 11% (at a 95% confidence level), in a study that aimed to determine the minimal clinically important difference for children for inpatient rehabilitation (Lyer 2003).

Gross Motor Function Measure

The GMFM is a clinical assessment tool that measures gross motor function and has been validated for use in children and young people with cerebral palsy aged 5 months to 16 years. There are two versions: the GMFM-88 has 88 items and raw scores are summarised on an ordinal scale; the more recently developed GMFM-66 contains a subset of GMFM-88 and is converted to an interval scale.

Parents and therapists assess five dimensions (A to E) of the child or young person's functioning: dimension A, lying and rolling; dimension B, sitting; dimension C, crawling and kneeling; dimension D, standing; and dimension E, walking running and jumping. Each item of the tool is scored from 0 (does not initiate) to 3 (completes) in each dimension. Research has identified that at least 13 items are needed to produce meaningful scores.

Clinically meaningful differences in scores (where 0 is ‘not important’ and 7 is ‘tremendously important’) have been researched from parents' and therapists' perspectives. The results obtained for parents were: 4.6 (2.7% increase in GMFM score), 5.8 (5.2% increase in GMFM score change), and 6.0 (1.6% increase in GMFM score). The results for therapists were: 3.8 (1.8% increase in GMFM score), 5.4 (7% increase in GMFM score), and 6.0 (24% increase in GMFM score).

Reduction of pain

The GDG considered all reported measures of pain.

Adverse effects of interventions

The GDG prioritised intervention-specific adverse effects (for example, drowsiness with benzodiazepines, antibodies raised to botulinum toxin, and urinary problems following selective dorsal rhizotomy (SDR)). The GDG also considered all outcomes indicative of deterioration (for example, development of contractures or increased pain).

Acceptability and tolerability

For evaluating acceptability and tolerability in children and young people, the GDG prioritised validated assessment techniques (for example, COPM-S) and also considered study-specific questionnaire.

Health-related quality of life

The Child Health Questionnaire (CHQ), a generic quality of life instrument designed for children and young people aged 5–18 years, measures dimensions in two domains: physical and psychosocial. The physical domain includes scales for physical functioning, role or social limitations, general health perceptions, and body pain. The psychosocial domain includes scales for role or social limitations, emotions and behaviour, self-esteem, mental health, general behaviour, impact on parents' emotions, impact on parents' time, family activities; and family cohesion.

There are four versions of the CHQ, these being parent forms with 98, 50 or 28 items (CHQ-PF98, CHQ-PF50, and CHQ–PF28, respectively), and a self-report form for children and young people aged 10-18 years, although the self-report form is rarely used. CHQ-PF50 is used most frequently, and there is conflicting evidence regarding its reliability for assessing the outcome of an intervention in children and young people with cerebral palsy.

Generic questionnaires tend to have lower responsiveness to change and less sensitivity than disease-specific measures. The GDG noted concerns regarding the relevance of some items in the CHQ for children and young people with severe cerebral palsy, particularly the items relating to physical functioning (for example, cycling and playing football). The GDG also noted that the CHQ does not address issues related to transferring or handling, limiting its value when applied to children and young people with cerebral palsy. The GDG was aware that floor and ceiling effects have been reported, with floor effects occurring more frequently in physical domains, and ceiling effects occurring more frequently in psychosocial domains. The GDG recognised that this suggests poor face validity of parent-reported CHQ data when applied to children and young people with cerebral palsy, but concluded that the CHQ was an acceptable tool for evaluating health-related quality of life in the guideline.

Clinically important differences in outcome measures

Clinically important differences in the outcome measures prioritised for consideration in the guideline were discussed by the GDG, as outlined above, but the possibility of defining clinically important differences for each measure was eventually discounted by the GDG for reasons including the following.

Some of the prioritised outcomes were primarily of clinical interest, and did not necessarily represent the child or young person's value judgements about clinical benefit. For example, reduction of tone would be of clinical importance only if associated pain was reduced, or if functional improvement (such as the ability to self-care, or to walk, or if body positioning was improved) was attained. Clinically important differences would be more appropriately assessed using GAS scores which reflect the child or young person's perspective of benefit, but such outcomes were not reported for all interventions considered in the guideline.

Some outcome measures were not assessed and recorded in a standard way in all the included studies. For example, Ashworth scores were generally analysed as continuous rather than categorical variables, they sometimes comprised a single estimate from a muscle group, and in other cases they comprised a composite score from assessment in muscle groups in upper and/or lower limbs. Therefore, a standardised measure of a clinically important difference could not be determined.

Treatment effect sizes vary according to the body part being examined for some outcome. For example, expected range of movement around the ankle is different to range of movement at the knee or elbow. Moreover, achieving a clinically important difference at one joint might not result in a clinically important difference in functioning overall. It was not always clear from the included studies whether an improvement in a particular outcome resulted in a corresponding improvement in another outcome (for example, whether an improvement in range of movement resulted in improved function where results were derived from different studies).

The expectations of treatment would vary according to the type and severity of spasticity in an individual child or young person. In the studies included in the guideline reviews, the participants had different severities of spasticity (for example, in the review question on oral drugs) or specific types of spasticity (for example, in the review question on orthoses, all the participants had hemiplegia in some studies, whereas all the participants had diplegia in other studies).

References

  • Haley S, et al. Boston: PEDI Research Group, New England Medical Center Hospitals; 1992. Pediatric Evaluation of Disability Inventory (PEDI) Development, Standardization and Administration Manual. 1992.

  • Lyer LV, et al. Establishing minimal clinically important differences for scores on the pediatric evaluation of disability inventory for inpatient rehabilitation. Phys.Ther. 2003;83(10):888–98. [PubMed: 14519060]