School readiness of children with cerebral palsy

Authors

  • Frances E Gehrmann,

    1. Queensland Cerebral Palsy and Rehabilitation Research Centre, School of Medicine, The University of Queensland, Brisbane, Qld, Australia
    2. School of Medicine, The University of Queensland, Brisbane, Qld, Australia
    Search for more papers by this author
  • Andrea Coleman,

    1. Queensland Cerebral Palsy and Rehabilitation Research Centre, School of Medicine, The University of Queensland, Brisbane, Qld, Australia
    Search for more papers by this author
  • Kelly A Weir,

    1. Queensland Cerebral Palsy and Rehabilitation Research Centre, School of Medicine, The University of Queensland, Brisbane, Qld, Australia
    2. Queensland Children's Medical Research Institute, The University of Queensland, Brisbane, Qld, Australia
    Search for more papers by this author
  • Robert S Ware,

    1. School of Population Health, The University of Queensland, Brisbane, Qld, Australia
    2. Queensland Children's Medical Research Institute, The University of Queensland, Brisbane, Qld, Australia
    Search for more papers by this author
  • Roslyn N Boyd

    Corresponding author
    1. Queensland Cerebral Palsy and Rehabilitation Research Centre, School of Medicine, The University of Queensland, Brisbane, Qld, Australia
    2. School of Medicine, The University of Queensland, Brisbane, Qld, Australia
    3. Queensland Children's Medical Research Institute, The University of Queensland, Brisbane, Qld, Australia
    • Correspondence to Roslyn Boyd at Queensland Cerebral Palsy and Rehabilitation Research Centre, School of Medicine, The University of Queensland, Level 7, Block 6, Royal Brisbane and Women's Hospital, Herston, Qld 4029, Australia. E-mail: r.boyd@uq.edu.au

    Search for more papers by this author

Abstract

Aim

To examine school readiness in preschool-age children with cerebral palsy (CP) on three of five domains compared with reported norms of children with typical development (CTD).

Method

A representative population of 151 preschool-age children with CP (87 males, 64 females; 131 [87%] with spasticity, 17 [11%] dyskinesia, 3 [4%] hypotonia) were assessed at 48 or 60 months corrected age. Children were functioning in the following Gross Motor Function Classification System (GMFCS) levels: I, 74 (49%); II, 17 (11%); III, 14 (9%); IV, 26 (17%); V, 20 (13%). Children's motor performance, self-care, and social function were assessed using the Pediatric Evaluation of Disability Inventory (PEDI) and communication using the Communication and Symbolic Behaviour Scales Developmental Profile (CSBS-DP). Results were compared with a reference sample of CTD (PEDI CTD n=412; CSBS-DP CTD n=790). Linear regression was used to compare these data by functional severity.

Results

Children with CP had significantly lower PEDI scores in all domains than CTD. Self-care scores ranged from 0.5 to more than 4SD below CTD, motor performance was 2 to >4SD below CTD, and social function between 0.5 and >4SD below CTD. Fifty-five per cent of children demonstrated significantly delayed communication skills. Non-ambulant children displayed significantly lower scores than ambulant children.

Interpretation

Preschool-age children with CP perform significantly below their peers in three of five key readiness-to-learn skill areas including mobility, self-care, social function, and communication abilities. Broader emphasis needs to be placed on multimodal screening and intervention to prepare children with CP for school entry.

Abbreviations
CSBS-DP

Communication and Symbolic Behaviour Scales Developmental Profile

CTD

Children with typical development

PEDI

Pediatric Evaluation of Disability Inventory

SES

Socio-economic status

Cerebral palsy (CP) is the most common physical disability in childhood (two out of 1000 live births)[1, 2] and has a significant cost impact on the Australian health sector of 1.47 billion Australian dollars (AUD) per year, with an average annual cost of AUD43 431 per individual.[3] Mobility and ambulation are factors of importance in individuals with CP, influencing levels of independence, performance in activities of daily living, communication skills, participation in education, vocational involvement, and quality of life.

School readiness is a framework for assessing profiles of strengths and vulnerabilities of the preschool child in the context of transition to school.[4] It considers a child's readiness to learn within five major skill areas: health and physical development; emotional well-being and social competence; approaches to learning; communication skills; and cognitive skills and general knowledge.[4, 5] School readiness is an important paediatric concept as preschool-age developmental deficits are linked with later academic performance.[6] Links exist between child development, school readiness, educational attainment, and future health behaviours.[7] Identifying the breadth and severity of difficulties in school-readiness domains can maximize a child's school-readiness and increase the likelihood of academic success, which may improve future social, economic, and health outcomes.[8]

Assessment and prediction of school readiness using the five skill areas mentioned above has been reported in the preterm infant population with evidence that neurodevelopmental assessment can both identify school readiness[5] and predict future school readiness from measures performed at 2 years of age.[9] A prospective, longitudinal study of preschool-age children with typical development demonstrated that language competence is a strong predictor of school readiness.[10] There are mixed reports about the impact of socioeconomic status (SES) on school readiness in preterm populations:[5] one study found no significant association[5] whereas another showed low SES was a powerful risk factor for not being school ready.[11] Although some of the children in the preterm studies were diagnosed with CP,[9, 11] those with marked CP were often excluded.[5] No known studies have assessed broad domains of school readiness specifically in children with CP.

Individual areas of the five domains of school readiness have been addressed in some cohort studies. Physical disability is often the most prioritized activity limitation in children with CP, with much focus placed on the ability to walk.[12] Other aspects of educational performance, such as social interaction and communication, are often overlooked despite links between reduced educational attainment and poor long-term social and health-related outcomes.[7, 8] When considering each school-readiness domain individually, the literature has demonstrated that in 5-year-old children with CP half present with one or more activity limitations in oromotor function and/or communication[13] and 16% are non-verbal.[13, 14] Cohort studies of preschool-age children with CP have also addressed the domains of physical performance across the spectrum of severity in self-care[16] and social functioning.[15] There is a paucity of cross-sectional data in this age group about cognitive skills and general knowledge. Evaluating difficulties in domains of school readiness in a child with CP may assist therapists to optimize that child's transition to formal education before school entry. By identifying the school readiness of children with CP compared with their peers with typical development before school entry, we can identify and implement appropriate additional assistance to maximize achievement at school.

The primary aim of this study was to identify if a representative population-based cohort of preschool-age children with CP has different patterns of school readiness than children with typical development. Assessment was based on three of the five readiness-to-learn domains: (1) health and physical development, (2) emotional well-being and social competence, and (3) communication skills.[5]

Method

Participants

Participants were recruited from a prospective, population-based cohort of children with physician-diagnosed CP born in Victoria, Australia, between 1 January 2004 and 31 December 2005 and in Queensland between 1 January 2006 and 31 December 2009. These children had been referred to tertiary referral centres with statewide catchment areas and expertise in CP surveillance and care as part of a nationally funded study (NHMRC 465128). The full study protocol, including ascertainment and recruitment methods for the entire prospective cohort study, has been published.[16] Children were eligible for inclusion in this study if they were 48 months or older. All motor types and functional abilities were included.[16] Children who had not attended an assessment at 48 or 60 months were excluded. Ethical review and approval were obtained from the relevant ethics committees at participating hospitals and universities in Victoria and Queensland, Australia.

Procedures

All children enrolled in the study received a clinical assessment at 48 or 60 months corrected age. Demographic data, motor type (e.g. spasticity, hypotonia, athetosis, or dystonia),[13] and levels in the Gross Motor Function Classification System (GMFCS)[14] and Manual Ability Classification System were recorded.[18] Children were assessed for motor type and GMFCS by two physiotherapists.[16, 19, 20] SES was calculated using the Socio-Economic Indexes for Areas Index of Relative Disadvantage, which uses postcode of residence to categorize into scores, with lower scores indicating areas of lower SES.[21]

The school-readiness markers of physical development, social competence and emotional well-being, and communication skills were assessed using the Pediatric Evaluation of Disability Inventory (PEDI) domains of self-care, mobility, and social function.[17, 18] The PEDI is a parent/carer-reported standardized and reliable assessment of functional ability in a typical (home) environment used in children with disability.[22, 23] The PEDI is a Rasch-analysed measure and has been standardized against a sample of 412 children with typical development (CTD) between the ages of 6 months and 7 years 6 months (49.3% males). Means and SDs of subdomain scores for different CTD age groups are standardized to 50 and 10 respectively.[22] The commonly accepted cut-off point of greater than 1SD below the mean as a definition of vulnerability was adopted for this study.[5, 11, 24] Health and physical development was represented by the PEDI subdomains of self-care and mobility along with GMFCS levels, as strong relationships have been shown between gross motor capacity, mobility performance and self-care abilities, and general health. Social competence and emotional well-being was represented by the PEDI social function subdomain using areas such as problem-resolution, social interactions, play, and community function. The PEDI social function domain also represented elements of communication ability with assessment of comprehension, functional use, and expressive communication.[22] The Competency domain is addressed in the Communication and Symbolic Behaviour Scales (CSBS) and PEDI social functioning, which have components that describe a child's regulatory responses to adult- and peer-interaction requests.

Previously, knowledge of colours, letters, and numbers based on the ability to communicate concepts and the use of fine motor skills in block construction and copying of shapes has been used for assessing the verbal and non-verbal cognitive components of school readiness. Both of these cognitive skill areas have high correlations with the PEDI domains of self-care and social communication.

Communication skills were assessed using the CSBS Development Profile (CSBS-DP) Infant-Toddler Checklist.[25] This was selected as the Communication Function Classification System had not been developed at the start of data collection.[26] The CSBS-DP is a parent-reported screening questionnaire, which includes three composites (social, speech, and symbolic communication) and a total standard score. The CSBS-DP has been standardized against younger children (aged 6–24mo, n=790, 50.2% male) with normative total standard scores developed for each age group (mean 100, SD 15); it can be used in older children who are functioning below a 24-month age level.[25] The CSBS-DP has strong reliability and strong predictive validity at 2 years.[27] When used in older children, the CSBS-DP screens for communication delay and identifies children requiring further comprehensive assessment by a speech pathologist. A total score at or below the mean for the 24-month-old age range indicates a potential communication delay and was considered a significant deficit in our cohort of preschool-age children with CP.

Statistical analysis

Demographic and classification characteristics are summarized as mean (SD) for continuous variables and n (%) for categorical variables. The performance of children with CP was compared with CTD by calculating mean and SD for all PEDI subdomain standardized scores and the CSBS-DP total score. The number of SDs below CTD mean was calculated for each child. As previous studies of school readiness in preterm children[5, 9, 11, 24] have all suggested 1SD as a cut-off for vulnerability across several outcome measures, we adopted the cut-off greater than 1SD below the mean, accepting that 16% of all children would be considered vulnerable. As the PEDI is typically used as a screening tool to identify children who would benefit from more intensive assessment, the use of less than 1SD is not unusual (e.g. Bayley III scales). Variation in performance depending on GMFCS level and ambulatory status was investigated by stratifying by GMFCS level; data were also stratified for term and preterm children and for 48- and 60-month assessments. Linear regression was used to compare the association between motor function (ambulant; GMFCS I–III vs non-ambulant; GMFCS IV–V) and PEDI and CSBS-DP scores. Association between performance and SES was calculated by stratifying into Socio-Economic Indexes for Areas tertiles and calculating mean and SD for all PEDI and CSBS-DP scores. Univariate linear regression was used to identify significant differences between Socio-Economic Indexes for Areas tertiles. All analyses used stata 11.2 software (StataCorp, College Station, TX, USA).

Results

Of the 292 children in the broader Australian CP child cohort, which included children from four birth years (NHMRC 465128), a total of 171 (109 males, 62 females) had reached the 48- to 60-month age range and were, therefore, eligible for this study. Of these, 151 children (87 males [57%]) satisfied the specific inclusion criteria for this study, 83 (55%) were included at their 48-month assessment, and 11 had not attended their 48- or 60-month appointment. Demographic and classification characteristics of patients are summarized in Table 1. All functional levels were included (GMFCS: I, 74 [49%]; II, 17 [11%]; III, 14 [9%]; IV, 26 [17%]; V, 20 [13%]; Manual Ability Classification System: I, 60 [40%]; II, 42 [28%]; III, 14 [9%]; IV, 11 [7%]; V, 24 [16%]). There were relatively more children classified in GMFCS level I than existing national population reports of an older age group (Australian CP Register: I, 36%; II, 32%; III, 12%; IV, 14%; V, 5%).[28] The predominant motor type and distribution were unilateral spasticity (44%) then bilateral spasticity (43%), dystonia/ataxia (6%), athetosis (5%), and hypotonia (2%;Table 1). This distribution of motor types is similar to the Australian CP Register.[27] SES was evenly distributed across three tertiles: most disadvantaged 39 (26%), middle tertile 61 (40%), and least disadvantaged 51 (34%).

Table 1. Characteristics of participants
Characteristicsn=151CP cohort (n=292)
  1. aMotor type for the CP child cohort is classified at study entry (between 18–30mo). bMotor type for the present study (n=151) is classified at 48–60mo. CP, cerebral palsy; GMFCS, Gross Motor Function Classification System; MACS, Manual Ability Classification System; SEIFA, Socio-Economic Indexes for Areas Index of Relative Disadvantage.

Sex, n (%)
Male87 (58)227 (62)
Female64 (42)110 (38)
Preterm birth (<37wks gestation)59 (39)132 (45)
Mean age (SD), range, mo53.7 (6.3); 45–65 
Assessment type
48mo83 (55) 
60mo68 (45) 
GMFCS level, n (%)
I74 (49)129 (44)
II17 (11)45 (15)
III14 (9)42 (14)
IV26 (17)31 (12)
V20 (13)45 (15)
MACS level, n (%)
I60 (40) 
II42 (28) 
III14 (9) 
IV11 (7) 
V24 (16) 
Motor type, n (%)
Unilateral spasticity66 (44)b93 (32)a
Bilateral spasticity65 (43)160 (54)
Dystonia9 (6)12 (4)
Athetosis8 (5)5 (2)
Hypotonia3 (2)12 (5)
Unclassifiable010 (3)
SEIFA disadvantage tertiles, n (%)
Most disadvantaged39 (26) 
Middle tertile61 (40) 
Least disadvantages51 (34) 

Children with CP at 48 to 60 months recorded lower PEDI scores in all domains compared with CTD standards (n=51; mean 50, SD 10; Table 2). Eighty-five per cent of self-care assessments were more than 1SD below CTD and 34% achieved a standardized score of 10 or less, equivalent to scoring more than 4SD below CTD. In motor performance, 88% of children scored more than 1SD below CTD and 51% scored more than 4SD below CTD. Sixty-nine per cent of children scored more than 1SD below CTD on performance of social function, 23% of these scoring more than 4SD below CTD. Mean and SD of each domain according to GMFCS level and preterm versus term birth are reported in Table 2. Children in GMFCS levels IV or V who could not ambulate even with gait aids scored significantly lower on the PEDI than those who could ambulate (GMFCS I–III). The mean difference (95% CI) for self-care was 19.0 (14.8–23.2; p<0.01), for mobility 13.3 (8.8–17.9; p<0.01), and for social function 20.8 (15.4–26.2, p<0.01; Table 3).

Table 2. Pediatric Evaluation of Disability Inventory (PEDI) subdomain scores according to Gross Motor Function Classification System (GMFCS) level and number of children scoring 1SD or more below normed standard scores (mean 50, SD 10; n=151)
PEDI subdomainMean scoreSDNumber of children ≥1SD below norm (total children) (%)
  1. aSignificant difference from GMFCS level I (p<0.001). bSignificant difference from term born (p<0.001).

Self-care, total23.914.9129 (151) (85.4)
GMFCS level
I33.413.855 (74) (74.3)
II23.1a13.815 (17) (88.2)
III18.7a9.013 (14) (92.8)
IV11.2a3.326 (26) (100)
V<10a020 (20) (100)
Preterm born32.114.750 (59) (84.7)
Term born30.012.980 (91) (87.9)
48mo assessment23.615.171 (83) (85.5)
60mo assessment24.314.858 (68) (85.3)
Mobility, total19.414.4133 (151) (88.1)
GMFCS level
I27.915.957 (74) (77)
II15.2a11.116 (17) (94.1)
III<10a014 (14) (100)
IV10.2a0.926 (26) (100)
V<10a020 (20) (100)
Preterm born28.715.652 (59) (88.1)
Term born29.815.580 (91) (87.9)
48mo assessment20.815.571 (83) (85.5)
60mo assessment17.612.962 (68) (91.2)
Social function total31.118.0104 (151) (68.9)
GMFCS level
I41.116.239 (74) (52.7)
II27.5a13.613 (17) (76.5)
III29.9a13.210 (14) (71.4)
IV20.7a15.322 (26) (84.6)
V11.4a4.720 (20) (100)
Preterm born42.115.734 (59) (57.6)
Term born33.6b13.672 (91) (79.1)
48mo assessment32.119.256 (83) (67.5)
60mo assessment29.816.648 (68) (70.6)
Table 3. Difference in Pediatric Evaluation of Disability Inventory (PEDI) and Communication and Symbolic Behaviour Scales Development Profile (CSBS-DP) scores between ambulatory children GMFCS levels I–III) and those who did not ambulate even with gait aids (GMFCS levels IV or V)
Outcome measureMean differenceStandard error95% confidence interval
  1. Positive mean difference indicates a higher score. GMFCS, Gross Motor Function Classification System.

PEDI self-care19.02.114.8–23.2
PEDI mobility13.32.38.8–17.9
PEDI social function20.82.715.4–26.2
CSBS-DP total41.23.733.9–48.5

Fifty-five per cent of preschool-age children with CP had CSBS-DP scores no greater than 1SD above 24-month-old CTD (Table 4). For communication, 22% were communicating at a 24-month level, 33% scored greater than 1SD below 24-month-old CTD, and 45% scored greater than 1SD above 24-month-old CTD. When considering communication by gross motor ability and term versus preterm birth, the mean (SD) for GMFCS level is reported in Table 4. Non-ambulant children (GMFCS IV or V) also scored lower on CSBS-DP than ambulatory children (GMFCS I–III), with a mean difference (95% CI) of 41.2 (33.9–48.5, p<0.01; Table 3). No significant difference was found between PEDI subdomain scores and CSBS-DP scores across Socio-Economic Indexes for Areas tertiles of disadvantage (Table 5). There was no significant difference found between PEDI subdomain scores and CSBS-DP scores across 48- and 60-month assessments (Tables 2 and 4).

Table 4. Total Communication and Symbolic Behaviour Scales Development Profile (CSBS-DP) standard scores according to Gross Motor Function Classification System (GMFCS) level and number of children scoring ≥1SD below normed standard scoresa
CSBS-DP scoreMean scoreSDNumber of children ≥1SD above 24mo norm (total children) (%)Number of children ±1SD of 24mo norm (total children) (%)Number of children ≥1SD below 24mo norm (total children) (%)
  1. aNormed standard scores: mean 100, SD 15 (n=151). bSignificant difference from GMFCS level I (p<0.001).

Total105.228.268 (151) (45)33 (151) (22)50 (151) (33)
GMFCS level
I122.718.555 (74) (74)14 (74) (25)5 (74) (9)
II99.b22.04 (17) (23)6 (17) (37)6 (17) (35)
III114.216.35 (14) (36)8 (14) (67)1 (14) (7)
IV84.6b26.44 (26) (15)2 (26) (11)18 (26) (69)
V66.2b3.70 (20) (0)0 (20) (0)18 (20) (90)
Preterm110.625.832 (59) (54)14 (59) (23.7)13 (59) (22)
Term101.829.536 (91) (39.6)18 (91) (19.8)37 (91) (40.7)
48mo assessment109.128.845 (83) (54.2)13 (83) (15.7)25 (83) (30.1)
60mo assessment100.527.023 (68) (67.7)20 (68) (29.4)25 (68) (36.8)
Table 5. Pediatric Evaluation of Disability Inventory (PEDI) subdomain scores (mean 50, SD 10; n=151) and Total Communication and Symbolic Behaviour Scales Development Profile (mean 100, SD 15) according to Socio-Economic Indexes for Areas (SEIFA) Index of Relative Disadvantage tertiles
 Mean scoreSD
  1. a

    No significant difference from SEIFA between tertile (p<0.001).

Self-care PEDI subdomain

SEIFA disadvantage tertilesa

Most disadvantaged106.026.7
Middle tertile106.828.8
Least disadvantaged102.829.1

Mobility PEDI subdomain

SEIFA disadvantage tertiles

Most disadvantaged19.814.1
Middle tertile20.515.4
Least disadvantaged17.713.5

Social function PEDI subdomain

SEIFA disadvantage tertiles

Most disadvantaged31.716.8
Middle tertile31.118.3
Least disadvantaged30.619.0

CSBS-DP score

SEIFA disadvantage tertiles

Most disadvantaged106.026.7
Middle tertile106.828.8
Least disadvantaged102.829.1

Discussion

In a prospectively studied population of preschool-age children with CP, we have found that reported performance of three of the five key school-readiness skills (health and physical development, emotional well-being and social competence, and communication) are mostly below the mean of peers with typical development. Self-care scores ranged from 0.5 to more than 4SD below CTD, motor performance was 2 to more than 4SD below CTD, and social function between 0.5 and more than 4SD below CTD. Fifty-five per cent of children demonstrated significantly delayed communication skills. Children with CP with severe motor deficits also demonstrate significant deficits in other neurodevelopmental fields, most functioning below their peers with typical development in performance of self-care (100% greater than 1SD and 87% more than 4SD below CTD), communication (91% at or below the 24mo level), and social function (91% greater than 1SD and 65% more than 4SD below CTD). Our data also show that of ambulatory children with higher motor function, who walked without gait aids (GMFCS levels I and II), 80% were still functioning more than 1SD below peers in mobility tasks and demonstrated deficits in communication and social function. Aside from some differences in social function, preterm children with CP demonstrated similar school-readiness profiles as children born at term.

In preterm infants a relation exists between neurodevelopmental deficits at 2 years of age and school readiness at 5 years.[9] To our knowledge, no other study has published data addressing school readiness in a population of preschool-age children with CP. In children with CP, the focus is typically on motor difficulties, with other domains of early development frequently not addressed. In this study, we have reported deficits in three of the five key readiness-to-learn domains. Research on cognitive skills and general knowledge domains of children with CP at school entry is required. These domains were not assessed in this study and there are some limitations in accurately assessing cognitive skills, such as executive function, in children under the age of 8 years.[36]

Multimodal early intervention may benefit all children with CP, and optimize the transition into the school environment by improving key development areas of communication, physical function, emotional and social development, and cognitive skills. In Australia, there are no intervention programmes available to promote social, communication, and cognitive skill development. In the Australian CP Child Study,[16] we monitored by formal questionnaire the use of all medical and allied health services and therapeutic recreational programmes that these children accessed every 6 months; we can confirm the lack of specific, tailored interventions to improve communication, social functioning, and cognition for children with CP. In Australia, we acknowledge that children with CP may access generic early child development programmes in their fourth year, provided by specialist, trained teachers, in the form of play-based programmes for 3 to 6 hours per week before starting formal schooling (30h per week) from the age of 5 years. Access to formal education services was not controlled for in our analyses.

Historically, the focus of CP management in Australia has targeted physical issues and population-based screening, whereas social function and communication skills have not been prioritized. A new national initiative implemented in Australia for developmental assessment is the Australian Early Development Index,[29] a population measure of children's development as they enter school, assessing key school-readiness areas.[29] Vulnerability across all domains of development in children with special health-care needs has been found, with these children over-represented in the bottom 10th centile of population-based data.[30] This measure may be a useful tool in children with CP as a reference of how they perform compared with the general population of school-age children. Difficulties in areas of school readiness may be identified with this assessment but other assessment tools would be required to define the extent of these difficulties. The Australian Early Development Index is designed for school-age children and as it does not offer early assessment, it may not assist in early intervention to optimize transition to school in younger children.

In the USA, the Individuals with Disability Education Act supports early intervention across domains of physical, communication, cognitive, social/emotional, and adaptive development.[31] However, internationally there is much variability in levels of intervention available to preschool-age children. In 2010, a new Australian federal government funding initiative, the ‘Better Start for Children with a Disability’ scheme, was implemented to provide early intervention services to children with a disability that has an impact on their development. This initiative allows early intervention funding of up to AUD12 000 for children aged up to 7 years with a qualifying disability (including CP).[32] Parents can choose to allocate this funding to services of their choice. For many parents of preschool-age children, priorities often include optimizing their child's self-care and mobility abilities, with school performance only becoming a priority as children reach school age.[32] Our study has shown that preschool-age children with CP already have substantial deficits in multiple development areas beyond mobility and self-care,[33] including communication and social competence. If readiness-to-learn skills are not considered until children reach school age, those with CP may be further behind their peers with typical development and less likely to optimize their educational potential. Multimodal interventions targeting key areas of early development would be a good use of the ‘Better Start’ funding to maximize school-readiness skills in children with CP.

A clear relation was found between GMFCS level and mobility, social function, self-care, and communication at 48 to 60 months corrected age. This is consistent with a population-based study of children with CP (mean age 58mo, SD 18) that found GMFCS level significantly predicted social function and self-care.[34] This study also showed that children with CP performed below standard scores of CTD in all PEDI subdomains, independent of GMFCS level.[34] A strong relation between GMFCS and mobility in preschool-age children (mean 3y 9mo, SD 1y) has been found, with moderate to good relationships between GMFCS level and self-care, dexterity, speech and thinking, and problem-solving.[34] Our group has previously reported the relation between GMFCS level and social function: a cohort of 18- to 30-month-old children with CP were performing significantly below peers with typical development in social function, with 67% of children functioning 1 or more SD below the mean.[15] Our present study shows similar results, with 69% of participants falling more than 1SD below the mean in performance of social function at 48 to 60 months, and 55% of participants having delayed communication. These data correspond with the Australian CP Register's 2009 report, which showed that 60% of school-age Australian children with CP had impaired speech.[28] Our findings of most children with CP scoring more than 1SD below peers in all learning domains corresponds to those identified in school-readiness studies in very preterm children: a group of very preterm children scored 0.5 to 1SD below age-matched comparisons in all domains of school readiness.[5] These findings have major implications for school attainment and may later have an impact on vocational attainment, SES, and health behaviours.

No relation was found between SES and performance in school-readiness domains for children with CP in our study. This is consistent with findings studying school readiness in very preterm children.[5] However, in another American study, SES did influence school readiness in a preterm population where only a small proportion had CP.[9] Our results indicate that screening for vulnerability in school-readiness domains should be included in assessment of preschool-age children with CP. This can inform clinical intervention and ensure children are receiving therapy targeting all areas of vulnerability. Inclusion of the Australian Early Development Index into assessment of school-age children with CP would also be informative for developmental status compared with CTD and areas of difficulty. By targeting therapy to areas of vulnerability within these domains, the transition to school may be maximized, having an impact not only on academic attainment but also on quality of life.

This study of preschool-age children showed a slightly different GMFCS distribution to that of the Australian CP Register (for which children are ascertained at 5y):[28] a higher proportion of children were classified in GMFCS level I,[35] potentially limiting the generalizability of our results. This may be due to the Australian CP Register collecting data on an older cohort of children (5–16y) as GMFCS classification can change over time, with children often reclassified to a less functional level.[16, 35] This correlates with the Register noting that 5-year-old children were predominantly (59%) in GMFCS levels I and II.[28]

A limitation of this study is its inability to assess all areas of readiness to learn. Absence of general knowledge, cognitive function, and approach to learning assessments restricts our ability to comment on global school readiness in preschool-age children with CP. In the present study, we focused on three of five domains of readiness to learn using reliable and robust measures. Measurement of cognition and general knowledge are less reliable under the age of 5 years.[36] The magnitude of deficits found in the measured domains suggests reduced school readiness compared with CTD despite failure to assess some areas. Another limitation is the ceiling effect of the CSBS-DP owing to standardized data only being available for CTD up to 24 months of age. This limits the accuracy of reported communication abilities of those children scoring above a 24-month level.

Previously, knowledge of colours, letters, and numbers, based on the ability to communicate concepts and the use of fine motor skills in block construction and copying shapes, has been used to assess the verbal and non-verbal cognitive components of school readiness. Both of these cognitive skill areas have high correlations with the PEDI domains of self-care and social communication.

We acknowledge the limitation of the CSBS-DP screening tool potentially not to detect children with milder expressive and/or receptive language difficulties. A future research opportunity would therefore be to follow up poor performance on the CSBS-DP with a formal speech pathology assessment to identify the predictiveness of CSBS-DP scores on communication in the older 4- to 5-year age group. Future research should include comprehensive assessments of the readiness to learn domains of cognitive skills, general knowledge, emotional well-being, and approach to learning in assessing school readiness in the population with CP.

Conclusion

This study has demonstrated that CP has an impact on children beyond pure motor disability. It has identified that preschool-age children with CP perform significantly below their peers in key skill areas of readiness to learn just before school entry. This indicates that multifaceted early developmental screening, focusing on all domains of readiness to learn, supported by multimodal interventions, may improve the transition to school and the children's developmental progression. The Australian federal government's ‘Better Start’ funding would be well used in this multimodal approach to early intervention for children under 6 years of age with CP.

Acknowledgements

We acknowledge the assistance of physiotherapists Rachael Jordan, Chris Finn, and Paula Beall in Queensland, and Anne Moodie and Belinda Luther in Victoria, for data collection. Ethical permission and recruitment were conducted through the Royal Children's Hospital and Health Service District Ethics Committee (HREC/07/QRCH/107, Queensland), The University of Queensland Medical Research Ethics Committee (2007001784), the Cerebral Palsy League of Queensland Ethics Committee (CPLQ-2007/08-10010), the Royal Children's Hospital Human Research Ethics Committee (25010E, Melbourne), the Mater Health Services Human Research Ethics Committee (1186C), and the Southern Health Human Research Ethics Committee (05077C, Victoria). Funding was provided by a National Health and Medical Research Council of Australia Project Grant (569605). R. Boyd is supported by a National Health and Medical Research Council Career Development Fellowship (1037220). The authors have stated that they had no interests that might be perceived as posing a conflict or bias.

Ancillary