The survival of babies born with Down syndrome has improvedmarkedly over recent decades in many countries.1,2 Childrenwith Down syndrome currently represent a significant proportionof all children with intellectual disability in Western Australia.3

Information about the functional strengths and weaknesses ofchildren with Down syndrome is of importance to many groups,such as prospective parents considering prenatal diagnosis; earlyintervention programmes; agencies currently and potentiallyproviding family support; and educational services.

A recognition of the need for a simple yet descriptive tool tomeasure disability led to the development of an innovativemeasure of functional ability in children, the Functional Inde-pendence Measure for Children (WeeFIM; Uniform DataSystem for Medical Rehabilitation, Buffalo, NY, USA).4 Thisinstrument was based on and adapted from the adult assessmenttool, the Functional Independence Measure (FIM), but with adevelopmental, habilitative and paediatric multidisciplinaryperspective.5 The intent of the authors of the present study was todevelop a measure of disability that was discipline-free;therefore, able to be administered by paediatric, health and reha-bilitation professionals, educators and community workers. Thistool should measure performance in daily activities, rather thanin the ‘consulting room’ type tasks of the psychological testsused routinely. However, it was not in any way designed toreplace existing comprehensive measurements of motor,communication, cognitive and educational skills.6

The WeeFIM includes measures of self-care (feeding,grooming, dressing and bathing), maintaining continency,

transfers, locomotion, communication and social skills. It wasanticipated that the WeeFIM could have multiple uses, including -practical applications (e.g. in evaluating family supports andeither biomedical or educational intervention strategies);7

comparing functional ability in different conditions;5 andcomparing the impact of health and neurological impairment in children with many types of disability in a variety of pop-ulations.8,9

The WeeFIM has been piloted on more than 2000 childrenwith developmental and genetic disorders, including limb def-iciencies, Down syndrome, spina bifida, cerebral palsy,extremely preterm infants,10,11 and also on children with braintumours,9 as well as on more than 500 children without a diag-nosed disability, and children with and without functional limi-tations after very low birthweight status and retinopathy ofprematurity.12–14 The tool has also been used specifically innon-Caucasian children.15

The WeeFIM has excellent test–retest, interrater and equiva-lence reliability, and excellent concurrent validity with theVineland Adaptive Behaviour Scale, the Battelle Develop-mental Inventory, the Amount of Assistance Questionnaire, andthe Pediatric Evaluation of Disability Inventory.16–18 Further-more, the WeeFIM has been demonstrated to be highly cor-related with clinical judgement.19

Here we report on the WeeFIM in a modified questionnaireformat (to be completed by the child’s parent or guardian) tomeasure functional ability in the areas of self-care, con-tinency, mobility, transfers, communication and social skills in

J. Paediatr. Child Health (2002) 38, 160–165

Functional status of school-aged children with Down syndrome

S LEONARD,1 M MSALL,2 C BOWER,1 M TREMONT2 and H LEONARD1

1TVW Telethon Institute for Child Health Research, Perth, Western Australia, Australia and 2Child Development Center,Rhode Island Hospital, Providence, Rhode Island, USA

Objective: To field test, in questionnaire format, the Functional Independence Measure for Children (WeeFIM, a scheduleusually administered by interview) on parents of a cohort of school-aged children with Down syndrome.Methods: The parents of 211 Western Australian children with Down syndrome participated in the present study, rep-resenting 79.9% of all children with Down syndrome in the State. Subjects were identified using two sources: (i) the BirthDefects Registry; and (ii) the Disability Services Commission.Results: The total WeeFIM score was 106.2 ± 17.0 (mean ± SD) out of a possible 126. Girls scored higher than boys(108.6 vs 103.6; P = 0.05). Scores increased across all age groups (P < 0.0001), even relative to normative data. Performancewas strongest in the transfer and locomotion domains and weakest in social cognition.Conclusion: We found that severe functional limitations are rare in school-aged children with Down syndrome. Somesupport and supervision are required for complex self-care, communication and social skill tasks. This study demonstratesthe feasibility of using the WeeFIM for collecting population survey data in children with developmental disability. This maybe useful for the longitudinal tracking of such populations, as well as the monitoring of response to interventions.

Key words: activities of daily living; developmental surveillance; Down syndrome; functional assessment; functionalstatus; WeeFIM.

Correspondence: H Leonard, TVW Telethon Institute for Child Health Research, PO Box 855, West Perth, Western Australia 6872, Australia. Fax: +61 8 9489 7700; email: hleonard@cyllene.uwa.edu.au

Accepted for publication 21 May 2001.

a population of school-aged children with Down syndrome inWestern Australia.

METHODS

The study population was a cohort of school-aged childrenwith Down syndrome living in Western Australia in 1997. Aspreviously described,20 cases were ascertained from twosources: (i) the Western Australian Birth Defects Registry(BDR) a population-based registry of birth defects;21 and (ii)the Disability Services Commission (DSC), a multidisciplinarygovernment body responsible for coordinating medical andtherapy services for people with an intellectual disability inWestern Australia. Parents (or guardians) of eligible childrenwith Down syndrome were invited to complete a postal ques-tionnaire (this instrument is available on request from HL),which, as well as the WeeFIM, included questions on thechild’s medical conditions, use of medical therapy and supportservices, educational placement and social issues.

The format of the WeeFIM was modified from the existinginterview model,5 to a parent-completed questionnaire versionby MM and MT who have extensive experience with theWeeFIM. The WeeFIM consisted of questions concerning 18tasks in six domains of functioning. For each task, parents wereasked to tick one box and one sub-box next to the statementsthat best matched their child’s level of competency at the task.Answers were scored (with the highest possible score on each question being 7) for a maximum total score of 126. Inaddition, WeeFIM quotients for total scores were calculatedusing the formula child’s WeeFIM rating divided by theexpected WeeFIM of a non-disabled peer and multiplied by100. Normative scores were obtained using tables and graphsin the WeeFIM version 5.0.4 The acceptability of the question-naire was pretested on a number of families on a Downsyndrome email list prior to being posted to all parents orguardians of children in the Western Australian cohort.However, we did not have the opportunity to compare dataobtained by conventional interview with the new questionnaireformat on individual subjects.

All information from the questionnaire was entered into aFileMaker Pro database,22 and exported into S TATA ,23 whichwas used to perform descriptive analyses of the data includingfrequency distributions. When comparing two continuouslydistributed variables t-tests were used and A N OVA for three ormore variables. Non-parametric tests (Mann–Whitney U andKruskal–Wallis) were used when the distribution was notnormal.24 The bivariate distribution of pairs of categoricalvariables was investigated using 2-D tables, and formal tests ofassociation were based on the Chi-squared test.25 A P value of0.05 or less was accepted as indicating statistical significance.Unless otherwise indicated, Chi-squared tests were based onone degree of freedom. Children were grouped into three agestrata for analysis (5–9 years, 10–13 years and 14–17 years).An adjusted WeeFIM score was calculated by weighting thecompleted WeeFIM score to account for missing values.

RESULTS

Response

Two hundred and ninety families were identified from the twodata sources (BDR & DSC) and one family was self-reported.

After exclusions due to migration from the State, recent deathor refusal of contact on the DSC register, 264 families wereeligible to participate. The questionnaire was returned by 211(79.9%) eligible families. The children of responders were sim-ilar in gender (P = 0.73), age (P = 0.18) and place of residence(P = 0.57) to the children of non-responders.

Demographic characteristics of respondents

The mean ± SD age of subjects was 11.4 ± 3.6 years (range5.8–17.7 years). All of the children lived at home, although 19children (9.0%) were not living with their birth parents. Themean ± SD number of siblings was 2.5 ± 1.7 (range 0–9).

WeeFIM

A total WeeFIM score was calculated for all children with a fully completed WeeFIM (n = 168). The mean ± SD scorewas 106.2 ± 17.0. For the 43 cases where one or more itemswere not completed, an adjusted score was calculated. If tradi-tional methods of scoring were used (where a score of 1 isgiven for missing values), the mean score for all children forwhom the WeeFIM was fully or partially completed was 103.4± 18.3. The commonest questions not completed were thosedealing with chair, wheelchair, toilet and bath/shower transfers.Some parents indicated that they thought that these questionswere ‘not applicable’ to their child and hence had not responded to them. In these instances a score of 1 seemedinappropriate, and instead we obtained an adjusted score byweighting the total score for these children under the assump-tion that the missing scores were of similar value to completedscores. The total WeeFIM score used for all children on whomit was available and the ‘adjusted’ score for the remainder ishereafter reported as the ‘WeeFIM score’. The mean WeeFIMscore was 105.7 ± 17.1.

Table 1 shows the WeeFIM quotients demonstratingcomparisons to normative data. The mean quotient for theentire group was 87.7. The mean quotients for the 5–9 yearsage group was 83.0, the 10–13 years age group was 88.3 andthe 14–17 years age group was 93.6.

The mean ± SD WeeFIM score for girls was 108.6 ± 14.8compared to 103.6 ± 18.5 for boys (P = 0.05). WeeFIM scoresincreased across age groups (P < 0.001) (Table 2). The WeeFIMscore for children living in the metropolitan area (105.6) wassimilar to the score for children in rural areas (106.1; P = 0.99).The mean scores for the lower (105.6) and higher (106.3)income groups were also similar (P = 0.99).

Table 3 shows the mean score for individual WeeFIMtasks. The maximum score possible for any task is 7. WeeFIMscores on each task were divided into three groups: 1, helpneeded (WeeFIM scores 1–4); 2, supervision needed (WeeFIM

161Functioning in Down syndrome

Table 1 Quotients for each age group

Age group Number Mean SD Minimum Maximum

5–9 years 83 83.0 14.7 20 10010–13 years 65 88.3 10.1 64 10014–17 years 59 93.6 10.8 32 100All 207 87.7 13.0 20 100

162 S Leonard et al.

score 5); and 3, no help needed (WeeFIM score 6 or 7) (Table4). Table 5 shows performance when tasks are grouped into thefunctional domains of self-care, sphincter control, transfers,locomotion, communication and social cognition.

Self-care

Over half of the children with Down syndrome (59.7%) neededno help with self-care tasks. There was no gender difference in self-care scores overall (P = 0.10), but there was an increasein score with age (χ2

2 = 39.3; P < 0.001).When the individual components of self-care were

examined, girls (mean score 2.2) performed somewhat betterthan boys (mean score 2.0) in dressing from the waist down (P = 0.06). There were significant improvements with age in

the areas of dressing from the waist up (χ22 = 39.8; P < 0.001),

waist down (χ22 = 37.8; P < 0.001), bathing (χ2

2 = 0.2; P < 0.001) and grooming (χ2

2 = 36.8; P < 0.001).

Sphincter control

Most children with Down syndrome (96.6%) needed either nohelp at all or supervision only with these tasks. Girls managedbetter than boys (P = 0.04). As would be expected, sphinctercontrol was better in older children (χ2

2 = 14.5; P < 0.001).When the components of sphincter control were examinedindividually, there was no gender difference in bowel orbladder control and the increase with age was only apparent inbowel control (χ2

2 = 18.3; P < 0.001). Furthermore, there wasno difference in sphincter control between those who had abowel condition (predominantly constipation),18 and those whodid not (P = 0.30).

Table 2 Adjusted WeeFIM scores for each age group

Age group Number Mean SD Median

5–9 years 83 98.9 18.8 10210–13 years 65 106.7 13.8 10914–17 years 59 114.3 13.7 118All 207 105.8 17.1 110

Table 3 Scores for individual WeeFIM tasks

Task Mean score SD

Bath shower transfers 6.8 0.8Bathing 5.4 1.8Bladder control 6.0 1.7Bowel control 6.4 1.4Chair/wheelchair transfers 7.0 0.5Comprehension 5.6 1.5Dressing waist down 5.2 1.8Dressing waist up 5.3 1.8Eating 6.4 1.2Expression 5.4 1.6Grooming 5.2 1.8Memory 4.9 1.5Movement 6.9 0.6Problem solving 5.0 1.5Social interaction 5.7 1.7Stairs 6.4 0.9Toilet transfers 6.8 0.9Toileting 5.9 1.7

Table 4 Scores for individual WeeFIM tasks – grouped into help,supervision and no help categories

Number and Number and Number andproportion proportion proportionrequiring requiring requiring

Task help supervision no help

Bathing 52 (25.4%) 55 (26.8%) 98 (47.8%)Bath shower transfers 6 (3.1%) 6 (3.1%) 182 (93.8%)Bladder control 29 (14.0%) 35 (16.9%) 143 (69.1%)Bowel control 18 (8.7%) 22 (10.6%) 167 (80.7%)Chair/wheelchair transfers 2 (1.1%) 0 (0.0%) 183 (98.9%)Comprehension 40 (19.3%) 36 (17.4%) 131 (63.3%)Dressing waist down 68 (37.6%) 26 (15.6%) 87 (46.8%)Dressing waist up 64 (31.1%) 43 (20.9%) 99 (48.1%)Eating 21 (10.1%) 16 (7.7%) 171 (82.2%)Expression 47 (22.9%) 43 (21.0%) 115 (56.1%)Grooming 65 (31.3%) 53 (25.5%) 90 (43.3%)Memory 82 (39.8%) 46 (22.3%) 78 (37.9%)Movement 2 (1.0%) 6 (3.1%) 186 (95.9%)Problem solving 61 (29.3%) 63 (30.3%) 84 (40.4%)Social interaction 42 (20.5%) 31 (15.1%) 132 (64.4%)Stairs 7 (3.4%) 8 (3.9%) 190 (92.7%)Toilet transfers 8 (4.2%) 1 (0.5%) 181 (95.3%)Toileting 42 (20.3%) 22 (10.6%) 143 (69.1%)

Table 5 Scores for grouped WeeFIM tasks

Number and Number andNumber and proportion proportionproportion requiring requiring

Task groups requiring help supervision no help Mean ± SD

Self-care 9 (4.5%) 72 (35.8%) 120 (59.7%) 5.6 ± 1.4Sphincter control 7 (3.5%) 67 (33.0%) 129 (63.6%) 5.9 ± 1.3Transfers 1 (0.6%) 6 (3.3%) 176 (96.2%) 6.9 ± 0.6Locomotion 1 (0.5%) 10 (5.2%) 183 (94.3%) 6.7 ± 0.7Communication 7 (3.5%) 89 (43.8%) 107 (52.7%) 5.5 ± 1.4Social cognition 10 (4.9%) 95 (46.8%) 98 (48.3%) 5.2 ± 1.3

Transfers

The majority of children with Down syndrome (96.2%) requireno help or supervision with transfer tasks. Girls performedbetter than boys (P = 0.03). When the components of transferswere examined individually, there was a gender difference intoilet transfers only (P = 0.04), with girls scoring slightlyhigher than boys.

Locomotion

As with transfers, the majority of children (94.3%) with Downsyndrome required no help or supervision with day-to-daymovement. There was no gender difference (P = 0.65).

Communication

More than half of the children with Down syndrome (52.7%)needed help or supervision with their communication tasks.Girls communicated better than boys (P = 0.02) and there wasan overall improvement with age (χ2

2 = 12.9; P < 0.01). Whenthe components of communication were examined individually,girls were better at expression (P = 0.02), but not at comprehen-sion, than boys. There was no association between ability atcommunication tasks by place of residence, income group orEnglish-speaking background. Children with a current hearingproblem had a lower total communication score (mean score5.2) than those without (mean score 5.6; P = 0.01), but therewas no difference in the ability of those who had ever had a glueear (5.4) compared with those who had not (5.6; P = 0.20).

Social cognition

Fewer than half of the children (48.3%) were able to carry outsocial cognition tasks with no help or supervision only. As withcommunication, girls were more able (P = 0.05) as were olderchildren (χ2

2 = 14.5; P < 0.01). When the individual compo-nents of social cognition were examined, girls scored betterthan boys on social interaction (P = 0.02) and memory(P = 0.04), but not with problem solving (P = 0.18). However,memory was the only parameter that showed a significantlyhigher score with age (χ2

2 = 24.8; P < 0.001). There was noassociation between ability at social cognition skills and placeof residence, educational level of parents, income group orEnglish-speaking background. There was no associationbetween social cognition abilities and the presence of a currenthearing problem (P = 0.10). Those who had ever had glue earwere no different from those who had not in terms of socialcognition abilities (P = 0.09).

Association between WeeFIM and other measures ofhealth and functioning

Educational placement

Children attended a variety of different types of schools andcombinations of educational settings. Fifty-eight children werefull-time in a mainstream class of a mainstream school.Seventy-two children attended an educational support unit

(ESU) in a mainstream school all or part of the time. Twenty-seven children attended an educational support centre (ESC)full-time or part-time. Forty-four children spent some or all oftheir time in a special school. The mean ± SD WeeFIM scorefor children attending mainstream schools was 106.7 ± 16.1,for those attending an ESU it was 108.2 ± 13.8, for thoseattending an ESC it was 105.1 ± 17.2, and for those attending aspecial school it was 101.0 ± 19.9. These scores were notsignificantly different (P = 0.26). However, if children agedunder 10 years were excluded, children attending both a main-stream school (mean WeeFIM score 114.7) and attending an ESU (112.1) had significantly better scores (P = 0.007; P = 0.03) than those attending a special school (105.8).

Parents of 122 children (60.7%) felt that their children’seducational needs were being met. Those who did not (n = 78),gave a variety of reasons for their unmet needs. These werecategorized into four groups: (i) concern regarding progresswith academic skills; (ii) need for more aide time and schoolresources; (iii) inappropriate programme or more teachertraining required; and (iv) other responses. There was nosignificant difference in WeeFIM scores according to what theparents considered were unmet needs.

Need for practical support

Seventy-eight families (38.1%) used some sort of practicalsupport. Although the proportion of children who required helpwith transfers was very low (3.8%), these families were signif-icantly more likely to use respite care (P = 0.001). The parentsof those children needing more help in the area of socialcognition also used more respite care (P = 0.04). The func-tional abilities of children whose families used domestic help,or sitter or child care services were no different from those whodid not.

Social issues

The parents of just over one-quarter of the children stated thattheir child with Down syndrome had no friends (26.6%), whilenearly two thirds (61.1%) said that they had two or morefriends. Of those with no friends, more required help withtransfers (P = 0.016) and in the area of social cognition (P = 0.05), compared with children with friends. Parents wereasked about the frequency of time the children spent with theirfriends. Just over one-quarter of the children (27.8%) spenttime with their friends outside school hours at least once aweek. There was no relationship between WeeFIM domainstatus and the frequency of seeing friends.

Medical effects

Neither the presence of a heart condition (P = 0.15), bowelcondition (P = 0.26) or current hearing or ear condition (P = 0.80) affected the total WeeFIM score.

DISCUSSION

The aim of the present study was to use a modified version ofthe WeeFIM to provide population-based information from aparental perspective about the functional skills of school-agedchildren with Down syndrome. This type of information ispertinent to those managing children with Down syndrome as well as to medical professionals and genetic counsellors. In

163Functioning in Down syndrome

addition, new parents seek this information so that they canhave a realistic perspective on what to expect from their childin terms of day-to-day functioning. Such information is alsovaluable for policy makers and service planners in consideringthe educational, vocational, long-term care and accommoda-tion needs for people with Down syndrome.

Information provided to both health care providers andfamilies must be based on data that are both current and repre-sentative. There have been major changes in the lifestyles,health care, and integration of people with Down syndrome inthe last two decades.26 These changes are likely to have had animpact on the health and well-being of this population, yetthere have been few recent population-based studies. The onlyAustralian study was not population-based,27 and a Britishstudy (Manchester cohort), although population-based, wassmall with only 117 subjects included.28 The widespreadincrease in uptake of maternal serum screening29 may well leadto a reduction in the birth prevalence of Down syndrome.Therefore, obtaining a large enough sample size for suchstudies in the future may be difficult.

In the present study, we have been able to achieve a current,large population-based sample through the linkage of twodatabases. The Disability Services Commission is the only orga-nization in Australia where there is regular contact throughmedical and specialist services for children with Downsyndrome. A recent study showed that 79% of children with anintellectual disability were registered with the DSC30; however,registration of children with Down syndrome is much higher. TheBirth Defects Registry has been able to achieve high levels ofascertainment by using multiple sources.21 Through the linkageof these two registers, we were able to obtain a nearly completeaccount of children with Down syndrome in Western Australia.

No differences were identified between responders and non-responders in terms of place of residence. However, children ofnon-English speaking parents may have been underrepresentedamong the responders, as participation in the study requiredcompletion of a written questionnaire. Unfortunately, informa-tion about ethnicity was not available. Nevertheless, the highlevel of case ascertainment and the high response fractionmean that the results of this study are likely to be both general-izable and precise.

The field-testing of a self-administered version of the WeeFIM(based on the interview format WeeFIM)5 has been an importantoutcome of the present study. It would have been valuable tocompare the responses from the questionnaire format to those ofan interview format for the same child, but this was not possiblewith our limited study resources. Future research could examinethe reliability between the questionnaire and the original version.Although there are many psychological tests available for themeasurement of intelligence, there is a need for simple, self-administered measures of functioning that are relevant to dailyactivities rather than performance in a formal setting. The devel-opment of this instrument is particularly valuable for obtaining aquick and accurate score where large numbers of subjects areinvolved, such as in the present study. This tool can also be usedfor other disabilities where identification of areas of strengthsand weaknesses is required. It could also be used to examine afuture cohort to determine whether functioning in tasks ofeveryday living have continued to improve in parallel with theimprovements we are seeing in physical health and communityintegration. However, we must caution that the WeeFIM does notreplace more detailed psychological adaptive, communicative, orvocational assessments.

The mean WeeFIM score of subjects in the population exam-ined in the present study is similar to the score achieved with theoriginal interviewer-administered instrument used in a clinicalseries of children with Down syndrome.31 However, the childrenwere slightly older in our study. Inspection of the quotients(comparison with normal scores) shows that functioningimproves with age. In our study there was a significant genderdifference in WeeFIM scores, with girls attaining a higher scorethan boys. Although this gender difference in functioning was notdocumented in the normative sample obtained for WeeFIMcomparisons,12 it has been reported in longitudinal studies ofadult Down syndrome populations.32

Children with Down syndrome characteristically have earlyhypotonia, which often results in significant input from physio-therapists in their early years. In terms of individual domains,we found that mobility skills (both locomotion through spaceand transfers on and off chairs and in the bathroom) of childrenwith Down syndrome were a functional strength. The score onthe mobility portion of the WeeFIM did not differ between agegroups, suggesting that walking skills are fully developed byschool age. Thus, community recreation participation becomesan important area to pursue for children and adolescents.

Performance in self-care skills overall revealed that there are many challenges for children with Down syndrome inachieving full independence in bathing, dressing andgrooming. We also found that there was a significant genderdifference in dressing in the youngest age group, with boysbeing more likely to require assistance. This may be a reflec-tion of volunteered parental assistance rather than of ability.Nevertheless, dressing and grooming skills appeared toimprove with age, so that by 14 years, more than half of thechildren needed no help or supervision to complete these tasks.Children coped relatively well with both bladder and bowelcontrol, and again older children had better control.

Boys with Down syndrome performed significantly less wellthan girls in the area of communication, although this may alsobe true of the general childhood population. Boys in generalhave high needs for assistance with communication andlanguage-related educational support (e.g. reading, writtenlanguage). Measurement of communication using the VinelandAdaptive Behaviour Scale supports this finding.33 It is oftenstated that the presence of ear and hearing conditions may havean impact on communication skills. However, in the presentstudy, although the communication level of those with ahearing loss was slightly lower, it was not significantlydifferent from those without a current hearing problem.

The gender difference seen in communication skills was notevident with problem solving. Social interaction and memoryskills were better in older children, as would be expected;however, ability at problem solving was similar across agegroups.

This study demonstrates the feasibility of using a parent-completed postal version of the WeeFIM for collecting popula-tion survey data in children with developmental disability. Thismay be useful for the longitudinal tracking of such populations,as well as monitoring responses to interventions, and planningservices for children with disabilities.

ACKNOWLEDGEMENTS

The authors thank L Tomov, for her assistance with data entry,the Down Syndrome Association of Western Australia for their

164 S Leonard et al.

165Functioning in Down syndrome

support of the study, and our colleagues at the TVW TelethonInstitute for Child Health Research for their advice and contri-butions to the development of the questionnaire. We also thankthe staff at the Birth Defects Registry and the administration,medical and specialist staff at the Disability Services Commis-sion for their assistance. The contributions made by thefamilies in this study is greatly appreciated.

REFERENCES

1 Mastroiacovo P, Bertollini R, Corchia C. Survival of children withDown syndrome in Italy. Am. J. Med. Genet. 1992; 42: 208–12.

2 Leonard S, Bower C, Petterson B, Leonard H. Survival of infantsborn with Down’s syndrome: 1980–96. Paediatr. Perinat.Epidemiol. 2000; 14: 163–71.

3 Bower C, Leonard H, Petterson B. Intellectual disability inWestern Australia. J. Paediatr. Child Health 2000; 36: 213–15.

4 WeeFIM System Clinical Guide: Version 5. State University ofNew York at Buffalo, Buffalo, 1998.

5 Msall ME, DiGaudio K, Rogers BT et al. The Functional Indepen-dence Measure for Children (WeeFIM). Conceptual basis and pilot use in children with developmental disabilities. Clin. Pediatr.1994; 33: 421–30.

6 Granger CV, Hamilton BB, Keith RA, Zielezny M, Sherwin FS.Advances in functional assessment for medical rehabilitation. Top. Geriatr. Rehabil. 1986; 1: 59–74.

7 Ottenbacher KJ, Msall ME, Lyon N et al. The WeeFIM instru-ment: its utility in detecting change in children with develop-mental disabilities [In process citation]. Arch. Phys. Med. Rehabil.2000; 81: 1317–26.

8 Azaula M, Msall ME, Buck G, Tremont MR, Wilczenski F,Rogers BT. Measuring functional status and family support inolder school-aged children with cerebral palsy: comparison ofthree instruments. Arch. Phys. Med. Rehabil. 2000; 81: 307–11.

9 Philip PA, Ayyangar R, Vanderbilt J, Gaebler-Spira DJ. Rehabilita-tion outcome in children after treatment of primary brain tumor.Arch. Phys. Med. Rehabil. 1994; 75: 36–9.

10 Msall ME, Tremont MR. Measuring functional status in childrenwith genetic impairments. Am. J. Med. Genet. 1999; 89: 62–74.

11 Msall ME, Tremont MR. Functional outcomes in self-care,mobility, communication, and learning in extremely low-birthweight infants. Clin. Perinatol. 2000; 27: 381–401.

12 Msall ME, DiGaudio K, Duffy LC, LaForest S, Braun S,Granger CV. WeeFIM. Normative sample of an instrument fortracking functional independence in children. Clin. Pediatr. 1994;33: 431–8.

13 Vohr BR, Msall ME. Neuropsychological and functional out-comes of very low birth weight infants. Semin. Perinatol. 1997;21: 202–20.

14 Msall ME, Phelps DL, DiGaudio KM et al. on behalf of theCryotherapy for Retinopathy of Prematurity Cooperative Group II.Severity of neonatal retinopathy of prematurity is predictive ofneurodevelopmental functional outcome at age 5.5 years. Pedi-atrics 2000; 106: 998–1005.

15 Liu M, Toikawa H, Seki M, Domen K, Chino N. Functional Independence Measure for Children (WeeFIM): a preliminarystudy in nondisabled Japanese children. Am. J. Phys. Med.Rehabil. 1998; 77: 36–44.

16 Ottenbacher KJ, Taylor ET, Msall ME et al. The stability andequivalence reliability of the functional independence measure forchildren (WeeFIM). Dev. Med. Child Neurol. 1996; 38: 907–16.

17 Ottenbacher KJ, Msall ME, Lyon NR, Duffy LC, Granger CV,Braun S. Interrater agreement and stability of the Functional Inde-pendence Measure for Children (WeeFIM): use in children withdevelopmental disabilities. Arch. Phys. Med. Rehabil. 1997; 78:1309–15.

18 Ottenbacher KJ, Msall ME, Lyon N, Duffy LC, Granger CV,Braun S. Measuring developmental and functional status inchildren with disabilities. Dev. Med. Child Neurol. 1999; 41:186–94.

19 Di Scala C, Grant CC, Brooke MM, Gans BM. Functionaloutcome in children with traumatic brain injury. Agreementbetween clinical judgment and the functional independencemeasure. Am. J. Phys. Med. Rehabil. 1992; 71: 145–8.

20 Leonard S, Bower C, Petterson B, Leonard H. Medical aspects of school-aged children with Down syndrome. Dev. Med. ChildNeurol. 1999; 41: 683–8.

21 Bower C, Rudy E, Ryan A, Cosgrove P. Report of the BirthDefects Registry of Western Australia. King Edward MemorialHospital. Centre for Women’s Health, Perth, 2000.

22 Anonymous. F I L E M A K E R P R O [program], version 3.0. ClarisCorporation, Santa Clara, 1996.

23 Anonymous. STATA STATISTICAL SOFTWARE [program]. Release 5.0Version. Stata Corporation, Texas, 1997.

24 Bland M. An Introduction to Medical Statistics. Oxford MedicalPublications, Oxford, 1987.

25 Armitage P. Statistical Methods in Medical Research. BlackwellScientific Publications, Oxford, 1971.

26 Carr J. Annotation: Long-term outcome for people with Down’ssyndrome. J. Child Psychiatry 1994; 35: 425–39.

27 Selikowitz M. Health problems and health checks in school-agedchildren with Down syndrome. J. Paediatr. Child Health 1992; 28:383–6.

28 Turner S, Sloper P, Cunningham C, Knussen C. Health problemsin children with Down’s syndrome. Child Care Health Dev. 1990;16: 83–97.

29 O’Leary P, Bower C, Murch A, Crowhurst J, Goldblatt J. Theimpact of antenatal screening for Down syndrome in WesternAustralia: 1980–94. Aust. N. Z. J. Obstet. Gynaecol. 1996; 36: 1–4.

30 Wellesley DG, Hockey KA, Montgomery PD, Stanley FJ. Preva-lence of intellectual handicap in Western Australia: a communitystudy. Med. J. Aust. 1992; 156: 94–102.

31 Msall ME, Roehmholdt SJ, DiGaudio KM, Duffy LC. Functionalindependence of school age children with Down syndrome. Pedi-atrics 1992; 31: 13A.

32 Shepperdson B. Two longitudinal studies of the abilities of peoplewith Down’s syndrome. J. Intellect. Disabil. Res. 1995; 39:419–31.

33 Dykens EM, Hodapp RM, Evans DW. Profiles and development ofadaptive behavior in children with Down syndrome. Am. J. Ment.Retard. 1994; 98: 580–7.