Measuring Developmental and Functional Status in Children with Disabilities, by K.J. Ottenbacher, M.E. Msall, N. Lyon, et al, Developmental Medicine and Child Neurology, 1999;41:186–194.
Functional skill assessment has become a priority in pediatric clinical, educational, and research settings. It provides valuable information about a child's ability to function within specific environments with a focus on the amount and type of support needed. Functional assessments, particularly ones that are fast and easy to administer, are necessary for age-appropriate program planning, evaluation, documentation, and reimbursement. In general, functional assessments focus on a child's ability to perform daily tasks. This stands in contrast to traditional developmental assessment tools that focus more on the process used to perform tasks. Although both types of assessments are currently used, the relationship between functional and traditional developmental assessments for children has not been studied systematically. The purpose of this study was to compare the scores obtained from a functional assessment tool, the Functional Independence Measure for Children (WeeFIM), and two widely used developmental pediatric assessment tools, the Battelle Developmental Inventory Screening Test (BDIST) and the Vineland Adaptive Behavior Scales (VABS).
The subjects were 205 children between the ages of 11 and 87 months who were receiving intervention for developmental disabilities. Subjects were recruited from three facilities for early childhood education and developmental disability rehabilitation in western New York. The most common medical diagnoses in the sample included cerebral palsy, prematurity, Down syndrome, spina bifida, epilepsy, and genetic disorders. Severity of disability ranged from mild to extreme as identified by scores from three standardized developmental assessment measures. A sampling plan was implemented to ensure an even distribution of type, age, and severity of disability within the sample.
The functional assessment tool used in this study was the WeeFIM, a measure developed to assess and track the amount of assistance a child needs to perform daily tasks. Eighteen items cover three general domains of self-care, mobility, and cognition and are ranked from complete independence (score = seven) to complete dependence (score = one). The WeeFIM can be administered either through direct observation or interview. It can be used in a variety of health, developmental, educational, and community settings. Advantages of the WeeFIM include an emphasis on consistent actual performance, usefulness for multiple disciplines, versatility in administration methods, minimal number of items, and inclusion of the use of adaptive equipment. Validity and reliability of the WeeFIM have been shown to be adequate.
The first of two developmental assessments used in the study was the BDIST, a shortened form of the Battelle Developmental Inventory (BDI). The BDI assesses the five developmental domains of personal-social, adaptive, motor, communication and cognition. The BDI is used to examine a child's skill level in performing functional skills but does not take into account the amount of assistance. The BDIST has been shown to be a strong predictor of BDI scores and was the form used in this study because of its shorter administration time of 30 to 35 minutes. The BDIST consists of 96 items and has been shown to be reliable and valid.
The second developmental assessment used in this study was the VABS. This tool was designed to assess personal and social sufficiency of individuals with and without disabilities by evaluating behavior in the domains of communications, daily living skills, socialization, motor skills, and an optional domain of maladaptive behavior. Each of the 301 items is ranked according to the child's ability to perform the skill successfully, ranging from zero points for never performing successfully to two points for usually performing successfully. Derived scores are figured based on information from national standardized samples of norm groups. The VABS must be administered by professionals with graduate degrees and takes 45 to 60 minutes to administer. The VABS has been shown to have internal consistency, test-retest, and interrater reliability, construct and concurrent validity.
The three assessments were administered according to their established protocols. The WeeFIM was administered to all 205 subjects. By random assignment, the BDIST was administered to 101 subjects and the VABS to the remaining 104 subjects. Either a parent or teacher who knew the subject well was interviewed for both assessments within three weeks of administering the assessment. A pediatric nurse with 20 years’ experience in developmental disabilities performed all the BDIST and VABS tests and a majority of the WeeFIM interviews. All other raters were health, developmental, or rehabilitation professionals with at least three years’ experience who had undergone the established WeeFIM training and testing protocol. All raters were blind to the child's diagnosis or scores on the initial developmental tests before administering the assessments.
Spearman correlation coefficients (r) were calculated to compare the various sub scores and total scores between the WeeFIM and the BDIST or the VABS. All of the correlations between total scores for the three instruments were above 0.70 and were statistically significant (p < 0.05). The correlation between the WeeFIM and the BDIST (r = 0.92) was slightly higher than that between the WeeFIM and VABS (r = 0.89).
Correlation coefficients between subscales ranged from 0.42 to 0.92. Only four of 36 correlations between the WeeFIM and BDIST were below 0.70. Nine of the 20 correlations between the WeeFIM and VABS were below 0.70. All of the correlations between the WeeFIM total scores and BDIST subscale scores were strong, ranging from 0.83 for the cognitive subscale to 0.94 for the adaptive behavior subscale. Similarly, all of the correlations between the WeeFIM totals and the VABS subscales were strong, ranging from 0.72 for communication to 0.91 for daily living skills.
No statistically significant correlation was found between age and severity of disability (r = 0.05) or between socioeconomic status and severity of disability (r = 0.21) as measured by the initial developmental assessments. As anticipated, significant correlations were found between age and the subscale and total scores, particularly in gross and fine motor skills.
When the WeeFIM scores were graphically plotted in scatterplots with the BDIST and the VABS, linear relationships were noted. In addition, there was evidence that a disproportionate number of subjects scored at the lower end of the scale. The lower scores were correlated as expected with younger ages. Previous research by Ottenbacher and colleagues, 1 however, showed that this limited range of ratings for younger children does not adversely affect the reliability of the WeeFIM.
The purpose of this study was to examine the relationship between children's performance on the functional assessment tool, WeeFIM, and two developmental assessment tools, the BDIST and VABS. The strong correlations between the WeeFIM total scores and the total and subscale scores of the other two tools suggest that the WeeFIM, BDIST, and VABS are measuring similar skill areas. It is unclear from this study, however, whether the specific skills that were assessed similarly using these three tools are actually subsets of the same construct. To assess the WeeFIM construct validity using these two assessment tools would require a comprehensive factor analysis of the combined items from all three instruments. Nonetheless, the results indicate that the WeeFIM provides basic information that is similar, although less comprehensive, than that supplied by the BDIST and VABS.
Limitations and Implications
There are several important implications of these results. First, the WeeFIM requires less time, effort, and training to administer than the BDIST or VABS. The WeeFIM can be administered in approximately 15 minutes by a trained interviewer. The BDIST takes 30 minutes to administer and must be administered in a specific order, a requirement that can be restrictive when dealing with young children. The VABS takes an hour to administer and requires that the tester has at least a graduate professional degree and considerable pediatric experience. Although the WeeFIM may be less comprehensive, it is a more practical way of attaining similar information in clinical settings.
A second implication is that the information provided by the WeeFIM may help to promote better understanding and agreement among different disciplines when planning interventions. In general, educational specialists as well as families tend to focus on function when planning programs and assessing progress, whereas medical and rehabilitation specialists have tended to focus on the physical impairments and etiology as measured by traditional developmental assessments. This difference may lead to unnecessary tension between disciplines. Because the WeeFIM supplies information that is similar to that in the BDIST and VABS and is framed within a functional context, it may help to eliminate unnecessary tension and decrease the number of assessments required overall.
This study has limitations. The first concerns the administration and scoring of the WeeFIM. It is unknown if an administrator's discipline has a significant impact on their scoring of items in the test, yet this study used raters from different disciplines. Moreover, there was no interrater reliability analysis of the raters to suggest that scoring was consistent. A second limitation concerns the age range of the subjects. This study included children from 11 to 87 months, yet the WeeFIM can be used for older children with disabilities up to the age of 18 to 21 years. It is unknown if similar correlations would be seen with this older population. Caution should be exercised when applying the results of this study to older children with disabilities.
Despite these limitations, this study has provided important insight into the value of the information provided by the functional assessment WeeFIM. The information attained is functional, similar to that provided by developmental assessments, and can be used by numerous disciplines. In addition, it is fast and easy to administer, a quality that is critical in today's healthcare environment. This study supports and expands upon past research that the WeeFIM has promise as a useful pediatric assessment tool in both healthcare and educational settings.
© 2003 Lippincott Williams & Wilkins, Inc.