|Home | About | Journals | Submit | Contact Us | Français|
Activity limitations are an important and useful dimension of disability, but there are few validated measures of activity limitations for adolescents and adults with developmental disabilities.
To describe the development of the Waisman Activities of Daily Living (W-ADL) Scale for adolescents and adults with developmental disabilities, and systematically evaluate its measurement properties according to an established set of criteria.
The W-ADL was administered among four longitudinally-studied groups of adolescents and adults with developmental disabilities: 406 with autism; 147 with fragile-X syndrome; 169 with Down syndrome, and 292 with intellectual disability of other or unknown origin. The W-ADL contains 17 activities and each is rated on a 3-point scale (0=“does not do at all”, 1=“does with help”, 2=“independent”), and a standard set of criteria were used to evaluate its measurement properties.
Across the disability groups, Cronbach’s alphas ranged from 0.88 to 0.94, and a single-factor structure was most parsimonious. The W-ADL was reliable over time, with weighted kappas between 0.92 and 0.93. Criterion and construct validity were supported through substantial associations with the Vineland Screener, need for respite services, caregiving burden, and competitive employment. No floor or ceiling effects were present. There were significant group differences in W-ADL scores by maternally-reported level of intellectual disability (mild, moderate, severe, profound).
The W-ADL exceeded the recommended threshold for each quality criterion the authors evaluated. This freely-available tool is an efficient measure of activities of daily living for surveys and epidemiological research concerning adolescents and adults with developmental disabilities.
Developmental disabilities are heterogeneous and life-long disorders that are often characterized by problems related to the functioning of the brain or senses and include genetic disorders that affect cognition, behavior, and multiple body systems. (1–3) The causes and consequences of these conditions may vary widely, even within a particular ‘type’ of disability. For example, some adults with autism experience severe limitations in basic tasks necessary for independent living, while others have no limitations in these areas (but may still struggle in social situations). (4) The nature and severity of these limitations are important for providing appropriate services and interventions, for research, and for informing public health policy. (5, 6) There are, however, few freely-available and high quality tools for measuring activities among adolescents and adults with substantial developmental disabilities. We sought to evaluate a new daily activities measure for this population that is feasible for surveys and epidemiological research.
Several widely-used theoretical frameworks of disability account for and describe aspects of disability beyond a medical classification. (7–9) The World Health Organization’s International Classification of Functioning, Disability, and Health (ICF) describes disability at three different levels: impairments (problems or deviations in the structure or functioning of a body part), activity limitations (difficulty executing tasks or actions), and participation restrictions (problems with involvement in life situations). (10) An innovative aspect of the ICF model is the recognition that both personal and environmental factors may affect the experience of disability. Importantly, the ICF distinguishes between “performance” (what an individual does in his or her current environment) and “capacity” (an individual’s ability to perform a task or action, the highest probable level of functioning) for activities and participation. (10)
A linear or consequential relationship between different levels of disability (e.g., from impairment to activity limitations) is neither presumed nor inevitable, however, adolescents and adults with developmental disabilities are likely to experience activity limitations. (11) In addition to the variability within diagnostic categories for many developmental disabilities, other co-occurring health conditions, social support, and the opportunities and constraints of the physical environment may contribute to an individual’s performance of activities. Measurement of the complex relationships between different aspects of disability is greatly aided through the use of precise language and a strong conceptual framework, such as that offered by the ICF.
Particularly for adolescents and adults with known developmental disabilities, a greater emphasis on activities offers an opportunity to acknowledge that “what people do is more important than the expectation that they do things ‘normally’,”(12) although this has not been the focus of most available measures. Other problems with available measures include “floor” effects and developmental-level anchoring which limit the usefulness of such measures. Specifically, some individuals with autism or intellectual disability are likely to score at the extreme minimum of population-normed measurement tools, and previous studies have suggested there are important differences among individuals scoring at the “floor” on standardized measures. (13–15) Some measures require raters to first determine an individual’s “mental” or “developmental” age, and then administer items intended for typically-developing individuals of that developmental level. This can result in assessment of adults with developmental disabilities with items that were designed for young children. For adults with substantial limitations, sensitive and age-appropriate measures are needed to detect meaningful changes over time and to distinguish between people with different needs.
Despite the widespread recognition that activity limitations represent an important dimension of disability, we are not aware of any freely-available and thoroughly-evaluated tools for measuring activity limitations in adults with developmental disabilities. Activities of Daily Living (ADL) scales have been used extensively in research and clinical applications, often for individuals recovering from surgery or a stroke, or to measure functional decline associated with aging and dementia. Some US states determine service eligibility and financial assistance to adults with limitations on the basis of ADL scores; in 2009, the state of Montana provided disability-related financial support at a rate of $34 per month for each point scored on a scale of ADL items. (16) ADLs are also considered durable indicators of activity limitations and consistent with the World Health Organization’s dimensional framework for disability. (17) Even though ADL-based measures are well-known and presumed to be useful, there is a paucity of rigorous research supporting the reliability and validity of ADL scales. (18) Establishing the quality of an ADL measure intended for populations with developmental disabilities could help articulate the needs of adults with developmental disabilities and facilitate better research and service provision.
The purpose of this paper is to describe the development and evaluate the measurement properties of the Waisman Activities of Daily Living (W-ADL) Scale for adolescents and adults with developmental disabilities. To guide our evaluation, we applied a widely-cited set of quality criteria for health measurement scales. (19)
This analysis makes use of data collected by three longitudinal studies comprised of four different disability groups (autism, Down syndrome, fragile X, and intellectual disability of unknown etiology), that together include 1,014 study participants. Although these studies were not prospectively designed for the development of the W-ADL, they contain a wealth of information related to the health and functioning of adults with developmental disabilities. This analysis utilized relevant items and measures when they were available in each of the studies. The study populations and selected measures used in the present analysis are briefly described here; detailed descriptions have been previously published and are cited below.
Data regarding 406 adolescents and adults with autism (referring to all autism spectrum disorders) initially were included in an ongoing longitudinal study including 9 points of measurement spanning a 12 year period. (20) The median age of the participants at Time 1 was 18 years (range: 10 – 52), and 73% of the participants were male. By parent report, 6% also had cerebral palsy and 23% had epilepsy or a seizure disorder. Independent autism diagnoses were confirmed with a research-administered Autism Diagnostic Interview-Revised. (21) The W-ADL was completed at Times 1, 4, 7, and 8; approximately ten years elapsed between data collection between Times 1 and 8. Data collection for Times 7 and 8 occurred approximately 18 months apart. The Vineland Screener, a measure of adaptive behavior, was administered at Time 4. The Vineland Screener is highly correlated with the full Vineland Scales (correlation coefficients ranging from 0.87–0.98). (22) Employment and education were coded according to Taylor and Seltzer’s Vocational Index for individuals with autism. (23)
Data regarding 147 adolescents and adults with fragile X syndrome were included in the first round of data collection in an ongoing longitudinal study. (24) The median age at the time of interview was 18 years (range: 12–48), and 82% of the participants were male. By parent report, 2% also had cerebral palsy and 11% had epilepsy or a seizure disorder. Fragile X was ascertained through genetic test results from the participants’ medical records. Data from the first time point are currently available. Mothers’ perceived caregiving burden for their adolescent or adult child with fragile X was assessed using the Zarit Burden Scale. (25)
Data regarding 169 adolescents and adults with Down syndrome and 292 adolescents and adults with other intellectual disabilities were included in a 10-year longitudinal study. (26, 27) The median age of the participants with Down Syndrome at Time 1 was 31 years (range: 15–55), and 61% were male. Down syndrome was ascertained via parent report. The median age of the participants with other intellectual disabilities at Time 1 was 35 years (range: 17–66), and 50% were male. Of the 292 individuals with intellectual disability (but not Down Syndrome), maternal report indicated that 23% had cerebral palsy and 13% had epilepsy (4% were reported to have both). Of the 169 with Down syndrome, 1% also had cerebral palsy and 2% had epilepsy or a seizure disorder. Data were collected at eight time points (approximately 18 months apart and more than ten years total between Times 1 and 8), and the W-ADL was given at each time. Mothers were the primary respondents for this study; however, a subset of fathers concurrently responded to W-ADL items at Time 6. Stanford-Binet intelligence tests were administered at Times 2 and 3 by professional clinicians and graduate students in educational psychology.
Terwee and colleagues proposed eight criteria to comprehensively evaluate the measurement properties of an instrument: content validity, internal consistency, criterion validity, construct validity, reproducibility, responsiveness, floor or ceiling effects, and interpretability. We summarized the purpose of each of the criteria and applied each criterion to the W-ADL scale using the methodology recommended by Terwee and colleagues.
This analysis supports the reliability and validity of the W-ADL in research on adolescents and adults with developmental disabilities. By evaluating this measure in the context of three large and established longitudinal studies--that together encompass four major clinical groups (autism, fragile X, Down syndrome, and intellectual disability of no known etiology)--we were able to demonstrate that the W-ADL exceeded Terwee and colleagues’ recommended measurement criteria. To our knowledge, this is one of the most comprehensive analyses of the measurement properties of an ADL-based questionnaire.
The 17-item W-ADL appears to be an extremely efficient and practical measure for assessing adults with relatively substantial disabilities. We observed a strong association with the Vineland Screener, which contains 90 items and displayed a floor effect in the autism group. Similarly, the W-ADL was associated with clinician-assessed Stanford Binet IQ scores among the adults with Down syndrome and/or intellectual disability, although many participants scored at the floor of the Stanford Binet or were deemed “untestable”. In contrast, the W-ADL was able to detect group differences between subjective or categorical “levels” of intellectual disability, including between severe and profound intellectual disability. Whereas many of the study participants scored at the extreme low end of the population distribution on other measures, substantial differences within these groups emerged when using an appropriate and adult-oriented tool.
The majority of the participants in these studies had significant intellectual disabilities or other serious impairments. Therefore, we have limited information related to the level of performance equal to the “ceiling” of the scale. There were 26 participants with autism that had IQs greater than 100 (at Time 4), but the mean W-ADL score for this group was 28.3 and only 1 participant scored the maximum of 34. Future work should consider items that may extend the “ceiling” of the W-ADL. The selection of additional activities should be informed through soliciting suggestions from individuals with the disability (when feasible) and their family members or caregivers. A planned future study to be conducted by our research group will directly interview participants about the determinants of their quality of life, and their responses will help inform future development of the W-ADL.
This study has some limitations, as the longitudinal studies were not designed for the purpose of evaluating the W-ADL. Although the weighted kappa (or ICC) for consecutive time points was high, real changes in the performance of activities between measurements were interpreted as part of the measurement error, possibly inflating our estimate of the SDC. Increasing the number of items and the range of responses for each item would likely increase the quadratic-weighted kappa and, in turn, decrease the SDC. (36) However, these improvements in measurement statistics would likely come at the expense of efficiency (by requiring more items) or the interpretability of the scale units (for instance, if items were scored on a scale from 0 to 5). Another limitation is that our assessment of the W-ADL was based almost exclusively on parent-report approaches. The vast majority of participants were living at home at the start of the study. Future work could examine W-ADL reliability between non-parent respondents.
Like other ADL-based measures, further work is needed to determine the minimum important change in scores. (18) One recommended method for determining the minimum important change is through an anchor-based approach. (37) However, because each point on the W-ADL represents partial or complete independence in a specific activity, any change in score could be perceived as important. The smallest important change may depend on the specific activities that change as well the context in which the changes occur. Additional work is also needed to develop inferences that could be made on the basis of W-ADL scores. We found that the W-ADL was associated with perceived caregiving burden, parent-reported need for services, and educational and vocational attainment; these associations could form the basis for a deeper analysis of the W-ADL’s predictive and discriminative properties. (e.g., Does a score at a certain age predict future employment or living situation? How well do scores discriminate between people that do and don’t qualify for certain services?)
A major strength of this analysis is that it utilizes some of the largest and longest running studies of the phenotype of autism, fragile X, Down syndrome, and intellectual disability in adulthood. In addition to the W-ADL items, numerous other instruments and questionnaires were administered throughout the studies affording the opportunity to examine different aspects of validity. Another important advantage of this study is the inclusion of adults with different categorical diagnoses. While the symptomatology and etiology between fragile X, Down syndrome, autism, and intellectual disability may differ, the W-ADL performed similarly across groups in terms of measurement properties.
An alternative measure to the W-ADL is the Scales of Independent Behavior-Revised (SIB-R) Short Form. (38) The SIB-R Short Form is a norm-referenced, 40-item measure of adaptive behavior used to determine a person’s level of functioning, and it has been regularly used to assess adults with intellectual or developmental disabilities. Each item is rated on a 4-point scale corresponding to either the proportion of time a person performs—or could perform—a task without help or supervision. Whereas the SIB-R focuses on capacity to carry out a task, the W-ADL focuses on the actual performance of activities which, as described by the ICF, is concerned with an individual in his or her current environment (which may include assistive devices or other support). The SIB-R Short Form also tends to contain more narrowly-defined tasks compared to the W-ADL. For example, the W-ADL contains an item on getting dressed, while the SIB-R Short Form contains multiple items about dressing, including an item on tying shoelaces. Someone wearing shoes with hook-and-loop fasteners (i.e., Velcro), could perform the activity of getting dressed without having the capacity to tie shoelaces, and thus would not be credited in the SIB-R. These conceptual differences could make one measure preferable over the other, depending on whether the goal is to measure specific skills or the lived experience of a person with disability.
The W-ADL may also have utility in disability screening or large surveys, particularly in low-resource settings. Activity- or ADL-based measures have been used in disability screening and their interdisciplinary and cross-cultural relevance offer practical utility in many situations. (17, 39–41). Additional work is needed to assess whether these items are relevant in different contexts and cultures, and to consider how specific activities may be more or less essential to—or representative of—typical life situations. Here, the ICF would serve as a useful framework for identifying important new activities, for understanding relationships between different levels of disability, and for identifying environmental barriers and facilitators for the performance of activities.
In summary, the W-ADL Scale is a freely-available (at www.waisman.wisc.edu/family/WADL) measure of activities of daily living for adolescents and adults with developmental disabilities. As a research tool, the W-ADL demonstrated excellent measurement properties across four well-characterized groups of developmental disabilities including autism, Down syndrome, fragile X, and intellectual disability.
This manuscript was supported by grants from the National Institute on Aging (R01 AG08768, M.R. Mailick, PI) and from the National Institute of Child Health and Human Development to the IDDRC at the University of North Carolina (P30 HD003100-S1) to support a Fragile X Research Center. The Fragile X Research Center has three additional sites (Research Triangle Institute International, the University of Wisconsin-Madison, and University of Kansas). The present analysis was based on data collected at the UW-Madison Waisman Center (M.R. Mailick, PI) site. We are extremely grateful to the 1,014 families who participated in this study; without their generous support and commitment, our research would not be possible. We also aclnowledge the support we received from the Waisman Center (P30 HD03352, M.R. Mailick, PI).
This study was presented in abstract/poster form at the 45th annual Gatlinburg Conference on March 7, 2012, and at the International Meeting for Autism Research on May, 18 2012.
Conflicts of Interest