Assessing Quality of Life of Children and Youth with Disabilities: A Review of Available Measures
What was the purpose of this study?
This purpose of this study was to systematically document and evaluate existing measures of quality of life for children in order to help guide service providers and others in determining which measures offer the best utility to centres and agencies that provide rehabilitation services to children with disabilities.
How did we define Quality of Life (QoL)?
The term "quality of life" has evolved over time; however, there is still no consensus on how QoL is conceptualized or measured, and there are many definitions. The early literature was mostly focused on objective indicators of children's health status with an emphasis on physical function. This idea broadened particularly with the introduction and widespread acceptance of the World Health Organization's definition of health as being more than the absence of disease; rather it encompasses physical, social and psychological dimensions of health. These three aspects became the major domains of the many measures of health-related quality of life (HRQL) that have been developed. There are generic HRQL measures that assess aspects of health deemed to be of universal importance, and there are disease-specific measures that address concerns or issues pertinent to a particular disease or disability. HRQL measures often assess the impact of the disorder or disease, and include items about limitations or restrictions in function.
It is recognized that QoL is fundamentally a holistic concept that goes beyond the health dimension. It has been suggested that it is impossible to differentiate what conditions or circumstances influence what aspect of one's life, and that QoL is affected not only by the disability but also by the person's experiences. The following definition by Wallander and colleagues (2001) shows this more comprehensive view "Quality of life is the combination of objectively and subjectively indicated well-being in multiple domains of life considered salient in one's culture and time, while adhering to universal standards of human rights" (p. 574). Another more concise definition is offered by Renwick and Brown (1996, as cited in Zekovic & Renwick, 2003): "[QoL is] the degree to which a person enjoys the important possibilities of his or her life" (p. 24). The scope of HRQL and holistic measures reflect that QoL is a multi-dimensional construct and may include dimensions of physical, emotional, social, productive and material well-being, interpersonal relations, personal development, self-determination, social inclusion, and individual rights (Felce & Perry, 1995, as cited in Wallander, Schmitt, & Koot, 2001; Schalock, 1996, as cited in Zekovic & Renwick, 2003).
What did we do?
The project consisted of several phases including:
- A comprehensive review of the literature;
- An in-depth evaluation of selected measures using structured guidelines; and
- Input from service providers in the children's treatment centres and agencies.
Comprehensive Review of the Literature:
A literature search was conducted online through several databases (MEDLINE, CINAHL, HAPI, PsycINFO and HealthStar from 1990-2004) to obtain articles pertaining to the concept of quality of life (QoL) and to identify assessments, instruments or tools to measure QoL and their related literature. The following search terms were utilized: quality of life, health-related quality of life, well-being, pediatrics, child, adolescent, review, assessment, and measurement.
Other strategies were used to ensure that related literature was not missed. These included: searching the above databases with the names of the authors of selected measures; examining all issues of the online Quality of Life Research journal; reviewing reference lists from all retrieved articles; searching the online MAPI Research Institute's free QoL measures database and their Quality of Life Newsletter; examining the All About Outcomes CD and other text resources; and reviewing conference abstracts.
A total of 107 generic and disease-specific measures were identified. Thirty-two measures were eliminated from the review if they were only applicable to children with diagnoses that would not typically be seen at children's rehabilitation centres in the province (e.g., asthma, cancer, celiac disease, cystic fibrosis, diabetes, dwarfism, skin conditions and weight issues). Another 39 measures were determined not to be relevant because they focused primarily on health status, were developed for adults, or were outdated in their concepts. Thus the list was narrowed to 33 measures that were reviewed. ( See Table 1 )
Evaluation of Selected Measures:
Four members of the research team participated in an initial review of two measures to establish procedures and consistency using the Outcome Measures Rating Form. This rating form covers information about the focus of the measure, clinical utility, scale construction, standardization, reliability, and validity. Based on pilot reviews and our reading of the conceptual articles, we modified the rating form by adding attributes and specific examples relevant to QoL. We also added an "overall utility" rating to judge the combination of clinical utility and psychometric soundness of the measure. Our interest in this project was to develop a resource not just of measures' descriptive features, which are essential in selecting a measure, but more importantly to provide an evaluation of these instruments in an easy-to-read tabular format for comparisons across measures.
As preparation for the reviews, the two rehabilitation science graduate students and the project coordinator, who was a research occupational therapist and instructor in evidence based rehabilitation, reviewed several of the same measures. Following the third measure, there was consistency in ratings and the students felt competent in their reviews to continue independently. No formal assessment of inter-rater reliability was performed. Students were encouraged to confer with the project coordinator if they were unsure of a rating so that a consensus could be reached. The relevant literature on 33 measures was reviewed and the Outcome Measures Rating Forms completed by either the students or the project coordinator. A database was developed to document the descriptive information about the measures and ratings for clinical utility, reliability and validity, and overall utility.
Input from Service Providers:
In October 2004, preliminary information from this study was presented in a workshop format at the annual conference of the Ontario Association of Children's Rehabilitation Services (OACRS). This session was well attended by approximately 40 participants including service providers, managers, parents of children with disabilities, researchers, educators and government decision-makers. We discussed the conceptual and practical issues in selecting and using QoL measures for children with disabilities. A handout was provided detailing eight measures that had been reviewed in depth. We also probed for information about measures the participants were familiar with since we wanted to understand the benefits and limitations of these measures from their perspective and to ensure none were missing from our database. There was keen interest in this topic and in our plan to provide an easy-to-read document summarizing all the measures we review. Such a document would allow users to assess the strengths and limitations of each measure and determine the one that would best suit their purposes.
What was found?
Breadth of Measures for Children
The 33 measures reviewed were primarily focused on aspects of children's lives within the physical, social, emotional/psychological, and school domains with the latter incorporating cognitive and productivity dimensions. The content of the items varied greatly from measure to measure. Of the 33 measures, 12 were developed to be completed by children and adolescents themselves, with 13 having parallel versions for parents or caregivers. Seven measures were proxy-completed for infants and young children. The majority of the measures are self-completed questionnaires while two are only administered through an interview, and five have the flexibility to be completed by either method. Eighteen of the measures are available in English only, the remaining 15 measures are available in multiple (particularly many European) languages.
Table 2 lists the descriptive features of the measures including what population and age range the measure was created for, and who the respondent is. The major domains covered in each of the measures and the numbers of items in the measure are in Table 3. The information about administration and scoring of the measures are found in Table 4 and Table 5. The psychometric ratings about reliability are in Table 6 and those about validity are presented in Table 7. Table 8 displays the ratings for overall clinical utility and overall utility. In some tables, columns contain an asterisk (*) to indicate "missing" data. This could mean that the information was not available at all or was unavailable at the time of our collecting the measures, not clearly specified in the documentation, not done, or currently under development.
Overall utility was rated on a 5-point scale (poor, fair, adequate, good or excellent). This rating took into account the measure's clinical utility and the psychometric properties of reliability and validity. For example, an excellent rating meant that the measure had excellent clinical utility, reliability and validity. A rating of good for overall utility meant adequate to excellent clinical utility and adequate to excellent reliability and validity.
Of the 33 measures, 11 had overall utility ratings of good; none had an excellent rating. Many of the measures reviewed hold great promise but are relatively new and have limited documentation to determine clinical utility or have only preliminary evidence of reliability and validity. Further reliability and validity information should enhance our understanding of their usefulness.
These ratings should be used as a starting point to guide you to a measure you think may be useful for your purpose. While we aimed for methodological rigour, and used the best materials to which we had access for this overview we were not always able to obtain primary sources or manuals for each measure. We were also limited to measures developed in or translated into English. As new materials are always becoming more available, we suggest that a user update this search and examine more thoroughly the literature on a measure selected for one's clinical or research purposes.
What did we conclude?
The field of QoL is expanding as evidenced by a variety of definitions and the number of measures available and under development. We concur with Wallander et al. (2001) for the need to reach conceptual consensus and to focus efforts on measures that have broad applicability across all children and youth that would allow comparisons across groups and settings.
Sound psychometric properties are prerequisites for using any measure. Selection of a QoL measure for clinical or research purposes is also dependent on one's purpose, and on one's definition of QoL and the domains of interest. Other considerations include the age of the child/youth and whether the measure is geared to their developmental level, how items were generated, whose perspective is sought, features of clinical utility, culturally appropriateness, evidence of acceptable reliability and validity, and ability to detect change. We hope that the attached tables provide a foundation for assessing and choosing a QoL measure. For an in-depth exploration of the conceptual and methodological issues or other reviews, there are several excellent articles noted in the Bibliography.
How are the findings useful?
The interest in QoL is consistent with changes in the field of pediatric health care and rehabilitation. There has been a shift from a more biologically-based approach where treatments were focused primarily on changing the child, towards interventions aimed at broader goals such as enhancing community participation and improving a child's QoL. We think the findings of this project will be useful in the following ways.
The interest of service providers in QoL demonstrates that children are viewed not solely by their disability but within the context of their whole lives. It should be reassuring that use of these measures attempts to better describe and understand a child's life experiences.
To Service Providers
QoL measures can be used for several purposes. They may be useful in understanding a child's needs and goals, and help one to consider how a treatment recommendation may impact the child's (and family's) QoL. Also, these measures could be used to evaluate specific interventions. The detailed tables are a resource to guide service providers in choosing measures based on their needs and purposes.
To Managers and CEO
QoL is seen as a valid area for clinical intervention and can assist in planning appropriate treatment. By looking at these aspects of the impact of treatment and services, it also addresses issues of quality of care.
To Policy Makers
QoL is a global outcome that can be used across ages and diagnostic conditions. For these reasons, it may be a useful tool to evaluate health policy interventions.
Depending on their needs and purpose, there are several measures available to conduct QoL investigations, which may include exploration of the factors that impact children's QoL.
Comprehensive methods to interpret meaningful differences in QoL data are crucial to the further development and use of this concept for children, youth, their families, service providers and decision makers. Marquis and colleagues (2004) add to the efforts to better understand the issue of clinically significant difference and provide guidance around how to approach this aspect of a measure's utility. A new and exciting aspect in the QoL field is the development of family QoL measures (see, for example, Park et al., 2003). This work is significant because it recognizes the dependency of a child's QoL on the context of his/her experiences with parents and siblings, as well as with others in environments such peer group, school and community (Matza, Swensen, Flood, Secnik & Kline Leidy, 2004; Renwick, Fudge Schormans & Zekovic, 2003).
We greatly appreciate the funding received from the Easter Seals Society to conduct this project. Many thanks go to Laura Shapiro and Omar Aboelela for their time and effort in conducting the reviews amidst their studies and clinical placements. We value the work of others: Trisha Stewart who helped in the beginning of the project with a variety of research and administrative tasks; Salina Jaffer who assisted with setting up the database of measures, Marjorie Stone who performed the data entry, and for the thoughtful feedback of several reviewers of this report. We also wish to acknowledge the many authors of measures who generously shared their materials with us.