- Top of page
- What this paper adds
- Overall purpose
- Supporting Information
Aim Our aims were to (1) describe the conceptual basis of popular generic instruments according to World Health Organization (WHO) definitions of functioning, disability, and health (FDH), and quality of life (QOL) with health-related quality of life (HRQOL) as a subcomponent of QOL; (2) map the instruments to the International Classification of Functioning, Disability and Health (ICF); and (3) provide information on how the analyzed instruments were used in the literature. This should enable users to make valid choices about which instruments have the desired content for a specific context or purpose.
Method Child health-based literature over a 5-year period was reviewed to find research employing health status and QOL/HRQOL instruments. WHO definitions of FDH and QOL were applied to each item of the 15 most used instruments to differentiate measures of FDH and QOL/HRQOL. The ICF was used to describe the health and health-related content (if any) in those instruments. Additional aspects of instrument use were extracted from these articles.
Results Many instruments that were used to measure QOL/HRQOL did not reflect WHO definitions of QOL. The ICF domains within instruments were highly variable with respect to whether body functions, activities and participation, or environment were emphasized.
Interpretation There is inconsistency among researchers about how to measure HRQOL and QOL. Moreover, when an ICF content analysis is applied, there is variability among instruments in the health components included and emphasized. Reviewing content is important for matching instruments to their intended purpose.
The effectiveness of health programmes and interventions has traditionally been decided by clinicians, administrators, researchers, and policy makers with little input from patients.1–4 With more children living with chronic, episodic, or progressive health conditions,5,6 health professionals need to assess the impact of care on day-to-day life, rather than restrict measurement to biomedical issues, such as morbidities, that might not resolve.7 In parallel, health care resources are becoming an increasingly precious commodity around the world and difficult decisions need to be made about the focus and allocation of health services.8 The impact of these services as reported by children and their families can provide insight that can help with these difficult decisions.
Scaled questionnaires that assess health, health-related quality of life (HRQOL), and quality of life (QOL) are valuable tools for capturing child and parent experiences. When these validated questionnaires are administered directly to children or their families without clinicians, they are called patient-reported outcomes (PROs).9 Patient-reported outcome instruments are essential patient appraisal tools, but they are often developed and implemented without a firm conceptual basis or a clear definition of what they are intended to measure. Many PROs were developed before there was clarity in the literature on what constituted functioning, disability, and health (FDH) rather than HRQOL or QOL.
This lack of agreement about definitions affects the ability to interpret the results collected from different instruments.7,10,11 The results obtained from one HRQOL instrument applied in particular context cannot easily be compared with results obtained with another HRQOL instrument if they are composed of different content. Envision the example of anti-spasticity therapy being provided to similar groups of children but measured with different instruments; one showing change following intervention, the other showing no change. Without detailed knowledge of the conceptual basis and content of both PROs, it is difficult to interpret whether the observed change or lack of change is due to the intervention or to the instrument (or possibly both).
Many clinicians and researchers choose to adopt the most popular instruments for their studies in the belief that use of a generic PRO will overcome measurement discontinuity between clinical situations. The reasoning behind this is that the results of one study will be comparable to another so long as the same generic instrument is employed. This approach may resolve the comparability problem between studies but what was being measured from the first place remains unclear (e.g. functioning, HRQOL, or QOL). The items, components, and domains found in the PRO, therefore, need to be assessed with a conceptual definition that clarifies the domain being measured.
In order to address this challenge, our group synthesized World Health Organization (WHO) definitions found in the International Classification of Functioning, Disability and Health (ICF) and the WHO-Quality of Life Instrument (WHOQOL) manual.12 We interpreted the definitions relative to the concepts of FDH, HRQOL, and QOL in order to delineate the constructs used within existing instruments (Fig. 1). According to this interpretation, FDH comprises the biopsychosocial components and interactions among body structures and function, and activities and participation in the context of the environment and personal factors.13,14 Quality of life is a person’s perception of their position in life… in relation to their goals, expectations, standards and concerns;12 therefore mention of perceptual or subjective elements of life must be explicit in order to measure a child’s QOL.11 Although some researchers and clinicians choose to define HRQOL as that portion of life affected by a health condition,15 this definition was not distinguishable from FDH as found in the ICF. Instead, HRQOL was interpreted to be subsumed under QOL such that it was defined by a child’s perception of his or her health and health-related states.
Figure 1. Organization of concepts based on World Health Organization definitions. aHealth-related quality of life is not a WHO definition but was subsumed as a component of WHO-defined quality of life in this analysis
Download figure to PowerPoint
These definitions form one part of a content analysis that has been validated specifically for child health PROs16 and applied in whole or in part in numerous studies of child health instruments, including, but not limited to, the areas of cancer, epilepsy, and cerebral palsy.11,17,18 The content analyses used in this study are based upon a method initially developed in 2002 by Cieza et al.19 followed by multiple iterations of validation in the subsequent revisions.16,20 Applying this analysis can help to decipher and code the differences between instruments that use an FDH approach to assessment according to the ICF or a QOL approach using the definition set forth by the WHOQOL group, with HRQOL subsumed as a branch of QOL focused on health (as interpreted by our group).
This method can also provide detailed information about the health and health-related content (if any) contained in FDH, HRQOL, and QOL instruments by mapping the content to the ICF. Functioning, disability, and health, as defined in the ICF, are conceptualizations of health that can be defined and measured; therefore, the classification provides an excellent base for the description of any FDH or HRQOL PROs.
Although there is general acceptance in the literature of the ICF emphasis on functioning as the basis for conceptualizing health,7,13,14,21 and of the WHOQOL definition as the basis for defining QOL,12,22 operationalization of these concepts into instruments measuring FDH, HRQOL, and QOL for children has been less readily observed. For example, many authors argue that HRQOL/QOL concepts require a child’s subjective evaluation of his or her life or health in order to be authentic.22–24 Yet assessments of health and life satisfaction or priorities are often omitted from so-called HRQOL or QOL instruments.
Previous reviews of PROs have discussed the conceptual inconsistencies of the measures, but few have offered a systematic approach to dealing with the problem.10,25 The challenges of measuring FDH and QOL/HRQOL with existing instruments have become increasingly evident, because of the inconsistencies between them; however, discarding existing PROs because they are ambiguously defined by today’s standards could be equally detrimental to understanding FDH and HRQOL/QOL outcomes in children. Therefore, this study seeks to make explicit the perspectives of FDH, HRQOL/QOL and the ICF-based health content of commonly used PROs so that the instruments currently in use can be more easily interpreted relative to each other, and so that they can be applied more consistently relative to WHO definitions.
- Top of page
- What this paper adds
- Overall purpose
- Supporting Information
A total of 709 studies yielded 1151 occasions of generic PRO instrument use applied to children. The characteristics associated with generic PROs are presented in Tables SI–SIII (supporting information online), which show that the PRO use is used in cross-sectional studies, with school-aged children using child respondents. Languages of PRO use, in order of descending frequency, were English, Dutch, German, Spanish, French, Swedish, Italian, Chinese, Norwegian, and Portuguese.
Part I: Functioning, disability, and health versus HRQOL/QOL perspectives
The perspectives found in the generic instruments, presented in Table I, indicate that the FDH perspective is the predominant approach in 9 out of 15 instruments. On the other hand, application of the same instruments in the literature showed researchers intended to measure an HRQOL/QOL approach for 12 out of 15 instruments (Table II).
Table I. Perspectives (%) found in generic instruments according to part Ib content analysis
| ||QOL/HRQOL ||FDH||Other or unknown|
|Child Health and Illness Profile||4.68||93.8||1.56|
|Child Health Questionnaire||33.3||59.9||6.8|
|Functional Disability Inventory||0.00||100.00||0.00|
|Functional Status Questionnaire||0.0||100.0||0.0|
|General Health Questionnaire||42.9||57.1||0.00|
|Health Utilities Index||12.5||87.5||0.0|
|Satisfaction with Life Scale (Diener’s)||100.0||0.0||0.0|
|SSWLS/MSSWLS (Huebner’s)||100 (65.0)||0.0 (30.0)||0.0 (5.0)|
|Youth Quality of Life Instrument||76.6||17.1||7.3|
Table II. Reported use of generic patient-reported outcomes (PROs) from included studies according to Part Ia content analysis
| ||Researcher reported aim of generic PRO use (%)|
|None||QOL||HRQOL||Functioning||Disability||General health||Mental health||Physical health|
|Child Health and Illness Profile||5.9||20.6||26.5||0.0||0.0||47.1||0.0||0.0|
|Child Health Questionnaire||4.8||29.8||38.9||7.1||3.6||13.1||2.0||0.8|
|Functional Disability Inventory||0.0||0.0||0.0||27.3||69.7||3.1||0.0||0.00|
|Functional Status Questionnaire||10.0||0.0||5.0||20.0||30.0||35.0||0.0||0.0|
|General Health Questionnaire||0.0||8.3||8.3||16.7||0.0||0.0||66.7||0.0|
|Health Utilities Index||2.0||8.2||59.2||2.0||0.0||28.6||0.0||0.0|
|Satisfaction with Life Scale (Diener’s)||0.0||83.3||8.3||0.0||0.0||8.3||0.0||0.0|
|Youth Quality of Life Instrument||0.0||100.0||0.0||0.0||0.0||0.0||0.0||0.0|
The agreement for the coding perspectives for each instrument is listed in Table SIV (supporting information online) and was the lowest (46.5%) for the KINDL27 and the highest (100%) for the Diener’s Satisfaction with Life Scale,23 the Functional Disability Inventory,28 the Functional Status II Revised,29 and the Huebner’s Student Satisfaction with Life Scale.30 Agreement on the perspective for each item was above 75% in 14 out of 15 of the instruments. Overall, percentage agreement for all items combined was 81.1%.
Part II: describing the health and health-related content according to the ICF
An overview of the distribution of item representation according to ICF components is found in Table III. Tables with more detailed information for the ICF content found in each instrument are available as supporting information published online (Tables SV–SXX), which show the specific ICF codes found in each instrument. Table SV is shown as a sample summary table in Appendix II. These supplementary tables can be consulted when in-depth consideration of an instrument’s content is needed relative to a particular assessment purpose.
Table III. Composition of ICF components found in each instrument according to part II content analysis
| ||Body function||Activities and participation||Environment||Personal factors||Not defined (but part of the ICF)||Not covered by the ICF (health condition)||Not covered by the ICF (QOL)||Not covered by the ICF (other)||Total items coded|
|CHIP (optional section)||1||2||16||2||2||39||0||0||62|
Table SXXI (supporting information online) shows the agreement with bootstrapped confidence intervals between assessors for each instrument using kappa and bootstrapped confidence intervals. Agreement was the lowest for the KIDSCREEN (0.73)31 and highest (1.00) for Diener’s Satisfaction with Life Scale,23 the Health Utilities Index32 and Huebner’s Student Satisfaction with Life Scale.30 Overall kappa for all items combined was 0.86.
Summary of parts I and II
Summaries of the dominant perspectives and health domains, as well as brief summaries of each instrument, are included in Table IV.
Table IV. Summary of instrument use and content
|Instrument name||Overall PRO use in included articles (%)||Dominant outcome targeted by PRO use in the literature (Ia)||Dominant perspective found in the PRO from content analysis (Ib)||Dominant ICF Component (II)||Instrument summary|
|Child Health and Illness Profile||3||Functioning||FDH (with an unknown subcomponent)||Body functions and activity and participation||FDH instrument|
|Child Health Questionnaire||22.1||HRQOL/QOL||FDH (with QOL/HRQOL and some unknown features)||Activities and participation||FDH with some QOL features|
|DISABKIDs||1.3||HRQOL/QOL||HRQOL/QOL (with some functioning features)||Body functions, activities and participation, and environment||HRQOL instrument with biopsychosocial components|
|Functional Disability Inventory||2.9||Functioning||FDH||Activities and participation||FDH instrument focused on functioning and disability|
|Functional Status Questionnaire II||1.7||Functioning||FDH||Body functions||FDH instrument with emphasis on body functions|
|General Health Questionnaire||1||HRQOL/QOL||FDH and QOL/HRQOL||Body functions||FDH instrument focused on body functions (emotions) with QOL features|
|Health Utilities Index||4.3||HRQOL/QOL||FDH (with one HRQOL/QOL attribute)||Body functions||FDH instrument with emphasis on body functions|
|KIDSCREEN||6.6||HRQOL/QOL||HRQOL/QOL (with some functioning features)||Activities and participation||HRQOL instrument with biopsychosocial components|
|KINDL||3.8||HRQOL/QOL||FDH (HRQOL/QOL subcomponent)||Body functions, activity, and participation||FDH instrument with some QOL features|
|PedsQL 4.0||22.8||HRQOL/QOL||FDH||Activities and participation||FDH instrument|
|SF_8_12_36||3.4||HRQOL/QOL||Functioning (with HRQOL/QOL subcomponent)||Activities and participation||FDH with minor QOL features|
|Satisfaction with Life Scale (Diener’s)||1||HRQOL/QOL||HRQOL/QOL||None||QOL instrument|
|SSWLS/MSSWLS (Huebner’s)||0.8||HRQOL/QOL||SLSS: HRQOL/QOL||None||Pure QOL instrument|
|0.5||MSLSS: HRQOL/QOL (health subcomponent)||Activities and participation and environment||HRQOL instrument with biopsychosocial functioning components|
|TNO-AZL Series||5.6||HRQOL/QOL||FDH (HRQOL/QOL subcomponent)||Body functions and activity and participation||FDH instrument with some HRQOL features|
|Youth Quality of Life Instrument||1.4||HRQOL/QOL||HRQOL/QOL (health facilitators/barriers subcomponent)||Activities and participation and environment||HRQOL instrument with biopsychosocial health components|
- Top of page
- What this paper adds
- Overall purpose
- Supporting Information
When WHO definitions of biopsychosocial health (FDH) and QOL health are applied to PRO instruments, inconsistency is revealed between the measurement perspective of the PRO and the purposes to which that instrument is being applied. Of the 15 instruments analyzed, 12 were (most frequently) applied to measure HRQOL or QOL but only four were coded as having mainly an HRQOL/QOL perspective using WHO definitions.
The largest discrepancy between an instrument’s content and its application was found for the Pediatric Quality of Life Inventory (PedsQL).33,34 Using WHO definitions, this measure was found to represent FDH, not HRQOL or QOL perspective for all of its items. It should be noted that analysis of the PedsQL showed that the items represented a broad biopsychosocial definition of health and health-related domains.11 Thus, conceptually, the PedsQL spans a wide definition of functioning, disability, and health (FDH) that includes some contextual factors, but the breadth of its health content should not be confused with WHO definitions of QOL.
The assumption of this analysis is that the difference between an FDH and QOL/HRQOL instrument is not so evident in what is measured, because both functioning and HRQOL instruments can measure the same (biopsychosocial) health and health-related domains. Both functioning and HRQOL instruments should address biological, psychological, and social components. What distinguishes the functioning approach from that of HRQOL/QOL is how the domains of life and health are measured. For example, a parent could be asked: ‘How difficult is it for your child to walk short distances?’ or ‘How satisfied do you feel with your child’s ability to walk short distances?’ These items both refer to the same domain (walking); however, where the first question targets functioning from a performance point of view the second targets satisfaction with a domain of functioning, which is a dimension of HRQOL.
In our sample of articles, researchers demonstrated greater consistency between instrument application and content when seeking to measure FDH than when seeking to measure QOL/HRQOL. Instruments that were named by developers as FDH measures (e.g. Functional Disability Inventory,28 Child Health Illness Profile,35 Functional Status29) were applied consistently in the literature relative to the ICF terms of functioning, disability, and health.
Although it is unclear why this mismatch between content and application occurred for QOL/HRQOL, we suspect that the time and context in which these instruments were developed played a role in the discrepancy. Instruments such as the Child Health Questionnaire and PedsQL were developed before conceptual differences between a biopsychosocial approach to FDH (as in the ICF/ICF-CY) and of QOL (according to the WHOQOL task-force) were made clear in the health literature. Our results do not indicate that these instruments should cease to be used; rather we argue that it is important to be aware of their perspectives relative to current standard definitions, as well as of the ICF domains in each instrument relative to the intended purpose.
In addition to this review, guidelines about health status measurement selection expressed in checklists such as COSMIN36–38 can be helpful because they focus on psychometric properties and instrument use. A conceptual understanding of FDH and QOL/HRQOL is still vital to the PRO selection process because it is crucial to have a firm understanding of the concept to be measured before selecting an instrument to measure it,38 irrespective of previously reported psychometric performance. These results are not intended to suggest that the importance of psychometrics be overlooked; rather we suggest that a good understanding of FDH and QOL concepts is a crucial to measurement validation and good psychometric performance.
The good news about the FDH and QOL instruments reported in this study is that the items can often be taken at face value, that is, the content of an item and what it is designed to measure are usually much more straightforward than when using a diagnostic assessment (e.g. depression questionnaires).39 If users possess a firm conceptual understanding of the difference between FDH, HRQOL, and QOL, then they will have the critical skills necessary to understand the face validity of these instruments. Thus, if an instrument user can properly identify (1) which approach/perspective they seek to measure (FDH vs QOL/HRQOL), and (2) the health or health-related domain (e.g. relative to the ICF) that describes or evaluates changes in a group of children, then the user has grounds for (3) assessing whether the content of an instrument overlaps with their intended purpose.
Another finding of this literature review was that the majority of the studies using PRO instruments for children did so in the context of cross-sectional research. We question whether instruments that have been useful to describe, distinguish, or differentiate populations of children will perform comparably well as the objective of child health research shifts towards more evaluative purposes. When evaluating interventions, the responsiveness of a PRO depends on its ability to detect change.40 As a first, step to selecting an instrument it is necessary to determine if a PRO includes domains that describe a population of interest or has domains that are expected to change following intervention. This information can be found in the supplementary Tables SI–SXXI linked to this article.
Finally, researchers should be clear about whether they are seeking to measure FDH (which includes the capacity, performance, presence/absence, frequency, severity, etc. of biopsychosocial domains), HRQOL (the expectations, standards, or concerns about those health domains), or QOL (the child’s personal assessment of their position in life). Such efforts will help determine whether interventions are being accepted or rejected based on whether the correct instruments were chosen, with appropriate content for what is being measured.