Skip to Content
eNews Join our Mailing List
Find us on facebook, like the CanChild fan page

Donate to CanChild

Keeping Current Heading

Keeping Current #93-1

Quality of Life

Peter Rosenbaum, Neurodevelopmental Clinical Research Unit (NCRU)*

©1993 Rosenbaum, P., Neurodevelopmental Clinical Research Unit (NCRU)
As of Oct. 1, 1998, the NCRU is called CanChild Centre for Childhood Disability Research

In contemporary health care delivery and evaluation, the decade of the 90s may well be subtitled the "Quality of Life" (QL) era. Few concepts have become as widely talked about in recent years, although the issues underlying QL are often confusing. This brief commentary will try to outline some of the territory - its terminology and potential usefulness in evaluating pediatric rehabilitation. In particular it is hoped that readers will become aware of the varied meanings of the phrase "quality of life", and recognize what an author or speaker means when they use it. For further reading in this area refer to Kaplan and Anderson (1990).

Health services outcome measures have traditionally included consideration of mortality (how many people die from a particular illness {or its treatment!}); and morbidity (what impact does the disease or its treatment {side effects} have on function, mood, behaviour, etc.). Both dimensions of outcome obviously remain important. What has been recognized for a long time however, are the limitations of these health indicators. Increasingly, clinicians and investigators have become interested in evaluating the experiential aspects of an illness or impairment - the so-called "psychosocial" variables. These often include assessments of feelings, self-image, behavioral consequences of illness or treatment, and role functions (as student, worker, spouse, etc.). These dimensions of health and illness are often referred to loosely as aspects of "quality of life" (or more accurately "health-related quality of life", to recognize the relationships between aspects of health or illness and these outcomes). Subjective and abstract concepts like these are increasingly being targeted as outcomes, every bit as important as the standard 'hard' (biological) measures such as blood pressure, serum sodium, or joint range of motion.

The reasons for interest in the personal and human (as well as biomedical) outcomes of illness are probably many. In our view, the fact that they have become recognized and the subject of study is an exciting direction in the evaluation of health outcomes. For one thing, these dimensions of "quality of life" may at times change much more than some "objective" outcomes. They may change after interventions directed at the person rather than at the biologic impairment. While not ignoring the need to address basic biologic dysfunction in order to improve human well-being, our attention to people's health-related quality of life suggests a redirection of our intervention priorities toward "whole people" and not just to their broken "parts". It also recognizes the importance of the human dilemmas imposed by impairments in biologic function, and of our potential ability to influence these issues.

A more formal QL approach in contemporary health care evaluation is the attempt to aggregate information simultaneously about several aspects of a person's health status. Here the need is to assess the morbidity of an illness, or the impact of an intervention, across multiple dimensions of a person's status and function; and in some systems of QL measurement, to place a "value" on these impacts.

A simple rehabilitation example might be the following treatment dilemma: should we recommend full-time wheelchair use for a highly motivated but severely disabled young adult whose gait training is exhausting for her, producing very limited functional gains for enormous effort, and distracting from other rehabilitation efforts that might (in this scenario) have a higher probability of success? How might we try to evaluate the effects of the use of a wheelchair as a permanent mobility aid? What aspects of "function" would we assess?

Our considerations might include the impact of the use of wheeled mobility (and the abandonment of "walking") on self-image, increased independent travel capacity, depression caused by the loss of hope to "walk", increased energy level for other activities, future "risks" (contractures, osteoporosis, etc.)... the list could be extensive. How does one accumulate information about these several "dimensions" of potential impact? More to the point, what values (and from whose perspectives) would one use in making a therapeutic recommendation, or evaluating its outcome?

At this point there may not be a simple or single answer to this rehabilitation problem. However, one approach of interest in child health has been studied by colleagues at McMaster University who have developed a Multi-Attribute Health Status (MAHS) assessment method to evaluate health-related quality of life in childhood (Feeny et al., 1992). Although the system was created specifically to assess morbidities associated with childhood cancer and its treatment, it may have much wider applicability across children's impairments. Several dimensions of function (including mobility, sensation, emotion, cognition, self-care and pain) are each described with several levels of function. Any particular child's functional status can then be captured across each dimension. With this unified and consistent system it is possible to contrast the health status of individuals receiving different types of treatment for the same disorder; to compare the health of people with different medical disorders; and to evaluate the health status of people over time, as their health status changes.

How did the McMaster investigators assess the "value" or "utility" of these health status profiles? A large number of randomly chosen parents independently provided information about their personal values or preferences for several health states presented to them with scenarios describing children whose function varied on the different dimensions outlined above. Perfect health on the MAHS system (where function is perfect on each dimension) is given a value of 1.0, while death is assigned a value of 0.0. All other morbidities are placed between these two cardinal values to provide a scale of values for health states. This method has proven quite robust and reliable.

Using mathematical methods, a formula (called a "utility function") can then be constructed that allows an observer to calculate a value or "utility" for any individual using the MAHS descriptions. In this way any person's functional status can be described on the MAHS system and the community's "valuation" of that health status can be determined as a single summary value (Barr et al., 1993). More sophisticated methods are also available to calculate years of "quality of life", using information on long-term prognosis.

Needless to say, despite enormous conceptual and technical advances in the field of measurement of health-related quality of life over the past fifteen years, many issues require further study and understanding. Some of the challenges involved in assessing QL in children have been outlined by Rosenbaum, Cadman, and Kirpalani (1990), and include issues of proxy respondents (parents answering for children, community parents answering for parents of "patients" etc.); the problem of assessing the "moving target" of the developing child; and the observation that parent and professional perspectives on a child's function or development may differ considerably.

What is exciting however, is the creation of systems that attempt to account simultaneously for the many factors impinging on health-related quality of life, using contemporary scientific measurement techniques and (in the childhood cancer example) framing the results of individuals' MAHS against a common set of community values. An extensive bibliography of current quality of life papers and instruments has been published (Spilker, White, Simpson, and Tilson, 1992; Spilker, Simpson and Tilson, 1992). It will become increasingly important for all of us in pediatric long-term care to learn about, understand, and begin to use measures of QL to assess the impact of what we do for children and families. Hopefully this summary of the issues has whetted people's curiosity!


Barr, R. D., Pai, M. K. R., Weitzman, S., Feeny, D., Furlong, W., Rosenbaum, P., & Torrance, G. W. (1993). Multi-attribute health status assessment: An approach to the measurement of global morbidity burden. A tool for clinical management - exemplified by brain tumors in childhood.Manuscript submitted for publication.

Feeny, D., Furlong, W., Barr, R. D., Torrance, G. W., Rosenbaum, P., & Weitzman, S. (1992). A comprehensive multi-attribute system for classifying the health status of survivors of childhood cancer. Journal of Clinical Oncology, 10, 923-928.

Kaplan, R. M., & Anderson, J. P. (1990). The general health policy model: An integrated approach. In Spilker, B. (Ed.), Quality of life assessments in clinical trials(pp.131-144). New York: Raven Press.

Rosenbaum, P., Cadman, D. & Kirpalani, H. (1990). Pediatrics: Assessing quality of life. In Spilker, B. (Ed.), Quality of life assessments in clinical trials (pp. 205-215). New York: Raven Press.

Spilker, B., White, W. S. A., Simpson, R. L., Jr., & Tilson, H. H. (1992). Quality of life bibliography and indexes: 1990 Update. Journal of Clinical Research and Pharmacoepidemiology, 6, 87-156.

Spilker, B., Simpson, R. L., Jr., & Tilson, H. H. (1992). Quality of life bibliography and indexes: 1991 Update. Journal of Clinical Research and Pharmacoepidemiology, 6, 205-266.