Letters
Defining and Measuring Quality of Life in Medicine
To the Editor.—Drs Leple ´ ge and Hunt
1
assert that variability
across cultures, between patients, and in the same patient
over time makes efforts to define the term quality of life im-
possible. It is an “idiosyncratic mystery.” They conclude that
physicians and health economists should avoid quality of life
assessment. At the same time, the authors assert that quality
of life is paramount to patients and is, indeed, the only concern
of the patient who seeks medical care. The unwelcome conclu-
sion is that outcomes—whether patients feel better and are
able to do more, whether they are spared subsequent treat-
ments, and whether they are glad they sought medical care—
are not a part of medicine.
This argument, however, rests on a faulty assumption. Vari-
ability among patient appraisal of quality of life is limited. No
one thinks severe abdominal pain is better than a runny nose,
as Fanshel and Bush
2
long ago pointed out (even if it is difficult,
as they recognized, to find a measurement unit that expresses
their relative quality-of-life impact). Even cultural variation
has limits. When asked about their ability to carry out common
tasks of daily living, the responses of !Kung and Herero elders
of southwest Africa (seminomadic pastoralists) were best un-
derstood in terms of a physical function component highly
correlated with age.
3
More generally, utilities elicited for
health states are highly correlated across populations and
across sociodemographic groups.
4
If health state utilities do
vary according to patient health status (as demonstrated for
dialysis patients
5
but not, however, for all patient groups), this
variability only shows that other distinct features of patient
experience must be considered by clinicians when recommend-
ing treatments or assessing outcomes. These include family
support systems, willingness to adopt prosthetic technologies,
and patient attachment to life.
Leple ´ ge and Hunt have exaggerated the idiosyncratic na-
ture of health-related quality of life (HRQL) by confounding it
with quality of life more generally. This distinction is critical:
HRQL measures are likely to be more highly correlated with
health status and more sensitive to changes in health than
general quality-of-life measures. When the authors suggest
that quality of life be viewed as “the best possible physical and
emotional state compatible with [a patient’s] medical condi-
tion,” they are talking about HRQL, which is already well
assessed by a variety of measurement tools. When they com-
plain that such measures of health status do not capture other
components of patient quality of life, such as the capacity to
love or have “a positive approach to everyday events,” they
are right, but the measures also were never intended to do so.
Steven M. Albert, PhD, MSc
Columbia University
New York, NY
1. Leple ´ ge A, Hunt S. The problem of quality of life in medicine. JAMA. 1997;278:47-50.
2. Fanshel S, Bush JW. A health status index and its application to health services
outcomes. Operations Res. 1970:1021-1066.
3. Draper P, Harpending H. Work and aging in two African societies: !Kung and Her-
ero. In: Bonder BR, ed. Occupational Performance in the Elderly. Philadelphia, Pa:
FA Davis Co Publishers; 1994.
4. Patrick DL, Sittampalam Y, Somerville SN, et al. A cross-cultural comparison of
health status values. Am J Public Health. 1985;75:1402-1407.
5. Sackett DL, Torrance GW. The utility of different health states as perceived by the
general public. J Chronic Dis. 1978;31:697-704.
To the Editor.—Drs Leple ´ ge and Hunt
1
provide an incomplete
view of the current state of the science of HRQL measure-
ment. We disagree with their pessimism about the value of
aggregated and normative outcome measures. Health out-
comes researchers must specify the conceptual model under-
lying an instrument; the patients’ perspective is critical in the
development of HRQL measures. We agree that patients are
the main source for information about the content and impor-
tance of domains to ensure that a quality-of-life measure ad-
equately reflects the impact of disease on functioning in ev-
eryday life and well-being. Most current instruments start
with eliciting concerns from patients (by qualitative methods
or focus groups) to determine the relevant domains.
Problems with confusing and unclear terminology have con-
tinued in HRQL research. The use of “subjective health sta-
tus” rather than “quality of life” will not resolve the problems
the authors have raised. We use HRQL when addressing qual-
ity-of-life outcomes that can be affected directly by health care
interventions, a stance consistent with Wilson and Cleary,
2
reserving quality of life for the global appraisal of life quality.
Patient outcomes describe the full range of measures used in
health evaluation, including clinical measures, symptoms,
functioning, and well-being.
Cultural differences present another challenge to HRQL
assessment. Studies have demonstrated that important do-
mains, such as physical, social, and psychological well-being,
are consistent across cultures. It is the expression of these
domains, including the range of activities and behaviors, that
varies within and between cultures and countries. The chal-
lenge is to design instruments to assess HRQL in culturally
meaningful ways and apply scientifically based methods for
linguistic and cultural validation.
Patients’ adaptation to and acceptance of their disease state
can lead to the apparent discordant finding that normative
health status measures demonstrate severe functional limita-
tions, while subjective ratings reflect patients’ satisfaction
with the quality of their life. Both perspectives are important
in evaluating the impact of health interventions. Idiographic
measures are useful for understanding the characteristics that
drive individuals’ assessments of their quality of life. Stan-
dardized, normative measures are useful for making compari-
sons of health care interventions and different populations.
Applications of techniques, such as 3-mode factor analysis,
3
that combine the normative and idiographic approaches can
be used to further understand responses to assessment in-
struments.
Methodological advances in health outcomes assessment
over the past 30 years have facilitated the introduction of pa-
tient outcomes in clinical trials, the monitoring of the health
Edited by Margaret A. Winker, MD, Senior Editor, and Phil B. Fontanarosa,
MD, Senior Editor.
Guidelines for Letters
Letters will be published at the discretion of the editors as space per-
mits and are subject to editing and abridgment. Letters will be
considered if they are typewritten double-spaced and do not exceed
500 words of text and 5 references. Please include a word count.
Letters discussing a recent JAMA article should be received within 4
weeks of the article’s publication. Letters must not duplicate
other material published or submitted for publication. A signed state-
ment for authorship criteria and responsibility, financial disclosure,
copyright transfer, and acknowledgment is essential for publica-
tion. Letters not meeting these specifications are generally not
considered. Letters will not be returned unless specifically requested.
Also see JAMA Instructions for Authors (January 7, 1998). Letters
may be submitted by surface mail: Letters Editor, JAMA, 515 N State
St, Chicago, IL 60610; e-mail: JAMA-letters@ama-assn.org; or fax
(please also send a hard copy via surface mail): (312) 464-5824.
JAMA, February 11, 1998—Vol 279, No. 6 Letters 429
©1998 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ by a Wayne State University User on 02/12/2021