| |
The evaluation
of dyspnea questionnaires in non-patients has shown that results
do not depend on age, sex, or levels of education (Harver,
Mahler, Schwartzstein, and Baird, 2000). Dyspnea questionnaires
also exhibit satisfactory test-retest reliability. For example,
the percent agreement between patient selections of descriptors
on two occasions held 4 to 15 days apart was 79%; the correlation
coefficient was 0.82 (p = 0.001) (Mahler
et al., 1996). Additionally, the percent agreement between
descriptors based on recall and as a result of moderate breathlessness
provoked by walking was 68% (r = 0.69; p = 0.004) (Mahler
et al., 1996). These results are equivalent to those obtained
for other dyspnea measures such as the Baseline Dyspnea Index
and the Oxygen Cost Diagram (for review, see Mahler,
Guyatt, and Jones, 1998).
In some
ways, the validity of dyspnea questionnaires is more established.
Current forms of the questionnaire exhibit good face and content
validity. For example, the selection of items was obtained through
peer review by both patients and experts (e.g. Simon
et al. 1990), and patients rarely offer alternative phrases
or descriptors suggesting that the questionnaires capture the
entire domain of relevant descriptors of breathing discomfort
(Mahler et al., 1996). Evidence
of criterion validity was provided in the analysis of the response
to bronchodilators of patients with asthma in an emergency department
(Moy et al., 1998). Relationships
among descriptors are consistent with predicted mechanisms underlying
dyspnea (Harver,
Mahler, Schwartzstein, and Baird, 2000).
Question
21.1
Verification
of the reliability and validity of dyspnea scales
 | Is
unattainable |
 | Is
almost complete |
 | Is
never ending |
|