Analysis of quality of life data from a randomized, placebo-controlled heart-failure trial

doi:10.1016/0895-4356(89)90160-1

Journal of Clinical Epidemiology

Volume 42, Issue 10, 1989, Pages 955-962

https://doi.org/10.1016/0895-4356(89)90160-1 Get rights and content

Abstract

In a randomized, double-blind, placebo-controlled, 3-months trial involving 111 congestive heart failure patients, one non-validated and three validated Quality of Life (QL) instruments were administered. Two randomized treatment groups were evaluated, one with 62 patients who continued on standard therapy and the other with 49 patients whose standard therapy was replaced by placebo. The data from Patient's Self-rating Scale (a non-validated instrument) and Spitzer's QL index showed a significant difference between the two treatment groups for an overall effect. There were no significant differences between two treatment groups for Sickness Impact Profile (SIP) and Quality of Well-Being (QWB).

For analyzing the multiple components in a QL instrument, the global statistics as suggested by O'Brien were applied to compare the two treatment groups. Univariate statistics complemented the global methods. The general use of global statistics in analyzing QL data is recommended.

References (14)

W.O. Spitzer et al.
Measuring the quality of life of cancer patients
J Chron Dis
(1981)
R.M. Kaplan
Validity of quality of well-being scale as an outcome measure in chronic obstructive pulmonary disease
J Chron Dis
(1984)
A.E. Fletcher et al.
Evaluation of Quality of Life in clinical trials of cardiovascular disease
J Chron Dis
(1987)
M. Bergner et al.
The Sickness Impact Profile: development and final revision of a health status measure
Med Care
(1981)
S.M. Hunt et al.
A quantitative approach to perceived health
J Epidemiol Commun Health
(1980)
R.M. Kaplan et al.
Health related quality of life measurement for research evaluation and policy analysis
Health Psychol
(1982)
P. Meier
Statistics and medical experimentation
Biometrics
(1975)

There are more references available in the full text version of this article.

Cited by (50)

Development of the EXACT-U: A preference-based measure to report COPD exacerbation utilities
2011, Value in Health
Citation Excerpt :
Changes in health states may not be accurately measured if the condition affects dimensions not captured by the MAU. The sensitivity of the EQ-5D to change associated with effective interventions has been questioned in several disease areas, including urinary incontinence, visual acuity loss, and heart failure, and, more relevantly, in measurement of COPD exacerbations [3–7]. COPD exacerbations have been documented to involve symptoms such as cough, sputum production, chest discomfort/tightness, and breathlessness, as well as systemic manifestations including disruption of sleep, psychological distress, fatigue, and activity limitation [8].
The exacerbations of chronic obstructive pulmonary disease tool (EXACT) is a condition-specific daily diary recently developed to evaluate the frequency, severity, and duration of chronic obstructive pulmonary disease (COPD) exacerbations. A preference-based algorithm for the EXACT would allow utilities to be reported from patients during an exacerbation when EQ-5D data are not available.
To develop the exacerbations of chronic obstructive pulmonary disease tool-utility (EXACT-U), a condition-specific preference-based measure to report utilities from the EXACT for use in cost-effectiveness studies.
Five items with three to five levels comprise the EXACT-U. Two groups of health states and respondents were constructed to allow for model development (Development group) and predictive validity testing (Validation group) using independent samples. Members of the UK general public each valued 11 randomized health states using time trade-offs (TTOs) scaled from full health/dead with 10-year durations. Regression models estimated from the Development group using individual data, mean data, and panel designs. Models assessed by number of inconsistent coefficients estimated and R² and tested against observed utilities from the Validation group using mean absolute error (MAE) and root mean squared error (RMSE).
A total of 55 health states, including the best and worst states, were valued in TTO interviews conducted with 400 respondents. Ten models were developed. The final preferred model contained no logical inconsistencies and found MAE = 0.04 and RMSE = 0.05 with a predicted utility range from 0.09 to 0.95.
The EXACT-U is a condition-specific preference-based measure with strong predictive validity to report daily utilities during an exacerbation.
Comparative responsiveness of generic and specific quality-of-life instruments
2003, Journal of Clinical Epidemiology
We assessed the relative responsiveness of generic and specific quality of life instruments in 43 randomized controlled trials that compared head-to-head 31 generic and 84 specific instruments. Using weighted effect size as the metric of responsiveness, we assessed the impact of instrument type, disease category, and magnitude of underlying therapeutic effect on responsiveness, and assessed the responsiveness of specific instruments relative to the corresponding domains of generic measures. In studies with a nonzero therapeutic effect, specific instruments (mean = 0.57) were significantly more responsive than generic instruments (mean = 0.39, P = .01), and than the corresponding domains of generic instruments (mean = 0.40, P = .03). Studies with low, medium, and high therapeutic effects showed a corresponding gradation in responsiveness differences between specific and generic instruments. We conclude that, overall, specific instruments are more responsive than generic tools, and that investigators may come to misleading conclusions about relative instrument responsiveness if they include studies in which the magnitude of the underlying therapeutic effect is zero.
Health-related quality of life measurement in heart failure: Challenges for the new millennium
2001, Journal of Cardiac Failure
Citation Excerpt :
We performed an extensive MEDLINE search of peer-reviewed publications covering the years 1966 to 1999 inclusive for citations of quality of life, health status, or questionnaires referenced with heart failure, drug therapy, angiotensin-converting enzyme inhibitor, angiotensin receptor blocker, vasodilator, β-adrenergic receptor blocker, digoxin, cardiac glycoside, diuretic, calcium channel blocker, inotrope, α-adrenergic blocker, and investigational agents (eg, tumor necrosis factor α-receptor antibody). We included only double-blinded, randomized, placebo-controlled trials related to drug therapy (19-56) and excluded studies that measured HRQL in the context of coronary artery bypass graft or valvular surgery, exercise rehabilitation or training, ventricular assist devices, pacemakers, implantable cardioverter defibrillators, or other nonpharmacologic interventions. We performed a second MEDLINE search using the names of the instruments in the trials included in this analysis and referenced them to heart, heart failure, and cardiac disease.
Health-related quality of life (HRQL), representing a patient-driven end point, has been increasingly emphasized in randomized clinical trials of new heart failure therapies. Measurement of HRQL depends on the use of validated instruments, with attention paid to the timing of administration and analysis of data in the context of conventional morbidity and mortality end points. In a review of HRQL measurement in heart failure drug trials published from 1966 to 1999, we found that important data, such as the number of participating subjects, are often lacking. HRQL is analyzed as a stand-alone end point without consideration of the underlying clinical trajectory of the disease. Improvements in trials methodology are warranted if quality-of-life data are to be meaningful in the determination of drug efficacy in heart failure.
Generic and specific measurement of health-related quality of life in a clinical trial of respiratory rehabilitation
1999, Journal of Clinical Epidemiology
The purpose of this study was to compare the performance of measures of health-related quality of life in a randomized controlled trial of respiratory rehabilitation versus conventional community care for patients with chronic airflow limitation. The study included 89 stable patients with moderate to severe chronic airflow limitation with measurement of health status at 12, 18, and 24 weeks. Outcomes included two disease-specific (the Oxygen Cost Diagram and the Chronic Respiratory Questionnaire [CRQ]) measures, a generic health profile (the Sickness Impact Profile [SIP]), and two utility measures (the Standard Gamble and the Quality of Well-Being index [QWB]). Of the measures, only the four domains of the CRQ (dyspnea, fatigue, mastery, and emotional function) showed statistically significant differences (P ≤ 0.05) between treatment and control groups. Correlation between change in the CRQ and change in other relevant measures, including the 6-minute walk test and global ratings of change in dyspnea, fatigue, and emotional function were generally weak to moderate (from 0.19 to 0.51). All correlations between change in the QWB, SIP, and Standard Gamble and other measures were very weak or weak (up to 0.30). Correlation between change in the three generic measures were all very weak (<0.15). The results suggest that unless investigators include responsive and valid disease-specific measures of health-related quality of life in controlled trials in chronic diseases, they risk misleading conclusions about the effect of treatments on health status.
Assessing the responsiveness of a functional status measure: The Sickness Impact Profile versus the SIP68
1997, Journal of Clinical Epidemiology
In this study, the Sickness Impact Profile (SIP) and the SIP68 are studied for their ability to detect changes in health-related behavioral status. Methodological approaches toward responsiveness are invented and discussed. Next, literature findings on the responsiveness of the SIP are presented and judged for their validity. The SIP appeared to be able to demonstrate changes in the expected direction and in accordance with changes detected by other instruments. Using data from seven different longitudinal projects in populations with different diagnoses, the responsiveness of both the SIP136 and the SIP68 are subsequently studied and compared. In all populations, changes in functional status were indicated by both instruments. In terms of effect sizes, the SIP136 and the SIP68 do not differ significantly in their responsiveness. Moreover, changes detected by both SIPs appear to be valid representations of changes in health-related functional status.
Clinical epidemiology and evidence-based health care
2014, Handbook of Epidemiology: Second Edition

View all citing articles on Scopus

View full text

Original articleAnalysis of quality of life data from a randomized, placebo-controlled heart-failure trial

Abstract

J Chron Dis

J Chron Dis

J Chron Dis

The Sickness Impact Profile: development and final revision of a health status measure

Med Care

A quantitative approach to perceived health

J Epidemiol Commun Health

Health related quality of life measurement for research evaluation and policy analysis

Health Psychol

Statistics and medical experimentation

Biometrics

Original article
Analysis of quality of life data from a randomized, placebo-controlled heart-failure trial