|
|
Health Status Measurements at
Diagnosis As Predictors of Survival Among Adults With Brain Tumors
Helen McCarter, William
Furlong, Anthony C. Whitton, David Feeny,
Sonja DePauw, Andrew R. Willan, Ronald
D. Barr
From The Juravinski Cancer Centre;
Department of Clinical Epidemiology and Biostatistics; Department of
Pediatrics; Centre for Health Economics and Policy Analysis, McMaster
University; McMaster Children's Hospital, Hamilton Health Sciences,
Hamilton; Health Utilities Inc, Dundas; Institute of Health Economics;
Departments of Economics and Public Health Sciences, University of
Alberta, Edmonton; Program in Population Health Sciences, Hospital for
Sick Children, Toronto, Canada; and Kaiser Permanente Northwest Center
for Health Research, Portland, OR -- Address reprint requests to
Ronald D. Barr, MD, Health Sciences Centre, Room 3N27B, 1200 Main St
W, Hamilton, Ontario, L8S 4J9, Canada; e-mail: rbarr@mcmaster.ca
|
|
|
Purpose. The intent of this
study was to determine whether baseline measures of
functional capacity and performance could be used to predict survival
in adults following the diagnosis of brain tumors.
Patients and Methods.
Comprehensive health status and health-related quality of life (HRQL)
were measured using the Health Utilities Index (HUI; McMaster
University, Hamilton, Canada) system by a self-assessment questionnaire
in a survey of 100 consecutive patients. The Karnofsky Performance
Score (KPS) and Folstein's Mini-Mental State Examination (MMSE)
scores were measured by a physician blinded to the HUI results.
The patients were observed for up to 5 years to recorded dates
of death.
Results. An HUI
questionnaire was completed for 93% of the patients and 69%
died within 5 years of assessment. The HUI revealed a burden of
morbidity and complexity of disability that far exceeded that
reported for the general population. KPS and MMSE correlated strongly
with each other (r = 0.52; P < .001). A decrease
of 0.1 units in HUI Mark 2 (HUI2) self-care single-attribute
utility score was associated with an increased hazard of death
of 30% (P = .023) for patients with low-grade tumors
(n=25). For patients with high-grade tumors (n=56), a 10
unit decrease in the KPS, a 5 unit decrease in MMSE, and a
0.1 decrease in HUI Mark 3 (HUI3) speech and dexterity
single-attribute scores were associated with an increased
hazard of death of 20% (P = .022), 26% (P =
.015), 36% (P = .021), and 18% (P = .035), respectively.
Conclusion. Scores
derived from the measurement of HRQL following diagnosis can
predict survival in adults with brain tumors.
|
