Home

Current Issue

Archives

Guidelines for Authors

Classified Ads

Links

Subscriptions

Subscriber Registration

Guidelines for Website Users

E-mail Alert Service

Contact Info

The Psychometric Properties of Patient Preferences in Osteoporosis

ANN CRANNEY, DOUGLAS COYLE, BA PHAM, JACQUELINE TETROE, GEORGE WELLS, ELAINE JOLLY, and PETER TUGWELL

ABSTRACT.

Objective. Osteoporosis is a chronic disease manifested by wrist, vertebral, and hip fractures that results in significant morbidity and burden to society. About 30% of postmenopausal women have osteoporosis according to the WHO criteria. Women with one vertebral fracture have a 4-fold increased risk of a subsequent fracture. The goal of treatment is to prevent fractures and improve quality of life. Preferences or utilities are now recommended for incorporating quality of life into evaluations of the cost effectiveness of new therapeutic interventions. We evaluated the psychometric properties of preference based measures in osteoporosis.

Methods. Preference scenarios were constructed with a health state classification system. The reliability and validity of the feeling thermometer and the standard gamble was assessed by interviewing 42 women from 4 different patient groups. The sensitivity to change of the feeling thermometer and standard gamble was compared with the Health Utilities Index Mark 2 (HUI2) and SF-36. All subgroups were reassessed about 2 months after their first interview.

Results. Preference measurement was feasible in women of different age groups. The reliability coefficients for health states ranged from 0.65 to 0.87. The preference scores for the marker states demonstrated content validity. Convergent validity of the feeling thermometer was supported by a significant correlation with the HUI2 (r = 0.38, p < 0.05) and the physical health summary of the SF-36 (r = 0.56, p < 0.005). The standard gamble did not correlate with the HUI2 (r = 0.15) or the feeling thermometer (r = 0.09), but correlated with 2 domains of the SF-36. The preference measures were sensitive to change, with the efficiency scores ranging from 0.78 to 1.0.

Conclusion. Preference measurements in the evaluation of osteoporosis are feasible. The feeling thermometer and standard gamble appear to be related to different aspects of health related quality of life. Both instruments were sensitive to change over a 2 month period. (J Rheumatol 2001;28:132-7)

Key Indexing Terms:

PREFERENCES
OSTEOPOROSIS
QUALITY OF LIFE
FRACTURES