The aim of this study was to compare the SF-36 and two disease-specific instruments--St. George's Respiratory Questionnaire (SGRQ) and Chronic Respiratory Disease Questionnaire (CRDQ) among chronic obstructive pulmonary disease (COPD) patients using healthcare utilization data. The three instruments were administered twice by telephone interview to 40 COPD patients separated by a 3-month interval. Healthcare utilization data were obtained for the 3 months immediately prior to each administration. Reliabilities for sub-scales on the three instruments ranged from 0.70 to 0.90 for the SF-36, from 0.78 to 0.85 for the SGRQ, and from 0.64 to 0.91 for the CRDQ. In general, the study sample had poor physical health, and somewhat better mental health. None of the demographic variables were significantly associated with either health status or utilization. Among the three instruments, the SF-36 explained the most variance in healthcare utilization. The aggregate scores such as the summary scores of the SF-36 and the total score of the SGRQ did not perform as well as the individual sub-scales of the respective instruments in explaining the variation in utilization. The conclusion that the SF-36 provides the best association to healthcare utilization data from among the three instruments studied should be interpreted with caution given the small sample size and relatively brief duration of the study.