How Well Do Smoking Cessation Quality Measures Predict Clinical Outcomes?'

Sherman SE, Yano EM, Lanto AB, Mittman BS. How Well Do Smoking Cessation Quality Measures Predict Clinical Outcomes?'. Paper presented at: Society of General Internal Medicine Annual Meeting; 2003 May 2; Vancouver, Canada.

Related HSR&D Project(s)

• CPG 97-002 – Implementing Guidelines for Smoking Cessation: A Randomized Trial of Evidence-Based Quality Improvement

Background A wide range of quality of care measures for smoking cessation have been proposed or are currently in use. We assessed how well these structure and process measures correlate with clinical outcomes, such as smoking cessation. Methods Our data are primarily from the Quality Improvement Trial for Smoking Cessation (QUITS), a recently completed trial of an organizational intervention to implement smoking cessation guidelines at 18 VA sites in the Southwestern US. We supplemented this with data from national VA databases. We measured the correlation between the 18 sites for structure quality of care measures (e.g., is there a system for screening), process quality of care measures (e.g., how frequently were patients counseled about smoking cessation), and clinical outcomes (quit attempts, smoking cessation). Our structure quality of care measure was the QUITS report card, a 25-item organizational checklist that assessed adherence to the system aspects of the smoking cessation guidelines. Our process quality of care measures included intensity scores for screening, counseling, referral, and treatment (from the QUITS administrative site survey); reported rate of smoking cessation counseling and medication prescription (QUITS patient survey); rate of smoking cessation clinic attendance (VA Outpatient Care File); and the actual rate of prescribing smoking cessation medications (VA Pharmacy Benefits Management database). Our outcome measures (from both the QUITS 12-month and 18-month patient follow-up surveys) were the percent who had tried to quit recently and the percent who had actually quit smoking. Results The percent who recently tried to quit at each site ranged from 27-59% at 12 months follow-up and from 27-52% at 18 months. The range of smoking cessation rates across the sites was 4-27% at 12 months and 3-20% at 18 months. Several of the potential quality measures correlated with each other to a moderate degree. However, none of them correlated with the rate of quit attempts or of actual cessation, either at 12-month or 18-month follow-up. Conclusion Higher performance on a wide range of quality of care measures did not translate into more quitters or even more attempted quitters. Better quality of care measures may be needed for smoking cessation.