Context: Public disclosure of quality data on health maintenance organizations (HMOs) might improve public accountability, inform consumer decision making, and promote quality improvement. But, because disclosure is voluntary, some HMOs could subvert these objectives by refusing to release unfavorable data.
Objective: To determine the association between HMO quality of care and withdrawal from public disclosure of quality-of-care data the subsequent year.
Design and setting: Retrospective cohort study of administrative and quality-of-care data on HMOs from the National Committee for Quality Assurance (NCQA) annual Quality Compass databases for 1997, 1998, and 1999, including Health Plan Employer Data and Information Set (HEDIS) quality scores.
Main outcome measure: One-year rates of HMO withdrawal from public disclosure of HEDIS scores for plans in the highest and lowest tertiles of HEDIS scores, adjusted for method of data collection and plan model type.
Results: Of the 329 HMOs that publicly disclosed HEDIS scores in 1997, 161 plans (49%) withdrew from public disclosure in 1998. Of the 292 HMOs that disclosed their scores in 1998 (including 130 newly participating plans), 67 plans (23%) withdrew from public disclosure in 1999. Plans whose scores ranked in the lowest-quality tertile were much more likely than plans ranking in the highest-quality tertile to withdraw from public disclosure in 1998 (odds ratio [OR], 3.6; 95% confidence interval [CI], 2.1-7.0) and 1999 (OR, 5.7; 95% CI, 2.7-17.7).
Conclusion: Compared with HMOs receiving higher quality-of-care scores, lower-scoring plans are more likely to stop disclosing their quality data. Voluntary reporting of quality data by HMOs is ineffective; selective nondisclosure undermines both informed consumer decision making and public accountability.