Laurent G. Glance, MD
Laurent G. Glance, MD, is Professor and Vice-Chair of Research and has a secondary appointment at RAND Health. He has extensive experience working with large clinical and administrative data sets, quasi-experimental methods, multivariable modeling, risk adjustment, simulations, and bootstrapping techniques. His research has focused on the following areas:
Understanding the Limitations of Risk Adjustments for Measuring Quality-of-Care. I was the recipient of a Research Career Development Grant (K08 HS11295) from AHRQ, which focused on the optimization of risk-adjustment methodologies for measuring Intensive Care Unit quality. We found that different ICU scoring systems frequently disagreed on which hospitals were quality outliers. We also showed that hospital performance is partly a function of the choice of statistical methodology. Most recently, we have shown that the prediction models recommended by the American Heart Association to risk stratify patients undergoing non-cardiac surgery frequently disagree on patients’ cardiac risk. These studies point to the necessity of identifying best-of-class prediction models for use in performance reporting and clinical decision-making.
The Impact of the Present-on-Admission Indicator on Quality Reporting. With funding from AHRQ (R01 HS 13617), we focused on the implications of the present-on-admission (POA) indicator in administrative data to differentiate preexisting conditions from complications. This work led to publications showing that the addition of the present-on-admission (POA) indicator enhances the ability of existing comorbidity algorithms to map ICD-9-CM codes to diagnostic categories accurately, and that the use of routine administrative data without POA information to construct hospital quality report cards may result in the misidentification of hospital quality outliers.
Optimizing Injury Severity Scoring. With funding from AHRQ (R01 HS016737), we used the National Trauma Databank as a platform to determine whether providing hospitals with trauma report cards will lead to improved population outcomes in trauma. This work has led to the development of a highly innovative empiric-based approach to trauma injury scoring using either clinical data or ICD-9-CM codes as an alternative to existing expert-based injury severity scoring systems. To address one of the primary limitations of trauma registry data, this project validated the use of multiple imputation for handling missing data in the context of quality measurement. This grant also examined whether higher-quality trauma care is associated with lower cost and whether failure-to-rescue is an important mechanism driving outcome differences across trauma centers.
Impact of Non-public reporting on Trauma Outcomes. As part of our AHRQ-funded R01, we conducted a prospective observational study to examine the impact of non-public reporting on trauma outcomes. We also examined the validity of trauma severity scoring by examining the ability of past hospital performance to predict future hospital performance. Finally, we examined the association between hospital structural characteristics and trauma outcomes.
Impact of Health Care Reform on Disparities. Healthcare reform aims to improve healthcare quality, expand access to healthcare, and lower the cost of health care. CMS created the Comprehensive Care of Joint Replacement (CJR) program to address the large variation in cost and quality for patients undergoing hip and knee replacements. The CJR program does not, however, adjust hospital payments for sociodemographic risk to compensate safety-net hospitals for the higher cost and worse outcomes of socially vulnerable patients. The CJR program also does not include disparity-sensitive metrics to incentivize hospitals to address the substantial gap in joint replacements between Black and White beneficiaries. Our work has shown that safety-net hospitals were less likely to receive rewards and more likely to be penalized by the CJR program. We also found that the CJR was associated with a widening of disparities in the utilization of total knee replacements.