The Hospital Medicine and Economics Center for Education and Research on Therapeutics at the University of Chicago

Center for Education and Research on Therapeutics (CERT)


Specific Aim 1: To perform a diverse set of descriptive and interventional research studies to improve the effectiveness and cost-effectiveness of hospital-based therapeutics.


Project 1: The Clinical and Economic Implications of Genetic Testing for Warfarin Management

Variation among individuals in the therapeutic dose of warfarin (coumadin) makes initiation of anticoagulation with warfarin a costly and sometimes dangerous aspect of hospital care. Genetic testing has greatly increased the accuracy of clinical dosing algorithms to predict therapeutic warfarin dosing among Caucasian patients but the effectiveness and costs of these tests in the hospital setting have not been systematically evaluated. It is also not known to what extent these tests are valid among other racial groups. The overall goal of this project is to develop and assess the effectiveness and cost-effectiveness of strategies that use genetic testing in the management of anticoagulation among racially diverse hospitalized patients.

For more information contact the project leader or other members of the research team:

Project 2: Validating Performance Measures for Patients Hospitalized with COPD Exacerbations

The primary goals of this study are to evaluate the validity (sensitivity and specificity) of identifying patients hospitalized for COPD exacerbations and evaluate the validity measuring the quality of hospital care for patients hospitalized for COPD exacerbations using administrative data compared to chart review. Secondary questions that will be examined in this project include: 

  • Do patient characteristics (e.g., presence of co-morbid conditions, severity of COPD exacerbation) affect the validity of administrative data for identifying patients hospitalized for COPD exacerbations?
  • Do patient characteristics affect the quality of care, as measured by administrative data or chart-review, for patients with COPD exacerbations?
  • What is the relationship between administrative-data-based and chart-based measures of the quality of hospital care to clinical outcomes (length of hospital stay, in-hospital mortality (all cause and respiratory-related), re-admissions within 30 days (all cause and respiratory-related), and mortality at 12 months (all-cause and respiratory-related)) in patients hospitalized with COPD exacerbations?

For more information contact the project leader or other members of the research team:

Project 3: Electronic Record Interventions to Address Off Label Prescribing

Recent quantitative estimates indicate that 21% of outpatient drug uses in the U.S. are off-label (by indication) and that the vast majority of these uses lack strong scientific support.  Similar estimates are not available for the inpatient setting.  Meanwhile, many newer drugs used off-label in the inpatient setting result in increased prescription costs and may not improve health.  However, off-label use may be clinically warranted in certain circumstances, and the practice of off label use can contribute to a more rapid evolution of clinical practice.  Unfortunately, even modern prescribing databases typically do not allow data driven analyses of off-label versus standard on label treatments because data on prescriptions are rarely linked to data on diagnosis or indication.  Moreover, little work has been done in the development of tools to collect the requisite data.


This is a pilot project to develop an electronic based intervention for gathering data regarding off-label drug use in an inpatient setting.  The project will identify and focus on specific drugs in an inpatient setting where there are particular clinical and economic concerns regarding off-label use.  Customized hospital  based clinical decision support (CDS) interventions at the time of computerized physician order entry (CPOE)  will be implemented to gather unique data linking prescribing decisions with clinical indications and diagnosis of the patient.  The project will evaluate the success of the intervention in terms of physician compliance and the ability of the intervention to generate useful and accurate information. 


For more information contact the project leader or other members of the research team:

Project 4: Using Social Network Analysis to Guide Quality Improvement Team Formation

Effective quality improvement in health care can require tools that can address challenges at patient, provider, and system levels. This simple fact has major implications for the formation of quality improvement teams. For example, it is well understood that the complex division of labor within hospitals implies that multidisciplinary teams are often needed to address systems issues.  No expert aiming to change prescribing behavior for a complex condition in the hospital would develop a team that would fail to span the relevant set of professions key to prescribing processes, for example physicians, pharmacists, and nurses. A rich scientific literature supports the value of such rationally constructed multidisciplinary quality improvement teams.


Analogously, it follows that an intervention that targets change among providers of a given type, e.g. physicians, should be rationally designed to affect key providers. The use of opinion-leaders to create change among peers is a well-established example of this approach, and is based on the premise that some individuals are likely to be more effective than others in producing change among their peers. Social network analysis is a well established tool for identifying who such individuals may be, but its potential for use in designing maximally effective health care teams in the hospital setting has not been fully exploited. The overall aim of this proposed project is to develop and pilot a highly innovative approach to use social network analysis to help hospitals develop more effective teams to improve quality. To accomplish this aim we will focus on the care of patients with community acquired pneumonia (CAP) among the Consorta network of hospitals.


For more information contact the project leader or other members of the research team: