Variation in Treatment and Outcomes in Sepsis
Working with faculty from the Critical Care Division at Baystate Medical Center, we are studying the care and outcomes of more than 200,000 patients hospitalized for the treatment of sepsis. Our work is focused on describing variations in treatment patterns and outcomes across hospitals, examining the effectiveness of common therapies, and exploring the relationship between cost and outcomes of care. In future studies, we hope to identify hospital characteristics and processes associated with high-value sepsis care.
VTE prophylaxis in Hospitalized Medical Patients
We are performing a comprehensive analysis of patient risk, physician prescribing, and outcomes for medical patients at high risk for venous thromboembolism at 400 US hospitals. Specific projects include developing a risk prediction model for developing VTE; describing patient, physician and system factors associated with VTE prophylaxis; and examining the comparative effectiveness and cost-effectiveness of low molecular weight and unfractionated heparins. This project is funded by a grant from the Doris Duke Foundation.
Readmission and Hospital Quality
Readmission of patients following hospital discharge is a clinically important, expensive, and often preventable adverse outcome. Working with colleagues at the Oklahoma Foundation for Medical Quality, we are using Medicare claims files and patient-level data from Medicare's Hospital Compare database to study the relationship between existing quality measures and risk-adjusted hospital readmission rates for pneumonia, heart failure, and myocardial infarction. This project is supported by a grant from the Baystate Medical Center Incubator Fund and is part of a larger effort to examine the effects of public reporting on readmission rates.
The Effectiveness of Antibiotics and Steroids in Patients Hospitalized for Acute Exacerbations of Chronic Obstructive Pulmonary Disease (COPD)
In collaboration with the Division of Infectious Disease at Baystate Medical Center, we are examining the effects of antibiotic therapy on the outcomes of patients hospitalized for acute exacerbations of COPD. In a related project, we are studying the comparative effectiveness of orally and intravenously administered corticosteroids. We previously reported that the vast majority of hospitalized patients are initially treated with intravenously administered steroids despite the fact that current guidelines suggest that oral therapy is equally effective, less expensive, and may be associated with fewer complications.
Potentially Inappropriate Medication use among Hospitalized Elders
The Beers' List identifies medications that should be avoided in persons 65 years or older because they are either ineffective or they pose an unnecessarily high risk and a safer alternative is available. Using administrative and billing data from over 300 hospitals nationwide, we are evaluating variation in prescribing of these medications; association of their use with patient, physician and hospital characteristics; and impact of their use on length of stay, falls, and hospital-acquired delirium.
The Association of Specific Antibiotics with Subsequent Clostridium Difficile Infection In a Large Multi-Hospital Database
The receipt of broad spectrum antibiotics has been shown to be a necessary precursor to subsequent infection with Clostridium difficile. However, no previous studies have identified the antibiotics most likely to be associated with this hospital-acquired infection. Using a large database of over 1,000,000 hospital admissions for common medical diagnoses, we are examining associations between specific antibiotics and the diagnosis of Clostridium difficile diarrhea.
Use of Inpatient Cardiac Telemetry Monitoring: Impact on Patient Safety and Costs
Telemetry is expensive and labor intensive for hospitals. In collaboration with University of Massachusetts Memorial Hospital, Beth Israel Deaconess Hospital, Mt Auburn Hospital and Tufts Medical Center, we are conducting a retrospective examination of telemetry utilization among inpatients. This study will help us to identify rates of appropriate and inappropriate uses of telemetry, find methods for limiting telemetry-associated expense, and improve efficiency of care for patients requiring telemetry.
Relationship between health information technology (HIT) and quality of care
As our country is facing the need to find strategies to improve quality of care while reducing costs, we are examining the relationship between use of health information technologies and in-hospital quality of care. Using a large retrospective database, quality scores from Medicare's Hospital Compare website, and the Health Information Management Systems Society Foundation survey, we are evaluating the performance of over 450 hospitals with various levels of implementation of health information technology. Our rich dataset will allow us to examine the relationship between HIT and many different performance measures, from Hospital Compare composite scores to measures of the daily care provided at patients' bedsides.
