Clinical Data Mining for Physician Decision Making and Investigating Health Outcomes: Methods for Prediction and Analysis

Writing with healthcare researchers in mind, Patricia Cerrito (mathematics, U. of Louisville) and John Cerrito, a retail and consulting pharmacist, describe how the investigation of healthcare databases can be used to examine physician decisions to develop evidence-based treatment guidelines that optimize patient outcomes. Focusing on patients with chronic illnesses who receive multiple treatments and data from the Medical Expenditure Panel Survey, the National Inpatient Sample, Medpar, and the Centers for Medicare and Medicaid, they cover preprocessing requirements; data and text mining tools like predictive modeling, market basket analysis, and clustering the patient observations; auditing data, cross-referencing variables, and the use of propensity scores; the extraction of subsamples by diagnosis or procedure; the use of matched samples for comparison; problems of changing the observational unit; the many-to-many relationship; large datasets; time series methods; patient compliance; coding and severity scores; estimating probabilities of treatment outcomes and adverse events; and examples of analyses for wound care, asthma and COPD, and diabetes. Medical Information Science Reference is an imprint of IGI Global. Annotation c2010 Book News, Inc., Portland, OR (booknews.com)