Secondary Analysis of Electronic Health Records
Reduzierte Artikel in dieser Kategorie
19,50 €
19,99 €
| Verlag | Springer |
| Auflage | 2016 |
| Seiten | 427 |
| Format | 16,3 x 2,7 x 24,1 cm |
| Gewicht | 872 g |
| Artikeltyp | Englisches Buch |
| ISBN-10 | 3319437402 |
| EAN | 9783319437408 |
| Bestell-Nr | 31943740A |
This book trains the next generation of scientists representing different disciplines to leverage the data generated during routine patient care. It formulates a more complete lexicon of evidence-based recommendations and support shared, ethical decision making by doctors with their patients.
Diagnostic and therapeutic technologies continue to evolve rapidly, and both individual practitioners and clinical teams face increasingly complex ethical decisions. Unfortunately, the current state of medical knowledge does not provide the guidance to make the majority of clinical decisions on the basis of evidence.
The present research infrastructure is inefficient and frequently produces unreliable results that cannot be replicated. Even randomized controlled trials (RCTs), the traditional gold standards of the research reliability hierarchy, are not without limitations. They can be costly, labor intensive, and slow, and can return results that are seldom generalizab leto every patient population. Furthermore, many pertinent but unresolved clinical and medical systems issues do not seem to have attracted the interest of the research enterprise, which has come to focus instead on cellular and molecular investigations and single-agent (e.g., a drug or device) effects. For clinicians, the end result is a bit of a "data desert" when it comes to making decisions. The new research infrastructure proposed in this book will help the medical profession to make ethically sound and well informed decisions for their patients.
Inhaltsverzeichnis:
Introduction to the Book.- Objectives of secondary analysis of EHR data.- Review of clinical database.- Challenges and opportunities.- Secondary Analysis of EHR Data Cookbook.- Overview.- Step 1: Formulate research question.- Step 2: Data extraction and preprocessing.- Step 3: Exploratory Analysis.- Step 4: Data analysis.- Step 5: Validation and sensitivity analysis.- Missing Data.- Noise vs. Outliers.- Case Studies.- Introduction.- Predictive Modeling: outcome prediction (discrete).- Predictive Modeling: dose optimization (regression).- Pharmacovigilance (classification).- Comparative effectiveness: propensity score analysis.- Comparative effectiveness: instrumental variable analysis.- Decision and Cost Effectiveness Analysis: Hidden Markov models and Monte Carlo simulation.- Time series analysis: Gaussian processes (ICP modelling).- Time series analysis: Bayesian inference(Motif discovery in numerical signals).- Time Series analysis: Optimization techniques for hyperparamete r selection.- Signal processing: analysis of waveform data.- Signal processing: False alarm reduction.