Pharmacoepidemiology and patient safety in the critically ill

Asad E. Patanwala, Jeffrey F. Barletta, Brian L. Erstad

Research output: Contribution to journalArticle

Abstract

Critically ill patients are at the greatest risk of experiencing preventable and non-preventable adverse drug events (ADEs) compared to other patient populations. Information on ADEs concerning these patients is derived from a wide spectrum of sources such as surveillance studies involving large databases to landmark clinical investigations. Historically there has been a lack of consistency with regard to definitions of commonly used terms such as medication error, ADE, and preventability, which is a fundamental to appropriately conduct and interpret results of epidemiological investigations in this setting. However, attempts have been made to standardize this terminology by national organizations. The increased risk of ADEs in the critically ill patient population is due to a variety of reasons, which include the routine use of the intravenous route of drug administration, medications with a low therapeutic index and high complexity, 'off-label' use and an environment that may be conducive to errors. Reporting of ADEs is largely voluntary and is generally underreported and skewed towards the most serious events. Other methods for surveillance of ADEs in the critical care setting include medical record review and direct observation. Each of these methods has their own limitations and the data obtained can vary depending on the method used. A combination of data from all of these methods is likely to produce the most comprehensive information to improve patient safety.

Original languageEnglish (US)
Pages (from-to)93-104
Number of pages12
JournalPharmaceuticals Policy and Law
Volume14
Issue number1
DOIs
StatePublished - Feb 20 2012

    Fingerprint

Keywords

  • Critical care
  • adverse drug events
  • pharmacoepidemiology

ASJC Scopus subject areas

  • Pharmacology, Toxicology and Pharmaceutics (miscellaneous)
  • Health Policy

Cite this