• Toxic Cascades: A Comprehensive Way to Think About Medical Errors

    One Pagers | Mar 01, 2001
    Susan Dovey, MD, MPH; Larry Green, MD; Ed Fryer, PhD; Robert Phillips, MD, MSPH

    Current thinking about threats to patient safety caused by medical errors is often focused in hospital on the immediate consequences of mistakes that affect specific aspects of care, such as testing procedures or medications. Some mistakes, however, become apparent distant from where they were committed and only after a lapse in time. The model of a toxic cascade organizes an approach to making U.S. health care safer for patients by locating upstream sources and downstream consequences of errors within a comprehensive, multilevel scheme.

    A Toxic Cascade conceptualizes four levels of threats to patient safety. This model's application is not limited to any particular health care setting. Each part of the cascade occurs in different ways in all parts of the health care system. The cascade can evolve entirely within a single health care location or cut across organizational boundaries.

    1. Trickles: Trickles make virtually no noise, leave behind little trace of their existence and are mostly absorbed unseen. The health system analogy is with errors, such as misfiled records, that clinicians in any setting recognize, that may or may not affect patients directly and that do not cause direct harm. Their immediate consequences are inconveniences, frustrations, and irritations that are frequently addressed as they occur, with no expectation that this will avert similar problems in the future. Downstream consequences are usually unknown.
    2. Creeks: Creeks are more obvious than trickles because they are seen, heard and create barriers to passage. The health care analogy is with mistakes such as prescribing drugs to patients who have allergy as a contraindication. These mistakes worry clinicians because of the potential seriousness of the harm they could cause patients. The immediate consequences of these threats to patient safety are often dealt with where they are detected. Their upstream sources are seldom explored, and their downstream consequences are often unknown.
    3. Rivers: Too big to ignore, rivers may be quiet, but they redefine the landscape. The health system analogy is with mistakes, such as undiagnosed fractures, that result in actual harm to patients. People who are responsible for making health care safer for patients often respond immediately by dealing with the specific type of problem in one specific part of the health system. Clinicians, patients and courts tend to place the blame for these events on individuals, who may be punished and removed. These individuals are followed by others at risk of making the same mistakes in the same systems, leaving upstream error sources unexplored.
    4. Torrents: So powerful that to stop them seems impossible, raging torrents make a noise that drowns out conversation and makes critical thinking difficult. The health system analogy is with errors that kill people. The Institute of Medicine drew attention to 44,000 to 98,000 patients who die from adverse effects of health care in hospitals each year. We do not yet know about torrents in other parts of the health system or enough about the upstream origins of lethal problems.


    From trickles to torrents, toxic cascades likely exist in every health care setting. The comprehensive safety effort that patients deserve will discover the consequence and source of errors, incorporating all locations of care. 

    The information and opinions contained in research from the Graham Center do not necessarily reflect the views or the policy of the AAFP. 

    Published in American Family Physician, Mar 1, 2001. Am Fam Physician. 2001;63:847. This series is coordinated by Sumi Sexton, MD, AFP Associate Medical Editor.