TABLE 2-2 Common Cognitive Errors in Clinical Decision Making
|Anchoring||Relying on initial impressions too early in the diagnostic process; failing to adjust initial impressions in light of new information|
|Availability||Judging a situation as being more likely or frequent if it easily comes to mind; judging based on the ease of recalling past cases|
|Framing bias||Tending to be swayed by subtleties in how a situation is presented (e.g., description of the risks and benefits of treatment options)|
|Premature closure||Accepting a diagnosis before it has been fully verified; believing in a single explanation of a situation without investigating other possibilities|
|Reliance on authority||Relying unduly on authority or technology|
SOURCE: Reprinted with permission from Redelmeier, 2005.
Timmermans, 1993; Tversky and Kahneman, 1973, 1974). These mental shortcuts range from overrelying on memorable past experiences to accepting data that confirm preexisting expectations and ignoring data that do not (see Table 2-2 for a summary of five of the most common cognitive errors). Several studies suggest that heuristics are used in health care settings and can have real impacts on patient care (Gandhi et al., 2006; Graber et al., 2005).
In most cases, the shortcut works well to solve the problem at hand (Redelmeier, 2005). Precisely because these shortcuts usually produce the desired outcome, however, most people are unaware of their own susceptibility to cognitive errors. While strategies to overcome cognitive errors in clinical decision making are beginning to be identified (Croskerry, 2002, 2003; Redelmeier, 2005), time and resource constraints, increasing stress among providers, and growing complexity are all barriers to overcoming the risks of these errors.
The volume of biomedical and clinical knowledge being produced has increased steadily over the past few decades. The number of journal articles in biomedical and clinical research fields has quadrupled since 1970, rising from more than 200,000 a year in 1970 to more than 750,000 in 2010 (see Figure 2-5).1 The pace of research now averages 75 trials and
1 The number of peer-reviewed journal publications was determined from searches of PubMed for MEDLINE articles published during a given year using the following MeSH terms: Guideline [V02.515], Journal Article [V02.600], Review [V02.912], Technical Report [V02.989] (National Library of Medicine; http://www.ncbi.nlm.nih.gov/pubmed/).