Friday, November 4, 2011

The Affect Heuristic

Source: XKCD.com
We've discussed conflicts of interest and bias frequently, particularly in regards to guidelines.  Bob Centor has a nice post today that discusses the affect heuristic and how it might impact clinical guidelines.

The affect heuristic suggests that a good feeling or emotion towards a situation (i.e., positive affect) would result in a person having a lower risk perception and higher benefit perception than supported by an unbiased look at the data. I think this is closely related to confirmation bias, which we have written about frequently, as well.

His recommendation for selection of guideline panels - choose strong methodologists who are non-experts - is one worth pondering.

2 comments:

  1. "His recommendation for selection of guideline panels - choose strong methodologists who are non-experts - is one worth pondering."

    As one living in a twilight zone run by PR bunnies,that really cheered me up.

    "If you pay for good statistics, that's exactly what you get: good statistics."

    --
    Regards
    Pat Gardiner
    Release and independently audit the results of testing British pigs
    for MRSA, C.Diff and Hepatitis E now!
    www.go-self-sufficient.com and http://animal-epidemics.blogspot.com/

    ReplyDelete