iSixSigma

Severity, Occurence, Detection scales

Six Sigma – iSixSigma › Forums › Old Forums › Healthcare › Severity, Occurence, Detection scales

Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
    Posts
  • July 7, 2005 at 4:06 pm #25201

    Srinivasan
    ★ 10 Years ★
    Member

    Hi,
    Can somebody give me examples of the severity, occurence, and detection scales used while performing a FMECA of a medical device? AIAG, MIL1629 etc give scales that are more relevant for automotive or electronic industries. I want something more applicable to the medical device (homecare) industry.
    Thanks for the help in advance.
     

    0
    July 13, 2005 at 5:13 pm #60591

    Perryman
    ★ 10 Years ★
    Participant

    How about this one.

    Severity Criteria
    Occurrence Criteria
    Detection Criteria

    Hazardous without warning
    10
    >= 1 in 2
     
    Absolute Uncertainty
     

    Hazardous with warning
    9
    1 in 3
     
    Very Remote
     

    Very High
    8
    1 in 8
     
    Remote
     

    High
    7
    1 in 20
     
    Very Low
     

    Moderate
    6
    1 in 80
     
    Low
     

    Low
    5
    1 in 400
     
    Moderate
     

    Very Low
    4
    1 in 2000
     
    Moderate High
     

    Minor
    3
    1 in 15,000
     
    High
     

    Very Minor
    2
    1 in 150,000
     
    Very High
     

    None
    1
    <= 1 in 1,500,000
     
    Almost Certain
     

    0
    July 13, 2005 at 10:21 pm #60592

    Heebeegeebee BB
    ★ 10 Years ★
    Participant

    I prefer a scale which will lend itself to a high level of discrimination in the data set.
    1
    5
    9
    The problem with a 12345678910 scale is you may get bogged down in “analysis paralysis” due to arguing/haggling over a point or two.
    1/5/9 allows for a quicker and more efficient use of time.
     
    my two cents.

    0
    July 14, 2005 at 3:31 pm #60593

    Ken Feldman
    ★ 10 Years ★
    Participant

    I also make a distinction about the “must be’s”.  In a hospital setting death might be classified as a 10 yet does not occur frequently and is pretty easy to spot so it might drop off of a FMEA.  By declaring any severity of 10 to be “must be” you don’t leave off critical issues.  As an alternative, I will use a scale value of 1,000 or infinity so that I make sure the RPN is noticeable.

    0
    July 20, 2005 at 3:46 pm #60600

    bunny
    ★ 10 Years ★
    Participant

    I highly recommend the rating scale developed by the VA National Center for Patient Safety.  It was published in an article in The Joint Commission Journal on Quality Improvement, May 2002 issue.  The title is “Using Health Care Failure Mode and Effect Analysis: The VA National Center for Patient Safety’s Prospective Risk Analysis System”.
    In a nutshell, it simplifies the severity rating scale into four categories: Catastrophic (10), Major (7), Moderate (4) and Minor (1) Events and the probablity rating scale into Frequent, Occasional, Uncommon and Remote.  Each of these categories has a good operational definition.  These are detailed on page 266 (Appendix 3) in the article.
    I do not have an electronic version of this article, but you can probably get a copy through The Joint Commission on Accreditation of Healthcare Organizations.

    0
  • Author
    Posts
Viewing 5 posts - 1 through 5 (of 5 total)

The forum ‘Healthcare’ is closed to new topics and replies.