Bad Performance Measurement on Tour (#2)

Just one day after Episode One of this tragedy series, one of my faithful readers told me about a target so ridiculous that it would be laughable if it wasn’t so serious.

During a hospital training session, those attending were informed that their NHS trust has a target of 2 MRSA cases this year. The trainer followed this up by saying, “We’ve already had 2, so that’s it, we can’t have any more”.

This means that they AIM for 2 cases of MRSA per year. It means that 2 cases per year is acceptable. It means that 2 cases per year is in the plan. It means that management can relax if ‘only’ 2 patients contract this horrible disease this year because the target will have been achieved. How strange.

These targets are like the train company in my previous post aiming for 10% of trains to be late. It’s akin to public satisfaction targets that aim for 12% of people to be dissatisfied, or crime detection rates that aim for 50% of crime to remain undetected. Think about it.

And this isn’t a case of misunderstood or misreported facts either. If you Google ‘MRSA targets’ you can see them for yourself. Here’s one:

“The Trust year to date performance is 2 cases of MRSA against a year to date target of 8.”

(It’s on page 2 of the document if you care to look).

Why 8? Or why 2? Why does the counter reset back to zero after 365 days (or 366 days if it’s a leap year)? As ever, these targets are completely arbitrary and do nothing to actually prevent MRSA.

How about this for an alternative approach – aim to totally prevent MRSA in hospitals (i.e. a ‘target’ of zero). Surely that’s what everyone really wants. If, for whatever reason, cases still occur, then use real measures to find out why, and take additional steps to tackle known causes.

MRSA ignores targets, as do all forms of data. If you do the right thing, the figures look after themselves.


About InspGuilfoyle

I am a serving Police Inspector and systems thinker. I am passionate about doing the right thing in policing. I dislike numerical targets and unnecessary bureaucracy.
This entry was posted in Systems thinking and tagged , , , , , , , , . Bookmark the permalink.

7 Responses to Bad Performance Measurement on Tour (#2)

  1. patently says:

    I’d hate to be the last patient admitted during a financial year in which they were below target for MRSA infections…

  2. tactical says:

    somewhat similar, when the SARS ‘explosion’ well, exploded, (don’t remember now if it was 2003 or 2004 or when exactly but I am sure you know) I was told pretty much the same. On this occasion, at that stage, there were ONLY 9 people who had died and when I mentioned that we should do something – without going into detail – , I was told that 9 people dying of a highly contagious desease is nothing compared to how many people there are in the world!!! Never mind decontamination, informaiton or communication across various agencies oh no, that would have been too much of a logical step. Needless to say, a couple of days later, they all got their skates on as the desease was spreading

  3. Pingback: British Policing: A Crime In Progress! « Dave's Bankside Babble

  4. yo mo says:

    Very bold use of the word ‘arbitrary’ some people get a right cob on if you do that. 😉

    Can I go back to wrapping data into a X-chart?
    What is the sample length? month? quarter?
    And how do you calc the centre line value? average? mode? median?


    The linked document is depressing, 90% bovine output, the rest meaningless piffle.

    • Morning. Yes, telling people that numerical targets are arbitrary can invoke reactions not unlike those expected if telling a roomful of young children that Father Christmas doesn’t exist. For a straightforward explanation about how to calculate where the centre line should go and how to put the chart together from a dataset, this is a good reference:
      Joiner, B. (1994) Fourth Generation Management. New York: McGraw-Hill (pp.148-149)
      Also, Wheeler, D.J. (2000) Understanding Variation: The Key to Managing Chaos. (2nd Ed.) Knoxville: SPC Press, and Wheeler, D. (1998) Avoiding Man-Made Chaos. Knoxville: SPC Press (The latter is not so easy to get hold of – although you might pick up a second hand copy – but it also contains a great article by Myron Tribus). For simplicity (i.e. stuff that I understand), I’d go for Wheeler’s 2000 book; for a quick and direct reference, those pages from Joiner should suffice.

      Oh, I like ‘bovine output’. I might use that. cheers.

  5. yo mo says:

    Cheers MrG. I found an on-line reference, and worksheet thanks to you pointing out Joiner.

    It seems to be the median of the intervals between data points for ‘R’ and the arithmetic mean of data points for ‘X’, (which gives the CL).
    (pi x R) + X = UCL and – X for the LCL.

    Now where did I put my crayons?

    Bit pointless to wish you a good weekend, you being on shifts, so instead TTFN.

  6. Targets in Education:
    After 25 years as a primary school governor (10 as Chair of Governors) I thought I’d seen it all when it came to targets, but now I’m not so sure…
    Latest edict from our local education authority is to measure performance and progress of children (5-11yrs) across a range of academic and physiological parameters (don’t ask) with an expectation that: ‘85% will be WELL ABOVE NATIONAL EXPECTATIONS’ [in a particular parameter. OK, “PSED”, since you ask]. If ‘national expectations’ can be considered a quasi-indicator of ‘average’ (my interpretation, tell me if I’m wrong) teachers are expected to turn out 85% above average in that parameter…??? Leaving aside the fact that our small village school = small statistical population (i.e. 1 child = 10%) and above-average (sorry!) numbers of SEN children on roll, how on earth do they achieve that?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s