Chance makes it impossible to assess reliably whether hospitals are meeting government targets to reduce MRSA infections, argues a statistics expert in this week's British Medical Journal.
The UK government has set a national target for reducing the rate of MRSA infection by 50% by 2008, but David Spiegelhalter, a senior scientist at the MRC Biostatistics Unit Cambridge, warns that setting these targets for individual hospitals is fraught with difficulties.
The basic problem, he says, is that it is unclear whether the targets refer to an observed rate reduction or a true reduction in underlying risk: this ambiguity is unimportant at the national level but, for individual hospitals, chance variation can make the observed rates extremely volatile and make simplistic notions of 'hitting targets' unreliable.
MRSA is an infectious disease and so tends to occur in clusters, making the volatility even worse.
Using data for financial years 2001-4, he found far more variability in the figures than would be expected by simple chance alone. For example, Aintree Hospitals NHS Trust had 34 cases in 2001-2, rising to 66 cases in 2002-3, and falling to 48 in 2003-4.
He therefore suggests that any attempt at ranking trusts into a detailed league table of change would be "entirely spurious."
He also shows that, since high or low rates are largely due to chance events that are unlikely to be repeated, rates in the subsequent year will tend to be closer to the overall average rate (a phenomenon known as regression to the mean). This immediately explains reports of hospitals slipping significantly down the league table from one year to the next.