If you're going for honours, read this thought piece: but do so with care. The authors argue that using outcome measures to measure quality of care is a waste of time: it's too difficult to adjust properly for casemix and anyway countable events like death are too rare to be informative. So we should use measures of process instead. This is quite a subtle argument. We need to start from the position that we must always measure outcomes before moving on to the refinement they suggest.
The point they don't perhaps make clearly enough is that the process you measure must be of proven effectiveness. Effective process is a good proxy for outcome; ineffective process is no proxy at all. So in public health we can measure a process such as polio vaccine uptake confident that this is a proxy for cases of polio prevented; but measuring number of grommets inserted doesn't tell you that population levels of auditory impairment have reduced.
The authors also, correctly in my view exempt highly specialised services from their argument. So outcome monitoring for paediatric cardiac surgery is ok.
Comments