Safety rather than quality

28 Apr
2009

 

I just received an interesting comment:

Nevertheless, quality tools (which include quality measures) can be useful. A recent example is preventing infections in central lines (NEJM 2006;355:2725-32). Here, an infection rate was observed (the quality measure), it was studied, steps were taken, and the infection rate decreased. The quality tool relies on measuring an error rate.

I agree with the intent of this comment, but do want to make a semantic differentiation.  When I read quality measurement, I think of performance measurement for the purposes of critique or P4P.  The example in this comment is really a safety marker.

So I favor collecting data, and using those data to improve safety.  Here, an increase in infection rate stimulated an investigation which found a treatable cause.  The concept herein is, I believe, a fundamental feature of the safety movement.  I distinguish that movement from the quality movement, just so that I can keep my movements straight.

Safety is very important.  Quality (performance measurement) as commonly used, provides a distraction from the whole of patient care.  We should champion a drive to safety – which is a systems concept.  We should look askew at performance measurement when it is linked to grades or rewards.

 

Related posts:

  1. Measurement – the good and bad
  2. Quality measurement – a delusion
  3. What is quality – a reader asks
  4. How to insure quality
  5. Comparative effectiveness research – more thoughts

Related posts brought to you by Yet Another Related Posts Plugin.

1 Response to Safety rather than quality

Avatar

rjhorn

May 1st, 2009 at 11:17 am

You’ve been too badly affected by the “War is Peace” crowd. Safety is one aspect of quality. In all the fields where quality processes are mature it is recognized that safety is one of the important quality metrics, and that there should never be any implied conflict between safety and quality.

The doublespeak crowd have confused matters by confusing process metrics with quality metrics. This is a commonplace problem and usually goes unmentioned. The difference generally does not matter. It matters in medicine primarily because decision makers are making bad decisions because of misunderstandings about the difference.

Automobile manufacture has a great many similar examples. One
quality metric of an automobile is rust resistance. The quality metric is years without rust while in normal use. This cannot be measured in the factory. So the factory must pick a process measurement, such as ounces of rust coating applied per car, that they can measure. From time to time someone finds a better process metric, such as measuring coating thickness.

As long as all involved remember that the goal and final arbiter is
the field experience of years without rust, this works well. The
factory uses process metrics and changes them when better metrics are found. The explanation to the factory workers is “we can’t measure years without rust, so we measure this instead. If you think of something better, let us know. We will change and use it.”

This needs to be the attitude for most of the healthcare metrics.

Unfortunately, the decision makers in government and senior
management do not understand this. Rather than understand that the real goal is patient health, they make the goal these process metrics and cast them in concrete. There needs to be the
realization that most patient health metrics can only be measured
retrospectively. This means that process metrics will be needed.
But they must be understood as process metrics, and participants
encouraged to find better process metrics that more accurately
predict patient health outcomes.

Comment Form