What percentage constitutes “enough” improvement?

Listening to a talk about HIV control… The speaker brings up a goal of “90-90-90” – 90% of the population knows their status, 90% of those are on anti-retrovirals, 90% of those have complete viral suppression. A laudable goal, to be sure, and it would be great for us to get there. But, why 90%? Why not 95% or 85%?

In our department’s quality scorecard, several of our measures have been “stuck” at 92 or 93%, generating some frustration…”Why can’t we move the needle here?” is the typical question we ask ourselves.

Once we’ve decided that metrics are an appropriate way to measure our quality (that’s a whole ‘nuther post or two), we want to construct our measures in a manner that is fair to the group being measured. We tell ourselves first that we are allowing for patient autonomy and choice as well as contraindications to the specific intervention. I can get behind those reasons. But then we add, “difficult to access”, “uninformed”, “they don’t show up to clinic”, etc., then we’re just making excuses for not trying to attend to population health.

So how much grace should we give ourselves for patient autonomy and contraindications? Seems to me we could study this for each intervention, but we don’t – we just SWAG it at 80%, 90%, etc.

Once we’ve defined the target, what do we do when we reach it? We can keep pushing – I see this the most. 93%, 95%, 10 percent more than we did last year, etc. What would certainly be more helpful once we’ve reached a desired level is to examine sustainability…to see if we’ve really achieved process improvement. I suspect lots of quality work is focused on meeting the measures at the last minute – achieving the short term goal, but missing the long term ideal of process change and relinquishing any chance at sustainability.

If we decide to use metrics to measure our quality, we should think through the life cycle of a quality intervention to set challenging but reasonable goals, and monitor for sustainability rather than focusing on higher-and-higher metric achievement. I think this would prevent some of the cynicism that the current metric-focused system is generating and might help with true quality improvement.