Sunday, March 27, 2016

An Analysis About What?

In a recent article in JAMA Oncology there is a study about the alleged usefulness of PSA tests. Simply stated they did the following:

1. Examined 18 variations on a theme, namely PSA levels and duration between tests.

2. Costs associated with procedures resulting from the tests.

3. Determined the cost of each life year saved compared to not testing.

The result if the QALY measure, that  generally useless, essentially banned by the ACA from use in any medical care. But alas, here it comes.

Their conclusion is simple:

For PSA screening to be cost-effective, it needs to be used conservatively and ideally in combination with a conservative management approach for low-risk disease.

We know that PSA is problematic. But we also know:

1. That PSA testing makes sense if and only if we examine it over time, specifically over a 5 year window of annual data at the very least. It goes up and down and it changes as the prostate grows. So like fasting blood sugar it all too often just tells us what happened last night.

2. Family history is also a significant factor. If you have no relatives with PCa then you have a good chance that you will not have it. Bayes to the rescue. But the corollary is not really true. Namely if you have a first degree relative you may have an increased risk. To be determined.

4. PCais not simple. It is a complex genetic disease and it is highly heterogenetic in its spread. 

5. PCa is all too often  not that serious but when it is it really is. Yet we do not know that boundary.

6. Calculating QALYs without taking the above into account is really really bad in my opinion.

Thus the more of these results that get published the more we confuse the patient. We just do not really know. Remember, if all else fails, listen to the patient. Let them be part of the informed decision process and please keep the QALYs in the UK whee they belong!