We Have to Do Better Than Statistically Significant

Researchers from the Bloomberg School of Public Health at Johns Hopkins have published a paper in the Journal of the American Medical Association looking at the impact of opioid prescribing in Florida after PDMP and pill mill reforms were put in place.  The results are "statistically significant." But statistical significance and public health impact are clearly two different things.

(You can read the abstract and purchase the study here.  You can read a synopsis here.)

The study compared opioid prescribing in Florida and Georgia, an attempt to measure one state's behavior vs. a "control" state that did not implement the reforms that Florida did (at least during the period of study from July 2010 to September 2012).  The results, in a nutshell:

• 1.4% decrease in opioid prescriptions
• 2.5% decrease in opioid volume
• 5.6% decrease in MED per transaction

While we should be celebrating any decrease in opioid use, these statistically significant reductions aren't altogether different than the data we're seeing from our industry's PBM drug trend reports measuring national changes in opioid use.  My reaction to both this study and the PBM drug trend reports is the same: We need to be doing much more to reduce medically unnecessary medications in the treatment of chronic pain.  

Leaving aside some of the shortcomings of this study (the data is from retail pharmacies only) and the frequent confusion among media and industry readers between correlation and causation (something the researchers address, but which is often ignored)... is this really the best we can do?  

It's possible the decreases will accelerate over time.  I'm hopeful that will be the case.  But the detailed results of the study show an interesting phenomena: the reduction in opioid prescriptions written by docs and the reduction in opioid use among patients is very much isolated to high prescribers (80th percentile and above) and high utilizers (80th percentile and above), respectively.  Docs and patients below the 80th percentile actually showed small increases in scripts and use.  

Maybe this shows we're striking the right balance.  Those with legitimate needs still have access while pill mills are being shut down?  Or maybe it shows we still have a lot of work to do... and that PDMPs and pill mill legislation, while absolutely necessary, will be nowhere near sufficient to fix the problem.  

Michael  

On Twitter @PRIUM1