Confirmation Bias: A Critique of Opioid Guidelines

If you've bothered to keep track of the drama that's unfolded since the election last month (no one would blame if you haven't...), you've no doubt heard the phrase "confirmation bias."  We tend to seek out, the theory goes, news and information that confirms our current view of the world. Opposing views create cognitive dissonance, making us feel less sure about ourselves and forcing us to confront the possibility that we might be wrong (perish the thought). Confirmation bias is something we should all strive to avoid.  Whether its a citizen consuming political news, a fund manager picking a stock, or a GM signing a player... when we pick and choose our data set, we're more likely to make bad decisions.  

I suggest we take a similar approach in the fight against prescription drug misuse and abuse.  

If one truly believes that the best available research, data, studies, and thinking should guide our approach to this public health issue, then one cannot be offended by alternative points of view offered by those who share the same goal.  If one wants to solve the problem, one must consider the other side's view.  There may be more in Health Affairs and JAMA than is dreamt of in our philosophies.  And it's in understanding the critique of our position that we find the nuanced, balanced, and sustainable solutions required to mitigate prescription drug misuse and abuse.  

With this in mind, I recommend reading "Neat, Plausible, and Generally Wrong: A Response to the CDC Recommendations for Chronic Opioid Use" by Stephen Martin, MD, a practicing family physician in Massachusetts who treats chronic pain patients (in other words, the very target of the new CDC guidelines).  Dr. Martin lays out a case against the CDC guidelines that is well written, well researched, and likely to be not well received by readers of this blog.

And that's the point.  If we're going to make progress, let's engage with the sharpest and most well-reasoned points our critics have to offer.  Dr. Martin's arguments boil down to three main bones of contention: First, that the CDC is inappropriately conflating public health initiatives and individual treatment decisions.  Second, that with respect to studies regarding long term use of opioids for chronic pain, "absence of evidence is not evidence of absence."  And third, that opioids can be used safely, even over the long term, in the context of what Dr. Martin calls a "skilled, longitudinal, patient-clinician relationship."

Disagree?  Good.  I mostly do, too.  But I'm not going to do your homework for you.  Read the article, think through his positions, examine his data.  Then develop rational, data-driven responses.  Be prepared to listen to an equally rational and data-driven response back.  And before you know it, you'll be engaged in a legitimate, fruitful dialogue that may, in fact, identify common ground and lead to better solutions than either position might have achieved on its own.

For those that perceive a broader theme to this post, I admit an ulterior motive.  Let's practice data-driven dialogue across our professional, personal, and political spheres and see if we can't mend some broken fences.

Michael
On Twitter @PRIUM1

The Patient Should Not Be Punished for the Sins of the Prescriber

Dr. Mitchell Katz of the Los Angeles County Health Department wrote an editorial that was published in last week's issue of the Journal of the American Medical Association (JAMA has made this set of editorials on opioid prescribing available free of charge).  The title caught my eye - Opioid Prescribing for Chronic Pain, Not for the Faint of Heart.  Indeed.

His intent is to offer a realistic appraisal of the new CDC opioid guidelines.  The tone of the editorial is best described as "great guidelines... but here's how the world really works."  While the pharmaceutical industry appears to have sheathed their swords for lack of argument, the physician community now has to figure out what to do with these new guidelines (if anything at all). Dr. Katz is supportive of the guidelines, but offers his view of their application through a lens of honest practicality.

A few key excerpts (emphasis added):

Even when seeing a patient who has not already begun taking opioids, we physicians have few alternatives for patients who have already tried nonsteroidal anti-inflammatory medications and acetaminophen without relief. Of the pharmacologic and nonpharmacologic options, none is likely to provide rapid pain relief, and none is very effective. Moreover, many of us work in resource-poor systems where arranging for someone to receive physical therapy or cognitive behavioral therapy—two useful therapies—is more difficult than weaning someone from long-term opioid use. To all patients, I give my well-rehearsed speech on why I believe opioids cause more harm than good for chronic pain, but ultimately I will prescribe them for a patient in pain for whom I see no other realistic option.

Embedded in this paragraph are two brutal facts that we must confront: 1) access to non-pharmacological modalities is not easy and cannot be assumed; 2) sometimes, even good doctors are faced with a choice between the lesser of two evils.  I would add that work comp payers should focus on alleviating issue #1 (by approving alternative modalities, going the extra mile to find practitioners, placing alternative non-pharm modalities in-network, etc.) in an attempt to relieve prescribers of burden #2.

Another cogent point from Dr. Katz:

One thing I am certain of: we need to engage patients in an honest and open way rather than quickly writing or refusing to write opioid prescriptions. Given that many patients may be defensive about using opioids, I always speak of my fears. I do not say “You are going through the pills too quickly.” Rather, I say “I am worried that at the doses of medication you are taking, the medicine will harm you.”

Language matters.  This is a subtle but critical point in the fight against opioid misuse and abuse: patient engagement isn't as complicated as we sometimes make it out to be.  Clinicians that are willing to have difficult conversations and who are willing to be honest and transparent with their patients will have more success treating pain than clinicians who find themselves, in Dr. Katz's words, "quickly writing or refusing to write opioid prescriptions."  

Finally, in the context of the complicated matter of opioids and benzodiazepines (you'll have to read his editorial for his full view on the matter), Dr. Katz offers this piece of advice: The patient should not be punished for the sins of the prescriber.  

That should serve as a guiding principle to all of us engaged in this fight.  Above all else, we need to focus on the health, safety, and functionality of injured workers.

Michael
On Twitter @PRIUM1

What the New CDC Opioid Guidelines Will Mean to Work Comp

Not much, I fear.  Allow me to explain...

What the CDC has done here is critically important to addressing the public health crisis that is opioid misuse and abuse.  While the CDC's process came under scrutiny (from, among others, me), the resulting guidelines are evidence-based, well-written, and carry the imprimatur of the preeminent public health agency in this country.  That's groundbreaking.  But I'm not sure we're going to see a significant impact in workers' compensation for three reasons.

Before we dive in, here's a link to the guidelines.
Here's a link to Dr. Tom Frieden's (CDC Director) letter regarding the guidelines in NEJM.
And here's perhaps the most practically valuable thing the CDC published earlier this week - a checklist for primary care physicians who prescribe opioids (this is excellent).

Reason #1 we're unlikely to see a significant impact in work comp: It takes a long time for new medical evidence to penetrate actual physician practice.  The guidelines have received a lot of press coverage over the last few days, but busy primary care physicians may not be immediately responsive to new medical evidence even if they see it on the front page of the local newspaper. First, not all physicians have the same faith in an agency of the federal government that I possess.  In fact, for some, the notion that the federal government published these guidelines may be the primary source of skepticism.  Second, there will armies of pharma reps to gently, diplomatically, but firmly push back, find holes, work arounds, etc. to increase the likelihood that current prescribing practices remain intact.  Third, there are obviously no enforcement mechanisms in connection with these guidelines.  A primary care doc who chooses not to follow them will face no immediate consequences (though, we should be clear, the long term consequences to the patients of such a doctor could be catastrophic).

Reason #2: Our primary cost drivers in work comp are long term, chronic pain cases.  The new guidelines offer precious little guidance for these types of cases.  Most of the guidelines focus on opioid initiation and to the extent chronic opioid therapy is addressed, the guidelines suggest avoiding it.  Well... what if we have an injured worker who has been on opioids for that past 10 years?  Whose dose has escalated regularly and dangerously over that period?  The guidelines suggest those opioids should be weaned.  Right.  Telling a primary care doc to simply wean a patient off of opioids in the midst of a long term, complicated, polypharmacy drug regimen is perhaps expecting too much.  I would have liked to have seen more detailed guidance on how to deal with such complex patients.  So why didn't the CDC go there?  Because it's really complicated, that's why.

Reason #3: Primary care docs, by and large, didn't create most of our pain management issues in work comp.  Granted, I'm dealing with a very skewed subset of cases here at PRIUM.  I recognize we suffer from adverse selection, so this might not be accurate for the entirety of the work comp universe.  But what we see is that surgeons and pain management specialists tend to initiate complex pain management drug regimens (after the profitable procedural work is done) and then, in perhaps the most unkindest cut of all, the patient is discharged back to the primary care doc... who is now overwhelmed by a monster of a drug regimen that he did not create.  Yes, primary care docs write almost 10 times more opioid scripts per year (28 million) than pain management and interventional pain management doctors combined (3 million), but will these new guidelines - aimed at primary care docs - help them much if they're not the ones making the initial prescribing decision?

I'm thrilled the CDC published these guidelines.  I think they represent useful, cogent, and practical thinking.  And I hope I'm wrong that we won't see a material impact in work comp.

Michael
On Twitter @PRIUM1 (just click the link to follow!)

Guidelines: How to Make Them Matter

There has been a lot of talk in the last couple of weeks about medical treatment guidelines.  North Carolina, Tennessee, Arizona, and several other states are looking into adoption of guidelines.

There is a lot of discussion about "evidence-based" vs. "consensus-based" guidelines as well.  While I'm an outspoken proponent of the evidence-based variety, I recognize that local political conditions can make adoption of such 3rd party guidelines difficult.  Politics is a poor excuse - states can benefit significantly not only from the credibility of 3rd party guidelines, but also from the fact that providers of such guidelines work constantly to ensure those guidelines reflect the most contemporary view of the medical evidence.  Consensus-based guidelines (where a state medical director or agency gathers a bunch of stakeholders and they all decide what's best for injured workers in a given state) tend to be static - they age... and they tend not to age well.  They also tend to be subject to influence from outside parties with vested economic interests (see Louisiana as an example).

But the debate between the two competing approaches to guidelines is actually a sideshow, a distraction that prevents regulators from focusing on what matters most: enforcement.

When reading proposed or adopted treatment guidelines for a given state, make sure you explore the extent to which you can successfully challenge treatment that falls outside of the guidelines.  For instance, let's say a medical treatment guideline states that prescribing opioid analgesics in excess of 50 mg MED daily is not recommended (as the ACOEM guidelines suggest).  And assume you have a case where the doctor is prescribing 250 mg MED daily... and that doctor has been unresponsive to requests for discussion regarding the case and has not provided the rationale for why he's prescribing outside of the evidence based guidelines.  Question: Now what?  

There are states that have good answers to this question.

In Texas, if those opioids are N drugs, the answer is required pre-authorization based on a 3rd party guideline (Official Disability Guidelines) and a utilization review process that is statutorily supported and has clearly defined dispute resolution mechanisms.

I was encouraged to read that in Tennessee, the goal is to ensure that any adoption of treatment guidelines is consistent with existing utilization review regulations (which include the ability for a payer to subject any schedule II-IV medication used for the purposes of pain management for more than 90 days to utilization review).

Those are guidelines with enforcement mechanisms.  If the adoption of guidelines doesn't come with enforcement mechanisms, they're still worth having (because good doctors will still be responsive to guidelines).  But those guidelines won't be as valuable as they could be if appropriate enforcement mechanisms are in place (because not-so-good doctors will ignore contemporary medical evidence and continue to administer sub-optimal treatment until the payer stops paying for it).

Michael
On Twitter @PRIUM1