Monday, July 10, 2017

Which Arm of the Trial Would You Pick?

For years, critics of opioid guidelines have argued that the absence of evidence on long term efficacy of opioids wasn't a sufficient reason to declare such long term use "medically unnecessary." Advocates of opioid therapy have been frustrated by the notion that because we didn't have a randomized clinical trial (RCT) showing the drugs to be helpful, we all assumed them to be harmful. While practical experience led to this totally plausible conclusion, detractors still saw hypocrisy in our dependence on documented medical evidence... that still didn't exist.

That RCT has finally arrived.

The study is to be published soon, but Dr. Erin Krebs recently presented her findings at the 2017 meeting of the Society for General Internal Medicine in Washington, DC.   You can read a synopsis and commentary in one of my favorite monthly medical newsletters, The Back Letter (subscription required).  Here's the key take-away:
Opioids provided no advantage in terms of function at the 12-month follow-up mark, and patients in the opioid wing of the study actually reported marginally more pain at 12 months than those in the non-opioid group.  
There are really two key points there: the first, that opioids don't appear to increase functionality, is obvious to most of us who have been living and breathing this issue for the last decade.  The second, though, refutes the most fundamental argument of opioid advocates - that the drugs at least control pain.  Turns out, they don't do a good job of that either.  To sum up, the opioid creates the exact opposite of its intended effect: it fails to increase function and instead actually increases pain.

Perhaps the most interesting characteristic of the study was its design.  It was randomized (patients are chosen by chance to participate in one or more clinical interventions - in this case, opioids for pain relief) and controlled (the clinical intervention was compared to a standard practice - sometimes placebo, sometimes not - as in this case - a non-opioid drug regimen).

But it wasn't blinded.  A blind trial indicates that the participants don't know which arm of the study they've been assigned to.  In this study, patients getting opioids knew they were getting opioids and patients not getting opioids knew they weren't getting opioids.  To me, that makes the results even more compelling.  Patients who knew they were receiving opioids to manage their pain still reported higher pain levels at the 12-month mark.  And patients who knew they were not getting opioids still reported increased functionality vs. the opioid group.

Apparently, there was some ethical debate regarding this study before its launch. It was assumed by some that such a trial would be impossible because people in chronic pain would simply refuse to join the non-opioid arm of the trial if that's where they were randomly assigned.  Further, some assumed it would be unethical to not provide opioids to people in chronic pain - that by instituting a non-opioid control group, patients would be deprived of necessary medical care.

As it turns out, the arm of the trial some ethicists assumed would be, in the first place, impossible to fill, and secondly, inhumane... well that's the arm of the trial you actually wanted to be in.

On Twitter @PRIUM1

Wednesday, July 5, 2017

Detox and Marshmallows: Dealing with Addiction

On Monday, NPR's Morning Edition ran two pieces back-to-back that while ostensibly not intended to be related, nonetheless struck me as providing important insights into opioid misuse and abuse.  

The first story (read and/or listen here) was about the potential perils of Do-It-Yourself detoxification from drugs of addiction. While there's plenty in this story for the schadenfreude crowd ("he says his stomach cramps felt like 'having Freddy Krueger inside you trying to rip his way out.'"), I was struck by the comments of a doctor interviewed for the story:
So can detoxing on your own be the solution? In most cases, the answer is no. 
In fact, a growing movement within the field of addiction medicine is challenging the entire notion of detox and the assumption that when people cleanse themselves of chemicals, they're on the road to recovery. 
"That's a really pernicious myth, and it has erroneous implications," says Dr. Frederic Baurer, president of the Pennsylvania Society of Addiction Medicine.
"Detox" does not equal "treatment."  Treatment may, of necessity, start with detox, but without counseling and the potential use of other medications, recovery is rare.  In fact, the relapse rate from detox alone is upwards of 90%.  We have to do more than just detox if we want long term results for injured workers suffering from addiction.  

The second story (read and/or listen here) was about marshmallows.  More accurately, it was about a child psychology experiment involving marshmallows.  In the the 1960s, a Stanford psychologist named Walter Mischel designed this experiment to study children's self-control.  Kids ages 3 to 5 have a marshmallow placed before them. Then researchers give the child the following instructions: You can eat the marshmallow now, but if you can wait for me to return, you'll get two marshmallows.

More than half of kids dig in.  And among those who don't gobble up the treat but instead exercise self-control, there appears to be a correlation to superior future academic performance and achievement.  (The story is interesting because new research deploys the experiment outside of western culture for the first time and the results are interesting, if not concerning).

The objective of the experiment is to study a psychological phenomenon called "delayed gratification."  Can one put off immediate gratification in return for greater, albeit delayed, reward?  Even among 3 to 5 year olds, the delay creates physical and emotional distress.  They whine, they squirm in their seats, their heart rates go up, they feel real stress.

Am I comparing a 4 year old who eats the first marshmallow to an opioid addict who can't go a day without a fix?  Absolutely not.  Addiction is a disease and it requires treatment.  It's not a simple failure of willpower.

Rather, I'm comparing the 4 year old who eats the first marshmallow to some insurers, employers, government regulators and politicians who want a simple, cheap, fast, pop-the-balloon solution for an injured worker who has been on opioids for 10 years.  Relying on detox alone is like gobbling up that first treat.  Instead, we need to squirm, face hard choices, make investments in sound treatment, and exercise patience.  

We could use a little more delayed gratification in the fight against opioid misuse and abuse.

On Twitter @PRIUM1

Wednesday, June 21, 2017

Will Opioids Kill the AHCA?

As a small group of Senators toil away in secrecy on an effort to recast a sixth of the American economy, one of the sticking points in the legislative negotiation is funding for the opioid crisis.

While substantial cuts in Medicaid seem destined to make it into the bill in one form (a basic cut in funding) or another (a shift to state-level block grants), several Senators from both parties are lobbying to include $45 billion over 10 years for the fight against opioid misuse and abuse, primarily aimed at availability and access to addiction treatment.  Moderate Republicans like Rob Portman of Ohio, Shelley Moore Capito of West Virginia, and Susan Collins of Maine have made this a central issue in work toward a Senate healthcare bill.  A few of their Democratic colleagues, namely Joe Manchin of West Virginia and Bob Casey of Pennsylvania, are arguing for even more funding - the $45 billion over 10 years isn't nearly enough in their view.

Note those states: Ohio, West Virginia, Maine, Pennsylvania.  These Senators are doing what Senators are supposed to do: represent their constituents. Portman has gone so far as to state publicly that he won't be able to vote for a bill that doesn't include this funding.  Keep in mind that to pass the American Healthcare Act (AHCA), Republicans can only afford to lose two of their 52 votes.  If they lose three, the bill won't pass.  

The opioid crisis has created one of the precious few areas of bipartisanship I can recall over the last several election cycles.  We might see legitimate arguments over appropriate funding levels, but the necessity of action is unquestioned and the focus on prevention and treatment is almost universally shared.  (Notably, one person who doesn't appear to share the view that prevention and treatment are superior tactics to criminal justice solutions is former Senator and current Attorney General Jeff Sessions: he'd rather return to a set of failed policies that have done nothing to stem drug-related crime in this country).

Personally, I'm torn.  I want to see substantial funding for prevention and treatment of addiction.  At the same time, if the AHCA dies in the Senate over this issue, it will serve to shine a very bright light on opioid addiction and simultaneously prevent a very bad bill from becoming law.

On Twitter @PRIUM1


Monday, June 12, 2017

Want to Decrease Disability by 53%?

According to a recent issue of Health Affairs, all we have to do is completely eliminate five risk factors: smoking, obesity, diabetes, high cholesterol, and hypertension.  

Easy, right?

I find it amazing that these risk factors contribute to (potentially, depending on the credibility you lend the study) more than half of all disability in this country.  And given that the study (Preventing Disability: The Influence of Modifiable Risk Factors On State and National Disability Prevalence) is written from a non-work comp perspective, I view this as more of challenge in our industry (where we accept the whole person and have relatively little influence over pre-injury behavior).

If the theoretical elimination of all five risk factors is a bridge too far for you, consider a more conservative analysis contained in the study: If each risk factor was reduced to the level of the "best performing" state (i.e., if all states mirrored the nation's lower obesity rate of Colorado), we would observe a decline in disability prevalence of approximately 7%.  And disability rates in regions where prevalence is highest (South, Appalachia, and Great Lakes) would drop more than 10% under such a scenario.

But our starting point is grim.  In the 18-54 age cohort, nearly 70% of US adults have more than one of the five risk factors.  In the 55-64 cohort, it's about 90%.  And in the 65-79 category, about 95%.

This isn't just clinical, it's cultural.

On Twitter @PRIUM1

Tuesday, April 11, 2017

Demanding Better Psychotherapy

Psychotherapy is undergoing a veritable revolution.  And if you're not paying attention, you could miss opportunities to change attitudes, spend less, and save lives.

For nearly a century, the field of psychotherapy (which includes cognitive behavioral therapy and other modalities common in the treatment of chronic pain) has been deemed a subjective and ethereal art based on human relationships, perception of progress, and patient self-reported outcomes. Therapists argued such things couldn't be measured, objectified, or codified.  This feels intuitive to most of us - how can the assessment and treatment of behaviors, thoughts, emotions, and reactions be reduced to mere data points, bar graphs, or pie charts.  Contemporary medical evidence appears to suggest these modalities are efficacious and that's sufficient for most of us to accept the status quo in the field.

The latest issue of The Atlantic Monthly contains an article by Tony Rousmaniere, What Your Therapist Doesn't Know.  Recognize that last name?  Only after reading the article and penning the first draft of this post did it occur to me to reach out to the estimable Peter Rousmaniere to ask if Tony was any relation.  In fact, Tony is Peter's oldest son.  I should have guessed.

In the article, we learn about feedback-informed treatment (or FIT).  In a space where 25 percent of patients drop out of therapy (likely higher among chronic pain cases, but that's just a guess on my part) and 5 to 10 percent of patients actually get worse during the course of treatment, wouldn't it be nice to be able to quantify, and perhaps even predict, patient progress?  Turns out, over 50 different (and, I suppose to some extent, competing) feedback systems have been developed over the past 20 years.  Most involve detailed questionnaires administered to patients and designed not only to measure progress, but also to help therapists identify blind spots (like when a patient might be offering less than truthful feedback directly to the therapist... or about to drop out of treatment... or getting worse).  One such feedback system was able to predict - with 85% accuracy and after only three sessions of therapy - which patients would deteriorate.  

New medical technologies, practice techniques, and methodologies can take a long time to be adopted into every day clinical application.  Rousmaniere discusses the history of the thermometer - at one point, taking a patient's temperature and using that data as a tool in diagnosis was considered heretical and potentially dangerous to the practice of medicine (in that it might make doctors lazy and dull their skills as diagnosticians).

The time has clearly come for injecting data, metrics, and objective performance feedback into psychotherapy.  And since payers are hearing the constant drum beat of "psychosocial... mental health... CBT...", the thought occurs to me that the least we can do, if we're going to pay for this apparently efficacious intervention, is demand feedback-informed treatment for injured workers.  We would demand no less in virtually every other area of medicine.

Why settle here?

On Twitter @PRIUM1

Tuesday, March 28, 2017

US System Lags in Back Pain Treatment

The US healthcare system is, in a word, dysfunctional.  We know this.  Many of us have experienced it first hand.  We spend a lot and have relatively little show for it:

There are many reasons for this that will require a plethora of future blog posts (political reasons, economic reasons, cultural reasons, etc.)  But today, we'll take a quick look at what happens in a country that adopts evidence-based guidelines in a rigorous, widespread, methodical way and tracks the results of those changes over time.

The Netherlands is a relatively small country northwest of Germany, making up the southern border of the North Sea.  Known for the laissez-faire reputation of its largest city, Amsterdam, the country's healthcare system has a few notable features that are anything but: health insurance is mandatory, the country has a well-established and well-organized network of 160 primary care centers, and both insurers and hospitals are mostly for-profit and compete for business among consumers.

Back in 2011, the Netherlands decided to address the rising cost and complications associated with low back pain.  They did so by essentially mandating adherence to evidence-based guidelines (through highly differential reimbursement - pay for the things that work, pay little/nothing for things that don't) and simultaneously incentivizing return-to-work within the social security and disability programs available to citizens.  Because of its relatively small size and well-organized primary care system, the country was able to gain real traction around evidence-based guidelines within the clinical community.  They reduced spend by 20% over 5 years (this included both direct spend - like clinical services - and indirect spend - like social insurance and disability payments).   

The closing paragraphs of a recent story addressing this study found in the pages of one of my favorite public health newsletters, The Back Letter, summarizes the potential obstacles and opportunities that the US healthcare system presents:
When these reforms were enacted, some providers and patients worried that they would negatively impact outcomes—by limiting their treatment options. But there has been no evidence that these changes have had a negative impact on the health and well-being of the general public. 
However, some of these policies might be difficult to enact in the US. Any suggestion of limiting treatment options in the US sparks protests about “rationing” and “denial of care.” And wholesale limitations on ineffective treatments for back and neck would draw huge opposition from drug and device companies, healthcare systems that employ ineffective treatments profitably, and professional societies whose members routinely offer ineffective or unproven care. 
How does one get a society with an entitlement mentality, with mistrust of any form of regulation, to accept evidence-based limitations on care? Perhaps the recent saga with opioids provides some insights – as the consumer tide has clearly turned against the excessive and deadly prescription of opioids for back and other chronic pain problems.
The key elements in bringing about this transformation in attitudes were vocal public advocates, compelling stories of bad outcomes, easily-understood information, and credible scientists taking strong public stands. These same elements might be brought to bear to reduce inappropriate levels of surgery, injections, pain interventions, and ineffective non-invasive treatments. 
However, the opioid debacle is an extreme situation, given its heavy death toll and destructive impact in every US state. Will healthcare providers, patients, policy makers, and payers be willing to make similar cutbacks to all forms of excessive back care? There is certainly a growing consensus that spine care has to be a primary target in efforts to dramatically reduce spending. However, how that process will play out remains to be seen.
 Who knew healthcare was this complicated?

On Twitter @PRIUM1

Monday, March 20, 2017

A Way to Fix a Drug Problem Without More Drugs?

Like many of my colleagues here at PRIUM and acquaintances outside of PRIUM who focus on chronic pain day in and day out, I get all manner of articles forwarded to me from friends about opioids and related medical treatment advances.  There's the spider venom that may hold the key to a more effective non-opioid painkiller.  There's the big-data, molecular-lottery approach that promises to identify medications capable of delivering pain relief without opioid side effects.  There's the on-going debate regarding medical marijuana's potential to stem opioid use.  This is just a sampling of the many studies, articles, ideas, and whims that appear in my inbox on a regular basis.

The theme that sticks out to me is the collective focus we seem to have on fixing a drug problem with more drugs.  

This reminds me of the unsettling moment at last year's National Prescription Drug Abuse and Heroin Summit here in Atlanta when Surgeon General Dr. Vivek Murthy offered some opening remarks prior to the arrival of President Obama at the conference.  In those remarks, he outlined a five-step strategy to combat prescription drug abuse and heroin and the first two steps were medication-based (#1 was expanding access to naloxone and #2 was expanding access to suboxone).  I think highly of Dr. Murthy and applaud his critical efforts to combat addiction in all its forms.  And I'm not even sure his plan isn't exactly what we need.  I was simply struck, again, by this theme of fixing a drug problem with more drugs.  It seems somehow counter-intuitive to me or, at least, not getting at root-cause issues.

So I was intrigued last week when I read about a study out of the University of Utah, Duke University, and Washington University in St. Louis.  The collaborating bioengineers have figured out a way to potentially manipulate our genetic code to suppress chronic pain.  For those rightly uncomfortable with the idea of messing with human DNA, the researchers are not editing or replacing genes.  Instead, they're using something called the CRISPR (Clustered Regularly Interspaced Short Palindromic Repeat) system to modulate the way genes turn on and off in order to protect cells from inflammation and tissue breakdown.  Early experiments point toward the possibility of eliminating the inflammation, cell death, and tissue damage associated with, say, low back pain caused by a herniated disc.  Sounds useful.  

Three quick observations:
1) This is at least 10 years away from human application.
2) Even then, it's not a panacea.  Surgery may still be required to fix underlying, biological causes of pain (though wouldn't it be nice to isolate that from non-biological, psycho-social contributors to pain?)
3) This work was funded by a National Institutes of Health grant, the likes of which could disappear if the current draft White House budget were to be adopted.

On Twitter @PRIUM1

Tuesday, March 7, 2017

What's Past is Not Prologue: Formularies Differ by State

I've recently heard predictions regarding California's forthcoming formulary that are based on the experience of Texas over the last five years.  Both are large states.  Both have well recognized, embedded utilization review processes.  And they have tertiary dispute resolution processes that, while not precise analogs, nonetheless look similar to one another (the Texas IRO process and the California IMR process).

So we might expect California, when it adopts it's drug formulary on or about July 1, to exhibit behaviors and results that are similar to the results Texas has achieved with its formulary (see my colleague Mark Pew's recent blog posts for insight into the CA timeline).  In Texas, prescriptions for "N" drugs fell by 81% and the costs for those same drugs dropped by 80%.  Total opioid scripts dropped by 8% and total drug costs in the state's work comp system fell by 15%.  All of this occurred with no discernible increase in loss adjustment expense (primarily, utilization review). So we might expect the same from California, right?  

I don't think so.  I offer three key data points in defense of that view:

First, the formularies are not the same.  Texas relies on Work Loss Data Institute's Official Disability Guidelines, Appendix A (which lists medications are either "N" or "Y" based on whether they are recommended for first-line therapy).  California has used as its starting point Reed Group's MDGuidelines in order to incorporate a formulary into MTUS.  The list of "non-preferred" drugs in these guidelines is a different, and frankly more restrictive, list than the ODG list of N drugs.  For instance, ODG has some opioid analgesics in the "Y" category.  The MDGuidelines categorized all opioid analgesics as "non-preferred." The two organizations take different approaches and I'm careful to avoid expressing preference for one or the other (PRIUM works closely with both sets of guidelines).  The important take-away here is that the two states are relying on different formulary approaches.

Second, about 38% of CA's lost-time claims have attorney representation.  That number for TX is 6%.  There are lots of reasons for this (attorney fee schedules come immediately to mind) that have nothing to do with formularies.  But adopting a formulary in a state with injured worker representation in the single digits is a fundamentally different proposition than adopting it in a state where 4 out of every 10 injured workers on indemnity have a lawyer.

Finally, loss adjustment expense in CA is already running at about 35% of total losses (which is 83% higher than the median in WCRI's latest study on LAE).  When the TX formulary came into full effect in 2013, the number of IRO decisions from 2013-2015 ranged in the 1200-1400 range per year over that three year period post-formulary implementation.  The number of IMR decisions in CA in 2016?  164,136.  And this is before the adoption of a formulary. Long term, as the prescriber community adjusts to the formulary list and its associated rules, the number of medication-related IMRs may in fact moderate.  But for the latter half of 2017, I wouldn't count on that happening.

Two different states, two different formularies, two different environments.  Beware of drawing conclusions regarding one based on the other.

On Twitter @PRIUM1

Monday, February 27, 2017

What if We Hypothesized Instead of Diagnosed?

The 19th century Romantic poet John Keats was also trained as a physician.  And it's his definition of a "man of achievement" that gives Arabella Simpkin and Richard Schwartzstein their jumping off point in a fascinating essay in the New England Journal of Medicine.  Wrote Keats regarding the necessary quality of such a person: "...when a man is capable of being in uncertainties, mysteries, doubts, without any irritable reaching after fact and reason."

How many physicians do you know who would agree with that sentiment?

Simpkin and Schwartzstein go on to make a noble and necessary argument for the place of uncertainty in modern medical practice.  While the proposal is made for the entirety of the profession and the full span of specialties, I was struck by the applicability of the argument to chronic pain management, specifically.
Too often, we focus on transforming a patient’s gray-scale narrative into a black-and-white diagnosis that can be neatly categorized and labeled. The unintended consequence — an obsession with finding the right answer, at the risk of oversimplifying the richly iterative and evolutionary nature of clinical reasoning — is the very antithesis of humanistic, individualized patient-centered care.
This is how non-specific low back pain turns into a 15 year old work comp claim.

The authors make several recommendations, each of which targets the very heart of medical education and clinical practice culture.  We need to cease viewing uncertainty as a threat, but rather embrace it as part of the iterative nature of care.  We need to move away from multiple choice tests in medical education that require definitive answers and instead focus on evaluating medical students' tolerance for uncertainty and ability to posit based on incomplete information.

Perhaps the most daunting recommendation they make, though, is the idea of moving away from concept of diagnosis and instead focusing doctor-patient conversations on the concept of hypothesis.
We can speak about “hypotheses” rather than “diagnoses,” thereby changing the expectations of both patients and physicians and facilitating a shift in culture. This shift may entail discussing uncertainty directly with patients, intentionally reflecting on its origins — subjectivity in the illness narrative, diagnostic sensitivity and specificity, unpredictability of treatment outcomes, and our own hidden assumptions and unconscious biases, to name a few. We can then teach physicians specifically how to communicate scientific uncertainty, which is essential if patients are to truly share in decision making, and we can reduce everyone’s discomfort by reframing uncertainty as a surmountable challenge rather than as a threat.
This requires treating the whole patient.  This requires recognizing the psychological and social contributors to pain perception and tolerance.  This requires seeing through the psychotropic effects of opioids and other addictive medications to get to the root cause issues of chronic pain.  This requires not just patient advocacy, but truly shared decision making.

Of course, this would also require us to move away from the dilapidated and counter-productive world of fee-for-service billing.  This would also require payers and employers to pay for quality, a genuine willingness to pay more dollars for less care (in the traditional sense) and more shared decision making (which will inevitably lead to better outcomes).

So there's a little slice of utopia for you, this Monday morning... with a heaping side helping of reality.

On Twitter @PRIUM1

Monday, February 20, 2017

Narrative Does Matter: Self-Guided Opioid Weaning

Perhaps it's obvious.  I write blog posts with regularity, I consume news voraciously, and I've never met a microphone I didn't enjoy speaking into.  But in the event it's not readily apparent, I'm happy to share that the single most important concept in contemporary communications is this: narrative.

"Narrative" is occasionally used as an epithet in political discourse (as in "you're just choosing facts that fit your narrative") and I'm as concerned as anyone else about the balkanization of modern media (which I describe as "choose your own narrative"), but the power of well-told stories to shape, change, or at least influence thinking is undeniable.  We live in a world of competing narratives and while the ability to identify such is critical, the ability to create such can be transformative.

Health Affairs understands this.  Whenever I get a new issue, I typically flip directly to my favorite section: "Narrative Matters."  Here, public health workers on the front lines share stories of what it means when platitudes turn into policy.  It's one thing to talk in the abstract about the CDC's Ebola response... it's quite another to listen to a doctor tell the story of running an Ebola clinic in Liberia.

This month's issue contains another in a long line of compelling stories, though this one hits close to home for those of us fighting to stem the tide of prescription drug misuse and abuse.  The story comes to us from Travis Rieder, a research scholar at the Johns Hopkins Berman Institute of Bioethics.  His journey, despite his role at Hopkins, doesn't have anything to do with his role in public health.  Travis likes to ride motorcycles... and his story begins with a horrific motorcycle accident.

I won't retell the story (you really should read it for yourself), but to summarize: Travis ended up deeply dependent on opioid painkillers.  Knowing he needed to stop taking them, he initiated his own weaning protocol (that was, in retrospect, far too aggressive - even thought it was suggested by one of his doctors).  He lived in agony for days, then weeks.  But he stuck to his plan.  At one point, it got so bad, he contemplated suicide.

Where were his doctors, you ask?  He found the medical profession to be some combination of afraid, inept, reluctant... perhaps all of the above... to assist in the weaning of his opioids.  And this was a motivated patient, asking to be weaned.  A highly educated, white collar academic who was begging for help... and got none.  "How could it be that my doctor's best tapering advice led to that experience?" Travis asks, "And how could it be that not one of my more than ten doctors could help?"  And think: this story found its way to Health Affairs because Travis is a known author in the field of bioethics.  How many non-bioethicists out there are suffering in this same opioid purgatory?

As my colleague Mark Pew has written about extensively, we've arrived at the hard work cleaning up the mess.  He even created a hashtag for it (#cleanupthemess), not because we're trying to score marketing points, but because we needed an organizing principle for the combined and coordinated effort its going to take to accomplish our collective goal.

Travis's story highlights the fact that the clean up may be harder than we imagine.

On Twitter @PRIUM1