A critical look at the opioid crisis and efforts to address it

The “opioid crisis” continues to grab headlines both nationally and internationally. Some of those headlines are quite fantastic, such as the January 2019 report from the U.S. National Safety Council that boldly claims “For the first time on record, your odds of dying from an accidental opioid overdose are greater than dying in a motor-vehicle crash”. It is a shame that such institutions feel the need to create such attention-grabbing headlines. While the opioid crisis is indeed a problem, and even one overdose death is too many, fear-mongering is simply not helpful. In this case, by ignoring some fairly basic rules of finite math like cumulative risk and exposure, that NSC report is such an irresponsible attempt to make a point that the NCS ought to be ashamed.

Moving from hyperbole and nonsense to more quantitative metrics, Google’s trends exploration platform (trends.google.com) reveals that searches using the term ‘opioid’ have continued largely unabated over the last 2 years. Opioid peaked as amongst the most searched terms on October 22, 2017 (the week President Trump was preparing to make the opioid crisis a national public health emergency), and since has remained largely stable around the 25th-40th percentile of ALL Google searches.

Google search term trends using the term ‘opioid’ from June 19, 2016 to June 19, 2019. The ranking is normalized to a score out of 100, with 100 being the most commonly searched for that day.

Google search term trends using the term ‘opioid’ from June 19, 2016 to June 19, 2019. The ranking is normalized to a score out of 100, with 100 being the most commonly searched for that day.

For another metric, we’ve been archiving every tweet that included any of #opioid #opioids #opioidcrisis or #opioidepidemic since November 20, 2017 (if you’re wondering, that’s 68,886 tweets through June 12, 2019). That database also seems to indicate that opioids remain a common topic on Twitter, averaging 447 tweet per day over that period. So clearly this continues to be a topic of considerable interest.

There seem to be no shortage of topics of considerable interest occupying the collective global consciousness. With so many big deals to address, perhaps it’s no surprise that the response to the opioid crisis has been swift and somewhat heavy-handed. New opioid prescribing guidelines for people in pain have now been produced on both sides of the Canada/U.S. border. Most include recommendations to attempt alternative strategies for pain management before trialing opioids, and tapering of people already on high-dose opioids to below a threshold of 85-90 mg morphine equivalents per day (MED). The former seems like a reasonable approach, though there are caveats to be discussed below. The latter however feels like far more nuance should have been built in, as they are essentially saying “we can no longer trust our doctors or patients to make good care decisions about these powerful medications, so us policy makers will now guide their practice”. This feels dangerous to me, and I propose the entire focus so far has been misdirected.

Regarding the recommendation for alternative pain management strategies, on the surface this seems like a logical response to a perception (perhaps warranted) that too often opioids are the first line of therapy offered to people with pain, or even prescribed for people who may feel pain in the future (e.g. post-surgery). By now most will be well-aware of the stats showing that on a per capita basis, the U.S. and Canada are the highest prescribers of opioids in the world and there is no compelling evidence that people with pain in other countries are suffering more due to their lack of opioid prescriptions. So perhaps other countries have figured out something about pain management that we have not. But I don’t think so. Another explanation could be that of cost and access – most of the alternatives that have been endorsed in new prescribing guidelines are not adequately covered by health benefit plans in North America, such as physiotherapy, massage, yoga, pilates, mindfulness meditation, acupuncture, cognitive behavioral therapy, or others. And that assumes that such options are locally available at all, which by most accounts is a fairly large assumption in many communities. Such alternatives present other challenges such as requiring patients to take time out of the day, to travel to regular appointments, and considerable variation in approach even between providers of the same discipline. Further, and I say this as someone trained as a physiotherapist, the scientific evidence for any of these as being effective pain management strategies is generally not strong. If you really look at the data, when authors report that there is no difference between pain relief achieved through opioids and that achieved through these alternatives, what we’re being told is that they’re equally ineffective. Or perhaps more appropriately and accurately, when we cast a critical lens on the issue we could say that the models, frameworks, and methods used to a) deliver and b) empirically evaluate pain management strategies have taken an overly narrow view of effectiveness for individual patients.

But we’ll leave that topic for another time.

If effectiveness is indeed equal, pills in most North American jurisdictions have been (and arguably continue to be) easier to access and less inconvenient than the other options, so perhaps it’s not surprising that medications continue to be the treatment of choice for many.

The line-in-the-sand-style upper limits for opioid dosing (85-90mg MED) in most guidelines is a second major issue. What it has led to is an equally (or more) problematic behavior from prescribers, often amounting to what has been described as non-consensual tapering. In the interest of harm reduction through reducing exposure, these types of practices appear to be increasing harm in other ways such as increased suffering and forcing otherwise good patients to risky or dangerous behaviours.

Would not a better option be to provide the estimates of risk of new opioid use disorder (which is actually quite low when the drugs are used properly) and the likelihood of meaningful improvement in pain (which admittedly may also be low for some people) and allow the patients and their families to make an informed decision? I can think of many patients who would likely accept a 1 in 400 chance of overdose death (the actual odds as I calculated them from CDC data) for a 1 in 20 chance of pain improvement. That’s difficult to write, but also reflective of the level of suffering of many people in pain.

As we cast a critical lens towards decisions and knowledge we start to ask ourselves challenging questions, like what are the underlying taken-for-granted assumptions that make us believe this is the ‘right’ way of doing things? Who do these actions privilege, and who might they harm?

Let me share excerpts from some of the emails I’ve received over the past couple of years after doing a media post or interview on this topic. Here are examples of what people are emailing me:

“…you mentioned ‘most people who I see nowadays with chronic pain will tell me that they do have a suicide plan somewhere’ – this is something I have also thought of, I am not proud of this Dave, but there was a time when I was in bed most of the day, curled up, sometimes with uncontrollable crying, thinking I was all alone, thinking why me, thinking no doctor would treat their family dog like the way my neurologist Dr. [name redacted for privacy] treated me – yes Dave, life was that bad, and I never want to be put in that situation ever again.”

“I am finding my many strategies to cope no longer match my pain levels. The tunnel vision and/or deafness of doctors and slowness of certain medical bureaucratic processes have made me consider joining the group you identify “who turn to the streets” to obtain possible relief.”

“I don’t consider suicide an option I WANT to act on, but when you face the complete hopelessness of getting appropriate medical care for as long as I have, along with the kind of pain that induces deep despair, it becomes a pragmatic alternative.”

“I do have a suicide plan, and it sits on a high shelf on one of my bookcases. I have discussed it with my husband and he is prepared to help me should it ever come to that.”

“Who is looking out for patients like my husband who wanted to die to end the pain in 1999, and then with proper medication gets his quality of life back to 80% and is living and enjoying life again with limitations from 2001 until 2012, only to have it taken away again a little bit at a time as if to torture someone and keep taking it away until their quality of life is back to wanting to die to end the pain.  The most frustrating part for me as a caregiver is that I'm losing the man I have loved and been married to for almost 40 years, and it is so unnecessary.”

For anyone who has ever sat in front of a person in pain, looked them in the eye and saw true suffering, I suspect that like me, you cannot read comments like these and believe that we are somehow on the right track with the strategies taken so far to address the opioid crisis. And while it is heartening in some ways to see that the medical, advocacy, and scientific communities are now starting to make their own noise about the injustices being done to people in pain (see for examples here, here and here), it seems as though much of the damage has now been done. Big pharmaceutical companies aside, the cause of the opioid crisis has too often been pinned on those who prescribe and consume opioids for pain. As such, the dual crises of chronic pain and opioid use disorder have become so enmeshed that it will take years to tease them apart again. Frighteningly, those like the people who send me those emails, may not have years.

What would it take for us to change the conversation from ‘people taking too many opioids have caused the opioid crisis’ to ‘people at risk of opioid use disorder have been given access to too many opioids’? What would it take to change that conversation even further, to ‘the reason people are at risk of opioid use disorder has little if anything to do with physical pain’?

I believe there are a number of concerning issues with the way opioid prescribing guidelines have been interpreted and implemented. For example, the current Canadian prescribing guidelines released in 2017 does allow for dosage higher than the stated limit of 90 mg MED: “Some patients may gain important benefit at a dose of more than 90mg morphine equivalents daily” (p. 60) though indicates a second opinion should be sought before doing so. In practice, from what I’m consistently hearing, physicians have been strongly encouraged (sometimes even threatened with regulatory discipline) to prescribe <90 MED, ideally <50 MED despite what the guideline describes as a ‘weak recommendation’ for the latter. It’s not worth going into the state of evidence for the purposes of this story, other than to say not all research or clinical evidence is created equal, though it is having very real effects on humans in pain.

Currently policies are missing the target in the opioid crisis

As an alternative I wish to forward a hypothesis that the focus to date on restricting opioid prescriptions has been more harmful than helpful in battling the opioid crisis. In order to understand my position, I need to explain the following perspective: the opioid crisis is not a problem of pain management. I suggest that we are dealing with two concurrent public health crises in North America, themselves embedded within far broader social contexts and drivers that will not be amenable to simply legislating away opioid prescriptions, more oversight of prescribing doctors, or even blocking the import of illegal China-made fentanyl into North America. I struggle to think of any time in history in which prohibition actually made a problem go away – with the possible exception of General Mao’s prohibition of opium in China in the mid-20th century. But since the strategy involved forced ‘treatment’ camps and executions, perhaps not the best framework for us to follow.

I do believe we are facing two public health crises in this conversation. The first is one of undermanaged chronic pain. If available survey data are to be believed, roughly 1 in 5 Canadian adults (~18%) live with daily chronic pain intense enough to interfere with function. Globally, the Global Burden of Disease studies supported by the World Health Organization reveals that musculoskeletal pain problems (e.g. neck, knee, and low back pain) have consistently been amongst the largest drivers of years lived with disability, and these have not changed appreciably since data have been collected starting in 1990. If you want to have fun with those data (because who doesn’t), you can check out the Global Burden of Disease interactive data visualizer at https://vizhub.healthdata.org/gbd-compare/. The take-home message here is that we’ve yet to really get much of a handle on chronic pain, using opioids or otherwise, and one would expect the burden of pain to continue to increase into the foreseeable future as the population ages.

The second public health crisis in my model is that of mental health disorders. Similar to the issue with chronic pain, according to the Canadian Mental Health Association, 1 in 5 Canadian adults also suffer from a mental health problem or illness. And this is again not simply a problem in North America, if you go back to that GBD data visualizer, you’ll see that amongst adults aged 15-49, musculoskeletal problems are second only to mental diseases in terms of years lived with disability globally. In fact if you look at all age ranges available in those data, you’ll see that mental disorders and musculoskeletal disorders have shared the top 2 spots for global burden since 1990. While the dedication of considerable resources to diseases like HIV/AIDS has seen a relative reduction in global burden in those communicable diseases, somehow we’ve yet to really crack mental health or chronic pain despite decades of research and treatment. My position is that what we’re currently seeing as an opioid crisis is in fact a collision of these two persistent, vexing, and highly prevalent public health crises.

One interesting question is this: is pain a musculoskeletal problem, a mental health problem, or both? The currently accepted definition of pain as ‘an unpleasant sensory and emotional experience’ (italics added) would suggest it’s at least partly both. I’m stopping short of stating that chronic pain is always an MSK and mental disorder, because that’s neither supported nor fair, though I will side with my colleagues in the field in endorsing a conceptualization of pain that cannot exist without the cognitive processes of perception, synthesis, and interpretation. So pain is, indeed, always ‘in your head’ in the same way that smell is in your head – they’re both an output of your brain’s processing of several bits of information. However to say smell, touch, or pain are purely manifestations of your psyche would be inappropriate.

Interestingly, addiction, which I’m more appropriately calling ‘opioid use disorder’ (OUD), is itself a mental health disorder. That is, there are diagnostic criteria for OUD in the Diagnostics and Statistics Manual – V, and dedicated diagnostic codes in the International Classification of Disease – 11th version (ICD-11), which for those interested is in the 6C43 category. “True” OUD (the term ‘true’ used with some hesitance), the result of which accounts for far too many overdose deaths, is very different from recreational use of opioids to get a ‘high’. OUD is characterized largely as an impulse control problem, demonstrated by continued use of a substance often despite knowing that it’s doing harm. Through this conceptualization you could easily (and I argue in most cases accurately) state that people with OUD do not want to get high any more than someone with depression wants to feel despair or someone with PTSD wants to live in fear. Rather, I (and others) would argue, OUD is a manifestation of a sense of despair, hopelessness, and isolation, perhaps starting as a means to cope with an otherwise intolerable mental state of unresolved trauma. In the absence of support and access to care, my empathic brain can understand how many are left with nowhere left to turn but to find ways to numb the pain.

And this is where things get interesting and complex, because there are most certainly touch points between chronic pain and mental disease. When opioids tend to be featured prominently in both, it’s easy to understand how the lay public and policy makers can conflate one with the other. Even many of the risk factors, to the extent we know them, are consistent between the two chronic conditions. These include lower socioeconomic status, lower educational attainment, poverty, early life adversity, trauma in adulthood, and possibly even some shared genetics (e.g. the mu opioid receptor OPRM1 at position rs1799971 has been associated with both the risk of opioid dependence and the experience of persistent pain). Then there’s a mutual risk to make the picture even more complex: the risk of chronic pain appears to increase in the presence of pre-existing mental disease, while the risk of mental disease appears to increase in the presence of chronic pain. It is estimated that 40% of people with chronic pain also meet the diagnostic criteria for a mental disease. But they are not the same thing.

So the story is indeed muddy. Yet muddiness should not be an acceptable excuse for harmful action.

Have North American doctors been too liberal when prescribing opioids in the past? Probably, though I’d argue it’s been largely driven from a benevolent desire to help when few alternatives exist, rather than malevolence, laziness, or some desire for illicit gain. Have pharmaceutical companies like Purdue Pharma been over-aggressive or even misleading in their marketing of opioids? Again, probably, though I’m no lawyer so I’ll let the courts make those calls (it seems they’ll have plenty of opportunities to do so). What I’m most interested in are the humans on the end of this – those in daily pain who have suddenly found themselves without a partner in their care, and those with unresolved emotional trauma who find few allies amongst the ‘healthy elite’. Those are the people who have been harmed in this crisis, yet too often their voices are silent as politicians and policy makers enact guidelines that are easy to flaunt yet do little to address the true problems. To wit, 2018 saw a clear reduction in overall volume of opioid prescriptions, yet record highs in overdose deaths. This should be a clear signal that prescribing policies have been aiming at the wrong targets.

As a thought experiment, consider this question: if opioid prescriptions dropped to zero, would OUD and overdose death cease to exist? OUD existed long before opioids were being prescribed for pain, and will continue to exist long after they’re gone.

Instead, I propose that the imposed caps have had the twofold undesirable effects of limiting access to potentially effective pain management for people who need it, while driving those with mental health disorders to far riskier alternatives.

It seems no one is winning.

For the more visual thinkers, I present a figure that may help explain my position. As I’ve stated, there is no doubt overlap between chronic pain and mental health/illness, that I’ve shown in a Venn diagram. I’ve then illustrated Substance Use Disorder (SUD) as a subtype of mental illness, with of course some overlap with chronic pain (people with chronic pain are not immune to mental illness, so of course there is overlap). Down from there, Opioid Use Disorder is characterized as a subtype of SUD. Again, overlap is expected here, and the fact that opioids span both conditions means the overlap with OUD should be significant.

Around the outside I’ve chosen to wrap the sociocultural context (I’m intentionally ignoring the biological diatheses in this image). I’ve listed just some of the very complex social and contextual/cultural influences that may increase the risk of chronic pain and mental illness, at least as far as I understand them. Poverty, poor housing, trauma, criminalization, lack of connection, missing sense of purpose or belonging are among them.

Opioid vs mental health image.jpg

And if we were to try to pinpoint where most of the resources and efforts have been dedicated in response to the opioid crisis, I suggest it would look like this:


This is not to understate the work of dedicated and passionate people who have been involved in several of those other social issues for several decades. Yet any time I hear or read about the opioid crisis, I can be almost guaranteed to find the words ‘highly addictive’ somewhere in that story, but it is rare that those same stories include any mention of the broader social or mental health issues of which, I somewhat pessimistically believe, the opioid crisis is simply a signal. Instead most media stories and political actions focus on opioids as strategies for pain management. Combined, these have the effect of marginalizing both those people with OUD as ‘addicts’ and people with chronic pain as ‘drug seekers’, both terms that are in bad need of replacement.

I encourage you to ask for example, who has in fact been privileged by firm prescribing caps? Engaging my inner cynic, healthcare funders seem like an obvious group who stand to benefit especially in regions where drugs are publicly funded. Who has been oppressed or marginalized? As I’ve argued here, those are clearly people in pain and people with unresolved emotional disorders. How could these strategies have been designed, implemented, and evaluated otherwise given the resources that have already been dedicated to a relatively (in my mind) small component of a much larger problem? While not a panacea, here’s a thought: how about better access to mental health services and improved access to alternative pain management strategies?

As a measurement theorist I am well aware that what gets measured is what receives the most resources. What would happen if ‘success’ in this war was not measured by number of prescriptions but by something like access to appropriate care, empathy, or some metric of human connectedness?

Two Huge Problems that will require patience, empathy, and resources to address

While the two crises should not be confused to be the same thing, similarly should we hold no assumptions that a means to address one will necessarily improve the other. An analogy would be the intentional introduction of the cane toad to Australian fields in 1935, with the idea that they’d be a ‘natural’ means of controlling the grey-backed cane beetles decimating sugar cane crops. As any Aussie will tell you, this did not end well, as the venomous cane toads have no natural predators in Australia which meant it could procreate and spread quickly. Worse still, those native predators like birds and snakes that did decide to have a go at a cane toad succumbed to the venom. As a result, not only did a non-native species invade an otherwise isolated and protected country, but it also led to the destruction of native species. By not adequately considering potential unintended consequences, a decision made to quickly address one problem had devastating downstream effects. The parallel to our current situation seems obvious.

For those who are in positions of power as far as decision-making, I encourage thoughtful pauses and reflection when making decisions that have real-world impact, even in the face of apparent crises. I like Tsoukas’ Conjunctive Theorizing approach for these purposes, encouraging decision makers to complexify rather than simplify when attempting to introduce new policies or actions (poetic praxeology is such a lovely concept). What is needed for both the pain and opioid problems must be big picture thinking, zooming way out to ask big questions like why is mental illness and opioid use disorder rising, and why is chronic pain increasing in prevalence and burden? Means to address these issues will not make good political sound bites, and will not allow a politician to wave a figure showing reduction in number of opioid prescriptions and claim victory, at least not within a 4-year electoral cycle. These are legacy decisions that must cross party lines.

It is far past time to unlink the crisis of opioid use disorder from chronic pain. Change the metrics for success, look beyond short term solutions and quick wins, and be willing to explore the difficult questions facing our societies like isolation, equity, despair, and healthcare access. Give doctors the power and respect to practice in the best interests of their patients. Continue to explore and provide access to different pain management strategies but stop saying they’ll be the solution to the opioid crisis.

Only through complexifying and casting a critical lens on these problems from a position of empathy and genuine compassion do we have even a chance of finding anything that even looks like an effective solution to these huge social and public health problems.