Continued Use Despite Harm: The Under-Utilized Criterion for an Opioid Use Disorder Diagnosis

Over the last few years, I have argued that we need to rethink the nature of opioid use disorder in the population of people who take opioids as prescribed for moderate-to-severe persistent pain. I’ve done so in various formats, including in presentations as well as here at the Institute for Chronic Pain, in both web pages (Should the Definition of Opioid Addiction Change? and Opioid Dependency & the Intolerability of Pain) and blog posts (The Central Dilemma in the Opioid Management Debate and Dreaded or Embraced? Opioid Tapering in Chronic Pain Management). 

I do so because I think that the fields of both pain management and addiction are overly focused on loss of control as the primary indicator of when a person on long-term opioids for pain management crosses the line into the problematic state of an opioid use disorder (OUD). The argument these fields tend to use goes something like the following: 

  • Physiological dependence on opioids, in the population of people who use prescription opioids on a daily basis for the management of pain, is largely considered as a condition that is both expected and benign when compared with the condition of an OUD. Physiological dependence, in other words, doesn’t necessarily rise to the level of alarm unless someone on such medications begins to evidence a loss of control over their use. 
  • Loss of control is largely considered a compulsive condition, which is neither expected nor benign, since it leads to distress and/or functional impairment, or even death by means of accidental overdose. 
    • Examples of loss of control tend to be readily observable: repetitive self-escalation of the use of opioids, leading to early refill requests; obtaining opioids from multiple providers at the same time or from friends or relatives or other illegal sources; use of opioids that aren’t prescribed; etc. 
  • At the current time, we tend to reserve the term “misuse” to refer to those instances of use which fail to be in accordance with how the medication is prescribed1; when done repetitively, misuse can cross the line into compulsive use, which is indicative of loss of control, and thus come to exhibit some degree of an OUD. 
  • Repetitive misuse and loss of control tend to go hand in hand and are seen as in contrast with those who may be physiologically dependent, but take opioids as prescribed. 

This line of reasoning has taken hold in the fields of pain management and addiction. In common practice, what it means is that the typical Image by Ahmed Zayan courtesy of Unsplash litmus test for whether someone on prescribed opioids for pain is addicted or not is whether they are taking their medications as prescribed or not. If they are taking their medications as prescribed, they may be physiologically dependent, but not addicted. If not, they are identified as both physiological dependent and addicted, since they have crossed the line into compulsive use behaviors and as such have lost control of their use. 

Misuse and loss of control are admittedly the clearest indicators of problematic use of prescription opioids. Typical examples, as already noted, tend to be readily observable. There’s relatively little doubt, for instance, when an opioid that isn’t prescribed shows up on a urine drug screen. Nonetheless, the fields of pain management and addiction have historically indicated an additional criterion for OUD, which is continued use despite harm.2, 3 This criterion tends to get short shrift in clinical practice. Perhaps, it is because it is not as clearly recognizable as what has come to be recognized as loss of control – the misuse of opioids or taking them in an unprescribed manner. Perhaps too, it is because it can occur in the absence of such misuse. In other words, continued use despite harm can occur when patients take opioids exactly as prescribed, and as such they aren’t as readily identified as addicted. 

Moreover, another reason that those in clinical practice tend to fail to recognize continued use despite harm is that the matter can be confused with patient-reported high pain levels and what constitutes a reasonable response to such pain levels. To put succinctly, it can be difficult to differentiate between continued use despite harm and appropriate medical decision-making in response to moderate-to-severe pain.

Let’s explain each of these two difficult-to-recognize aspects of opioid use disorder one at a time.

Addiction to opioids when taking opioids as prescribed

As the phrase continued use despite harm suggests, the criterion of OUD under consideration occurs when a patient on prescription opioids for pain management insists on their continued use even if it places the patient in danger, such as risk of accidental overdose or exacerbating a co-occurring life-threatening condition. 

The criterion is most clear when it occurs in the context of illegal use of opioids, such as those that are bought off the streets or other non-medical sources. The problematic nature of insisting on the use of opioids when having no knowledge or assurance of their true nature is clear. In some ways, it is a variant of impaired control: you use what you are told are hydrocodone pills, for instance, but you actually have no knowledge of whether they are truly hydrocodone; they may be hydrocodone, but they might also be illegally manufactured fentanyl; as such, the use of these pills place you at considerable risk of harm (e.g., accidental overdose, since fentanyl is exponentially more potent); a more reasonable decision in response to this lack of knowledge would be to forgo their use; one who continues to insist on their use is acting with impaired control. 

But what of the use of this criterion for OUD in a population of people who are taking legally obtained, prescribed opioids for the management of pain?

Say, for example, a person who takes high-dose daily opioids for moderate-to-severe persistent pain also has severe sleep apnea. He reports that he is intolerant to the use of a C-PAP and so his continued use of high-dose opioids places him at significant risk of accidental overdose. His healthcare providers have cautioned against continued use of opioids, especially at his current high dose levels, and have gone so far as to encourage him to reduce his current opioid dose, but he remains adamantly against it and refuses. 

Now, to be clear in our example, the patient is using opioids as prescribed and he takes only those opioids prescribed to him. As such, he is not misusing them and so most providers and patients in this scenario don’t tend to consider use of prescription opioids of this kind as meeting criterion for having lost control. As such, most wouldn’t consider him to have an OUD.

Nonetheless, it is a problematic scenario. He is refusing, so to speak, to come off the ledge of a dangerous precipice. He might die with hisImage by Loic Leray courtesy of Unsplash current use, but nevertheless refuses to change his current use. 

Suppose further that his healthcare providers have cautioned him against continued use at the high dosing schedule and have supportively encouraged him to reduce many times. Maybe his spouse or family have joined in on the encouragement to reduce his dose. He subsequently knows that his continued use may cause him harm, if not death. Were it not for the fact that the substance in question is a prescribed opioid for pain, it would be clear to all that he suffers from impaired use. For instance, suppose in this example the substance wasn’t opioids but alcohol: he continued to consume alcohol after having been told in similar circumstances and by similar people that a pre-existing liver condition is increasingly made worse and so continued use poses considerable risk of harm, if not death. The two instances of impaired control are essentially similar. It’s just that when opioids are used in these ways under the auspices of a prescribing provider it seems to cloud the recognition of an OUD.

This example is not an uncommon scenario in the field of pain management. Day-to-day clinical experience is replete with additional examples of those who continue to insist on using opioids in high risk scenarios: 

  • patients who take exceptionally high doses because of the tolerance that has developed over years of taking opioids on a daily basis as prescribed
  • patients on moderate (or high) doses of opioids who have taken them for years as prescribed and are increasing in age, thus their current dose is increasingly dangerous with each passing year
  • patients on moderate (or high) doses who also take benzodiazepine medications or sedative hypnotic medications
  • obese patients on moderate (or high) doses, with or without sleep apnea, who also take such latter medications
  • elderly patients on opioids who continue to use opioids following a fall or following the onset of cognitive impairment as a side effect of opioids
  • patients with a history of addiction who take long-term, daily opioids as prescribed
  • patients who continue to use opioids following an accidental overdose.

The list could go on.

It’s rare for providers in actual clinical practice of pain management to recognize these behaviors as an OUD.

Through the course of my career, I have found it uncommon among my addiction medicine colleagues as well. Consults related to patients like those described above tend to come back that the patients are using their medication as prescribed and so do not have an OUD. At best, the consult comes back with a recommendation to the prescribing provider to reduce the opioids, but even this helpful recommendation obscurs the fact that it is the patient, not the provider, who is refusing to reduce their opioid dosing and thereby their risk.

Succinctly, the fields of pain management and addiction medicine need to change their perspective on the criterion that we ourselves have advocated for using when diagnosing an OUD. Continued use despite harm can occur even when patients are using opioids as prescribed.

These examples of continued use despite harm are indicative of impaired control over the use of opioids. It is because of the addictive nature of opioids that such patients insist on their continued use under high-risk circumstances.

Suppose, for example, someone with a severe depression develops serotonin syndrome due to the use of antidepressants and the healthcare providers’ recommendation is to stop the use of the medications. We’d be hard pressed to imagine a scenario in which the patient becomes so sensitive and threatened by the recommendation that he or she becomes argumentative and insistent on the continued use of the medications that cause such risk. While all things are possible, such a scenario is not common. In most scenarios of this kind, the patient is open to the recommendation to reduce the use of antidepressants and open to pursuing alternative therapies for depression. 

It is much more common, though, in the population of those with persistent pain who have been taking moderate to high daily doses of opioids despite the above risk of adverse events.

Suppose, to take another example, the long-term use of a proton pump inhibitor is now thought to be contributing to certain health risks and the recommendation is to stop the use of the medication and seek alternatives. It would be uncommon for patients in this scenario to become argumentatively insistent on its continued use despite the associated risks.

Again, it is common in the population of those taking long-term opioids for pain with concomitant risk factors.

Suppose, to take one last example, someone develops a GI bleed from the long-term use of an anti-inflammatory for moderate-to-severe arthritis pain and as a consequence the recommendation from healthcare providers is to discontinue the use of an anti-inflammatory and seek alternative therapies for the management of pain. While it may be common to have misgivings in this scenario, it would be uncommon for individuals to become so threatened by the loss of the medication that they are argumentatively insistent on its continued use despite the GI bleed. 

As has been mentioned, it is fairly common in similar high-risk scenarios when taking long-term opioids.

The difference, of course, between all these examples and that of opioids is that opioids are highly addictive. With repetitive exposure to addictive substances, brain changes occur that lead to compulsive use even in high risk scenarios. In the absence of such brain changes, people maintain the ability to control their behavior, making more or less rational decisions, in response to risk. Antidepressants, proton pump inhibitors, and anti-inflammatories simply do not foster such changes to the brain and so these capacities for rational decision-making are maintained. Opioids, however, do foster such changes to the brain, thus leading to impaired decision-making, or control, and continued use despite associated risks is the result.

Continued use despite harm & confusion with appropriate responses to high pain levels

Patients who insist on the continued use of opioids under conditions of risk to their life commonly maintain that pain relief is more important than life itself. They argue that without pain relief their life would be insufferable and so, when compared to a life of intolerable pain, the risk of catastrophic events such as death through accidental overdose or relapse of a prior addiction (for those on opioids with a prior history of addiction) is preferable.

On countless occasions in clinical encounters or public forums, such as in public policy debates, patients on long-term opioids, who have a history of taking opioids exactly as prescribed, maintain such sentiments: life with their level of pain wouldn’t be worth living were it not for opioid medication management. In other words, opioids are literally their lifeline.

Indeed, such sentiments are often perceived as immediately and obviously true: living a life of moderate-to-severe daily pain seems an intolerable prospect without opioids, and so their use along with their associated risks, seems the preferable option. Any expression of doubt by others is met with affront and accusations of stigmatizing those who have the unchosen life circumstances of living with moderate-to-severe persistent pain. The common litmus test for understanding is living under such circumstances or not: “If you had my level of pain, you’d understand!” The litmus test shuts down the possibility of managing pain well without opioids.

To make clear, the implied corollary to this assertion is that managing moderate-to-severe pain is impossible without the use of opioids. Some pain, in other words, simply requires opioids. There is no other choice between the use of opioids and intolerable pain and suffering. (“You think I like taking opioids?!? There’s nothing I’d like better than to not have to take them!”) Self-management of certain pain levels seems just not possible.

Healthcare providers who prescribe long-term opioids often make a similar calculation: reduction of pain with opioids is a greater value than any of the afore-mentioned catastrophic adverse events. Moreover, such providers typically never think twice about the calculation, perceiving it similarly as their pain patients – the risk-benefit ratio seems to immediately and obviously fall on the side of the ledger involving use of opioids. (See, How Important is Pain Reduction with Opioids?)

Indeed, the reader of this article may have been saying something similar when I used the analogy to someone who continues to use alcohol despite a life-threatening liver condition. It’s easy to argue that the two scenarios are different because the patient using opioids is using opioids for pain, while the patient using alcohol is not using alcohol for a legitimate medical condition. In this context, we might observe the almost countless frequency of people who use alcohol despite high risks because they are using alcohol to medicate latent depression or past trauma. The reader might counter that while depression and trauma are legitimate health conditions, the use of alcohol to medicate them are not legitimate medical responses to them. True, but that is exactly what is at question: Is the use of opioids despite the risk of catastrophic harm, such as death or the exacerbation of a pre-existing addiction, an appropriate response to moderate-to-severe pain?

With the empathy and compassion of those who care for people with the unchosen life circumstance of living with moderate-to-severe pain, it is time to question whether this risk-benefit calculation is warranted. We know, for instance, that most people in the general population with moderate-to-severe pain do not take opioids for pain.4, 5 This fact is the norm. The norm is not the continued use of opioids despite risk of harm. 

Indeed, it is time to even take it a step further: the unquestioning, steadfast belief that some pain is simply so great that no other choice is possible but for to take opioids at the risk to life is indicative of a problematic state of addiction, even if the use of opioids is exactly as prescribed. The perception that pain is so severe that it is prima facie intolerable without opioids in people taking daily opioids for years is a function of neuroplastic changes to the brain induced by repetitive exposure to opioids. The personal affront with which other peoples’ doubts of these unquestioning perceptions are met is the shame-based defensiveness that so often accompanies addiction. When those suffering from an addiction are initially approached about their addiction by others, the common response is denial and affront.

Again, most people with moderate-to-severe persistent pain do not take opioids for pain, let alone do so despite life-threatening risks. They do not perceive moderate-to-severe pain as insufferably intolerable. Without repetitive exposure to opioids, they haven’t undergone neuroplastic changes to their brains that influence their perceptions and abilities to make rational decisions in response to pain. They do not feel compelled to take opioids in response to moderate-to-severe pain despite life-threatening risks. In other words, they do not have an OUD. 

Continued use despite harm has long been advocated for use in the identification of those with an OUD by the fields of pain management and addiction. It has, however, been long under-utilized in the population of people who take long-term opioids as prescribed for the management of persistent pain. For the welfare of those for whom we care in these fields, it is time for this under-utilization to change.


1. Volkow, N. D., Jones, E. B., Einstein, E. B., & Wargo, E. M. (2019). Prevention and treatment of opioid misuse and addiction: A review. JAMA Psychiatry, 76(2), 208-216. doi: 10.1001/jamapsychiatry.2018.3126

2. American Academy of Pain Medicine and the American Pain Society. (1997). The use of opioids for the treatment of chronic pain: A consensus statement. Clinical Journal of Pain, 13, 6-8.

3. American Academy of Pain Medicine, American Pain Society & American Society of Addiction Medicine. (2001). Definitions related to the use of opioids for the treatment of pain: Consensus statement of the American Academy of Pain Medicine, American Pain Society & American Society of Addiction Medicine. Wisconsin Medical Journal, 100(5), 28-29.

4. Nahin, R. L., Sayer, B., Stussman, B. J., & Feinberg, T. M. (2019). Eighteen-year trends in the prevalence of, and health care use for, non cancer pain in the United States: Data from the Medical Expenditure Survey. Journal of Pain, 20(7), P796-809. doi: 10.1016/j.pain.2019.01.003

5. Toblin, R. L., Mck, K. A., Perveen, G., & Paulozzi, L. J. (2011). A population-baed survey of chronic pain and its treatment with prescription drugs. Pain, 152, 1249-1255.

Date of publication: 10-17-2022

Date of last modification: 10-17-2022

About the author: Murray J. McAllister, PsyD, is a pain psychologist and consults to clinics and health systems on improving pain care. He is the founder and editor of the Institute for Chronic Pain.

This website is certified by Health On the Net Foundation. Click to verify. This site complies with the HONcode standard for trustworthy health information:
verify here.

Search only trustworthy HONcode health websites:


© 2017 Institute for Chronic Pain. All rights reserved.

To improve your experience on our website, we use cookies to examine site traffic and enable additional capabilities such as social media interaction and marketing.