When harms outweigh benefits of long-term opioid therapy for pain: Need for a new diagnostic entity, research and improved treatments

Given myriad harms and modest or absent benefit of long-term opioid therapy (LTOT) for chronic pain [1], guidelines recommend prescribers frequently reassess LTOT and, in a patient-centered fashion using shared-decision making, taper when harms outweigh benefit [2, 3]. When tapering is indicated, a subset of patients are unsuccessful; some experts suggest that struggling to taper represents impaired control over opioid use, a hallmark of opioid use disorder (OUD), described by the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5) as a problematic pattern of opioid use leading to problems or distress. Specifically at issue is the DSM-5 criterion of ‘persistent desire or unsuccessful efforts’ to reduce use. We posit that interpretation of this criterion is unclear and leads to confusion about when to diagnose OUD in the context of LTOT, causing negative consequences for patients and stymying treatment and research. We believe the best path forward is to revise DSM-5 to define a condition specific to LTOT with difficulty tapering (without otherwise meeting criteria for OUD).

Regarding unsuccessful efforts to cut down and its corollary ‘use for longer than intended’ in the context of LTOT, several points must be considered. First, ‘efforts’ may be driven by the prescriber, conducted at their pace, with prescribers having control rather than patients. Patients may lack desire to cut down, a wholly different scenario than that intended to be captured by the DSM-5, wherein individuals want to curtail substance use, but are unable to. Second, although tolerance and withdrawal were excluded from the DSM-5 definition in the context of LTOT because they are ‘normal, expected’ symptoms, this is also true of difficulty tapering. Prolonged exposure to opioids downregulates μ-opioid receptors, upregulates sympathomimetic hormones and alters in the hypogonadal-pituitary-adrenal axis. Disturbing homeostasis through tapering elicits predictable responses including fear, anxiety and other withdrawal symptoms. Therefore, as DSM-5 asserts tolerance and withdrawal symptoms during LTOT should not be counted when diagnosing OUD, we propose that neither should difficulty tapering.

A new diagnosis is needed that better aligns with patients' experience on LTOT; by defining and naming it, the field could more effectively conduct research to measure, prognosticate and treat it. To begin the process of consensus-generation on the diagnostic criteria for a new diagnosis, we engaged 38 multi-disciplinary experts in a Delphi study [4, 5]. Although this project will soon expand—because of a recent United States (US) National Institutes of Health award—to include persons with lived experience and a broader array of researchers, policymakers and clinicians, preliminary criteria include (1) benefits not outweighing harms of LTOT; (2) difficulty tapering; and (3) absence of loss of control, marked by overusing medications and running out of prescriptions early. Presence of loss of control would fulfill an OUD criterion.

Another advantage of a new diagnosis for this scenario is to avoid over-diagnosis of OUD. In a chart review study, Howell et al. [6] found that 29% of OUD diagnoses were likely invalid with a major source of misclassification being clinicians' use of OUD among patients on LTOT without documented issues of taking medications other than how prescribed. Consistent with Centers for Disease Control and Prevention guidelines, assessing LTOT's impact should revolve around comparing improved pain-related function (i.e. benefits) versus harms and should be collaborative and transparent. Should a taper falter, pushing an OUD diagnosis—by asserting the presence of loss of control when the patient has been largely adherent to the treatment plan—can be alienating for the patient and may derail the opportunity to make needed changes. Of course, if clear evidence of loss of control of opioid use emerges (e.g. self-escalating dose) an OUD diagnosis should be strongly considered. As such, we do not advocate for a new diagnosis to spare patients on LTOT the stigma of being correctly diagnosed with OUD; we advocate for a new diagnosis to avoid the stigmatizing and confusing experience of being incorrectly diagnosed with OUD.

Finally, we need a new diagnosis for this scenario to design studies that will generate evidence to improve quality of care. Guidelines recommend optimizing non-pharmacologic and non-opioid therapies and working with patients to taper LTOT, but these broad concepts need greater detail to help practicing clinicians. One approach, often helpful to improve pain and function while enhancing opioid safety is switching from LTOT to sublingual buprenorphine [7]. However, as sublingual buprenorphine is only approved by the US Food and Drug Administration (FDA) for OUD, prescribers seeking to use it to treat this scenario often face regulatory or insurance coverage barriers. Patients may reject it, not recognizing it as a treatment for their condition or may not be able to afford it.

One potential solution is for the FDA to expand the indication for sublingual buprenorphine to this scenario, a move we strongly support. It is encouraging that the US Department of Health and Human Services, which oversees the FDA, included switch to buprenorphine as a recommended approach among patients on LTOT for whom tapering is indicated, but not tolerated [8]. However, the FDA cannot approve a medication for an undefined indication; another reason to name this diagnosis and develop criteria for it. Beyond buprenorphine, other treatments, including non-pharmacological interventions such as mindfulness-based stress reduction [9] or behavioral therapies [10], also need to be examined.

In our experience, for patients on LTOT for whom harms outweigh benefit, but tapering is stalling, misapplying a diagnosis of OUD rarely leads to improvements in patient function or the patient-provider relationship. This condition needs a different treatment approach. Often, such patients will not receive the care they need within OUD treatment programs where clinicians may not be trained to address chronic pain. We believe a new diagnostic entity would align well with guidelines encouraging continuous reassessment of benefits and harms of LTOT and help generate evidence for how to proceed when that balance becomes unfavorable.

William C. Becker: Conceptualization (equal); writing—original draft (supporting). Joseph W. Frank: Conceptualization (equal); writing—review and editing (supporting). Sara N. Edmond: Conceptualization (equal); writing—review and editing (supporting). Joanna L. Starrels: Conceptualization (equal); writing—review and editing (supporting).

None.