Along with the increasing prescribing of opioid analgesics for chronic pain have come growing concerns about patients misusing or developing addiction to those medications. Some believe that the phenomenon of craving in patients with chronic pain treated long-term with opioids may not only predict medication misuse but may also signal addiction. Several investigations have explored opioid craving and its effects in such patients, but the research has been of low quality and critical questions remain unanswered.
In the most recently reported study, researchers affiliated with Harvard Medical School, Boston, MA, conducted an investigation to characterize self-reported craving in patients prescribed ongoing opioids for chronic noncancer pain and how this might be affected by an opioid-compliance intervention [Wasan, Ross, et al. 2012]. They hypothesized that craving for prescription opioids would be significantly correlated with the urge for more medication, preoccupation with the next dose, and current mood symptoms.
Writing in the Journal of Pain, Wasan, Ross, and colleagues report on craving during a 6-month period among 62 patients treated with opioid analgesics. Data for this prospective cohort study had been collected while subjects were enrolled in a randomized controlled trial to improve prescription opioid medication compliance [also from a team at Harvard and reported separately in Jamison et al. 2010]. Participants had been divided into 3 groups:
- High-Risk Experimental (HRE) group — these patients (N= 21) were determined to be at high risk for medication misuse based on their responses on a battery of assessment questionnaires. They participated in a structured behavioral training program for preventing substance misuse.
- High-Risk Control (HRC) — these subjects (N=21) similarly were determined to be at high risk but received “usual treatment” rather than participating in the behavioral therapy program.
- Low-Risk Control (LRC) — participants in this group (N=20) had no identified drug-use problem potential and received “usual treatment.”
All patients were monitored via periodic urine toxicology screening. Additionally, using electronic diaries, patients completed ratings of craving at monthly clinic visits and daily during a 14-day take-home period. In all cases, “craving” for prescription opioids during the prior 24 hours was assessed via 4 questions rated on a 0-to-100 visual analog scale (VAS):
- How strong was your urge to take more opioid medication than prescribed?
- How much did your mood or anxiety level affect any urge to take more opioid medication?
- How often have you found yourself thinking about the next opioid dose?
- How much have you craved the medication?
Results indicated that most patients (61.3%) denied any craving at baseline (0 on the 1-100 VAS), and there were no differences between those who did not and those who did report experiencing craving on demographic variables (eg, age, gender, pain relief from medications, anxiety/depression, and other factors). In subjects reporting craving initially, both high risk (HRE and HRC) and low risk (LRC) subjects consistently experienced craving during the course of the study. Craving was significantly (P < 0.001) correlated with the urge to take more medication, fluctuations in mood, and preoccupation with the next dose. However, it was not significantly associated with the rate of drug misuse during the study.
It is important to note that craving was only weakly associated with current levels of pain or average pain over 24 hours, so craving seemed independent from a perceived need for greater pain relief. The intervention to improve opioid compliance in the high-risk group was significantly associated with decreased craving over time as compared with the high-risk control group (P < 0.05).
The researchers conclude their findings indicate that craving is a potentially important psychological construct in patients with chronic noncancer pain prescribed opioids, regardless of their level of risk for misusing opioids. Targeting craving may be an important intervention to decrease misuse and improve prescription opioid compliance.
COMMENTARY: Craving appears to affect a significant minority of patients taking long-term opioid analgesics, and it might have important influences on therapeutic non-compliance. However, opioid craving in this context also seems to be a poorly defined concept that research to date has not elucidated.
An earlier and much larger observational study by Wasan and colleagues from Harvard [Wasan, Butler, et al. 2009] found that 55% of 613 patients taking long-term opioids for chronic noncancer pain never felt any craving for their medication; whereas, the remaining 45% reported some degree of craving on a scale from 0=never to 4=very often. This latter group — who were significantly more often male and unmarried — also exhibited a greater incidence of aberrant drug-use behavior and more frequent abnormal urine toxicology screens.
However, this and the more recent studies — the latest trial by Wasan, Ross, et al.  and the original study by Jamison et al.  from which data were drawn — have serious limitations making them of questionable validity:
- First, these studies were small and statistically underpowered. Although Jamison’s team reported doing a power analysis to justify that only 21 subjects per group was adequate, our own assessment suggests that such small groups resulted in a mismatch of groups in terms of craving at baseline and a high degree of variance in the point estimates of outcomes; therefore, even if there were very large effect sizes they still could not be trusted as being accurate relative to true estimates in a larger population of patients.
As it is, ratings of craving in the Wasan, Ross, et al.  study were modest (mean ratings ranging from 8.7 to 26.2 on the 0-to-100 VAS for the 4-items), and effect sizes between groups were at most moderate (d-scores ranging from 0.2 to 0.53 maximum, not calculated by the authors). In short, the outcomes are interesting but tentative, at best, and represent a very low level of evidence. [These statistical concepts have been explained in our Series, “Making Sense of Pain Research” — here.]
- Second, and most importantly, both studies by Wasan and colleagues do little to further our understanding of craving and its impact among opioid-treated patients with chronic pain. Most recently, Wasan, Ross, et al.  concede that “we lack data as to whether the concept of craving is clearly understood by the persons taking opioids for pain.” It is possible that the questions in their study purportedly assessing craving were measuring an entirely different construct as interpreted by the subjects.
Similarly, in the earlier study, Wasan, Butler, et al.  acknowledged the possibility that “craving could have been interpreted in a number of different ways — as a desire or urge for pain relief as well as a dependency on the drug due to its opioid effect. Unfortunately we were unable to differentiate between those who interpret craving to be directly related to their pain or due to a reliance on the euphoric effect of the drug. …there [also] is the possibility that medications other than opioids may have been brought to mind when considering craving.”
The presence of craving among both high-risk and low-risk subjects in the most recent study [Wasan, Ross, et al. 2012] suggests that craving may be experienced by a proportion of all patients taking long-term opioids independently of any inclination toward substance misuse. As the authors admit, “The relationship between craving and risk for opioid misuse among those with chronic pain is unclear. Further studies are needed to determine whether a certain threshold of craving is useful as an indicator for development of prescription opioid dependence and/or addiction.”
However, it seems necessary to first clearly define what craving is, clinically, and how to accurately measure it. This is vital because craving is considered as an important diagnostic criterion of opioid addiction, according to current standards, yet so many patients without a substance-use disorder also report some amount of craving. If clearly defined, opioid craving might be an early warning signal of an impending problem, which then might be alleviated by appropriate counseling or other interventions. At present, however, we know little about the qualities of opioid craving, whether it changes over time, and how it affects patient behavior — and the research to date in persons with chronic pain has been largely unhelpful.
> Jamison RN, Ross EL, Michna E, et al. Substance misuse treatment for high-risk chronic pain patients on opioid therapy: A randomized trial. Pain. 2010;150:390–400 [abstract here].
> Wasan AD, Butler SF, Budman SH, et al. Does report of craving opioid medication predict aberrant drug behavior among chronic pain patients? Clin J Pain. 2009(Mar-Apr);25(3):193-198 [article here].
> Wasan AD, Ross EL, Michna E, et al. Craving of Prescription Opioids in Patients With Chronic Pain: A Longitudinal Outcomes Trial. J Pain. 2012(Feb);13(2):146-154 [abstract here].
Don’t Miss Out. Stay Up-to-Date on UPDATES!
Register [here] to receive a once-weekly eNotification of new postings.