There have been ongoing concerns among healthcare providers and regulators about the addictive potential of opioid analgesics during pain treatment. Some claim patients prescribed these medications readily become addicted to them, while others believe that therapy-induced, or iatrogenic, addiction is relatively uncommon. Recently, the most complete systematic review of research evidence to date attempted to put the rumors to rest on this controversial subject.
Silvia Minozzi, MD and colleagues — from the Department of Epidemiology, Lazio Regional Health Service, Cochrane Drugs and Alcohol Group, Rome, Italy — conducted a comprehensive systematic review to assess the incidence and prevalence of dependence syndrome (ie, addiction) associated with opioid therapy for pain relief in adults with and without a previous history of substance abuse [Minozzi et al. 2012]. Their literature search included Medline, Embase, CINHAL and the Cochrane Library up to January 2011.
Systematic reviews and primary studies were included if they reported data about the incidence or prevalence of opioid addiction in patients receiving opioid-class drugs for treatment of acute or chronic pain due to any physical condition, whether cancer- and noncancer-related. Any type of clinical research design qualified, as long as sufficient data were available for assessment. Of 2,871 potentially relevant studies identified (excluding duplicate studies), data were extracted from 17 investigations that qualified for inclusion, involving a total of 88,235 patients.
Writing in an early online edition of the journal Addiction, the researchers report that qualifying studies for analysis included 3 systematic reviews, 1 randomized controlled trial (RCT), 8 cross-sectional studies, and 4 uncontrolled case series. Most studies involved adult patients with chronic noncancer pain, two also included patients with cancer pain, but only one included patients with a previous history of dependence.
Minozzi et al. found that the incidence of addiction reported across the various studies ranged from 0% to 24% (median 0.5%), while prevalence ranged from 0% to 31% (median 4.5%). However, there was a great amount of variation among the studies, or heterogeneity, in terms of design, definitions of addiction, data collection, and other factors, so a data meta-analysis could not be conducted. Overall, the researchers rated the evidence as being of very low quality.
The researchers observe that fears of patients developing opioid addiction are greatest when prescribing opioids to treat chronic noncancer pain conditions, as both the course of treatment and the life expectancy of patients are often expected to be long — allowing time for addiction to develop. However, despite the many limitations encountered in their review, the authors conclude that, “The available evidence suggests that opioid analgesics for chronic pain conditions are not associated with a major risk for developing dependence [addiction].”
COMMENTARY: During the 1990s it was believed that addiction developing in persons prescribed opioids for pain was a relatively rare occurrence, largely based on observations of such treatment in cancer patients. Slowly, the pendulum has swung in the other direction and today there is great concern among practitioners and government agencies that such addiction in patients with all types of chronic pain is much more common than previously believed. However, as this review by Minozzi and colleagues attests, the truth in this matter has yet to be discovered and confirmed by valid and reliable research studies.
Systematic reviews and data meta-analyses are high on the evidence hierarchy [See the recent UPDATE — Part 13 in our series on “Making Sense of Pain Research” — explaining systematic reviews and meta-analyses here]. However, in this case, the researchers make it abundantly clear that available evidence on this subject of addiction during pain therapy with opioids is grossly inadequate and a data meta-analysis could not even be attempted.
“The most impressive finding of the present review is the deficiency of good-quality studies,” Minozzi et al. write. “This seems to stand in contrast to the widespread concern of doctors and authorities relating to the prescription of opioids for pain management.”
The wide ranges for addiction incidence (0%-24%) and prevalence (0%-31%) reported by the researchers are indicative of the confusion and lack of precision surrounding this subject. It further seems uncertain that the median values — 0.5% incidence and 4.5% prevalence — can be reliably used to denote the extent of addiction in patients receiving opioid analgesics for pain.
At that, there is uncertainty in this article by Minozzi and colleagues regarding use of the terms “incidence” and “prevalence,” and, in a table describing study characteristics, they also use the term “frequency.” The authors do not explain the meaning of these terms, as used in the studies examined or in their article.
Prevalence usually refers to the proportion (eg, percentage) of patients in a population affected by a disorder compared with those not similarly affected at a point in time. In this case, prevalence could be inclusive of all patients who came into pain treatment with current or prior opioid addiction plus those who newly developed addiction (de novo) during therapy with opioid analgesics [see Wu et al. 2003].
Incidence normally is used to denote the proportion of new, or de novo, cases developing during a particular time period [Wu et al. 2003]. For example, it might indicate the percentage of patients receiving opioid therapy for pain who newly develop addiction specifically as a result of treatment (also referred to as iatrogenic). The term “frequency” in the review article by Minozzi et al. seems to be a nonspecific measure of occurrence, denoting either incidence or prevalence.
Distinctions between prevalence and incidence can make a big difference, and it makes sense that the incidence would be much smaller than prevalence, as Minozzi and colleagues suggest. However, the authors may have just blindly adopted the terms as they were vaguely used to report data in the various studies reviewed, and this is another indication of confusion surrounding this whole area of research. There were many other limitations described by Minozzi et al., such as:
- A heterogeneous array of opioids was reported in the studies — including weak and strong opioids in short- and long-acting formulations — as well as oral, intrathecal, and intravenous modes of administration. So, no conclusions could be asserted regarding these factors differentially affecting development of addiction during opioid therapy for pain.
- Only 11 of 17 studies reported on length of opioid treatment, which ranged from 3 days to 81 months, but most studies were of short duration during which time development of addiction might be unlikely. None of the studies reported the time since the start of opioid therapy after which addiction became apparent.
- Only one study reported separate data for addiction in patients with a prior history of substance abuse or addiction, which was probably an important contributor to addiction prevalence. Indirect evidence from logistic regression analyses in 3 studies found that prior substance abuse/dependence was a strong predictor of addiction during opioid analgesic therapy.
- Data regarding addiction developing in patients with cancer who were treated with opioids were inadequate for deriving any conclusions. The single study of any consequence, involving only 100 patients, found no (0%) incidence or prevalence of addiction.
- A diversity of definitions and measurements were used to assess addiction, with relatively few studies applying standardized DSM or ICD criteria. At that, only one study has directly addressed the fact that DSM/ICD criteria need to be adjusted for patients with pain, since behaviors alleged to be addiction-related may actually be driven more by needs for pain relief rather than drug seeking and taking.
In sum, despite having conducted the most exhaustive evidence gathering and review process on this subject to date, and their contention that addiction is not a “major risk” of opioid therapy for chronic pain conditions, Minozzi et al. concede later in their article, “The present data on the incidence and prevalence of dependence following the prescription of opioids to treat chronic and acute pain cannot be considered conclusive.”
Indeed, we similarly concluded in an earlier UPDATE — “When is Rx-Opioid ‘Addiction’ Something Else?” [here] — the following:
“Given the difficulties of definition, observation, and measurement — and potential influences of bias and prejudice — it is not surprising that estimated risks, prevalence (which includes pre-existing substance-use disorders), and incidence rates of de novo iatrogenic opioid addiction in persons treated for chronic pain have varied widely.”
We further observed that, if nothing else, the extreme range of alleged “addiction” reported in all of the various studies to date — from 0% to 50% — suggests that little is known about the true risks and occurrence rates, whether prevalence or incidence, of this disorder in persons with chronic pain receiving opioid therapy. At the same time, there is a possibility that addiction could be prevalent in the population of persons with chronic pain in treatment to an extent consistent with that of the general populace, which some authorities estimate to range from 13% [Jones et al. 2012] up to 61% [Sussman et al. 2011]. Again, however, the wide variance of such estimates may reflect a lack of precision and confusion over accepted definitions and inclusiveness — eg, possibly involving prescription or illicit drugs (including but not only opioids), alcohol, and/or tobacco — or even other, non-chemical addictions.
Taking all of this into account, there remains considerable uncertainty about both the true prevalence and incidence of addiction to opioid analgesics prescribed as a component of therapy for any type of pain condition, at any dose, and for any specific period of time. So, the most reasonable, evidence-based answer at this time to questions on this topic would seem to be, “We still do not know.”
> Jones JD, Mogali S, Comer SD. Polydrug abuse: A review of opioid and benzodiazepine combination use. Drug Alcohol Depend. 2012;125:8-18.
> Minozzi S, Amato L, Davoli M. Development of dependence following treatment with opioid analgesics for pain relief: a systematic review. Addiction. 2012(Oct18); online ahead of print [abstract here].
> Wu L-T, Korper SP, Marsden ME, et al. Use of incidence and prevalence in the substance abuse literature: A review. Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies; 2003 [PDF here].
> Sussman S, Lisha N, Griffiths M. Prevalence of the addictions: A problem of the majority or the minority. Eval Health Prof. 2011;34(1):3-56 [abstract here].
Don’t Miss Out. Stay Up-to-Date on Pain-Topics UPDATES!
Register [here] to receive a once-weekly e-Notification of new postings.