Commentary on Ling et al. (2013): Is there a role for psychotherapy in the treatment of opioid dependence?

Authors


Ling et al. [1] describe results from a carefully executed study comparing the addition of two efficacious psychotherapies alone and in combination with buprenorphine for treating opioid dependence. Although psychosocial services are required to be provided with opioid maintenance, none of the enhanced interventions in this study improved outcomes relative to medication management alone. Results from this study may question whether or not clinicians should be providing psychotherapies concurrently with opioid maintenance medications.

In considering this issue, it is important to reflect upon the study design and treatments provided. The pharmacotherapy studied, buprenorphine, is highly efficacious in reducing opioid use [2-4]. Because of buprenorphine's efficacy and safety profile, it provides a reasonable alternative to methadone for some opioid-dependent patients.

The two enhanced psychotherapies evaluated, cognitive–behavioral therapy (CBT) and contingency management (CM), are efficacious in reducing drug use. A meta-analysis of all psychotherapies for substance use disorders found CM to be the intervention with the largest effect size [5]. CBT also has demonstrated efficacy [6], but the two interventions combined do not necessarily or reliably provide additive benefits [7-10]. One possible explanation for the lack of additive benefits is that it is far more difficult to demonstrate differences between efficacious interventions than between an intervention and a control or no therapy condition.

Not only did the Ling et al. [1] study find the combined psychotherapies to not have additive benefits, neither CBT nor CM yielded benefits over medical management. The lack of any benefit of psychotherapy may be surprising at first glance, but numerous explanations must be considered.

First, and as noted by Ling et al. [1], the medical management in this study exceeded the standard care that is usually provided in community-based settings. Patients receiving medical management left urine samples twice weekly for 18 weeks (for which they were compensated), and results were discussed with them weekly. This frequency is in stark contrast to what typically occurs in community settings. Most probably, in the Ling et al. [1] trial, the frequent urine toxicology results were utilized to increase the probability that patients in all four treatment conditions were on optimal doses of buprenorphine. Thus, patients in this study probably received well-tailored and optimal doses of a medication known to reduce opioid use.

Secondly, it is difficult to demonstrate an additive effect of an intervention when a highly efficacious background treatment is applied, because there is less overall room for improvement. This study was powered to detect medium effect sizes between the psychosocial interventions, and it is rare to achieve medium effect sizes between two active interventions. Nevertheless, a small-to-medium effect size is noted for CM in this study when either of the conditions applying CM is compared with medication management in terms of the longest duration of opioid abstinence achieved: over 14 weeks versus 10.9 weeks. The effect sizes achieved with CM in this study are within the range of effect sizes obtained from CM interventions [11, 12], but the present study did not contain sufficient numbers of subjects to detect statistically significant differences of a small-to-medium nature between groups.

Thirdly, this study was designed to assess reductions in opioid use. Most studies of CBT and CM in opioid maintenance populations were developed to reduce the use of other non-opioid drugs such as cocaine [13-16], because the best method to reduce continued opioid use in maintenance populations is to increase the dose of prescribed opioids [5, 17]. Given the frequent urine testing and face-to-face meetings provided, it is likely that all patients in this study had their doses adjusted in relation to their illicit opioid use, diminishing the ability of psychotherapy to improve upon the benefits of buprenorphine in terms of their opioid use.

Although one could argue that the results from this study do not support a role for enhanced psychosocial care in the pharmacological treatment of opioid dependence, one could also contend that this study speaks directly for the need for enhanced psychosocial care of opioid-dependent patients receiving pharmacotherapies. The medical management provided in the context of this study did so well that it suppressed illicit opioid use by more than 50% using a conservative index assuming that all missed samples are positive; the rates of opioid-negative samples submitted (i.e. of patients attending care) were probably more than 70%. If all opioid-dependent patients were provided with the level of individualized care described in this study, perhaps 70% of patients would be retained in maintenance therapy long term and levels of opioid use would decrease to this extent. Conversely, novel psychotherapies and/or pharmacotherapies are still needed to assist the 30% or more of opioid-dependent individuals who do not remain engaged in care or reduce illicit opioid use (and polydrug use) in response to buprenorphine and medical management.

Declaration of interests

None.

Ancillary