While buprenorphine has been prescribed for many patients over the last 10 years, there’s still controversy about whether healthcare professionals should be allowed to work while on buprenorphine.
In an article in March 2012 Mayo Clinic Proceedings, Hamza and Bryson cite studies that support their conclusion that medical professionals should not be allowed to work while taking buprenorphine as maintenance for opioid addiction. The authors say studies show that people taking buprenorphine have some impairment when performing safety-sensitive tasks that are required in practice as a physician. (1)
I read this article with great interest, since I have been prescribing buprenorphine and telling my patients they won’t be impaired while taking a maintenance dose. Wanting to know if I am misleading patients, I scrutinized the studies cited in this paper.
I’m not sure the authors’ conclusions are backed up by the studies they cite.
The most worrisome misinterpretation was the Schindler et al study. The Mayo study by Hamza and Bryson interpreted the Schindler study thusly: “significant differences were found between them [methadone and buprenorphine groups] and the controls.” But when I read the original study, the authors’ conclusion was really the opposite: “The synthetic opioid-maintained subjects investigated in the current study did not differ significantly in comparison to healthy controls…” (2)
When I looked at other articles cited by Hamza and Bryson, I discovered that what I read didn’t match Hamza and Bryson’s conclusions of what I read.
Three of the studies cited in the Mayo article (Pickworth et.al., Jensen et. al., and Zacny et.al.) all looked at healthy volunteers who were given buprenorphine, then tested to see if they were impaired. In other words, these test subjects weren’t opioid dependent. All three studies showed impairment, and I don’t doubt it, because opioid-naïve subjects would be expected to feel a great deal of opioid effect with their first dose of buprenorphine. But studies of opioids-naïve subjects given buprenorphine don’t seem applicable to opioid-addicted patients on buprenorphine for maintenance.
The Rapeli et al study looked at methadone and buprenorphine patients in early recovery, so these groups would be expected to be different than those on established maintenance therapy.
Soyka et al compared opioid addicts on buprenorphine and methadone at 2 weeks, then at 8-10 weeks. This study also had a control group. The patients on methadone and buprenorphine had impaired cognition on testing compared to the controls, but they improved with length in treatment. This study was randomized but not blinded. This means patients and researchers knew who was on methadone, buprenorphine, and who was a control subject. Interestingly, in a later letter to the editor defending their conclusions, Hamza and Bryson mistakenly claimed the study was double-blinded, but clearly it was not. Also the study was relatively small, since only 46 patients completed the study. The purpose of the study was to see if methadone was more impairing than buprenorphine. The authors of the Soyka study didn’t conclude the buprenorphine group was impaired to the point they were unable to work, only that they performed better than methadone patients.
One study, by Messinis et al, did compare abstinent heroin addicts on naltrexone with opioid addicts on maintenance buprenorphine, and showed the buprenorphine group had more cognitive impairment than the naltrexone group in cognitive functions. To me, this is the main study that speaks to the actual issue of impairment. It gives a basis to require more studies be done. However, the small size of the study, 18 patients, limits the impact of this study. (3)
The ideal study to resolve this issue would be a double blinded prospective study of opioid-addicted healthcare professionals who are randomized either to abstinence-base treatment or buprenorphine maintenance treatment. Then cognitive abilities can be compared at various times during recovery, like 3 months, 6 months, 1 years, and 2 years. Such a test is unlikely to be done, since most addicted professionals enter abstinence-based recovery, and have a high rate of success.
I do think medication-free recovery is the ideal. I acknowledge that’s my bias, even though I strongly believe medication-assisted treatment is a life-saving option. But then, medication-free treatment is the ideal for all diseases. If a patient can achieve good blood pressure control by changing her diet and exercise, I think most of us would agree that’s a superior outcome to taking blood pressure medication to achieve the same result.
Most doctors and dentists have the resources to afford the prolonged inpatient treatment needed for medication-free recovery. The monitoring required for continued licensure is additional leverage and accountability that most opioid addicts don’t have after leaving inpatient treatment. These factors produce excellent recovery rates in these healthcare professionals, much better than that achieved by the average opioid addict.
But no recovery works for everyone. If a healthcare professional has failed traditional abstinence-based recovery, but is able to do well on medication-assisted recovery with buprenorphine, is the data strong enough to say such a recovering person on a stable dose of buprenorphine can’t work in healthcare?
We must be careful about this decision. If the decision is going to be based solely on patient safety, and not on a bias against medication-assisted recovery, then healthcare professionals on opioids for acute or chronic pain must also logically be removed from the workforce, unless we can prove they don’t have cognitive deficits from prescribed opioids. And what of other medications, like benzodiazepines, which are more likely than opioids to cause impairment?
If professional monitoring boards rely on the evidence cited by this study to refuse to allow healthcare professionals on buprenorphine to return to work, they leave themselves open to accusations inconsistent safety standards if they allow other healthcare professionals to work while being prescribed opioids or benzodiazepines.
It would be a mammoth task to monitor every healthcare professional who is prescribed a controlled substance. But if a professional on stable a dose of buprenorphine can’t work safely, how can we assume a surgeon who takes legitimately prescribed opioids for back pain is safe to work?
Frankly I suspect most of the posturing about the dangers of healthcare workers on buprenorphine is really an attempt to remove medication-assisted recovery as a treatment option for healthcare professionals. I don’t know if the mayo article authors, Hamza and Bryson, have any underlying bias against medication-assisted treatments, or perhaps biases favoring abstinence as the only worthy treatment goal. I don’t know these two people at all. But my impression is that they have taken a sweeping position supported by shaky evidence. The studies they cite are evidence enough to call for larger studies, but don’t seem adequate in themselves to deny a potentially life-saving treatment to a healthcare professional.
- Hamza H, and Bryson E, “Buprenorphine Maintenance Therapy in Opioid-Addicted Health Care Professionals Returning to Clinical Practice: A Hidden Controversy, Mayo Clinic Proceedings., 2012, 87(3);260-267
- Schindler SD, et al, “Maintenance therapy with synthetic opioids and driving aptitude, European Addiction Research, 2004; 10(2):80-87acol.
- Messinis et al, “Neuropsychological functioning in buprenorphine maintained patients versus abstinent heroin abusers on naltrexone hydrochloride therapy”. Hum. Psycholpharm. 2009;24(7):524-531