American Medical News
By — Posted Jan. 14, 2013
Scenario: A great deal of research suggests that physicians are not good at predicting who will misuse pain medication. To avoid making treatment decisions based on stereotypes, some physicians use such agreements for all their patients with chronic pain.
Reply:We have spent a fair amount of time thinking about whether to use behavioral agreements with all patients who are on opioids. Note that we don't call them contracts; that's intentional. Contracts are legal documents — and this is, instead, a way for us to share with patients our concerns about the overuse of opioids, their potential for abuse and our commitment to try to relieve their pain in the face of these issues. (The terms “opioids” and “opiates” are used synonymously here to cover a broad range of pain-relief drugs.)
There are no data supporting behavioral agreements, and little data supporting the use of urine drug testing or other efforts to decrease misuse of opioids/opiates. An interesting and exciting advance in the field from the chronic pain group at Harvard University in Massachusetts is an exception to this. The group did a randomized controlled trial of monthly urine screens, compliance checklists, and individual and group motivational interviewing therapy with patients at high risk of opiate misuse. Using a standardized assessment for drug misuse (the Drug Misuse Index) they found that the members of the experimental group were less likely to have positive DMI scores but similar pain scores.
We decided to use behavioral contracts because we worry that patients otherwise will not understand why we are concerned about opioids and why we treat them differently from other medicines. For example, when a patient loses some of his HIV or antihypertensive medicine, we refill it immediately and do not question him or her about how it got lost or ask for proof of its being lost. On the other hand, we want patients to know that we care about their pain and that, whether or not we are going to use opioids, we will not abandon them but will continue to work with them to try to relieve their pain.
Therefore, we use behavioral agreements with all of our patients. (We work in an HIV clinic in which we assume there is a relatively high pre-test probability of drug misuse.) We do this because we cannot tell who is going to misuse or abuse opioids.
There is a great deal of data suggesting that physicians are not very good in their predictions about drug misuse, and, therefore, we do not want to make this decision based on stereotypes about who misuses opiates. It seems safer and more just, we believe, to enter into the agreement with anyone who is using opioids for chronic pain. We take the same position on urine drug testing; we test all of our patients on a frequent, although irregular, basis.
In general, patients respond relatively well to being asked to sign the agreement. If they complain, it is usually to the ancillary staff — or they just do not come back. The agreement is part of a longer conversation about the use of opioids and their potential misuse, as well as our concern about accidental overdose. All of our patients also are trained on how to use nasal Naloxone.
If a patient is offended, we explain to him or her that this is done for all patients in an attempt to balance the known prevalence of misuse in a clinical practice such as this one with our commitment to all patients to attempt to relieve their pain, regardless of what drug they use.
We try to make our medical decisions based on what the data suggest is best for the individual patient. Thus, while we do consider the risk of abuse and its consequent side effects, we are less likely to weigh the risks of drug diversion or misuse of opiates by family members. Although some authors have argued that individual patient decisions should take these public health considerations into account, we believe that our primary obligation is to the individual patient.
At the same time, our use of opiates has decreased over the last decade in response to data suggesting that their risk-benefit ratio is less positive. That is, opiates do not seem to improve functioning or pain scores in many patients with chronic nonmalignant pain. The risk of medicine misuse is higher than previously suggested, and growing data indicate the importance of physical medicine, rehabilitation and psychiatric interventions for these patients. Our dosing of the medicines also is based on the best available evidence and individual patient responses. Although opiate management agreements do not influence these decisions, they do help clarify what needs to be done in case of misuse.
We do not think behavioral agreements have a legal benefit; we think practicing good medicine is the best legal strategy. On the other hand, to the degree that talking through and signing the agreement promotes shared decision-making, it may be legally useful.
If patients engage in behaviors that we specifically mention in the behavioral agreement, we institute behavioral consequences. Depending on the reason, we will have the patient come in more frequently, order urine drug tests more frequently, stop using short-acting opioids or stop using opioids all together. We do not believe this is a simple rule for every agreement violation behavior.
Our behavioral consequences need to match the frequency and severity of the misuse. For example, if a patient reports that medications were stolen, we insist that he or she provide a police report before considering whether or not to replace the medication. If the patient's urine drug screen is positive for illicit drugs, we may ask that the patient come in weekly for additional urine drug screens. If the patient is using fentanyl patches, we may ask that he or she date them and bring in the used patches before providing a refill.
Or in situations that require more intense monitoring, we may ask that the patient come in every three days to pick up his or her next patch. If surgery is the prescribed treatment to relieve the pain syndrome, the patient is expected to follow through and is monitored to see if he or she proceeds with it. Pain medications may be tapered and discontinued if the patient repeatedly cancels a required surgery.
We are not under the illusion that patients will never make mistakes. The data suggest that one-third to one-half of patients at some point will do something that we had asked them not to do in the behavioral agreement. Our job is to try to discern how dangerous a mistake they made, and to respond appropriately and clearly with the patient.
— Robert Arnold, MD, Leo H. Criep Chair in Patient Care, Institute for Doctor-Patient Communication Section of Palliative Care and Medical Ethics, University of Pittsburgh
— Carol Heape, RN, psychiatric nurse clinician, Pittsburgh AIDS Center for Treatment, University of Pittsburgh Medical Center