Opioid Treatment Agreements: Helpful or Hurtful?

A 46-year-old man with chronic back pain and alcohol use disorder, which was moderate and in sustained partial remission, was referred for a pain management evaluation due to his primary care doctor’s concern after the patient requested early refills of hydrocodone two consecutive months. After an extensive interview and physical examination, the findings and treatment plan were discussed with the patient. He was presented with an opioid treatment agreement for twice daily extended-release morphine and instructed to read through it carefully. The patient angrily stated, “I’m sick and tired of you and all my other doctors not trusting me and not treating my pain. You’re already trying to find a way to fire me!”

Prescription opioid misuse is a clinical and public health problem. Chronic nonmalignant pain management with opioids requires particular attention to personal and public health risks, such as side effects, dependence, dose escalation, overdose, and diversion (1). Opioid treatment agreements (OTAs) were developed to mitigate these risks.

An OTA is a written document that explicitly lists the expectations for opioid therapy. These documents, alternatively referred to as “contracts” or “informed consents,” are used to provide patient education, facilitate mutual agreement about treatment course, and establish terms of monitoring for safe opioid use (e.g., preventing diversion and addiction) (2). For these reasons, OTA use is recommended by professional organizations, including the American Academy of Pain Medicine and state medical boards (1). In the present article, we summarize OTA content, efficacy, and concerns associated with OTA use (Table 1).

The content of OTAs varies greatly. Two studies systematically examined the content of OTAs, one in academic medical centers (3) and the other in private practice (4). Both studies identified inconsistencies in the length and topics addressed. Most OTAs included the name, dose, and refill frequency of the prescribed medication, as well as patient expectations, prohibited behaviors, and reasons for termination of treatment. Refill procedures, expectations for urine drug screen monitoring, and risks of addiction were also discussed in most OTAs. Treatment goals were described in 56% of the OTAs used in private practice and 38% of those used in academic settings, and approximately 20% in both settings included a description of the responsibilities of the treating physician. Less common topics included the risks of driving while taking opioid medications, encouraging family involvement in care, and guiding principles for a healthy lifestyle. While all required patient signature, these documents differed in the need for physician signature. Our review of the literature did not yield any studies examining the efficacy of OTAs based on content, and thus the included topics and semantics are based solely on clinical judgment of the OTA author and treating physician. Sample OTAs are available online from the National Institute on Drug Abuse (www.drugabuse.gov/sites/default/files/files/samplepatientagreementforms.pdf) and the American Academy of Family Practice (http://www.aafp.org/home.html).

The efficacy of OTAs for enhancing patient adherence to treatment and limiting adverse outcomes (opiate misuse, diversion, or overdose) is unclear. A 2010 meta-analysis found low quality evidence (as determined by an amalgam of previously published quality rating scales and the Grading of Recommendations Assessment, Development, and Evaluation quality assessment criteria) that opioid misuse decreased with OTA use (5). This conclusion was based on four studies of varied designs with control group comparisons. A different study involving retrospective chart review examined the outcomes among patients with OTAs in a resident-run internal medicine clinic (6). The OTA in this study was a single-page document consisting of 16 statements outlining expected patient behavior, described as conditions for treatment. The decision to have a patient sign an OTA was made by the physician, and a total of 4% of all clinic patients signed OTAs during the 5-year observation period; the authors did not provide the total number of patients prescribed opioids during the study period. The most common medical diagnoses associated with OTAs were chronic lower back pain and fibromyalgia, and the most commonly prescribed medication was oxycodone/acetaminophen. Among these 330 patients, contracts were discontinued in a total of 122 patients (37%): 27 (8%) ended due to positive urine drug screen, 14 (4%) due to prescription drug abuse, 65 (20%) due to discontinued opioid therapy, and the remaining due to transfer of care or contract rules violation. Less than half of patients with OTAs had completed a urine drug screen during the retrospective observation period. These results highlight problems with the use of OTAs in common practice in which only a subset of patients treated with opioids are placed on OTAs, and the monitoring proscribed by the OTA is not followed.

The efficacy of OTAs may also be limited by patients’ misunderstanding of their existence and purpose. A 2012 study by Penko et al. (7) compared patient and physician understanding of OTAs. Patient response to the question “do you currently have a pain medicine agreement with any healthcare provider, clinic, hospital, or emergency room?” demonstrated a 61% sensitivity and 64% specificity, with physician reports of OTA use set as the gold standard. Thus, while the majority of patients were aware of the presence of an OTA, a significant minority were not.

It remains unclear whether specific populations benefit from OTAs. Some argue that OTAs should be used with all patients on chronic opioid therapy as a form of universal risk management (8), while others argue that OTAs should be limited to those at high risk for substance misuse (9). Unfortunately, no clear evidence guides clinicians on this matter.

Ideally, OTAs should help patients on chronic opioid therapy better understand the risks/benefits of opioid use to make educated decisions consistent with their principles and goals. One barrier to achieving this objective is that the literacy demands of OTAs are too high for some patients (10). A 2007 study found that the mean reading grade level of several OTAs was greater than or equal to 14th grade (11), well above the recommended 6th-grade reading level for health-related documents. A low literacy OTA version may address this concern, although this document has not been validated (12).

Another concern is that OTAs may be more punitive than beneficial for patients. Certain OTAs may violate ethical principles of autonomy and erode patient-physician trust, despite stated goals of enhancing patient-physician mutuality. Requiring a patient to sign an OTA can be coercive (10), particularly when presented as a mandatory condition of treatment, and can lead to a paternalistic imbalance in power in the patient-physician relationship (10). Because patients with chronic pain are particularly vulnerable, they may more readily agree to an OTA to which they otherwise may have not (10, 13). This pitfall is enhanced by the adversarial and intimidating language found in OTAs (10). In the content review discussed above (3, 4), it is important to note that reasons for treatment termination were a more common topic in OTAs than treatment goals, thus calling into question the stated goal of OTAs to establish mutual agreement regarding treatment course.

OTAs may also contribute to the practice pattern of “defensive medicine” (10), which is medical intervention performed to protect the clinician from medical liability, rather than in the best interest of the patient. Some question whether the true purpose of OTAs is to relieve physician anxiety rather than improve patient care (14). Specifically, in cases of patient and physician disagreement around opioid prescriptions, a signed OTA allows the physician to relieve anxiety by appealing to the guidelines set forth in the OTA and avoid difficult conversations (14). Overall, this suggests that OTA use may be putting the clinicians’ interests before the patients.’

The U.S. Food and Drug Administration (FDA) created a multidisciplinary work group composed of physicians, pharmacists, nurses, patients, and plain language experts to develop and validate the Model Patient Prescriber Agreement (model PPA) (15). The model PPA was designed to facilitate shared provider-patient treatment decision making and provide education regarding potential risks and benefits of, as well as alternatives and limitations to, prescription opioids with a more cooperative tone. In contrast to commonly used OTAs, consequences of patient misbehavior (early refill requests and PPA violation) were minimized. A survey of 209 FDA employees revealed that the model PPA provided acceptable patient information, supported shared decision making, and was at least somewhat easy to understand with a neutral tone. Thus, the model PPA has potential for good acceptability and usability in clinical practice (15). Work examining the function of the model PPA in clinical settings is currently ongoing.

Clinicians and policymakers are understandably concerned about risks associated with chronic opioid therapy. Proponents of OTAs argue that these documents reduce these risks while providing informed consent to patients. However, the lack of evidence regarding best content and the questionable efficacy to mitigate risks and improve patient care, along with potential pitfalls, make it difficult to fully support OTA use for patients on chronic opioid therapy. Further studies demonstrating OTA benefits for patient care and establishing the incidence of diversion and iatrogenic opioid use disorders may help to substantiate their use.