Circulation 94,429 • Volume 26, No. 1 • Spring 2011   Issue PDF

Opioid Prescribing: REMS Sleep, Need Reawakening

Gregory W. Terman, MD

Letter to the Editor

On September 27, 2007, the Food and Drug Administration Amendments Act (FDAAA) was signed into law. This law signaled a very significant addition to FDA authority including additional requirements, authorities, and resources with regard to drug safety both pre- and post-marketing—including the authority to require Risk Evaluation and Mitigation Strategies (REMS). These REMS may be demanded from the sponsor (usually a drug company), either before or long after the drug has been approved, to ensure that the benefits of a drug outweigh its risks. On February 6, 2009, the FDA sent letters to 16 manufacturers of 24 opioid analgesics informing them that they would be required to submit REMS on their products ( These products primarily included extended release opioids as well as methadone. Statistics from 2007 showed that these 24 products were prescribed 21 million times for 3.7 million unique patients. This was a big deal. In this copy of the APSF Newsletter, Dr. Joan Christie gives a nice summary of a risk mitigation strategy for prescribing one of these products—methadone—including patient evaluation, informed (patient and doctor) prescribing, and appropriate monitoring. I will take a slightly different tack in this article—describing what progress has been made in implementing prescription opioid REMS during the last 2 years. For those short on time, the grim answer is still, “not much.

The risks of prescription opioids are real. The roughly 600% increase in opioid prescribing from 1997-2007, based in large part on pain doctors such as myself, entreating physicians to not be afraid of treating pain patients with opioids, was accompanied by a 300% increase in accidental deaths related to prescription opioids according to the Centers for Disease Control (CDC). Although it is not an exact science, to be certain that people dying with opioids in their bloodstream died of those opioids, there is no denying that mortality thought to be caused by prescription drugs is one of the few causes of death still on the upswing in the CDC Health US 2010 report. Neither can we deny the data that someone dies from causes labeled “unintentional drug overdose” every 19 minutes in this country—making it the leading cause of accidental death in 17 states (even “besting” the number of motor vehicle accident deaths in those states). Further, although other sedatives (including alcohol, benzodiazepines, and illicit drugs) are also often associated with these deaths, the numbers of prescription opioid-associated deaths have, for most of the last decade, been greater than the number of heroin- and cocaine-associated deaths combined. It is little wonder then that in May of 2009, at an FDA public forum discussing opioid REMS, I found myself, on behalf of the American Pain Society Board of Directors, “pledging to help the FDA at every opportunity to develop, deploy, and determine efficacy of (opioid) REMS.” Of course, we also stipulated that these REMS should not endanger patient access to opioids necessary to treat their pain. I reminded the forum that opioid treatment and pain treatment have never been synonyms and that, indeed, many American Pain Society members have never written an opioid prescription for pain (in their jobs as nurses, psychologists, and basic scientists). We, and hundreds of other groups and individuals, submitted comments concerning REMS to an online “docket,” which was open at least intermittently from May 2009 till October 2010. Throughout this process the FDA has been nothing if not responsive to its “stakeholders.”

A proposal for opioid REMS was published by the FDA in June 2010. Shortly afterwards, on July 22-23, 2010, a joint meeting of the Anesthetic and Life Support Drugs and the Drug Safety and Risk Management advisory committees discussed the proposed REMS but voted them down, 25 to 10 ( Nevertheless, the role of advisory committees is advisory only, and the FDA is expected to come out with their final opioid REMS requirements in the next few months (if not sooner). The manufacturers of the covered products will then have 120 days to respond. Thus, one can say that the REMS sleep at the moment, and when they will wake is not yet known.

While the FDA is deciding on their course of action, it is important to note that other government agencies (both federal and state) have also noticed the prescription opioid overdose problem. Most of these actions will likely restrict, or at least “guide,” prescribers and dispensers in treating their patients in pain, and I admit that I am always nervous when legislators dabble in medicine. Despite this bias, however, it was even clear to me last July (when I voted FOR the REMS as a reasonable first step) that many of the concerns advisory committee members had with the proposed REMS were outside the FDA’s jurisdiction to correct (even under FDAAA). In the last part of this article I will discuss some of these concerns—largely replicating my May 2009 comments to the FDA, but with updates of some of the non-FDA activities taking place in the country that are relevant to the topic.

  1. REMS should cover the entire class of opioid medications—not just long-acting opioids. This has certainly been a concern about the way the FDA views opioid REMS from the start. People can die from any opioid. Past experience has shown us that any attempt to regulate only a few opioids will drive prescribers, users, and particularly misusers of these medications to other, less stringently regulated, opioids that may be less effective therapeutically and actually may pose greater addictive or toxicologic risks. Indeed, it may be said that it is only for the opioid abuser that any opioid will do! Whether such class-wide REMS would drive prescribers or smaller manufacturers to completely abandon opioids is not known, but would likely depend on other details of the REMS. For example:
  2. There should be NO REMS registry requirements for patients using opioids. Registries have historically been an important mechanism by which the FDA attempted to improve safety (an “element to assure safe use” or “ETASU” in REMS). By the word “registry” I refer to a database filled with names of patients, prescribers, dispensers, or distributors who are allowed to be involved in the medical use of a particular drug and which is populated solely on the basis of a willingness to sign up to be in the database. In the REMS accompanying the approval of buccal fentanyl (Onsolis) in July 2009 for example, registries were huge components. In order for a patient to receive the drug, the prescriber, the patient, the pharmacy, and the distributor will all be in separate databases (registries) maintained by the manufacturer. Obviously, if this was done for every opioid on the market, access to opioids for patients in pain would be compromised. In fact, no evidence exists to suggest that a federal or state patient registry diminishes abuse or misuse of medications. In contrast, evidence DOES exist suggesting that such an approach can stigmatize patients and impose significant burdens on all parties, resulting in stilted prescribing and, perhaps, inadequate pain management. Enhancements in the current state Prescription Monitoring Programs (PMPs), including integrating these into a national program, would be a better system for helping to monitor where drugs are going—providing real-time information for prescribers concerning other prescriptions their patients have filled recently. Improved PMPs would allow identification of patients receiving opioids from several prescribers (a risk factor for opioid associated overdoses, along with a history of drug abuse and psychiatric comorbidities). The Office of National Drug Control Policy (ONDCP) has also consistently encouraged expansion of PMPs (e.g., but at least 8 states still have no PMP, and of those that do, most are only rudimentary and slow, and even the best programs have no way as yet to share information across state lines. Indeed, after my talk at the FDA, encouraging expansion of PMPs, in May 2009, I returned home to my own state (Washington) to hear that our PMP had been shut down due to lack of funds (a decision that has now, thankfully, been reversed).
  3. Demonstrated prescriber and dispenser knowledge concerning opioid pharmacology should be expected of all who seek DEA licensure. It seems appropriate that clinicians be required to demonstrate competence in safe and effective prescribing or dispensing of opioids for therapeutic purposes. Broad participation and compliance would be most likely if demonstrated competence were a prerequisite for DEA registration. The content of competency testing should be developed with appropriate expert input—independent of drug company influences. Indeed, over the last year a number of pain treatment curricula/guidelines (including appropriate opioid prescribing) have become available through state mandates and public and private funding mechanisms. The use of such education as a pre-requisite for DEA licensure would likely require new congressional legislation, but this may still be the fastest route to consistent and effective change compared with waiting for changes in all prescriber Boards in all 50 states. In my state of Washington again, for example, opioid guidelines for chronic pain treatment were crafted by a panel of “pain experts” in the state supported by all of the state health care payers. These were published in 2007 (and updated in 2010) as an “educational initiative” ( However, in July 2010 a bill (WA HSB 2876) was passed into law requiring all health care boards in the state whose licensees prescribe opioids to develop “rules” for chronic pain treatment. These rules are due to go into effect in July of this year and at present show considerable similarity to the previous educational guidelines. How these rules will affect the efficacy and safety of opioid prescribing for pain in Washington State, not to mention how this approach might compare to 49 other approaches to this problem in other states is, of course, unknown. This leads to the next issue:
  4. All implemented REMS should be measurable and, when necessary, easily reversible. Frequent intentional evaluations of all REMS components for their positive and negative impacts must be tied to their implementation. Indeed, in the FDA’s initial REMS proposal, frequent evaluations of REMS effects were mandated. However, many advisory committee members were concerned that there was no scientifically valid method detailed for collecting appropriate statistics and no current baseline data from which to compare changes. Again, the FDA is not a funding mechanism for appropriate studies in this area, and unfortunately, the NIH has thus far demonstrated minimal interest in supporting research on either the mechanism or prevention of prescription opioid-associated deaths.
  5. REMS education programs should also be aimed at the public. Most surprising to me in my initial introduction to this field was learning that much of the problem with opioid overdosing is due, not to prescribing per se, but to drug diversion. My initial presumption was that poor dosing practices were causing the problem, but in some studies of prescription opioid-related deaths more than half of decedents had NO prescription for the opioids that probably killed them. Worse, more than half of those patients (labeled “non-medical” users) received the drug from someone that they knew who HAD gotten the drug by prescription. No amount of education of prescribers or dispensers about opioid pharmacokinetics will save these people. However, that does NOT mean that there is nothing we can do! Product-specific patient education materials were stipulated in the proposed REMS and are available already in many instances. This information can and should be discussed with patients by providers. In particular, focusing drug education on appropriate use (it is only for them), storage (keep under lock and key), and and disposal (per FDA guidelines) of the drugs are key safety interventions we can perform for our patients and their families. In addition, public education programs must be intensified concerning the dangers of sharing opioid prescription drugs and the urgent need for treatment of opioid-induced sedation. Such public education programs should be combined with new and creative “give back” and/or “buy back” programs enabling collection and appropriate disposal of unused prescription opioids to further reduce availability and diversion. Again the ONDCP is supporting these efforts and last September’s National Take Back Day drew 121 tons of drugs in 4 hours. Unfortunately, all of the sites in my area were police stations, where some patients might be reluctant to take their excess drugs. The next Take Back Day is scheduled for April 30. In short, we must be aware of these and other steps to avoid diversion. If every realtor knows that the most common visitor to an open house is one who wants to check the medicine cabinet, doctors probably should, too.

Gregory W. Terman, MD, is a Professor in the Department of Anesthesiology and Pain Medicine at the University of Washington School of Medicine, Seattle, WA.


While this article was in press the Office of National Drug Policy released a comprehensive plan for combating Prescription Drug Abuse ( As a part of their 4-prong plan (including increased provider and patient edutcation, Prescription Drug Monitoring Programs, prescription drug takeback programs, and DEA efforts to shut down so-called “pill mills”), the FDA sent letters to manufacturers of long-acting optiates updating their general, primarily educational, requirements for REMS for these drugs (e.g., Readers are referred to these ONDCP and FDA plans to compare them to the suggestions mentioned in this article.