By Richard A. Lawhern, Ph.D.


A CDC Guideline for prescribing opioids in chronic pain was published in March 2016.  It has become clear that this Guideline is generating horrendously negative results for both chronic pain patients and their doctors.  Many doctors are choosing to leave pain management rather than face possible prosecution by State or Drug Enforcement Agency authorities for over-prescription of pain relieving opioids.  Tens (if not hundreds) of thousands of patients are being summarily discharged without referral and sometimes without management of opioid withdrawal.  There are increasing reports of patient suicides.

Many professionals in medicine have published sharply critical reviews of the problems of the CDC Guideline.  Also of deep concern are proposals by the US Centers for Medicare and Medicaid to make the Guideline a mandatory standard for insurance reimbursement.   Many believe that the Guideline was deeply influenced by an anti-opioid bias on the part of key writers on the CDC Working Group which authored it.  Even more damaging have been revelations that medical evidence assembled in support of the work was manipulated to discredit opioid reliability and over-magnify risks.  As one group of medical professionals phrased the issues, the CDC Guidelines are “neat, plausible, and generally wrong.”   Inarguably, so also is most of the dominant public narrative on opioids and chronic pain.  As remarked in a widely viewed TED talk  on U-Tube, “Most of What You Think You Know About Addiction is Wrong.”

It is now clear that CDC Guidelines must soon be withdrawn for a major rewrite.  This is the only ethically and morally sound way forward.  The public narrative has become mired in an avalanche of hype, conflicting claims and financially self-interested posturing on the part of addiction treatment specialists, insurance company partisans and pharmaceutical company marketers.  Especially important in any rewriting process must be the inclusion of stakeholder voices that were largely unheard in the first writing of the Guidelines.  Chronic pain patients themselves should be among this list, as should board certified pain management specialists active in community practice outside hospitals.

It is thus appropriate to ask what pain patients might write if they were tasked to revise prescription guidelines themselves.  The following is not a “standard” of medical practice. The present state of validated medical evidence is inadequate to produce such a standard with reliability.  This summary of principles is instead, a good faith effort to capture both the state of medical evidence and the experience of many thousands of patients who support each other online and in social media — too often in the absence of support from medical professionals and government policy communities.

Guidelines on Prescription of Opioids in Chronic Pain

  1. There should be no distinction in principle between the objectives of treating pain which is acute, chronic, or associated with advanced medical conditions assessed to be terminal. The objectives in all three cases are to alleviate suffering, promote patient functioning and improve quality of life.  In this sense, treatment of pain is always “palliative”.  To deny treatment of pain when effective means exist to manage it, is a fundamental violation of human rights.
  2. In the practice of general medicine, prescription opioids are not considered to be drugs of first choice in the control of long term pain. Opioids are more frequently used with moderate to severe acute (short term) pain associated with injury, surgery or dental procedures, where pain is expected to resolve in a few days. When prescribed for periods over 60 days, opioids are most often used for pain that is found by a physician to be unresponsive or intractable to other therapies. In the context of chronic pain and palliative care for terminal conditions, managed access to opioid prescriptions is indispensible at the current state of medical knowledge.
  3. Insurance organizations and some State legislatures have placed increasing emphasis on STEP (fail first) therapy as an alternative to opioids or surgical interventions and as a measure of cost control. However, the potential benefits of such therapy must be explicitly weighed in each patient, against the risks of disease progression and development of more intractable pain.  Likewise needed is a thorough review of the outcomes of interventional spinal surgeries and injections, to establish for what conditions such procedures are most likely to be effective and with what long term medical risks to patients.
  4. In opioids as with any other medications, best medical practice is that treatment should be accomplished with the minimum dose that is effective and for the minimum duration appropriate for the intended purpose. In acute pain associated with injury, surgical or dental procedure, this may mean that a maximum of seven days medication is initially prescribed, in doses and of types deemed adequate for the patient to sleep and to function in daily life.
  5. The experience and training of individual physicians are centrally important in determining appropriate types and strength of medications. However, provision must also be made for integrating patient reports, particularly when pain control is experienced as inadequate. Extension of an opioid prescription for acute pain must be founded upon the prognosis for full resolution of the medical issues which prompted the prescription.
  6. Treatment of pain in America takes place in a social environment where addiction to opioids and overdose deaths involving them as contributing factors are a major public health issue. There is widespread concern that unmanaged or casual availability of opioids may lead many people to become addicted, no matter what the “source” of such drugs may be.  While ready availability of prescription opioids before 2010 may have contributed to their diversion by adolescents or by addicts shamming pain, evidence from a Cochrane Review in 2008 suggests that risk of addiction in patients who have no previous history of opioid use is very small – possibly less than one percent.  Opioid overdose deaths in 2016 were dominated by street drugs (heroin, fentanil illegally imported from China and Mexico, methadone diverted from treatment programs).  Public policy on opioid prescriptions must acknowledge this context.
  7. The most reliable risk factors associated with addiction are status as an adolescent, a history of family trauma, and/or long term unemployment. None of these factors is addressed by restriction of opioids prescribed to patients in pain.  Education of patients to control others’ access to drugs at home and to turn in or destroy unused medications may help at the margins to reduce exposure of non-patients to potentially addicting drugs.  But there is no evidence that restriction of opioid medications reduces overdose deaths.  To the contrary, it is clear that reformulation of Oxycontin in 2009 to reduce its abuse potential was accompanied by a sustained increase in overdose deaths involving street drugs.
  8. Before starting opioid treatment for chronic pain, clinicians and patients should mutually establish treatment goals and expectations for pain and function. Criteria and a transition plan should be established for how opioid therapy will be discontinued if observed benefits do not appear to justify risks or untoward side effects.  However, opioid therapy should not be discontinued in the absence of evidence that the underlying disorder is resolving, or that effects and side effects of treatment outweigh the benefits in the individual patient.
  9. Before starting and periodically during opioid therapy for chronic conditions, clinicians should discuss with patients known risks and realistic benefits of opioid therapy as well as patient and clinician responsibilities for managing therapy. Patients should be reminded of their responsibility for controlling access to powerful opioid drugs by friends or family members who have not been prescribed them.
  10. Clinicians should not dismiss patients from their practice solely because of a substance use disorder. Such action can adversely affect patient safety and could represent patient abandonment. Identification of substance use disorder represents an opportunity to initiate potentially life-saving interventions.   It is important for the clinician to collaborate with the patient regarding their safety, to increase the likelihood of successful treatment. Although identification of an opioid use disorder can alter the expected benefits and risks of opioid therapy for pain, patients with both pain and substance use issues require ongoing pain management that maximizes benefits relative to risks.  Consultation with an addiction treatment specialist is in order for general practitioners and others who lack current training in the specialty or who are not appropriately licensed.
  11. Multiple physiological and genetic factors should influence physician prescription of opioid types and doses to the individual patient. These factors include the patient’s history of previous opioid exposure, any history of alcoholism or drug abuse, the types and causes of pain being treated, the physical weight of the patient, and the individual’s ability to metabolize opioids or non-opioid medications. When pain is unresponsive to initial prescription of opioids, genomic testing should be considered to establish whether the patient is a poor metabolizer of opioids or other medications.  Long term medication management should integrate the findings of such testing.  Research is needed to assess the reliability of opioid-related genomic testing and to reduce its costs.
  12. Although it seems plausible that higher opioid dose levels might be associated with higher risk of dependency or addiction in some patients, medical evidence is presently inadequate to generalize numerical thresholds on opioid dose levels considered dangerous. An important criterion for evaluating risks and benefits to long-established patients is whether the patient has been on a stable opioid dose, without progression of the underlying medical conditions which generate pain.  When the patient is stable — even on doses over 400 MMED — there should be no reason to change opioid dose levels, absent indications of progression in the underlying medical disease or disorder, or identification of an alternative treatment with fewer potential side effects.  Physical dependence on opioids is not a sufficient criterion for forced withdrawal of these medications in stable chronic pain patients, regardless of dose.
  13. It must be recognized that established addicts may attempt to doctor-shop for physicians who are incautious in prescribing opioids to patients with a history of either medical or non-medical use. To reduce hazards from diversion or misuse of prescribed opioids, doctors and pharmacists must have timely access to a US national opioid prescription database, cross-referenced to reliable real-time identity data. Online use of such a database must be optimized for rapid access by physician and pharmacy office staff, with a goal of completing data entry and system response in a maximum of 60 seconds online.
  14. When there are indications that a patient has requested treatment by three or more doctors in a year, the patient’s medical records should be reviewed to assess circumstantial factors beyond the patient’s control, before taking action to more closely monitor opioid access, alter treatment plans, or engage a substance abuse management specialist in coordinated treatment.
  15. US Government authorization for production and distribution of opioid medications should be optimized to ensure that all patients with legitimate medical needs receive their prescriptions in a timely manner. Measures must also be implemented to detect drug diversion before, during or after distribution. Pharmacists who believe that a prescription exceeds accepted medical practice should contact the prescribing physician’s office immediately, to confirm the appropriateness of the prescription.  In no instance should a patient be refused refill on a valid prescription solely on the basis of pharmacy corporate policy or the individual “comfort levels” of the pharmacist.
  16. Urine testing should be implemented in initial and periodic follow-up physician appointments, to assess whether patients are using medications other than those prescribed or in quantities different than prescribed. When a patient must travel more than one hour to attend medical appointments, use of short-notice urine testing should be implemented by public health nurses or local clinics in the communities where patients live. There is presently no conclusive evidence that drug use “contracts” between patients and doctors actually reduce the number of hospitalizations for opioid toxicity or the number of overdose deaths. Research investment may be needed to lower costs and improve reliability (reduce false positive rates) of such testing.
  17. Clinicians should not test for substances for which results would not affect patient management or for which implications for patient management are unclear.  For example, experts have noted that there may be uncertainty about the clinical implications of a positive urine drug test for tetrahyrdocannabinol (THC). In addition, costs of urine testing can be reduced by restricting confirmatory testing to situations and substances for which results can reasonably be expected to affect patient management.  Before ordering urine drug testing, clinicians should have a plan for responding to unexpected results.
  18. Just as there is a potential for opioid misuse among patients, there is valid concern that some physicians licensed to dispense opioids may do so inappropriately or with inadequate patient oversight. However, the volume of opioids prescribed in a medical practice is not a one-size-fits-all criterion to justify confiscation of a doctor’s medical records or imposition of sanctions prior to review by State medical boards or courts. Prescribing patterns must be assessed in the context of the numbers of patients served, the nature of their underlying medical issues, and the area availability of practitioners and centers formally accredited to prescribe.
  19. A careful and non-stigmatizing distinction must be applied in the assessment of opioid risks and side effects among long-term pain patients. It is established that many if not most chronic pain patients do not get a euphoric “high” from opioid use. Nor do these patients commonly display the spectrum of compulsive and self-destructive drug seeking and escalating dose levels displayed by addicts. Some pain management specialists believe that neurochemical and genetic markers exist which identify patients in whom chronic pain has produced permanent changes in the brain. Such markers might in principle offer means of differentiating some addicts from chronic pain patients. Further research may be needed to confirm the science and reduce such testing to common and affordable practice.
  20. Physical dependency on opioids for pain control may cause withdrawal symptoms when medications are suddenly reduced. However, drug dependency is not the same medical entity as addiction. In patients for whom resolution (“cure”) of underlying medical conditions is deemed unlikely, dependency may be an acceptable and manageable side effect in maintaining the best possible quality of life, in the absence of effective alternative means.   When dependent patients are withdrawn from opioids for whatever reason, physicians are ethically required to provide assistance with withdrawal symptoms.
  21. Despite many articles in popular press and medical literature, there is little validated medical evidence for what is called “opioid-induced hyperalgesia” (increasing sensitivity to pain over time, requiring sharply increasing doses of opioids for effective pain control). There are no documented diagnostic protocols for this supposed medical entity, and there is no generally accepted protocol for treatment, other than tapering down present medications and tapering up others. Some pain management specialists go so far as to suggest that if opioid-induced hyperalgesia exists at all, it may be associated primarily with constant exposure of the brain to opioids delivered by intrathecal pain pumps.  Given these observations, when drug tolerance or increased pain are reported by patients, medical assessment is warranted for progression in the underlying processes believed to have produced pain. Development of opioid abuse disorder should not be a default assumption. It is also well established that initial diagnosis of many chronic pain conditions is complex and frequently incorrect.
  22. Opioids have both short and long term side effects which must be actively managed by the prescribing physician. These may include constipation, dry mouth, nausea, confusion, lethargy, sedation, suppression of sexual libido, situational or chronic depression, and anxiety.  Drug interactions are also of concern in developing a pain management plan for the long term.  It is known that a significant fraction of patients who die from opioid-related overdose have also been prescribed Benzodiazepine drugs.  Physicians should avoid dual prescription of Benzodiazepine drugs and opioids wherever possible.  When benzodiazepines are deemed medically necessary, drug metabolism and interactions should be monitored frequently to avoid accidental death by respiratory suppression.  Alcohol is also a major contributor in drug overdose and patients should be counseled on its dangers.
  23. There is much public discussion of Naloxone as an interventional drug to counter the immediate effects of opioid overdose. The CDC has proposed to make this drug widely available to emergency medical technicians and police as an aid in saving the lives of drug overdose victims. However, there seems to be little medical evidence that Naloxone can be used similarly to Methadone, to suppress cravings for other opioids over the long term.  A few published reports have suggested that Naloxone therapy may merely postpone the decline and death of addicts who do not enter a more sustained program of community based non-drug treatment.  Investment will continue to be needed in community programs of addiction treatment. There is no easy one-size-fits-all solution for addiction.
  24. There is also much public and professional discussion of non-opioid medications (e.g. steroids, NSAIDS, anti-seizure medications, anti-depressants) and behavioral therapies (rational cognitive therapy, operant behavioral therapy, creative visualization, acceptance therapy) as alternatives in chronic pain management. The present state of medical evidence is not sufficiently developed to generalize practice standards for the use of such therapies in preference to opioids.  Based on published papers and tens of thousands of patient reports in social media, it is clear that these non-opioid approaches to pain control do help some patients, some of the time.  But no patient should be coerced to accept such therapies as replacements for opioids, particularly if the patient hasn’t experienced improvement in pain during trial of the therapy.
  25. Due to its status as a Schedule I “narcotic”, marijuana-related research has largely been stifled in the US even as it continues elsewhere. In recent years, there has been an increasing trend toward legalization of medical uses and in some US States, of recreational use. While much remains to be learned about the mechanisms and reliability of marijuana and its components (e.g. CBD oil, THC) as medical treatments, one fact is entirely clear: there is no valid evidence supporting the designation of marijuana or its component products as a “gateway” drug for opioids or other addictive drugs.  Moreover, data are emerging from States where medical marijuana has been legalized, suggesting that where marijuana is available, opioids are less often used both medically and recreationally.  Immediate rescheduling of marijuana is warranted to facilitate further medical research to investigate and quantify its potential benefits,  reported by tens of thousands of chronic pain patients in social media.