Last updated: Apr 20, 2008
"Only about 5% of people with chronic pain ever see a specialist."

Russell K. Portenoy, MD, is chairman of the Department of Pain Medicine and Palliative Care at Beth Israel Medical Center in New York City.

Q: Would you say that opioids are a last resort?

A: No. Opioids should be considered for every patient with chronic, moderate to severe pain, but in every case, you would only prescribe the opioid after carefully considering the responses to several questions.

Q: What are those questions?

A: First, what is typical treatment with respect to this pain? Second, is there some other therapy that has as good or better efficacy and safety? Third, is this person at relatively high risk of opioid side effects for whatever reason? And fourth, is this patient likely to be a responsible drug taker, or is there a history of substance use problems?

So in some cases, for example a patient with severe pain who has not done well with several steroid or other drug injections and physical therapy, and who presents to the doctor with back pain so severe that he can't walk—that patient might be considered a candidate right then for a trial.

Q: What is an example of that review process with a typical patient who has arthritis of the knees and hips.

A: Everybody would agree that the first-line therapies typically would include acetaminophen, physical therapy, or a TENS unit, or maybe—if there's a single joint that has some swelling—an injection.

The next-line therapy would be an NSAID. But if that person has a history of an ulcer or a history of bad heart disease, the NSAID risk gets to be relatively high. So that patient might be considered for a trial of an opioid at that point.










Q: If I'm that patient and I'm put on a trial, how will I use the drugs?

A: Almost everyone with chronic pain appears to benefit more from regular, fixed, scheduled use as opposed to PRN [when needed] use. There is a general perception, two decades old, that patients do better if they have pain medicine in their blood 24/7. It's done in a sustained way, so that the blood levels don't fluctuate much.

Q: In the whole range of treatments for chronic pain, where do opioids fit in?

A: The chronic use of opioid therapy to treat noncancer pain syndromes, such as headache and low-back pain, and arthritis, continues to be controversial. Most pain specialists nowadays would say that opioids might be considered in any patient who has chronic, moderate to severe pain, but generally should only be implemented if there are no other treatment options that have a favorable and safe effect. The shortest way of saying this is that most pain specialists would not consider opioids first-line treatment for chronic noncancer pain except in highly selected patients.

But we have accumulated clinical experience that suggests the following: There is a sub-population of patients with chronic pain, who can be given access to long-term opioid therapy, and they will experience sustained and meaningful control of pain in the absence of intolerable side effects and without the development of tolerance or the need for dose escalation. And they will not develop any aberrant drug-related behaviors consistent with abuse, diversion, or addiction.

Q: What about the use of opioids for breakthrough pain?

A: It looks like about 60% of patients with chronic pain have flairs that can be called breakthrough pain, and in the cancer population, the use of a short-acting opioid co-administered with a long-acting drug is the standard of care.

With noncancer pain, it's a moving target. People are trying to figure out if it should be the standard of care or not. I think it should not. I think it should be a case-by-case decision.


Q: What are some of the risk factors when opioids are being considered? Do they all relate to addiction?

A: No. Suppose you have a patient with very bad lung disease who might be at risk for the respiratory effects. (Opioids can suppress breathing.) Or you have a patient who has severe gastrointestinal problems—where the constipation induced by the opioid might become very problematic. Or you have an elderly person with arthritis who has a mild dementia: In that case, the bias would be to try an NSAID because the opioid has a higher likelihood of causing cognitive impairment.

Q: Is the ultimate concern, though, addiction?

A: No, it's broader than that. It's responsible drug use, a term I use purposely because for clinicians, addiction is an uncommon problem—a very, very serious problem, but it's an uncommon problem.

Q: So there are irresponsible uses that do not involve addiction?

A: What's much more common for clinicians than addiction is what has been called aberrant drug-related behavior. Behaviors like doctor shopping or frequent visits to the ED [emergency department], or increasing the dose during pain flare-ups without permission. Or taking an opioid to help you get to sleep at night, or taking it when you're feeling anxious. Or in some cases using an illicit drug, like smoking marijuana on the weekend, without telling you.

A clinician who is trying to prescribe these drugs safely ought to be monitoring all of those behaviors and trying to work with the patient so that the behavior regarding these drugs is responsible—meaning take the drugs as prescribed.

Q: It's not as simple as saying that opioids deliver a "high," is it? What "benefits" do abusers get from the drugs?

A: There are studies that have been done that show that in the usual person—with no history, and no family history of addiction—the typical mood response produced by opioids is dysphoria, not euphoria. But in some cases, they might be driven by co-morbid psychiatric disease—they may have anxiety disorder and realize that these drugs produce some reduction in anxiety. Or they have a depressive disorder—these drugs were used in the 1950s as antidepressants before we had any real antidepressants.

Or the patient may have a co-morbid psychiatric disorder associated with impulsive drug use—they would take any centrally-acting drug, any drug that alters their consciousness, impulsively.

There are also people who have an addiction biology, and it's profound. I talked to a physician who became addicted to opioids, and he told me that the first time he took an opioid, it was like he had discovered something very magical about life. He said, "I knew this was my substance, this was something that I needed." With a single dose.


Q: What is the risk of actual addiction?

A: Most scientists who work in this area think that about 10% of the population in developed countries have the biological predisposition, the genetic predisposition, to potentially become addicted. Truly addicted. Which is a huge number, 10%.

Q: If a chronic pain patient passes your various tests and is a good candidate for an opioid, what happens then?

A: At the present time the professional community is telling doctors that they have two obligations whenever they prescribe a controlled prescription drug.
  • Number 1: To know the pharmacology so that the patient's outcomes—meaning the pain relief they get, and the side effects they experience—those outcomes are optimal.
  • Number 2: They need to do risk assessment and management to ensure that the patient takes the drugs in a responsible way, and there is minimal risk of abuse, diversion, and addiction.


Q: What does that mean for the patient's experience?

A: Every patient should undergo a comprehensive assessment and risk stratification. The doctor takes a history and then makes a decision: Is this person at high risk or at low risk of developing problematic drug-related behaviors?

The most accepted factors that put a person into a high-risk category is a personal history of substance abuse now or in the past, a family history of substance abuse now or in the past, or a history of major psychiatric disorder. And there are many, many other factors: Current smoking, history of physical or sexual abuse.

Q: Give an example of a high-risk patient.

A: A young man who injures his back at work and has pain for six months, sees a doctor, and the history reveals that the patient binge drinks on the weekend, uses marijuana three nights a week, and has a brother who has been through detox. If an opioid is being considered for that patient, then the structure of the therapy should be very defined and very rigid, it might include any or all of the following.



  • An opioid agreement that is used to educate the patient about responsibilities and consequences of bad behavior
  • A small number of pills prescribed
  • The requirement that the patient returns with the pill bottle so that a pill count can be done
  • The requirement that the patient gets urine drug screens periodically
  • A requirement that the patient gets a consultation with an addiction-medicine specialist
  • The requirement that the patient uses only one pharmacy, so that you can track what has been dispensed









Q: What about a low-risk example?

A: A patient 70 years old develops bad knee and hip pain from arthritis, and the history reveals no personal history of substance abuse, including no use of alcohol, no family history, and no known psychiatric disease—that patient has very, very low risk of developing problematic behaviors. For that patient, a structure might be to come back in a month and provide a phone call in the middle.

Q: Sounds complicated. Should chronic pain patients seek out a specialist?

A: Only about 5% of people with chronic pain ever see a specialist. This is a type of therapy that, for 20 years, people like myself had been promoting as needing to be done by primary care doctors.

Q: What advice do you give patients who are looking for possible opioid treatment?

A: I would like patients to think, "Opioids may or may not be appropriate. But I need to see a physician who's comfortable with prescribing opioids and also knows how to do it in a way that's safe and effective for me. When I go into that physician, I know that I'm going to have to be honest and let that person do a good assessment. And I'm going to have to provide my records to that person. If that means that I have to have urine drug screens, so be it. If I have to sign an opioid agreement, if it's reasonable and educational, I'll sign it. If I have to go and get treated by a psychologist at the same time and I can afford it, I'll do it."

There has to be a recognition that this is a controversial therapy that takes a lot of effort on the part of the clinician, and the patient has to not only adhere to the therapy, but also has to communicate and be willing to be monitored.


Q: Given all that, do you believe that opioids are underused in the treatment of chronic pain?

A: Absolutely. I've seen this controversy in the U.S. going back and forth for about 25 years. This is a pendulum that swings back and forth depending on how frightened people are of addiction and abuse, and depending on how much the advocacy community gets the word out about undertreatment.

There's a whole political and social context here that is not based on any known science. And in the 2000s we seem to have the pendulum shifting toward more denial that the therapy can be useful, more reluctance to prescribe, more concern about regulation.

Q: That's an unfortunate swing for those people who would benefit from these drugs.

A: No question. But I want to acknowledge what my colleagues would say, many of them—that 25 years of research has yet to show the evidence that long-term opioid use is effective for chronic pain.

There have been a large number of good clinical trials, but they've all been either short-term or in very selected populations, or didn't measure all the issues.

But the bottom line is that we have about 9,000 years of clinical experience showing that they can work. And you also have a consensus in the professional community of pain specialists—not just in the U.S., but also in Canada and England and other countries in Europe. You have a consensus that has evolved based on the data that do exist and the observations that exist.

The real issue is, let's stop arguing about should patients ever get opioids and start arguing about who should get them and how you prescribe in a way to optimize the outcomes.

Q: Of course, even when drugs work, patients don't always take them.

A: In the past 20 years, there's been all of these new modified-release formulations, so now there are once-a-day drugs, twice-a-day drugs, patches that last three days, all for the treatment of chronic pain.

So you would think that compliance would be easier because it's more convenient, and in some respects that's true. But we just did a little study here, which we haven't fully analyzed yet or published, and what we discovered in our group was this: In almost 100 patients, about 50% were non-adherent, and the vast majority of that group was undertreating.

It raises questions: Why are they undertreating? Are they afraid? Or do they have side effects? Is it money?

The bottom line is, most patients are not out there acting like [drug addicts], most patients are pushing you to give less, or not taking everything you prescribe. They're not interested in abuse, they're interested in getting off this stuff!