Prescribe opioids, but with caution

Use of opioids for chronic pain quadrupled in a 10-year span, despite scant evidence that they are effective for noncancer pain and ample evidence of harm. Learn more about how to prescribe them safely.


There's scant evidence that opioids are effective for chronic noncancer pain, just a single 38-patient cohort study from 1986 suggesting that these patients were doing well with them. Yet, said Molly A. Feely, MD, FACP, that single study was used to launch the “Pain as the 5th Vital Sign” campaign in 1995.

The campaign and the subsequent release of OxyContin brought the use of opioids for chronic pain to the forefront, Dr. Feely said. “The reality is, we have minimal data other than that 38-patient cohort study that actually suggests that opioids help chronic pain,” she noted. And there is ample evidence of harm.

Dr Feely Photo by Kevin Berne
Dr. Feely. Photo by Kevin Berne

Dr. Feely, who is an assistant professor of medicine at the Mayo Clinic in Rochester, Minn., analyzed these issues during her talk “Pain Management: Strategies for Safe Prescribing” at Internal Medicine Meeting 2015.

From 1999 to 2010, sales of opioids quadrupled, Dr. Feely reported. From 1997 to 2011, admissions to inpatient addiction rehab facilities increased 900%. By 2010, prescription opioid overdose deaths exceeded motor vehicle accident deaths as well as heroin overdose and cocaine overdose deaths combined. Eighty percent of current heroin users report that prescription opioids were their gateway drug to heroin.

The lack of data doesn't mean that opioids should never be prescribed for noncancer pain. But physicians need to recognize the caveats, Dr. Feely said. The only evidence for this use is graded poor to fair quality, and most recommendations are expert opinion.

“That doesn't mean that we should never ever use opioids for chronic noncancer pain,” she said. “We all in our practices have patients who have done really, really well with these medications, and these medications have helped improve their function.”

Fortunately, among the 12 guidelines on opioid prescribing available, there is substantial concordance. (“If you are going to stray from the guidelines, you ought to have a very good reason and you ought to have that reason very well documented,” Dr. Feely said.)

Among the first steps to properly managing prescription opioids is performing a risk assessment, she said. Start with a comprehensive medical assessment, discuss what's causing the pain, and rule out conditions such as cancer.

Next, document the medical need for opioids having tried other alternatives first, such as nonopioid pain relievers, physical therapy, behavioral changes, and complementary and alternative medicine. Then assess the risk of misuse, including a history of addiction, sexual abuse, or comorbid psychiatric disorders.

Dr. Feely strongly recommended using a formal assessment tool, which helps physicians remember to screen for every risk factor and automatically supplies proper documentation for the medical record. Many risk assessment tools are available online, and no particular tool is better than the others, she said.

Take advantage of state-by-state prescription monitoring programs to track whether patients have sought prescriptions from multiple clinicians. Many states can search by patient across neighboring states, a useful feature for physicians who practice near borders. Only Missouri does not have a prescription monitoring program, a gap that will likely be legislatively corrected by the end of the year, Dr. Feely said.

Use treatment agreements, which highlight the expectations of treatment and of patient behavior. These agreements used to be called drug contracts, but Dr. Feely urged physicians to consider them informed consent documents to remove the stigma, since informed consent needs to be performed anyway.

“How many of you would do a flex sig [flexible sigmoidoscopy] in your office without doing informed consent?” she asked. “In our office ... we have to do informed consent for trochanteric bursa injections. The death rate from flex sig is less than 1 in 100,000. The death rate from prescription opioids is 5 times that.”

The treatment agreement and informed consent process also defines expectations from treatment. “I want to see them improve function. If we get improvement in pain, that's a bonus,” she said.

If a patient can return to work, or perform functions of daily living, then opioids are considered effective. This will depend upon the patient. One rule of thumb is a 30% improvement in function, she said. If Dr. Feely is treating a 25-year-old patient with mechanical back pain, she might seek substantial improvement, such as 50%, or the ability to return to work. “If I have an 85-year-old woman with bad degenerative arthritis, I may want to see that she is getting to and from the bathroom by herself,” she said.

Once the physician has decided to prescribe opioids, adherence monitoring should be done to assess side effects, abuse or misuse, and the possibility of diverting medications.

Dr. Feely wants to see an acceptable side effect profile but doesn't expect that there will be no side effects. She advised to check for impairments to cognition, the presence of nausea, constipation, significant sedation, pulmonary compromise, and the development of hypogonadism. “Chronic opioids are one of the leading causes for hypogonadism in men,” she said.

Maintaining the ability to drive is important for many patients, Dr. Feely said. There is little evidence to guide physicians in terms of safe driving on chronic opioids. The guidelines all agree that when opioid doses are started or titrated, patients shouldn't be driving. However, there may be evidence to suggest that as long as patients don't have significant sedation, aren't in severe pain, and their dose is stable, they are probably OK to drive, she said.

Abuse or misuse of opioids needs to be documented, Dr. Feely said. In her community, police believed that 50% of prescription drugs were diverted directly to street sales. Such abuse or misuse might be detectible through the need for early refills or lost prescriptions, without any reason given for that circumstance. A single episode may not be reason to stop therapy. But if there are repeated episodes, opioids may not be the best choice for that patient, Dr. Feely said.

She uses pill counts in only a handful of her very high-risk patients or as part of a remediation plan in patients who have exhibited some form of aberrant behavior, she said.

But, she added, urine drug testing should be done in every patient. It ensures that prescription drugs that were prescribed were used and that drugs that shouldn't be taken aren't. Perform it at baseline (she reminded the audience that every opioid prescribing guide agrees on this step) and do it in every patient at least once a year, more frequently in high-risk patients or as the clinical situation warrants.

While urine drug testing seems simple, it quickly gets complex, she said. First, there are variations in the commercial screening products. Next, physicians should know what prescribed drugs metabolize into. For example, codeine metabolizes into morphine, while hydrocodone metabolizes into hydromorphone. Also, make a friend in the lab, Dr. Feely advised, so you can routinely ask that person about the latest tests to screen for abuse or misuse.

Documentation in the setting of chronic opioids is critically important, Dr. Feely said. Use the treatment agreement or informed consent as documentation, scan the risk assessment tool into the medical record, and use a template to document follow-up visits, she said.

At every follow-up visit, document the following:

  • consultation of the state drug prescription monitoring program;
  • absence of aberrant behavior,
  • functional status, and
  • management of side effects.

Patients who present at high risk for overdoses, such as those who take more than 100 mg of morphine-equivalent doses daily or those who take opioids and a sedative, warrant a second opinion from a pain specialty clinic.

“A second opinion ... does not mean that the clinic is going to take over her care,” Dr. Feely said. Instead, it can provide a second opinion that helps mitigate risk for both the physician and patient. Get a second opinion even if the specialty clinic is a day's drive away, she advised. If there are insurance problems or some other barrier, at least seek a second opinion from a practice partner.

Dr. Feely's final advice was that there is no evidence whatsoever that long-acting opioids help resolve any of the prescribing issues that she outlined. Long-acting opioids are as abused as short-acting and provide no improvement in pain control over short-acting opioids. Long-acting opioids also have an increased risk of death, she noted.

“If you're contemplating long-acting opioids in a patient, I would encourage you to think twice about your rationale and document thoroughly,” she said.