America’s opioid epidemic has reached crisis proportions, enough so that last week the United States Surgeon General took the unprecedented step of sending a letter to 2.3 million American health care professionals asking them to take a pledge to “turn the tide” on the opioid crisis.

Relieving pain is an essential element of good care, and we are appropriately reminded that the use of these powerful medications requires precision, caution, and perspective.

But something very important is missing from this prescription: a recognition of the needs and health challenges of older adults.

As I discussed in a recent Huffington Post blog, older adults aren’t immune from the scourge of addiction. However, they also contend with frustratingly high rates of undertreated pain. As many as half of all older adults suffer from chronic pain caused by degenerative arthritis, nerve damage (from fibromyalgia, shingles, and diabetes), and cancer, and many more require pain relief after injury or surgery.

For additional expert perspective on older patients and pain, I turned to my colleague, Cary Reid, MD, PhD , a longtime grantee of The John A. Hartford Foundation, Associate Professor and Director of the Office of Geriatric Research in the Division of Geriatrics and Palliative Medicine at Weill Cornell Medical College, and director of Cornell’s Translational Research Institute on Pain in Later Life. The following questions and answers have been edited and condensed.

Terry Fulmer (TF): In his letter, Surgeon General Vivek Murthy suggested that clinicians refer to the CDC Guidelines for Prescribing Opioids for Chronic Pain. What do these guidelines offer for the treatment of older adults, many of whom live with chronic pain?

Cary Reid (CR): The CDC guidelines are pretty explicit that, in prescribing, we should pay attention both to whether pain is reduced by a clinically meaningful amount, and also to whether there is an improvement in function. If not, you get rid of the medicine. On paper, that is reasonable. But for older populations, it can be problematic.

For example, I see people in their 90s, some of whom are homebound from arthritis, back pain, or other conditions. The likelihood of them experiencing meaningful improvement in function is pretty low, yet they may have a life expectancy of several years.

The CDC guidelines also say it’s OK to give opioids to people receiving palliative or end of life care, but they are not preferred therapy for individuals with chronic noncancer pain. The reasoning here is that the risks associated with long-term opioid use is substantial (e.g., falls, hospitalizations, risk of addiction), while the benefits of this type of care remain uncertain. But we need to recognize that the line between palliative care and chronic disease care can be blurry in older patients.

That is an aging issue: recognizing that some proportion of the pain that geriatricians treat is better thought of in a palliative care context. The goal is to mitigate pain so patients’ quality of life is improved.

TF: How great is the risk of opioid addiction among older adults?

CR: Generally, age is a protective factor. The older you are, the less likely you are to end up in an emergency department with an overdose. It is true that people are speculating that aging Boomers, who felt comfortable experimenting with drugs earlier in life, may be different, i.e., more willing to experiment with this class of medications, than the current cohort of older adults.

TF: I suspect that might well be the case with Boomers and it’s something we’ll need to watch. On the other side, are you concerned that the tightening of the opioid supply will result in even more undertreated pain for older adults?

CR: Yes. As we add more policies like drug monitoring, clinicians have to go onto online databases to review whether their patients are receiving medications from other providers. This task can be very time-consuming and operate as a barrier to physicians prescribing this class of medication. I think the policies and procedures we are putting in place are going to increase the well-established and sizable number of older people who do not have pain adequately treated.

TF: From your perspective as a geriatric researcher, what further investigation is most needed now?

CR: I would say – as a researcher – that researchers and the funders who make research funding decisions have failed, in this way: we have not developed an evidence base to understand the long-term risks and benefits associated with the use of opioid medications as well as other classes of pain medications. We’re talking about people who may live with pain for decades, and it’s critically important that we know if these medicines are, on average, providing more benefit than harm to this population.

TF: What is your view of the advice to use more non-opioid alternative medications for older adults?

CR: In terms of non-opioid medications, it turns out that what we have available is quite limited. Tylenol is considered the safest medication but also the weakest, and not terribly helpful with persistent pain. For other classes of medications, such as antidepressants or anticonvulsants, many of the long-term consequences of these medications are unknown. Their long-term efficacy is also not known.

We do know something about the safety issues of NSAIDs, like Ibuprofen, and they are quite profound. The American Geriatrics Society released guidelines in 2009, and the British Pain Society and British Geriatrics Society followed in 2013, calling for limited use of nonsteroidals for noncancer pain, because of well-known side effects in older people. These include cardiovascular risk, gastrointestinal complications, and potential damage to the kidneys. Yet these are, in part, what the CDC is promoting.

We put clinicians in a bind. We have not given them a strong evidence base to guide their decision, and instead have said, here are two guidelines, which are in conflict.

TF: As the New York Times pointed out in a recent editorial, our health care and insurance systems often make it easier to write a prescription for an opioid than to recommend evidence-based services such as physical therapy, acupuncture, or cognitive behavioral therapy. As a clinician, this has always frustrated me, but what’s your perspective on non-pharmacologic alternatives?

CR: Non-pharmacologic therapies for pain are, in principle, a good thing, but we do not have adequate ability to deliver them in our health system. We don’t have the workforce to deliver evidence-based therapies like chronic disease self-management or yoga. Medicare does not pay for long-term treatment using these therapies. So it’s fine to encourage people to employ non-drug approaches and it is certainly safer for a geriatric population. But the downside is what I’m hearing all over the country: “we don’t have access”.

TF: This access issue is something we at The John A. Hartford Foundation are working on through initiatives that are helping community-based organizations partner with health care systems. But your point is well taken and more certainly needs to be done.