How are you feeling? A smiley-face scale that spans a grinning pain-free face (0 on the pain scale) to a crying, excruciating-pain grimace (level 10) is an overly simplistic way to measure a person's pain. Yet this and similar self-report scales are currently the best tools that researchers, doctors and other medical personnel have to determine how much discomfort a person is feeling.

Some researchers believe there is a better way. Now, the National Institutes of Health (NIH) is hoping to make that improvement a reality sooner rather than later. This summer, the NIH's National Institute on Drug Abuse (NIDA) announced funding opportunities for the development of a device to objectively measure pain, through the Small Business Innovation Research Grant and Small Business Technology Transfer Grant programs.

The first phase of the grants will provide up to about $200,000 for research to develop and test various prototypes, says Dave Thomas, PhD, a program officer at NIDA and a member of the NIH pain consortium. If that preliminary work proves promising, additional funding will be available for more in-depth research. A "pain-o-meter" could improve the way scientists study pain and the way health-care providers treat it, Thomas says.

"Chemists don't stick their fingers into their test tubes to see how warm the solution is. You can't do good science without good measures of what you're trying to study."

How soon such a device could make its way to the clinic is debatable. Yet scientists are closing in on objective measures of pain from several angles, including brain activity patterns, immune biomarkers in the blood and patterns of facial movements.

A nation in pain

According to the National Health Interview Survey in 2012, 17.6 percent of the U.S. adult population experienced severe levels of pain in the preceding three months. And more than 11 percent reported having daily pain in that period. Chronic pain is the most common cause of long-term disability in the United States.

Pain is a costly problem, too. According to an analysis by Johns Hopkins University health economists Darrell J. Gaskin, PhD, and Patrick Richard, PhD, the national cost of pain ranges from $560 billion to $635 billion annually, including both health-care costs and lost productivity (Journal of Pain, Vol. 13, No. 8, 2012). To treat that suffering, many patients have turned to prescription opioids, which carry a risk of dependency and overdose. According to the Centers for Disease Control and Prevention, sales of prescription opioids quadrupled from 1999 to 2010— and deaths from overdoses of these medications quadrupled.

Even for physicians who specialize in pain, it often takes trial and error to find the right treatment or combination of treatments to help a given patient, says Sean Mackey, MD, PhD, chief of the division of pain medicine at Stanford University and past president of the American Academy of Pain Medicine. "We're frustrated, our patients are frustrated and we spend huge amounts of healthcare dollars until we find things that work. If we can shortcut that approach, it will transform health care."

Identifying the presence or absence of pain could be useful in some cases, such as in infants, the very infirm or people who otherwise aren't able to communicate their discomfort, Mackey says. But ultimately, pain researchers agree that the broader goal is to develop a system that can identify the underlying cause of a person's pain—and suggest the treatments that are best suited to treat it.

Pain is a complex phenomenon, caused by several mechanisms. Imagine that your foot hurts. It might be due to nociceptive pain, which arises directly from damaged tissue such as a broken bone or a joint inflamed by arthritis. Or it could be neuropathic pain, which occurs when the nerve fibers themselves become damaged or dysfunctional. Neuropathy is common in the feet and legs of people with advanced diabetes, for instance, and can cause numbness, burning, sharp pains and increased sensitivity to touch. Then again, the tissues and nerves of your feet might be perfectly healthy. Instead, the discomfort might be due to central pain syndrome, a neurological condition in which the pain arises from dysfunction in the brain and spinal cord. The syndrome can be caused by problems such as stroke, epilepsy, trauma or Parkinson's disease.

"Too often, we define pain by what hurts: your knee, your back. But the system causing the pain doesn't always relate to a specific part of the body," Thomas says. "If we could understand the origin and mechanisms of the pain, it could guide treatment."

Not a 'pain lie detector'

Researchers are exploring a variety of mechanisms that could be useful for diagnosing pain. Some are using neuroimaging tools to look for telltale patterns of brain activity. Mackey and colleagues used fMRI and machine learning to determine a pattern of brain activity associated with the pain caused by applying heat to the arms of eight healthy volunteers. Analyzing brain activity patterns from eight new participants, the computer was able to predict the presence of thermal pain 81 percent of the time (PLOS ONE, Vol. 6, No. 9, 2011).

In a similar study with 114 participants, Tor Wager, PhD, director of the Cognitive and Affective Neuroscience Laboratory at the University of Colorado, Boulder, and colleagues used fMRI to discriminate nonpainful warmth from painful heat more than 90 percent of the time (New England Journal of Medicine, Vol. 368, No. 15, 2013).

Measuring this type of evoked pain in healthy volunteers in a laboratory is very different from detecting patterns of chronic pain in people suffering in the real world, Wager cautions. Still, there are hints the research could move in that direction. Recently, he and postdoctoral researcher Marina Lopez-Sola, PhD, found a signature of brain activity that corresponds to fibromyalgia, the disorder marked by musculoskeletal pain. That signature could identify fibromyalgia sufferers with 93 percent accuracy, they showed (Pain, Vol. 158, No. 1, 2017).

Others are exploring the connection between the immune system and chronic pain. Nerve cells and immune cells have been shown to signal back and forth, and there's evidence that immune activity can contribute to pathological pain, as Mark Hutchinson, PhD, a neuroscientist at the University of Adelaide, and colleagues have described (Nature Reviews Immunology, Vol. 14, No. 4, 2014).

Hutchinson and his colleagues are developing a system that uses immune biomarkers to diagnose pain states. Using a rat model, they were able to identify pain-related patterns of immune cells in the blood. Applying that model to humans, they were able to predict the presence of pain in 28 of 34 people with chronic pain (PLOS ONE, Vol. 8, No. 10, 2013).

The process takes up to 48 hours to get results, he says. But his team is attempting to create a system that can accurately diagnose pain in minutes. He envisions a scenario in which physicians could take a quick pain reading to determine how well patients are responding to their current treatment. In theory, such a test could also identify people whose immune biology puts them at increased risk of developing chronic pain. "Imagine at your yearly checkup, you have your pain priming test done. If you test positive you know that if you start having pain, you should treat it quickly so you don't end up with persistent pain," he says.

Still other researchers are exploring pain-detecting systems based on mechanisms such as pupil response or facial movements. Dianbo Liu, a doctoral student at the Massachusetts Institute of Technology, and colleagues created an algorithm that analyzes people's winces and grimaces to determine how much pain they're in (Journal of Machine Learning Research, Vol. 66, No. 1, 2017).

Moving forward

Most pain researchers agree that pain-detection devices will likely incorporate information from multiple sources, such as a brain scan along with a blood test or facial analysis. Tools that would allow physicians to pinpoint specific pain mechanisms and zero in on appropriate treatments remain further afield. But NIDA's Thomas and researchers such as Mackey suggest that's where the field is headed. "Can you use a pattern of brain activity to determine whether somebody will respond to a specific treatment, whether a drug or a procedure or cognitive-behavioral therapy? Can we predict whether someone will develop chronic pain, or opioid dependence? That's moving the dial towards precision medicine," Mackey says.

Researchers are optimistic about the promise of such a system, but they also have reservations. Imagine, for instance, that insurance companies deny people pain treatment because their brain scans don't confirm their self-reports. There could also be legal implications for people who claim that an accident caused pain and suffering.

"People might have real pain but may not have a typical brain or show typical blood biomarkers," Wager says. "I believe we need research to corroborate and understand the source and mechanism of pain, but we should never use it as a pain lie detector."

Mackey urges his colleagues to consider the policy and ethical implications before the science jumps ahead. "While I see the potential for good, there are all sorts of ways this technology could go awry," Mackey says. "We desperately need objective biomarkers of pain, but we need to tread carefully and thoughtfully in moving forward."

Additional reading

Brain Imaging Tests for Chronic Pain: Medical, Legal and Ethical Issues and Recommendations
Davis, K.D., et al. Nature Reviews Neurology, 2017

Imaging Pain
Martucci, K.T., & Mackey, S.C. Anesthesiology Clinics, 2016

Multivariate Classification of Structural MRI Data Detects Chronic Low Back Pain
Ung, H., Brown, J.E., Johnson, K.A., Younger, J., Hush, J., & Mackey, S. Cerebral Cortex, 2014