Artificial Intelligence May Decide Whether You Get Rx Opioids

By Andy Miller and Sam Whitehead, KFF Health News

Elizabeth Amirault never heard of NarxCare until last year, when she learned its software was tracking her medication use. During a visit to a hospital in Fort Wayne, Indiana, Amirault told a nurse practitioner she was in severe pain and received a puzzling response.

“Your Narx Score is so high, I can’t give you any narcotics,” she recalled the man saying, as she waited for an MRI before a hip replacement.

Tools like NarxCare are used to help medical providers review controlled substance prescriptions. They influence, and can limit, the prescribing of painkillers, similar to a credit score influencing the terms of a loan. NarxCare’s overdose risk ratings – known Narx Scores -- are produced by health care technology company Bamboo Health (formerly Appriss Health).

NarxCare’s software uses artificial intelligence to analyze data about prescriptions for controlled substances to identify patterns of potential problems involving patients and physicians. State and federal health agencies, law enforcement officials, and health care providers have enlisted these tools, but the mechanics behind their algorithm formulas are generally not shared with the public.

Artificial intelligence is working its way into more parts of American life. As AI spreads within the health care landscape, it brings familiar concerns of bias and accuracy and whether government regulation can keep up with rapidly advancing technology.

The use of systems to analyze opioid-prescribing data has sparked questions over whether they have undergone enough independent testing outside of the companies that developed them, making it hard to know how they work. Lacking the ability to see inside these systems leaves only clues to their potential impact.

Some patients say they have been cut off from needed care. Some doctors say their ability to practice medicine has been unfairly threatened.

Researchers warn that such technology — despite its benefits — can have unforeseen consequences if it improperly flags patients or doctors.

“We’re concerned that it’s not working as intended, and it’s harming patients.”

— Jason Gibbons, Health Economist

“We need to see what’s going on to make sure we’re not doing more harm than good,” said Jason Gibbons, a health economist at the Colorado School of Public Health at the University of Colorado’s Anschutz Medical Campus. “We’re concerned that it’s not working as intended, and it’s harming patients.”

Amirault, 34, said she has dealt for years with chronic pain from health conditions such as sciatica, degenerative disc disease, and avascular necrosis, which results from restricted blood supply to the bones.

The opioid Percocet offers her some relief. Amirault had been denied the medication before, but never had been told anything about a Narx Score, she said.

In a chronic pain support group on Facebook, she found others posting about NarxCare, which scores patients based on their supposed risk of prescription drug misuse. She’s convinced her ratings negatively influenced her care.

“Apparently being sick and having a bunch of surgeries and different doctors, all of that goes against me,” Amirault said.

Database-driven tracking has been linked to a decline in opioid prescriptions, but evidence is mixed on its impact on curbing the epidemic. Overdose deaths continue to plague the country, and patients like Amirault have said the monitoring systems leave them feeling stigmatized as well as cut off from pain relief.

The Centers for Disease Control and Prevention estimated that in 2021 about 52 million American adults suffered from chronic pain, and about 17 million people lived with pain so severe it limited their daily activities. To manage the pain, many use prescription opioids, which are tracked in nearly every state through electronic databases known as prescription drug monitoring programs (PDMPs).

The last state to adopt a program, Missouri, is still getting it up and running.

More than 40 states and territories use the technology from Bamboo Health to run PDMPs. That data can be fed into NarxCare, a separate suite of tools to help medical professionals make decisions. Hundreds of health care facilities and five of the top six major pharmacy retailers also use NarxCare, the company said.

The platform generates three Narx Scores based on a patient’s prescription activity involving narcotics, sedatives, and stimulants. A peer-reviewed study showed the “Narx Score metric could serve as a useful initial universal prescription opioid-risk screener.”

NarxCare’s algorithm-generated “Overdose Risk Score” draws on a patient’s medication information from PDMPs — such as the number of doctors writing prescriptions, the number of pharmacies used, and drug dosage — to help medical providers assess a patient’s risk of opioid overdose.

Bamboo Health did not share the specific formula behind the algorithm or address questions about the accuracy of its Overdose Risk Score but said it continues to review and validate the algorithm behind it, based on current overdose trends.

Guidance from the CDC advised clinicians to consult PDMP data before prescribing pain medications. But the agency warned that “special attention should be paid to ensure that PDMP information is not used in a way that is harmful to patients.”

This prescription-drug data has led patients to be dismissed from clinician practices, the CDC said, which could leave patients at risk of being untreated or undertreated for pain. The agency further warned that risk scores may be generated by “proprietary algorithms that are not publicly available” and could lead to biased results.

(Editor’s note: A citizen’s petition filed with FDA earlier this year alleged that NarxCare software “altered the practice of medicine in the U.S. to the detriment of patients,” and sought to have the software declared a misbranded medical device and recalled. The petition by the Center for U.S. Policy was rejected by FDA on technical grounds because it was “not within the scope” of the agency’s petition process.)

Impact on Patients

Bamboo Health says NarxCare’s rating system should never replace decisions made by physicians. But some patients say the risk scores have had an outsize impact on their treatment.

Bev Schechtman, 47, who lives in North Carolina, said she has occasionally used opioids to manage pain flare-ups from Crohn’s disease. As vice president of the Doctor Patient Forum, a chronic pain patient advocacy group, she said she has heard from others reporting medication access problems, many of which she worries are caused by red flags from databases.

“There’s a lot of patients cut off without medication,” according to Schechtman, who said some have turned to illicit sources when they can’t get their prescriptions. “Some patients say to us, ‘It’s either suicide or the streets.’”

The stakes are high for pain patients. Research shows rapid dose changes can increase the risk of withdrawal, depression, anxiety, and even suicide.

Some doctors who treat chronic pain patients say they, too, have been flagged by data systems and then lost their license to practice and were prosecuted.

Lesly Pompy, a pain medicine and addiction specialist in Monroe, Michigan, believes such systems were involved in a legal case against him.

His medical office was raided by a mix of local and federal law enforcement agencies in 2016 because of his patterns in prescribing pain medicine. A year after the raid, Pompy’s medical license was suspended. In 2018, he was indicted on charges of illegally distributing opioid pain medication and health care fraud.

“I knew I was taking care of patients in good faith,” he said. A federal jury in January acquitted him of all charges. He said he’s working to have his license restored.

One firm, Qlarant, a Maryland-based technology company, said it has developed algorithms “to identify questionable behavior patterns and interactions for controlled substances, and for opioids in particular,” involving medical providers.

The company, in an online brochure, said its “extensive government work” includes partnerships with state and federal enforcement entities such as the Department of Health and Human Services’ Office of Inspector General, the FBI, and the Drug Enforcement Administration.

In a promotional video, the company said its algorithms can “analyze a wide variety of data sources,” including court records, insurance claims, drug monitoring data, property records, and incarceration data to flag providers.

William Mapp, the company’s chief technology officer, stressed the final decision about what to do with that information is left up to people — not the algorithms.

Mapp said that “Qlarant’s algorithms are considered proprietary and our intellectual property” and that they have not been independently peer-reviewed.

“We do know that there’s going to be some percentage of error, and we try to let our customers know,” Mapp said. “It sucks when we get it wrong. But we’re constantly trying to get to that point where there are fewer things that are wrong.”

Prosecutions against doctors through the use of prescribing data have attracted the attention of the American Medical Association.

“These unknown and unreviewed algorithms have resulted in physicians having their prescribing privileges immediately suspended without due process or review by a state licensing board — often harming patients in pain because of delays and denials of care,” said Bobby Mukkamala, chair of the AMA’s Substance Use and Pain Care Task Force.

Even critics of drug-tracking systems and algorithms say there is a place for data and artificial intelligence systems in reducing the harms of the opioid crisis.

“It’s just a matter of making sure that the technology is working as intended,” said health economist Gibbons.

KFF Health News is a national newsroom that produces in-depth journalism about health issues.

The Tangled Mess of Prescription Opioid Guidelines

By Roger Chriss, PNN Columnist

The opioid overdose crisis has impacted medical practice in unanticipated and unfortunate ways. A recent JAMA study warned that efforts to reduce opioid prescribing through tapering raises the risk of overdose and mental health crises in pain patients on stable, long-term opioid therapy.

This study is the latest to find that opioid tapering is fraught with risks. Amid this, the American Medical Association has issued a call to revamp the CDC’s problematic 2016 opioid prescribing guideline because of its “devastating” impact on pain patients.

“The CDC should remove arbitrary thresholds, restore balance and support comprehensive, compassionate care as it revises the guideline,” wrote AMA news editor Kevin O’Reilly.

But revising the CDC guideline may not have much effect. The guideline is voluntary and doesn’t have the force of law, but many states have implemented their own guidelines in ways that make them enforceable. They are often paired with requirements and regulations covering everything from daily dose and prescription duration to drug testing, pain management agreements, and tapering. These state guidelines do not necessarily follow the CDC guideline on even basic issues of dose, duration or recommended use.

Some states, including Minnesota and Oregon, have adopted the CDC’s recommended threshold of 90 morphine milligram equivalents (MME) as a maximum daily dose not requiring consultation with a pain management specialist or a special exemption. Other states make their own rules. Washington has kept to 120 MME in its latest guideline update, as has Tennessee.

State policies also differ on the merits of using opioids for chronic pain. The Medical Board of California recommends that physicians and patients “develop treatment goals together” for long-term use of opioids, while Arizona’s opioid guideline flatly warns physicians: “Do not initiate long-term opioid therapy for most patients with chronic pain.”

On tapering, states do not agree much at all and generally do not follow federal HHS guidelines that tapering be individualized and “slow enough to minimize opioid withdrawal symptoms.”

Minnesota’s opioid guideline recommends that physicians “routinely discuss tapering with patients at every face to face visit” and allows for forced, rapid tapers or discontinuation under some circumstances.

Tennessee’s guideline notes that there are “many reasons to discontinue chronic opiate therapy” and “several different weaning protocols outlined by various sources.” It does not recommend any specific one, leaving it up to individual doctors to decide how to taper their patients.

The VA and Department of Defense have their own guideline, which contains a complex set of treatment algorithms that span several pages and effectively exclude almost all patients from long-term opioid use. Further, according to a separate algorithm, the VA is clearly aiming to taper or discontinue opioids in as many patients as possible. The guideline states "If prescribing opioid therapy for patients with chronic pain, we recommend a short duration.”

The Trouble With Algorithms

Many of the state guidelines are paired with a prescription drug monitoring programs (PDMPs) and use NarxCare, a private analytics system that gives individual risk scores to every patient based on their medical and prescription drug history. PNN first covered NarxCare in 2018, noting that patients can be automatically “red flagged” by the system for seeing too many doctors or using multiple pharmacies.

Maia Szalavitz recently wrote about Narxcare in Wired, noting that legitimate patients were being denied medications or abandoned by doctors because of their Narxcare scores.

“A growing number of researchers believe that NarxCare and other screening tools like it are profoundly flawed,” Szalavitz wrote. “None of the algorithms that are widely used to guide physicians’ clinical decisions — including NarxCare — have been validated as safe and effective by peer-reviewed research.”

A similar problem exists for data from PDMPs. A well-documented analysis by Terri Lewis, PhD, found that “machine learning” algorithms are often based on untested assumptions and financial incentives for providers, not on patient care.

“The worst part of machine learning (ML) snake-oil isn’t that it’s useless or harmful — it’s that ML-based statistical conclusions have the veneer of mathematics, the empirical facewash that makes otherwise suspect conclusions seem neutral, factual and scientific,” wrote Lewis. “What the PDMP is NOT designed to do, is detect patients who are using their opioids correctly from patients who are misusing their medications.”

All of the above imposes a significant risk and burden on patients, in particular if they relocate for work or school, or seek medical care outside of their state of residence.

In essence, patients are subjected to a set of federal recommendations from the CDC that may inform some state laws or regulations that are then implemented in a privatized process with little transparency or accountability. Patients simply cannot tell what is happening at the time of implementation, and if they see a problem after the fact, it is usually too late to fix it.

The AMA’s current effort to improve the CDC guideline for opioids is a laudable step forward. But the mess is far larger and more complex, and the role of the CDC is smaller than is generally appreciated amid an abundance of contradictory guidelines and regulations.

Roger Chriss lives with Ehlers Danlos syndrome and is a proud member of the Ehlers-Danlos Society. Roger is a technical consultant in Washington state, where he specializes in mathematics and research. 

Is Walmart Blacklisting Doctors?

By Pat Anson, PNN Editor

Carolyn Eastin has been a nurse practitioner in Arizona for nearly a decade. She is the only licensed prescriber at a busy pain clinic in Scottsdale, where many patients are on high doses of opioid medication.

Eastin says she was “mortified” by a letter last month from Walmart’s corporate headquarters that said the retail giant’s Walmart and Sam’s Club pharmacies would no longer fill her prescriptions for opioids.

“In reviewing your controlled substance prescribing patterns and other factors, we have determined that we will no longer be able to continue filling your controlled substance prescriptions,” the letter states. “We regret any inconvenience this may cause you or your patients.”

“It was very humiliating. I was upset about it,” says Eastin. “We’ve already had patients who can’t get prescriptions there.”

The unsigned Walmart letter has all the appearances of a form letter. It makes no mention of any complaints against Eastin, whether any of her patients have overdosed, or if her prescribing is medically inappropriate.

Eastin writes prescriptions for nearly 500 chronic pain patients at the Absolute Rehabilitation and Pain Medicine clinic. Most are on high doses of opioids, at levels well above those recommended by the CDC in its 2016 opioid guideline. Eastin believes she was red flagged by Walmart after a prescription database search and is now on a “blacklist” of prescribers.

“I’m the only one who writes prescriptions here. So I would have higher numbers than the average nurse practitioner,” she told PNN. “I’m not doing anything wrong. I’m tapering my patients. I’m trying to do my part to bring people into compliance.”

Does Walmart blacklist doctors who prescribe high doses of opioids? Is Walmart practicing medicine without a license? We posed those questions to Walmart and received a one sentence reply.

“Walmart does not comment on our patients or their prescribers,” wrote Erin Hulliberger, Walmart Corporate Communications in an email.

Corresponding Responsibility

What Walmart is doing isn’t illegal. Under federal law, pharmacists have a “corresponding responsibility” when filling medications -- a legal right to refuse to fill prescriptions they consider unusual or improper. Most pharmacists will call the prescribing doctor to double-check before turning away a patient, but in some cases there’s a “hard edit” – pharmacy jargon for when a company database tells a pharmacist not fill a prescription.

“It may be appropriate for a pharmacy to not fill a prescription because of a dangerous dose or even an inappropriate drug,” says Lynn Webster, MD, a pain management expert and past president of the American Academy of Pain Medicine.

“However, if they are refusing to fill a prescription because the doctor has met some arbitrary dosage threshold by the company without understanding the circumstances, it would be inappropriate and potentially harmful to the patient.”   

A former Walmart pharmacist told PNN the company closely monitors prescriptions and doctors at every store.

“They had assembled prescription numbers for every doctor who had filled prescriptions at my store. They knew the exact number of medications ordered and sold down to the tablet. They knew what drugs the doctors wrote for and what percentage of the total each drug they wrote for," said the pharmacist, who asked to remain anonymous.

Walmart recently began using Narxcare, a private healthcare database developed by Appriss Health, which analyzes the prescriptions and medical claims of millions of Americans. Every patient evaluated by NarxCare is assigned a “risk score” based on the past two years of their prescription drug use, medical claims, electronic health records and even their criminal history.

Is Walmart using Narxcare data to blacklist doctors? A spokesman for Appriss denies it.

“No, we do not provide provider risk scores to Walmart, or anyone else,” said David Griffin, Vice President of market and communication for Appriss Health.

Prescription Data Mining

The data mining of opioid prescriptions has become increasingly common in healthcare and law enforcement. Doctors, pharmacies, insurers, and federal and state regulators are all using prescription drug databases to look for signs of opioid “overprescribing.”

In addition to Walmart, the Absolute clinic has drawn the attention of the Arizona Medical Board. Last year the clinic’s owner, Dr. Steve Fanto, signed an agreement with the medical board to stop practicing medicine after he was accused of overprescribing Subsys, a potent fentanyl spray.  

Although Fanto is no longer involved in the day-to-day management of the Absolute clinic, it remains under scrutiny. Last December, the clinic was raided by DEA agents, although no charges were ever filed as a result of that investigation. The clinic has also received over a hundred letters from insurer United Healthcare warning about the high dose prescriptions Carolyn Eastin is writing.

“It’s a bullying technique,” says Jessica Webb, the office manager at Absolute. “We’re afraid if we don’t taper down, the insurance companies will file a complaint on Carolyn. If Carolyn’s license is revoked, we’ll have to shut down. She’s the only reason we’re still open.”

Caught in the middle of this tug of war are Absolute’s patients – who could be faced with the difficult task of finding new doctors if the clinic closes. Their painful medical conditions -- and the human suffering that comes with them – appear to be non-factors in Walmart’s decision not to fill Eastin’s prescriptions.

“I can’t tell you the number of patients that made the comment to me, ‘I might as well die. I don’t have any other options but to take my own life because I can’t live without this medication,’” Eastin said. “It’s very difficult to look these people in the eye every single day and say, ‘I want to help you but I’m going to take more away from you.’”

‘I Hope It Doesn’t Harm Patients’

Pain management experts say its unethical to use data mining alone to judge whether a prescription should be filled.

“It appears that Walmart is trying to reduce their risk of being charged with criminal conduct and paying more fines. It is hard to know. But whatever their reason I hope it doesn't harm patients,” said Webster.  

“Without further explanation about the basis for their decision, and without information about any previous interventions attempted by Walmart, this seems remarkably punitive and capricious,” said Bob Twillman, PhD, Executive Director of the Academy of Integrative Pain Management.

“I would hope that companies that review prescriber records would attempt a series of less drastic interventions with the prescriber before reaching this point. There may be issues of concern, but the appropriate step is to discuss those with the prescriber and try to reach an understanding about what is going on, and about what needs to happen to ensure that everyone is prescribing and dispensing safely. I’d be interested to hear from Walmart about whether they have such a graduated policy and if not, why not.”

Carolyn Eastin says there’s only so much she can do. Abruptly tapering patients or cutting them off from opioids could cause more harm than good.

“We’re trying to follow the CDC guidelines. We’re trying to comply with the insurance letters asking for tapers,” she says.  “I’m doing my part to taper my patients. But these things have to be done delicately. We just can’t go, ‘Okay, give me all your meds.’”  

“We’ve pushed them to get physical therapy and cognitive behavioral therapy, and patients are having good results with that,” says Jessica Webb. “But we have some patients, if I had to take their meds away, I couldn’t go to sleep at night. They’re that sick.”