Database tech takes aim at the opioid crisis
Prescription drug monitoring programs try to aid doctors, but the jury’s out on their effectiveness.
As the healthcare industry continues to look for solutions to the opioid epidemic, the tools providers are increasingly leveraging are state-run prescription drug monitoring programs—electronic databases meant to track controlled substance prescriptions by flagging suspicious patient prescribing activities.
The good news, according to proponents, is that physician use of PDMPs is dramatically growing while doctors have decreased opioid prescribing nationwide.
The bad news, also according to supporters of the technology, is that these databases need to be better integrated into clinical workflows and to provide more comprehensive, up-to-date data at the point of care.
The American Medical Association’s Opioid Task Force 2018 Progress Report notes that the number of opioid prescriptions decreased by more than 55 million—a 22.2 percent decrease nationally—between 2013 and 2017, with a 9 percent decrease—more than 19 million fewer prescriptions—between 2016 and 2017. In fact, all 50 states have seen a decrease in opioid prescriptions over the past five years, reports the AMA.
Last year, clinicians across the country queried PDMPs more than 300.4 million times—a 121 percent increase from 2016 and a 389 percent increase from 2014, according to AMA. Further, both states with and without mandates to use PDMPs saw large increases, reported the physician group.
“Today, more than 1.5 million physicians and other healthcare professionals are registered in state-based
PDMPs—between 2016 and 2017, more than 241,000 individuals registered,” according to the AMA. “As PDMPs improve, America’s physicians and healthcare professionals are using state PDMPs more than ever.”
For its part, the Centers for Medicare and Medicaid Services has been encouraging states to integrate PDMP data into electronic health record systems to minimize provider burden and improve interstate health information exchange.
“This integration removes the requirement for providers to log in to a separate system, manage a separate log in and disrupt their workflow to query the PDMP,” wrote Tim Hill, acting director for the Center for Medicaid and CHIP Services, in a letter this summer to state Medicaid directors. “Single sign-on interoperability between EHR and PDMP—such that PDMP results are displayed when the EHR indicates a controlled substance is prescribed—could be supported, as an example.”
In addition, the CMS guidance recommends that states consider integrating PDMPs into health information exchanges, “where further integration with pharmacy data, shared care plans, drug utilization review (DUR) programs, Emergency Medical Services (EMS) data, Medication Assisted Therapy (MAT) data, advanced directives and other EHR data might assist clinical decision making.”
Need for alert system
However, according to Joel White, executive director of Health IT Now’s Opioid Safety Alliance, what’s missing from PDMPs is an alert system that would offer a much needed capability not currently provided by these databases.
“Today, clinicians rely on PDMPs to flag fraudulent opioid transactions and, while these systems hold great promise, significant blind spots remain,” White says. “Too often, PDMPs are not updated in real time, do not include prescriptions filled across state lines and do not include fill attempts—leaving clinicians with only a partial view of a patient’s true opioid history.”
The Opioid Safety Alliance has been calling for a nationwide Prescription Safety Alert System, based on a model developed by the National Council for Prescription Drug Programs (NCPDP) that leverages existing ANSI-accredited standards widely adopted by the industry.
Toward that end, bipartisan congressional legislation has been introduced to create a nationwide Prescription Safety Alert System to enable pharmacists to better protect patients from opioid overuse.
The Analyzing and Leveraging Existing Rx Transactions (ALERT) Act, introduced by Reps. Tom MacArthur (R-N.J.), Ann Kuster (D-N.H.) and Barbara Comstock (R-Va.), would require the Department of Health and Human Services to work with the private sector to establish a system that analyzes the transaction data pharmacists and payers—such as health insurers and Medicare—generate when prescriptions are filled.
“We absolutely have to get smarter about how we use technology and data analysis to fight this crisis,” says MacArthur, who is co-chair of the Bipartisan Heroin Task Force. “By giving pharmacists, insurance companies, and programs like Medicare a new tool to understand the data they already have, we can help prevent further harm.”
According to MacArthur, the data analysis would provide real-time feedback to pharmacists at the point of sale and would be included in their normal workflow.
“A pharmacist will receive an alert that someone might be at risk of overuse based on their prescription history, or might be doctor-shopping to feed their addiction,” adds MacArthur. “Instead of filling that unnecessary prescription, pharmacists will have an extra tool to detect and prevent these dangers.”
The ALERT Act is endorsed by Health IT Now, which launched the Opioid Safety Alliance in January—along with IBM, Intermountain Healthcare, McKesson, Oracle and Walgreens—to advance a health IT-centric policy agenda to combat the abuse of opioids.
“The Prescription Safety Alert System delineated in the ALERT Act will arm clinicians with the proper tools to thwart opioid misuse in real-time and prevent undue delays in access for those with a legitimate medical need,” says White, who sees the alert system as complementing PDMPs by providing in-workflow clinical data at the point of dispensing, using already existing transaction information—including capturing transactions that may occur across state lines and unsuccessful fill attempts.
“It is gratifying to have NCPDP’s model supported by the HITN Opioid Safety Alliance and Representatives MacArthur, Kuster and Comstock,” says Lee Ann Stember, NCPDP’s president and CEO. “Our members representing diverse stakeholder perspectives developed the model to provide a sustainable solution that conforms to provider workflows and can complement existing PDMPs to prevent diversion, ensure appropriate access to medications for patients with a valid medical need, and protect patients.”
Benefits of PDMPs questioned
However, Leo Beletsky, associate professor of law and health sciences at
Northeastern University, is not convinced of the overall benefit of PDMPs. He believes that the recent scaleup in the number, scope, funding and legal mandates of these state-run databases has led to unintended harms that have not received sufficient attention.
“By collecting information on who is prescribing, dispensing, and receiving scheduled drugs, PDMPs are intended to detect—and deter—problem patients, rogue prescribers, and pharmacists who may be diverting potentially addictive and otherwise risky drugs,” wrote Beletsky in the Indiana Health Law Review. “The success of PDMPs has been measured primarily by their impact on suppressing medication supply with little regard for truly meaningful metrics. But, when it comes to improving patient care and addressing drug-related harms, the evidence of PDMP benefit is far from clear.”
In fact, Beletsky’s 2018 narrative review of the existing empirical evidence on PDMPs indicates that these programs may be doing more harm than good. He reviewed 34 peer-reviewed studies evaluating PDMPs, of which only 11 (32 percent) considered any overdose outcomes. In addition, of studies assessing overall mortality, three found PDMP deployment to be associated with reduced overdose rates, four reported a null result, and three reported PDMPs to be associated with an increase in overdoses.
“These findings stand to challenge the kind of unbridled enthusiasm, generous investment, and cavalier policy emphasis that has buoyed PDMPs since the onset of the overdose crisis,” according to Beletsky. “Given evidence of mixed impact, the unintended harms of these systems warrant urgent examination. This includes their potential role in deterring proper prescribing practices; chilling help-seeking among patients, especially those made vulnerable by a history of trauma in the healthcare settings and criminal justice involvement; further fraying the fabric of provid-
“Given evidence of mixed impact, the unintended harms of these systems warrant urgent examination.”
er-patient trust; and facilitating patient transition from prescription to black market drug supplies.”
Similarly, David Fink, a doctoral candidate in epidemiology at Columbia University’s Mailman School of Public Health, has co-authored a systematic review published this summer in the Annals of Internal Medicine, which showed that there is insufficient evidence to confirm whether implementing these PDMPs actually increases or decreases overdoses.
At the same time, he contends that some evidence has demonstrated unintended consequences of implementing these programs. For example, three studies showed an increase in heroin overdose deaths after PDMPs had been implemented.
Likewise, a 2017 study by University of Pennsylvania and Pennsylvania State University researchers— published in the American Journal of Managed Care—found that PDMPs do not drive down opioid overdose death rates, but might have the unintended consequence of adding to them by driving users to black market drugs like fentanyl and heroin.
“PDMPs were not associated with reductions in drug overdose mortality rates and may be related to increased mortality from illicit drugs and other, unspecified drugs,” concluded the authors of the Penn-Penn State study. “More comprehensive and prevention-oriented approaches may be needed to effectively reduce drug overdose deaths and avoid fatal overdoses from other drugs used as substitutes for prescription opioids.”
Last year, a paper by a Purdue University researcher discovered that while PDMPs drove down the prescription rate of oxycodone, they significantly drove up the rate of heroin use.
“It is crucial to determine if these programs are helping to reduce opioid overdose,” Fink says. “So far, the definitive conclusion we can draw from our evaluation is that the evidence is insufficient and that much more research is needed to identify a set of ‘best practices.’”
According to Fink, if population health is the focus, researchers need to understand all effects of PDMPs, including increased rates of fentany land heroin-related overdose. ☐