(CNN) Prescription drug monitoring programs -- believed to be an important tool in the fight against the opioid epidemic -- may be only marginally effective in reducing overdose deaths, according to a new study. But certain features, such as mandatory reviews by providers, could still be useful.
The statewide programs aim to tackle the epidemic by reducing the supply of opioids. They provide physicians with access to a statewide database that includes all of their patients' previous opioid prescriptions, including those from other physicians or hospitals. This, in turn, allows providers to flag patients who may be abusing their medications.
There has been little research into the programs' actual effectiveness, according to David Fink, a social epidemiologist at Columbia University's Mailman School of Public Health and a leading author of the study, published Monday in the journal Annals of Internal Medicine.
"These programs can differ substantially between states, and some states change over time," Fink said. "One of the biggest challenges when you're doing this research is that you oftentimes say, 'What is the effect of prescription drug monitoring programs?' But it's tough to say what we mean by that."
The study looked at changes in fatal and non-fatal opioid overdoses in states after the implementation of prescription drug monitoring programs between 1999 and 2016. It combined the results of 17 previous studies and included data from all 50 states and the District of Columbia.
The researchers ultimately found that the studies' results were inconsistent. Of the 10 studies that looked at fatal overdoses after implementation of these programs, three reported a decrease in overdose deaths, one reported an increase, and six reported no change. Of the three studies that looked at non-fatal opioid overdoses, one showed a drop in overdoses after program implementation, and two showed no change.
"What it really offers is this cautionary tale that says a few things. One, the amount of money and resources that are going into these programs need to be evaluated to understand what their ultimate costs are," Fink said.
"Studies have come out in the past looking at reduced prescribing behavior, but we're taking it to the next level -- looking at fatal or nonfatal overdoses. And what we're seeing is that when you actually look at the literature, it isn't that strong to support" the programs.
The study also identified some unintended consequences of the drug monitoring programs. For example, three of the six studies that looked at heroin overdose rates found a significant increase in heroin-related overdoses after implementation of these programs.
"This result highlights what many would call the substitution effect: when people substitute one drug for another," said Sheila Vakharia, policy manager for the Drug Policy Alliance, who was not involved in the study.
"According to the substitution effect, people who have become dependent on prescribed opioids and lose convenient access to them may substitute with heroin. However, illicit heroin is often of unknown quality and purity ... so it can increase risk of accidental overdose and poisoning."
Prescription drug monitoring programs have existed in some form since 1918, when New York state mandated that pharmacists copy certain prescriptions to a state database, according to Fink.
The most recent wave of the programs occurred in the 1990s and early 2000s, when the rise of electronic databases eased the burden of reporting, he said.
Today, all 50 states and the District of Columbia have passed legislation enabling prescription drug monitoring programs. All states but one have fully functioning programs.
"Every single state has a prescription drug monitoring program or has implemented legislation. We say 'implemented legislation' because Missouri has been dragging their feet. They have implemented legislation and will hopefully be enacting it next year," Fink said.
Because the programs vary significantly between states, the researchers also looked at those features of the programs that were most effective in reducing opioid overdose deaths, Fink said.
For example, some states require that physicians review the database before prescribing new opioid medications, and some proactively report patients' prescription histories to law enforcement and licensure boards to identify illegal prescribing activity.
"Each state has its own set of policies and practices, and each one has its own set of interventions," said Dr. Wilson Compton, deputy director of the National Institute on Drug Abuse of the National Institutes of Health, who was not involved in the study.
"So classifying features like requiring clinicians to use (the programs), how quickly data is shared, whether they are embedded in electronic health records, whether they include data from nearby jurisdictions -- these are the kinds of issues that are useful to have collected uniformly and consistently."
The researchers found that mandatory reviewing of the database by physicians prior to writing new opioid prescriptions had the strongest association with reduced opioid deaths. In 2018, 37 states relied on mandated reviewing, according to Fink.
Other features linked to a decrease in overdose deaths included frequent (at least weekly) updates of the database, sharing of prescription data between states and monitoring of noncontrolled substances such as medications used to treat high blood pressure and diabetes.
"The ones that seemed to be the most effective were the ones that are typically turned to, which are mandatory registration, state authorization for prescribers to access the data and frequency of reporting," Fink said.
"And these are the ones that are oftentimes thought to be the most effective. It makes sense that a program is going to be more effective if it requires the prescriber to actually use the data, and it's probably going to be more effective if the data is as up-to-date as possible."
According to Vakharia, focusing on the demand, rather than the supply, side of the opioid epidemic could also be an effective next step in reducing opioid-related overdoses in the United States.
"Restricting supply generally has little impact on demand and can be a costly waste of resources when demand-side interventions deliver more meaningful results and changes," she said.
These interventions can include "increases in access to mental health treatment, increases in access to evidence-based substance use treatment like medication-assisted treatments (methadone and buprenorphine) ... and increased investments in other community level supports and social services which can help improve quality of life."
The study was limited by the variety of different types of prescription drug monitoring programs, according to Vakharia.
"It makes a lot of sense that they had a hard time drawing clear connections in this review," she said. The programs "differ widely from state-to-state, so it's difficult to make generalizations about how these policies work as a whole. ... You're not comparing apples to apples here."
The paper also highlights the need to implement these programs within a suite of other programs, according to Fink.
"Every time we're thinking about opioid policy, we need to think about the broader picture," he said. "The prescription drug monitoring programs are here, so we want to maximize their utility."
Looking ahead, the new research provides valuable insight into the potential role that these programs can play in the larger effort to tackle the opioid epidemic.
"The programs are not a panacea for addressing the opioid crisis," Compton added. "Even if a physician identifies what looks like problematic prescribing patterns, that's not the end of your work; that's the beginning."