Rx Monitoring Doesn’t Stem Opioid Overdose Deaths

Prescription drug overdoses and fatalities, largely involving opioid analgesics, have been called an American epidemic. One measure expected to stem the problems of opioid misuse, abuse, diversion, overdose, and death has been the Prescription Drug Monitoring Program, or PDMP. While PDMPs have been implemented in the U.S. by a majority of states during the past decade, a new research report from the CDC raises significant questions about their effectiveness.

Great Expectations
State-run PDMPs (sometimes simply called prescription monitoring programs, or PMPs) typically require retail pharmacists to enter data from prescriptions for controlled substances into a centralized electronic database. These data identify the prescriber, dispenser, and patient, as well as the drug, dose, and amount dispensed. Such information potentially allows authorities to identify individuals who might be prescribing, dispensing, or using prescribed controlled substances inappropriately. Depending on legally-sanctioned uses of the data, state authorities might then employ interventions aimed at reducing abuses and/or diversion of controlled substances and associated health consequences, such as drug addiction and fatal drug overdoses.
Beginning in 2003, development of PDMPs have been supported in part by federal funding from the U.S. Department of Justice and the Department of Health and Human Services, and with passage in 2005 of the National All Schedules Prescription Electronic Reporting Act, or NASPER, state PDMPs have proliferated. As of spring 2010 a total of 35 states had operational programs and 6 states had enacted legislation authorizing such programs [data can be accessed here].

Disturbing Research
Despite the considerable efforts and funding devoted to PDMPs, and their subsequent growth, there have been few assessments of their effects on the problems they were designed to address, and no studies have evaluated the impact of PDMPs on inadvertent lethal overdoses due to opioid analgesics. Writing in the February 2011 online edition of the journal Pain Medicine, a team from the U.S. Centers for Disease Control and Prevention (CDC) in Atlanta, Georgia, led by Leonard Paulozzi, MD, MPH — a widely-respected medical epidemiologist in the Division of Unintentional Injury Prevention — quantify the impact of PDMPs on mortality rates from drug overdose and how these rates compare with quantities of opioid drugs distributed at the state level [Paulozzi et al. 2011].

For their retrospective observational study the investigators accessed U.S. mortality data by state and by year for 1999 through 2005. Overdose deaths, excluding suicides, involving opioid analgesic poisoning were identified, and other data were gathered covering the time period of interest, including U.S. demographic data and distribution data for the 7 most commonly prescribed opioids: the Schedule III drug hydrocodone, and the Schedule II drugs fentanyl, hydromorphone, meperidine, methadone, morphine, and oxycodone. During at least some portion of the 6-year observational period, 19 states had operational PDMPs.

The investigators found that during the study period the nationwide rates for drug overdose mortality approximately doubled overall and rates specifically for opioid-related overdose mortality tripled. At the same time, average per-person opioid analgesic consumption throughout the U.S. roughly tripled. Surprisingly, the operation of PDMPs were NOT significantly associated with any lower rates of drug overdose or opioid overdose mortality, or with declining rates of opioid drug consumption.

In fact, overall, the operation of PDMPs resulted in increased trends in overdoses and mortality, and states with PDMPs consumed significantly greater amounts of hydrocodone (Schedule III) and non-significantly lower amounts of Schedule II opioids. The trends were not influenced by data relating to race, age, urbanization, education, or income within the states examined. However, in 3 PDMP states (California, New York, and Texas) that required use of special prescription forms, increasing rates of overdose mortality and analgesic consumption during 1999–2005 were lower than other states with PDMPs. The authors do not provide any definitive reasons for what made the programs in these 3 states somewhat more successful.

Startling Conclusions
Paulozzi and coauthors conclude that, “…it can be said unequivocally that PDMP states did not do any better than non-PDMP states in controlling the rise in drug overdose mortality from 1999 to 2005.” Not only were PDMPs a failure in stemming drug overdose mortality but their expected effect on overall consumption of opioids appeared to be minimal. The authors recommend that “PDMP managers need to develop and test ways to improve the use of their data to affect the problem of prescription drug overdoses.”

This first-of-its-kind study appears to be a condemnation of PDMPs as not being the hoped for solution to stem the tide of opioid-related overdoses. Still, the authors contend that, “in theory, PDMPs have the potential to address the problem of prescription drug overdoses, but to do so, their use of the information they collect will need to be enhanced.” And, as with any investigation of this nature, there are limitations, not the least of which is that the operation and performance of PDMPs may have changed (perhaps improved) substantially since 2005, in our opinion.

It is disappointing that, more often than not, government reports of this sort rely on historic data and trends, at least several years old, which may not reflect present-day realities. In this computerized age, is there a reason for not having more current data on hand to provide timely and accurate analyses? Government reporting on opioid-related problems also can be burdened by data that are gathered from diverse sources, using different definitions of the problems — such as, what constitutes nonmedical use, misuse, abuse, or a drug-related death — and sometimes portray negatively-biased perspectives [eg, discussed in UPDATE here].

Puzzling Data
In a recent presentation at a CDC Grand Rounds program — “Prescription Drug Overdoses: An American Epidemic,” February 17, 2011

— Paulozzi said that by 2007 drug-induced deaths, amounting to 1 death every 19 minutes, exceeded fatalities associated with firearms, homicides, and suicides, and were second only to deaths in motor vehicle crashes. Unintentional overdose deaths involving opioid analgesics exceeded those associated with cocaine and heroin, and such fatalities paralleled increases in opioid analgesic sales.

Paulozzi further stated that between 1999 and 2007 opioid-analgesic overdose deaths increased nearly 4-fold, from 2,901 to 11,499. Along with that, he observed that in 2007 there were 698 mg/person of opioids distributed; enough for every American to take 5 mg of hydrocodone every 4 hours for 3 weeks. However, he did not calculate that, with such enormous potential opioid consumption (the denominator, 300 million citizens) and the relatively small number of associated fatalities (the numerator, 11,500 deaths), the opioid-analgesic overdose death rate would be only about 0.004% [a very modest figure in comparison with other substance abuse problems, such as the 0.18% death rate in American society associated with alcohol misuse; also see prior UPDATE on ‘probability blindness’ here].

Paulozzi also presented data that opioid-overdose deaths increase dramatically as the daily dose is increased, yet the research study he cited as evidence for this (Dunn et al., Ann Int Med, 2010;152) does not actually demonstrate such relationships with certainty [as discussed in UPDATE here]. Moreover, while Paulozzi mentioned that fatalities due to opioid overdose are more common in persons without legitimate prescriptions for the analgesics or in those visiting multiple prescribers for the drugs (eg, “doctor shopping”), he did not mention during his presentation the new research demonstrating the failure of PDMPs to detect and curtail such problems.

Hurdles Ahead
Meanwhile, there have been very recent reports in the news [here and here] of Florida governor Rick Scott’s controversial plan to repeal the state’s approved but yet-to-be-implemented PDMP that was expected to help crack down on “pill mills” and widespread opioid abuse running amok in the state. The governor’s position was that the electronic monitoring program was delayed by funding difficulties and that “[PDMPs] may not be as effective as advocates claim” (without making any reference to the Paulozzi et al. study). There also were concerns expressed by the governor that a PDMP might infringe on patients’ rights to privacy.

Nearby states with PDMPs were enraged, fearing that their residents would travel to Florida for illicitly prescribed pain medications to bring back home. The possible averting of a PDMP in Florida also drew sharp opposition from the Florida Academy of Pain Medicine and various citizens’ groups. The medical practitioners stated that such a system would be “the single most effective weapon in the battle to shut down Florida’s so-called ‘pill mills.’” However, the research by Paulozzi and colleagues suggests that they may be placing too much faith in a single program of this sort for attacking drug problems that are complex and appear to be endemic — at least until better ways of effectively using PDMPs can be realized.

Progress in maximizing the potential of PDMPs for solving opioid-related problems may be slow in coming. According to a news report [here], last year there were only two grants awarded nationally by the CDC for projects examining unintentional drug poisoning. One of them, awarded to Traci Green, PhD, MSc, at Rhode Island Hospital, will explore the use and effectiveness of PDMPs in Rhode Island and Connecticut for reducing the number of accidental overdose deaths involving prescription opioids. Results from the project are not expected until 2012.

REFERENCE: Paulozzi LJ, Kilbourne EM, Desai HA. Prescription Drug Monitoring Programs and Death Rates from Drug Overdose. Pain Med. 2011(Feb 18); online ahead of print [abstract here].