ADVERTISEMENT

Taking PDMPs Up a Notch

1 Comment

Three simple ways to optimize Prescription Drug Monitoring Programs

According to the Centers for Disease Control and Prevention (CDC), 78 Americans die daily from opioid overdoses, a number that’s quadrupled since 1999 [1]. Paralleling this rise, the volume of opioid prescriptions has similarly quadrupled. Emergency physicians prescribe low quantities of opioids per prescription, yet we also prescribe high volumes of individual prescriptions [2]. In many cases, short-term opioid prescriptions are entirely appropriate to help control acute pain associated with illness and injury, however, a national focus on opioid overdoses should prompt us to pause and think before reflexively prescribing. While “drug seeking” is a constant ED concern, research suggests that emergency physicians are actually not very effective in sniffing out these patients [3].

One proposed solution to improve clinician situational awareness when making prescribing decisions are “prescription drug monitoring programs” (PDMPs) [4]. PDMPs are statewide databases used by physicians, pharmacists, and law enforcement to obtain data about controlled drug prescriptions. Historically, PDMPs date back to the 1930s in California and Hawaii where duplicate or triplicate prescription forms were mandated for controlled substances [5].  Today, while 49 U.S. states have PDMPs (sorry Missouri!), each PDMP can vary tremendously in their function and in who can access them. Most PDMPs exist as online web portals where clinicians can look up patient controlled substance prescription drug histories [6]. In this article, we describe the development and evolution of PDMPs and describe three ways to increase their utility in the ED.

ADVERTISEMENT

Do PDMPs work in the ED? 
Despite their ubiquity, there are conflicting studies on the value of PDMPs in EDs. Some studies show that PDMPs are highly impactful on emergency physician prescribing behavior. For example, a 2010 Ohio study showed a 41% decrease in opioid prescriptions when emergency physicians reviewed the state’s PDMP [7]. Other studies have shown no impact: a cross-sectional study of more than 500 emergency physicians across multiple states found no difference in prescribing patterns for opioid analgesics based on whether physicians used PDMPs or not [8]. The reason for conflicting studies is clear. The variation in PDMP effectiveness depends heavily on the wide variability in design and implementation of these information portals, specifically the process of accessing and using them in the ED [9]. In addition, 22 of 49 states mandate that prescribers query the system before prescribing controlled substances, yet those mandates also vary considerably. For example, Connecticut mandates PDMP use if more than three days of a controlled substance is prescribed – a policy has little impact on emergency physicians who typically prescribe low quantities of medications [10]. Even though the majority of states have implemented PDMPs, what remains are a variety of programs with different design and implementation strategies functioning in different legislative environments which have predictably produced variable results.

How to fix PDMPs in the ED
Given the broad variation in PDMP design, we think are three ways that PDMPs could be optimized for emergency physicians.

1. Improving identification of high-risk patients through unsolicited reporting
Today, when clinicians log into a PDMP, they find lengthy reports of prescriptions and are often difficult to interpret [7].  For some patients, a long history of drug-seeking behavior may be obvious when reviewing information from a PDMP; for others, it may be more difficult to determine.  A potential solution for this problem is “unsolicited reporting”, where high-risk behavior is explicitly defined and automatically reported to clinicians. Unsolicited reporting is recognized by the CDC as a best practice, yet is not commonly used. National guidelines, for example, from the CDC could be used to filter a patient’s history and provide a quick notification to emergency physicians when an individual with high-risk prescription behavior is assigned to them. For instance, when a patient is registered in an ED, an alert to a physician (similar to a sepsis alert) could be generated to identify a high-risk prescription drug history. The clinician could then further investigate these concerns. However, there are many barriers to unsolicited reporting include legislative restrictions, resource limitations, and concerns about unintended consequences. Some states do not allow for unsolicited reporting and many legislatures are unaware of the importance of this issue. Even when legislation permits the use of unsolicited reporting, many states do not have sufficient resources to tackle this added dimension of their PDMP program. In addition, even though PDMPs tend to be conservative in setting thresholds for identifying questionable behavior, concerns remain that inaccurate reporting, or leaks could lead to unintended damage to the patient or clinician [11]. In Florida, for example, the PDMP database of more than 3,000 people was released to defense attorneys – which included thousands of patients who were not under investigation. Patients have also expressed concern that these programs might postpone obtaining needed medications and also increase fear of being labeled as an addict. Some physicians, who treat cancer patients or other patients in need of long-term opioids, panic that PDMPs may harm their reputation or licensure [12]. Twenty states already provide unsolicited reports, and some have proven valuable [13]. For example, a Massachusetts study found that unsolicited reporting of PDMP data was useful and easy to understand [14]. However, unsolicited reporting must be carefully calibrated in order to balance the competing concerns of usability, privacy, and fears about “labeling”.

ADVERTISEMENT

2. Better integrating Electronic Health Record (EHR) and PDMP data 
In 2011, the Office of the National Coordinator of Health IT (ONC) and the Substance Abuse and Mental Health Services Administration (SAMHSA) identified several ways that health IT could optimize PDMPs [15].  One of the major problems was that PDMPs require access through a separate web portal outside the hospitals EHR. This clunky process requires logging into multiple systems, maintenance of separate passwords, can be time consuming and may not immediately fit into regular ED workflow. An integrated approach has proven successful in an Indiana pilot study where a PDMP report was inserted directly into the patient’s EHR. CareWeb, an ED information system, automatically queries the PDMP upon the patient’s arrival and formats a viewable PDMP report as an additional tab that the clinician can click on within the platform. The results of this program have been impressive: 58% of prescribers reduced the number of prescriptions and quantity of pills and in 72% of cases, the prescribers reported that the PDMP provided information they otherwise would not have known. Moreover, greater than 97% found the integrated system easier to use than the prior separate login approach [16]. ONC is leading other pilots in collaboration with SAMHSA, the Office of National Drug Control Policy (ONSCP), and the CDC to standardize an approach to make PDMP data available through health information exchanges and EHRs.  SAMHSA is now funding 16 states to advance this integration and many of these states are focusing their pilots in the ED including Oklahoma, Nebraska, Illinois, and Indiana [17].

3. Integration Between States 
Another problem affecting PDMP use is that many PDMPs only report data at the state-level, which can provide an incomplete picture of a patient’s drug prescriptions particularly if patients see providers in multiple states. To address this issue, some states have created interstate sharing hubs such as one that exists between Michigan, Indiana, and Ohio where providers are able to query data outside of their own state. Maryland and DC hospitals also share information through The Chesapeake Regional Information System for Our Patients (CRISP).

One technical barrier to implementing interstate PDMPs is that many PDMP systems have been developed using different tools, software, and security standards. The Bureau of Justice Assistance (BJA), a component of the Office of Justice Programs, is leading the effort to facilitate policy and technology solutions to make interstate sharing achievable. BJA has assisted in the creation of a national Prescription Drug Monitoring Information Exchange (PMIX) architecture which has established a formal set of standards that must be met in order to share data between states. The PMIX technology is available at no cost to states or software vendors that want to adopt this system [18].  This is a tremendous incentive given that the annual cost of state PDMPs can range from $125,000 to $1 million with most financed through a variety of sources including the state general fund, prescriber and pharmacy licensing fees, state controlled registration fees, and state/federal grants [19]. The PMIX program is being rapidly adopted by states and may lead to a nationally integrated system over time (http://www.pdmpassist.org/pdf/PDMP_interoperability_status.pdf). While these efforts need to be further evaluated as they continue to expand, they represent an important health IT advancement in fully utilizing the value of PDMPs [20]. 

ADVERTISEMENT

Realizing the full value of PDMPs in the ED
The good news for emergency physicians is that many states and other groups are working hard to make PDMPs more accessible and useful in practice.  Emergency physicians should encourage their leaders, including policymakers, to implement and evaluate PDMP strategies, and provide a complementary legislative framework to enhance our ability to fight the opioid epidemic and will improve patient care and efficiency.


PDMP Case Study: Florida Leads The Way
In 2009, after realizing their state employed 90 of the nation’s top 100 oxycodone-dispensing physicians and that 1 in 8 citizens were dying from prescription drug overdose, Florida passed a bill to create a prescription drug monitoring program (PDMP) and database. “The purpose of the PDMP,” said Florida Department of Health Deputy Press Secretary Brad Dalton, “is to provide the information that is collected in the database to health care practitioners to guide their decisions in prescribing and dispensing highly abused prescription drugs.” Florida Statutes Section 893.055 requires health care practitioners to report to the PDMP within seven days of prescribing a controlled substance. The program is considered one of the most comprehensive prescription drug monitoring programs in the nation when compared to its 48 counterparts (Missouri does not have a program).

In 2015, the department released its Electronic Florida Online Reporting of Controlled Substances Evaluation, known as the E-FORCSE Annual Report, in which it touted the positive implementation of its systems by most state health facilities and correlated how prescription drug dispensing and abuse had fallen sharply. The data shows improvement. There has been a 7.2 percent increase, from 6,288 to 6,741, in the number of dispensers reporting controlled substance information to the E-FORCSE database since the last reporting period. From 2010 to 2012 there was a more than 50% decrease in oxycodone deaths. At last report, the rate for deaths due to oxycodone overdose had decreased to 2.4 per 100,000. This statistic is the lowest number of reported overdoses since 2006, and continues to fall.

graphs

However, the success of Florida’s PDMP is anything but simple; several issues attempting to complicate its progress. Palm Beach County lists an uptick in prescription drug and heroin use in its latest study from 2013-2014. The E-FORCSE report itself says that deaths associated with heroin use has risen sharply (111.4 %) statewide in the same time frame. The Agency for Health Care Administration did a study on Neonatal Abstinence Syndrome (NAS) among children of opioid-addicted pregnant mothers in hospital discharge data. According to the study, the number of infants born addicted to opiates has been on the rise every year in the past decade-from 338 in 2005 up to 2,487 in 2015.

ADVERTISEMENT

Only time will tell how the Florida PDMP will impact prescription drug dispensing and abuse in the state. Whatever happens, Florida will continue to be a bellwether for larger national drug
-Sonya Swink


REFERENCES

  1. Centers for Disease Control and Prevention. Increases in Drug and Opioid Overdose Deaths — United States, 2000–2014. MMWR 2015; 64;1-5.
  2. Pomerleau AC, Nelson LS, Hoppe JA. The Impact of Prescription Drug Monitoring Programs and Prescribing Guidelines on Emergency Department Opioid Prescribing: A Multi-Center Survey. Pain Med. 2016 Mar 19. pii: pnw032. [Epub ahead of print]
  3. Grover CA, Elder JW, Close RJ, Curry SM. How Frequently are “Classic” Drug-Seeking Behaviors Used by Drug-Seeking Patients in the Emergency Department? West J Emerg Med. 2012 Nov; 13(5): 416–421.
  4. Compton WM, Jones CM, Baldwin GT. N Engl J Med. 2016 Jan 14;374(2):154-63. doi: 10.1056/NEJMra1508490.
  5. Todd KH. Pain and prescription monitoring programs in the emergency department. Ann Emerg Med 2010;56:24-26.
  6. Hildebran C, Cohen DJ, Irvine JM, Foley C. How clinicians use prescription drug monitoring programs: a qualitative inquiry. Pain Med. 2014 Jul;15(7):1179-86. doi: 10.1111/pme.12469. Epub 2014 May 16.
  7. Baehren DF, Marco CA, Droz DE, Sinha S, Callan EM, Akpunonu P. A statewide prescription monitoring program affects emergency department prescribing behaviors. Ann Emerg Med. 2010;56:19–23.
  8. Pomerleau AC, Nelson LS, Hoppe JA, Salzman M, Weiss PS, Perrone J. The Impact of Prescription Drug Monitoring Programs and Prescribing Guidelines on Emergency Department Opioid Prescribing: A Multi-Center Survey. Pain Med. 2016 Mar 19. pii: pnw032. [Epub ahead of print]
  9. Smith RJ, Kilaru AS, Perrone J, Paciotti B, Barg FK, Gadsden SM, Meisel ZF. How, why, and for whom do emergency medicine providers use prescription drug monitoring programs? Pain Med. 2015 Jun;16(6):1122-31.
  10. Haffajee RL, Jena AB, Weiner SG. Mandatory use of prescription drug monitoring programs. JAMA. 2015 Mar 3;313(9):891-2. doi: 10.1001/jama.2014.18514.
  11. Prescription Drug Monitoring Program Center of Excellence at Brandeis. Guidance on PDMP Best Practices. Options for Unsolicted Reporting. Available at: http://www.pdmpexcellence.org/sites/all/pdfs/Brandeis_COE_Guidance_on_Unsolicited_Reporting_final.pdf
  12. Islam MM, McRae IS. An inevitable wave of prescription drug monitoring programs in the context of prescription opioids: pros, cons and tensions. BMC Pharmacol Toxicol. 2014 Aug 16;15:46. doi: 10.1186/2050-6511-15-46.
  13. Deyo RA, Irvine JM, Millet LM, Beran T, O’Kane N, Wright DA, McCarty D. Measures such as interstate cooperation would improve the efficacy of programs to track controlled drug prescriptions. Health Aff (Millwood). 2013 Mar;32(3):603-13. doi: 10.1377/hlthaff.2012.0945. Epub 2013 Feb 13.
  14. Thomas CP, Kim M, Nikitin RV, Kreiner P, Clark TW, Carrow GM. Prescriber response to unsolicited prescription drug monitoring program reports in Massachusetts. Pharmacoepidemiol Drug Saf. 2014 Sep;23(9):950-7. doi: 10.1002/pds.3666. Epub 2014 Jun 12.
  15. Enhancing Access to Prescription Drug Monitoring Programs Using Health Information Technology: Connecting Prescribers and Dispensers to PDMPs through Health IT: Six Pilot Studies and Their Impact. The MITRE Corporation. Available at: https://www.healthit.gov/sites/default/files/pdmp_pilot_studies_summary_0.pdf
  16. Enhancing Access to Prescription Drug Monitoring Programs Using Health Information Technology: Integrating an Emergency Department with a Prescription Drug Monitoring Program: A Pilot Study. Available at: https://www.healthit.gov/sites/default/files/indiana-1_pdmp_pilot_study_0.pdf
  17. Haegerich TM1, Sugerman DE2, Annest JL2, Klevens J2, Baldwin GT2. Improving injury prevention through health information technology. Am J Prev Med. 2015 Feb;48(2):219-28. doi: 10.1016/j.amepre.2014.08.018. Epub 2014 Nov 6.
  18. Prescription Drug Monitoring Programs:
Critical Information Sharing Enabled by National Standards. Available at: https://www.bja.gov/programs/pmixarchitecture.pdf
  19. Finklea KM, Bagalman E, Sacco LN. Prescription drug monitoring programs [Internet]. Washington (DC): Congressional Research Service; 2013 Jan 3 [cited 2013 Jan 25]. Available from: http://www.fas.org/sgp/crs/misc/R42593.pdf
  20. Enhancing Access to Prescription Drug Monitoring Programs Using Health Information Technology: Connecting for Impact: Integrating Health IT and PDMPs to Improve Patient Care. Available at: https://www.healthit.gov/sites/default/files/connecting_for_impact-final-508.pdf
ABOUT THE AUTHORS

Dr. Elder is a practicing emergency physician and a Robert Wood Johnson Clinical Scholar at the Yale University.

HEALTH POLICY SECTION EDITOR Dr. Pines is a practicing emergency physician and a Professor of Emergency Medicine and Health Policy at the George Washington University.

1 Comment

  1. Your article incorrectly states Indiana to be the first to integrate a PDMP report to an ED record in 2016. This integration was achieved in Washington fully 2 years earlier. In November 2014 WA PDMP reports were integrated to existing Emergency Department Information Exchange (EDIE) reports. The article does not address a definition of “integration”. It may be argued that providing a PDMP history report that is separate or in addition to the existing EHR record isn’t actually integrated; rather to integrate PDMP (or any other data) to an EHR requires the data to be pulled into the EHR existing screens and workflows. To do otherwise exposes care providers to a system that would allow infinite additions to the time spent accessing screens/views in the EHR system. “Integration” requires the new data be 1) pulled into the EHR system and 2) presented in existing views in the proper native workflows ensuring the healthcare provider need provide no extra effort or time to consume the new and additional data.

Leave A Reply