Impact Evaluation of A Prescription Drug Monitoring Program (PDMP) Use Policy Based on Guidelines and Best Practices Matthew White Micaiah Fifer A thesis submitted in partial fulfillment of the requirements of the Masters of Healthcare Administration Program Pacific University
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Impact Evaluation of A Prescription Drug Monitoring Program (PDMP) Use Policy Based on
Guidelines and Best Practices
Matthew White
Micaiah Fifer
A thesis submitted in partial fulfillment of the requirements of the
Masters of Healthcare Administration Program
Pacific University
i
Abstract
Introduction From the outset of the implementation of pain management drugs, aberrant behavior relating to controlled substances such as opioids has led to public safety concern, abuse, addiction, and increased risk of overdose and mortality. Beyond the prescribing physicians, some pharmacies have implemented a Prescription Drug Monitoring Program (PDMP) use policy and started reviewing patient records at the point of prescription filling, which rely on data from Oregon’s PDMP database. There are several sets of guidelines for prescribers to use when prescribing these medications that are supported by research, and can be used to develop PDMP use policies by pharmacies and regulatory bodies. Methodology This study has utilized the Oregon PDMP by applying a set of criteria to a de-identified sample of opioid-prescribed patients in an effort to evaluate effectiveness in reducing the following aberrant behaviors. These criteria include: (1) overutilization by maximum morphine equivalent daily dose or maximum methadone daily dose, (2) multiple long-acting or multiple short-acting concurrent opioid prescriptions, (3) multiple concurrent prescribers, (4) multiple concurrent pharmacies, (5) concurrent prescribing of an opioid and benzodiazepine or opioid, benzodiazepine and carisoprodol, and (6) early refills from those filling opioid prescriptions at a retail pharmacy. Results Of these criteria, we found that implementation of the PDMP use policy resulted in a statistically significant decrease in the proportion of patients with a maximum daily morphine equivalent dose >120 mg, maximum daily methadone dose >40 mg, concurrent benzodiazepines, opioids, and carisoprodol, filling of prescriptions from multiple concurrent providers, and filling of prescriptions at multiple concurrent pharmacies. Recommendations Based on the results of this study and due to the risks of aberrant behaviors as summarized in the literature in Chapter 2, which indicate a public safety concern, abuse, addiction, or increased risk of overdose and mortality, we recommend requiring pharmacy PDMP use policies for all pharmacies that include the following criteria: (1) maximum daily morphine equivalent dose >120 mg, (2) maximum daily methadone dose >40 mg, (3) concurrent benzodiazepines, (4) opioids, and carisoprodol, (5) filling of prescriptions from multiple concurrent providers, and (6) filling of prescriptions at multiple concurrent pharmacies.
Data Collection ................................................................................................................................................ 20
iii
Data Analysis ................................................................................................................................................... 23
Data Security and Handling ........................................................................................................... 25
Since the introduction of pain management drugs, aberrant behavior relating to misuse of
controlled substances such as opioids has led to increased dependence which ultimately can lead
to deaths due to overdose. This development is reflected in the fact that “unintentional
prescription opioid overdose death rates (not including methadone) have increased by 5.5 times
from 0.5/100,000 [people in 2000] to 2.5/100,000 [people in 2011]” (Oregon Center for Health
Statistics, 2013) where the denominator represents the population of individuals utilizing opioids
from years 2000 to 2011 per 100,000 Oregonians. The increase in overdoses highlights a need
for processes that address the issue of opioid aberrant behavior. One potential solution lies in the
implementation of more stringent PDMP use policies directed towards those utilizing opioids as
a form of pain management.
The Oregon PDMP is essentially a tool meant to give providers and pharmacists the
ability to review statewide patient prescription history and assist them in managing prescribing
practice. The PDMP utilizes “information provided by Oregon-licensed retail pharmacies”
(Oregon PDMP Public Portal, 2014). Pharmacies submit prescription data to the PDMP system
for prescriptions that have been dispensed to Oregon residents. This information is protected
health information and is collected and stored securely.
“Oregon-licensed healthcare providers and pharmacists and their staff may be authorized for an account to access information from the PDMP system. Bordering state licensed healthcare providers may also be authorized for access accounts. By law their access is limited to patients under their care” (Oregon PDMP Public Portal, 2014). The purpose of creating the PDMP was to support the appropriate use of
prescription drugs. “The information is intended to help people work with their healthcare
providers and pharmacists to determine what medications are best for them” (Oregon PDMP
IMPACT EVALUATION OF A PDMP USE POLICY
2
Public Portal, 2014). Research has been done by many groups, and there are several sets of
guidelines (payers, prescribing groups, and professional associations) for prescribers to use when
prescribing these medications. Guideline specifics are covered in the literature review section of
this study (chapter 2). Beyond the prescribing physicians, some pharmacies in the state have
begun implementing PDMP use policies at the point of prescription filling, which rely on data
from Oregon’s Prescription Drug Monitoring Program (PDMP) Database. The utilization of
pharmacists as the initiator of this PDMP review makes sense due to their position in the
patient’s healthcare journey and ease of access to pertinent information at the point of
dispensing. Physicians have access to the same PDMP information as pharmacists; however it
would take constant tracking of the patient after leaving the medical visit to discover aberrant
behavior. Once discovered, the physician would be in the place of reacting to prescriptions
already filled, whereas the pharmacist can screen for all prescription filling activity prior to
dispensing, and act on the spot to alert the provider and possibly avoid dispensing a prescription
that would put the patient at risk.
Purpose Statement
This study evaluates the impact of a pharmacy PDMP use policy. The researchers
compared a pharmacy that implemented a PDMP use policy with a control pharmacy given other
factors being similar (i.e., type of prescriptions filled, number of prescriptions filled) to evaluate
the impact of pre- and post-implementation of a pharmacy PDMP use policy.
This is a quantitative study of patient records from the PDMP, which models the
following over time and between intervention and control groups: (1) overutilization by
maximum morphine equivalent daily dose or maximum methadone daily dose, (2) multiple long-
acting or multiple short-acting concurrent opioid prescriptions, (3) multiple concurrent
IMPACT EVALUATION OF A PDMP USE POLICY
3
prescribers, (4) multiple concurrent pharmacies, (5) concurrent prescribing of an opioid and
benzodiazepine or opioid, benzodiazepine and carisoprodol, and (6) early refills from those
filling opioid prescriptions at a retail pharmacy. This data has been analyzed for differences in
each category between pre- and post implementation of the PDMP pharmacy use policy.
Research Question
At the point of opioid prescription filling, should: (1) overutilization by maximum
morphine equivalent daily dose or maximum methadone daily dose, (2) multiple long-acting or
of an opioid and benzodiazepine or opioid, benzodiazepine and carisoprodol, and (6) early refills
from those filling opioid prescriptions at a retail pharmacy (OHA, 2013). These criteria are
supported by the literature as parameters to identify individuals at risk for abuse and dependence
of opioid medications and can be accessed via the PDMP at the point of dispensing by
pharmacists to monitor patient utilization of opioids. Identifying these behaviors in this way is
one way to keep these patients safe.
IMPACT EVALUATION OF A PDMP USE POLICY
4
Rationale for Thesis
Overutilization by dose, multiple long-acting or multiple short-acting concurrent opioid
prescriptions, multiple prescribers during overlapping time periods, multiple filling pharmacies,
concurrent prescribing of an opioid and benzodiazepine, and early refills are all criteria that can
be screened for in the Oregon Prescription Drug Database (Oregon Health Authority, 2013;
Opioid Prescribers Group, 2013). These criteria are also determinants of whether a patient is
adhering to their pain contracts, “doctor shopping,” or receiving higher than average doses of
pain-relieving opioids.
Pharmacists can access the PDMP database online at www.orpdmp.com to query
prescription filling information for controlled substances, including the patient name, patient
address, date of birth, drug filled, dose, quantity, date written, date filled, prescriber information,
and dispensing pharmacy. With this information, pharmacists have the potential to identify
individuals at risk for abuse and dependence of opioid medications by reviewing PDMP records
for these specific and identifiable criteria at the time a prescription is dispensed (Oregon Health
Authority, 2013; Opioid Prescribers Group, 2013).
Rationale for Project
Unintentional prescription-drug-related hospitalizations and deaths have risen steadily
since 2000 indicating that there is a clear safety issue that corresponds to opioid abuse
(Unintentional Prescription Drug Overdose in Oregon, 2013). With 14.7% of Oregonians ages
18-25 self-reporting that they used prescription pain relievers for non-medical purposes, there is
also a concern for misuse (Oregon Health Authority, 2013). In addition to non-medical use, the
number of Oregonians receiving “substance misuse disorder treatment” increased five-fold
between 2000 and 2011 (Oregon Health Authority, 2013). With the increased non-medical use,
IMPACT EVALUATION OF A PDMP USE POLICY
5
substance misuse treatment, and unintentional safety-related events, most of which are opioid
related, there is a need to look at ways to prevent misuse of opioid prescriptions.
Pharmacists are the last check before the patient receives a prescription. This presents
tremendous opportunity to implement interventions that can increase patient safety. At the
pharmacy level, data provided by the PDMP is available to help identify patients at risk
(Prescription Monitoring Program, 2013). Data supporting a PDMP use policy can result in
policy changes in the profession of pharmacy.
Summary
Since the implementation of pain management drugs, aberrant behavior relating to
controlled substances such as opioids has led to increased dependence, which can lead to deaths
due to overdose. Research has been done by many groups, and there are several sets of
guidelines for prescribers to use when prescribing these medications. Beyond the prescribing
physicians, some pharmacies have implemented a PDMP use policy, and started reviewing
patient records at the point of prescription filling, which relies on data from Oregon’s
Prescription Drug Monitoring Program (PDMP) Database. This study utilized the Oregon PDMP
by applying a set of criteria to a de-identified sample of opioid prescribed patients in an effort to
evaluate their effectiveness.
IMPACT EVALUATION OF A PDMP USE POLICY
6
CHAPTER 2 – LITERATURE REVIEW
Opioid Abuse
The issues of dependence and aberrant behavior pertaining to opioid use are current
concerns in journals and scientific literature. Dependence is defined in the literature as including
“a maladaptive pattern of substance use leading to clinically significant impairment or distress, as manifested by three (or more) of the following: 1) tolerance (requires increased amounts of the substance to desired effect or diminished effect with use of the same amount of the substance); 2) withdrawal (characteristic withdrawal syndrome for the substance or taking the substance to relieve or avoid withdrawal symptoms); 3) the substance is often taken in larger amounts or over a longer period than intended; 4) unsuccessful efforts to control substance use; 5) time spent on activities necessary to obtain, use, or recover from effects of the substance; 6) reduced psychosocial functioning; and 7) continued use despite known risks” (APA, 2000, p. 110).
In reference to opioids, healthcare professionals recognize that the “increased use of these
drugs has resulted in an increase in overdoses and other problems associated with drug misuse”
(Booze, 2006, p. 1). Concern in this area of healthcare has only increased as time has passed,
since prescription “misuse, abuse, addiction, overdose, and other health and social consequences
of inappropriate [prescription] use are taking a rapidly growing toll on individuals and
institutions in the United States” (Katz et al., 2013, p. 295). Something that seems consistent in
most opioid related studies is the recognition that “the behaviors of some chronic pain patients
on opioid therapy put them at greater risk of consequences often associated with addiction,
including elevated use of healthcare services, crime, and death” (Katz et al., 2013, p. 296).
There appears to be a parallel between the increase in prescribing and the increase of
abuse as is noted in multiple studies. One such study points out that “along with the increase of
prescriptions for controlled drugs from 1992 to 2003 of 154%, there was also a 90% increase in
the number of people who admitted to abusing controlled prescription drugs” (L. M.
Manchikanti, Fellows, Ailinani, & Pampati, 2010, p. 419). Similar findings from the Centers for
IMPACT EVALUATION OF A PDMP USE POLICY
7
Disease Control noted that the trend of increasing prescribing of opioids has closely paralleled a
rise in the numbers of overdoses (Methadone-associated overdose deaths factors contributing to
increased deaths and efforts to prevent them: Report to congressional requesters, 2009). Illegal
use of opioids has begun to outpace the abuse of non-legal drugs like cocaine in some capacities.
One study points out the “nonmedical use of prescription pain relievers is greater than
stimulants, sedatives, and tranquilizers” (Moore, Jones, Browder, Daffron, & Passik, 2009, p.
1426). Opioids can be acquired illegally by utilizing multiple pharmacies or “doctor shopping”.
Individuals can sometimes “fill prescriptions at multiple pharmacies, making them more difficult
to track. These prescriptions may be used by addicted individuals themselves, diverted to family
members or friends, or sold on the street” (Katz et al., 2013, p. 296). A screening tool can be
utilized to prevent these two activities since
“at the time of dispensing an opioid analgesic at a pharmacy, a fraud and abuse screen [can be] automatically conducted, including checks of PMP [Prescription Monitoring Program] data, validity of Drug Enforcement Administration (DEA) registrations of prescriber and pharmacist, vital status registries to ensure that both prescriber and patient are alive, and federal debarment and exclusion databases” (Katz et al., 2013, p. 296). Unsurprisingly, this healthcare issue has a substantial monetary cost as well. Studies
acknowledge the idea that “prescription opioid abuse and addiction are serious problems with
growing societal and medical costs, resulting in billions of dollars of excess costs to private and
governmental health insurers annually” (Katz et al., 2013, p. 295). The study asserts further that,
“although difficult to accurately assess, prescription opioid abuse also leads to increased
insurance costs in the form of property and liability claims, and costs to state and local
governments for judicial, emergency, and social services” (Katz et al., 2013, p. 295). The
existence of opioid related addiction, aberrant behavior, death, and cost is well documented in
the literature.
IMPACT EVALUATION OF A PDMP USE POLICY
8
Prescribing guidelines for chronic pain
In recent years, there has been a number of opioid treatment guidelines published. The
American Pain Society and the American Academy of Pain Medicine published a set of
guidelines in 2009 that address the use of opioid therapy for chronic non-cancer pain. At that
time the panel concluded that there was weak evidence to formulate recommendations and relied
on an expert panel. Regarding methadone, the guidelines recognize that “methadone is
characterized by complicated variable pharmacokinetics and pharmacodynamics and should be
initiated and titrated cautiously” (Chou et al., 2009, p. 118). “Starting methadone doses should
generally not exceed 30 to 40 mg a day even in patients on high doses of other opioids.” The
guidelines recognize that “more research is needed on how policies that govern prescribing affect
clinical outcomes” (Chou et al., 2009, p. 118).
Reinforcing the need for further research, the American Pain Society and American
Academy of Pain Medicine reviewed research gaps in the 2009 guideline, partly looking at
methods for monitoring opioid use and detecting aberrant drug-related behaviors. The study
confirmed the need for methadone research and reinforces the lack of evidence on benefits and
harms of high-dose opioids. It pointed out that the highest doses of morphine in trials are 240 mg
per day with the highest average dose being 120 mg per day (Chou, Ballantyne, Fanciullo, Fine,
& Miaskowski, 2009, p. 118).
Approaching the lack of evidence from another direction, the American Pain Society
points out “the United States carefully monitors controlled substance use, and probably has
liberties and privacy in the U.S. prevents the U.S. from collecting comprehensive healthcare
data across the population at large.” The study also points out that “we need evidence, not only
IMPACT EVALUATION OF A PDMP USE POLICY
9
to guide practice, but to inform healthcare policies and drug regulations” (Ballantyne, 2010, p.
830).
In response to the documented need for opioid related research, an Opioid
Pharmacotherapy Research Guideline was published by the American Pain Society. Among the
many areas for research that are pointed out, tool validation is essential in clinical research. The
study identifies that “no tools exist at present for specifically predicting aberrant medication
behaviors” (Chapman et al., 2010, p. 822).
In (Victor, Alvarez, & Gould, 2009, p. 1051) the emphasis on need for research is
replaced by an analysis that suggests that the “availability of pain-treatment guidelines,
recommendations, and education alone may not be enough to influence opioid-prescribing
practices in the treatment of chronic pain.” This is an important distinction that corresponded
with the idea that tools for detecting aberrant behavior are lacking (Chapman et al., 2010),
suggesting a need for intervention other than just having guidelines.
The American Society of Interventional Pain Physicians (ASIPP) advises that PDMP
screening for opioid abuse is recommended as it will identify opioid abusers and reduce opioid
abuse despite limited evidence for reliability and accuracy. The ASIPP further recommends that
prescription monitoring programs be implemented as they provide data on prescription usage,
reduce prescription drug abuse or doctor shopping (Manchikanti et al., 2012). Prescription Drug
Monitoring Program
The development a Prescription Drug Monitoring Program (PDMP) is not new to the
United States. The idea of developing a program state-wide as well as nationally comes from a
variety of studies that have been encouraged by the Obama administration in recent years. Some
studies have observed the need for such programs in populations where
IMPACT EVALUATION OF A PDMP USE POLICY
10
“the majority of opioid prescriptions among [a] large commercially insured population might have been appropriate, a substantial number were prescribed in a manner that suggests potential patient misuse or inappropriate prescription practice by providers” (Liu, Logan, Paulozzi, Zhang, & Jones, 2013, p. 657). In response to these types of situations, it has been largely observed that by “using
medical as well as drug claims data, it is feasible to develop models that could assist payers in
identifying patients who exhibit characteristics associated with increased risk for opioid abuse”
(Rice et al., 2012, p. 1). The idea of PDMPs are quickly gaining steam as is evident in that “48 of
the states have prescription drug monitoring programs (PDMPs), but not all of the programs are
up and running” as is outlined by journalist Kuehn. Furthermore, there are “at least 40 states now
that have operational PMPs or have enacted PMP legislation, covering 87% of the US
population, and the White House plan has committed federal resources to expanding them” (Katz
et al., 2013, p. 299). He goes on to impress the value upon his readers in stating, “such programs
can help physicians, pharmacists, and insurers identify patients who are inappropriately
accessing opioid medications, such as through obtaining prescriptions from multiple physicians
simultaneously” (Kuehn, 2012, p. 20). There is evidence to support value in both provider and
patient observation in implementing a PDMP. One study observes that “health plans may be able
to use these methods to identify members with chronic pain who exhibit similar problematic
behaviors in order to effectively intervene and optimize resource allocation” (Tkacz et al., 2013,
p. 872).
Screening categories like previous use of opioids, previous documented addictions, and
prevalence of mental health issues has largely been observed to help determine an individual’s
propensity for abuse of opioids. Some suggest that
IMPACT EVALUATION OF A PDMP USE POLICY
11
“using currently available screening tools designed for use in populations on or considering opioid therapy is recommended as there is evidence that patients with a prior history of drug or alcohol abuse or psychological problems are at increased risk of developing opioid related use/abuse problems” (Opioid overuse a "public health emergency", 2013, p. 103). However, the literature cautions those who would use PDMPs for anything other than
treatment decisions. One study highlights this idea in stating that “the information obtained about
drugs patients have been prescribed should be used to make therapeutic decisions but not to
punish the patients or to terminate their treatment, according to the guidelines” (AATOD: OTPs
should use PMPs but not input data... American Association for the Treatment of Opioid
Dependence... opioid treatment programs... prescription monitoring programs, 2012, p. 7).
What of the state of Oregon’s needs in this arena? In regards to previous utilization of
Prescription Monitoring Programs (PMPs) in other states, “the body of evidence, though sparse,
is robust for one assertion: PMPs reduce the use of targeted prescription medications” (Fornili &
Simoni, 2011, p. 80). Other studies are in agreement and have come to the conclusion that “even
early versions of today’s PMPs have been shown to reduce prescription drug diversion” (Katz et
al., 2013, p. 299). This assessment does not assume that all PDMPs are created equal, which they
are not. One journalist notes differences, and articulates that
“the tremendous state variability in PMP design, scope, and operationalization imposes tremendous difficulty in assessing the effectiveness of PMPs in reducing prescription medication abuse and diversion. Although several studies have empirically investigated the effectiveness of paper-based PMPs on the intended and unintended influences on controlled prescription medication use, none have evaluated electronic monitoring PMPs” (Fornili & Simoni, 2011, p. 79).
The idea that all PDMPs are not created equal gives credence to the idea that there needs
to be much more evaluation of programs on an individual basis. One author asserts, “future
studies to evaluate the impact of PMPs on medical use and access, abuse and diversion, and
clinical and economic outcomes are imperative.” They continue along that vain in stating
IMPACT EVALUATION OF A PDMP USE POLICY
12
“PMP outcome measures are needed to fully evaluate the impact on patients along with drug abuse and diversion. The goal of balancing appropriate treatment with reductions in prescription drug abuse and diversion should remain in the forefront of new and alternative drug policy strategies” (Fornili & Simoni, 2011, p. 81).
Prescribing Guidelines – Dosing Limits
The American Society of Interventional Pain Physicians (ASIPP) recommends
stratification of patients into low-, medium-, and high-risk categories for dosing. “Up to 40 mg of
morphine equivalent [per day] is considered as low dose, 41-90 mg of morphine equivalent [per
day] as moderate dose, and greater than 91 mg morphine equivalent [per day] as high dose”
(Manchikanti et al., 2012, p. S68).
The Southern Oregon Opioid Prescribing Guidelines are in agreement with the ASIPP,
with a slightly different dose/risk stratification. The Opioid Prescriber’s Group (OPG)
recommends keeping doses of opioids below 120 mg Morphine Equivalent Dose (MED) per day.
Doses of greater than 100 mg MED per day have a nine-fold increase risk in death over low-dose
morphine. The guidelines recommend that methadone doses generally not exceed 40 mg daily.
The guidelines classify patients as high risk if they take greater than 100 mg MED per day or 40
mg methadone per day, moderate risk if they take 41-100 MED per day, and low risk if they take
40 MED per day or less. The guidelines point out that many experts advise against concomitant
use of benzodiazepines and opioids because of synergistic effects resulting in respiratory
depression (OPG, 2013).
McCarthy (2012) is in agreement with the Opioid Prescriber’s Group (OPG, 2013). The
State of Washington, when revising the guidelines issued in 2007, designated an “opioid dose
threshold of 120 mg/day” morphine equivalent dose. “At that dose, the guidelines advised, if the
patient’s pain and functionality had not substantially improved, the physician should consult with
a pain specialist before increasing the dose” (McCarthy, 2012, p. 1).
IMPACT EVALUATION OF A PDMP USE POLICY
13
Another study builds on previous research in Washington looking at the workers
compensation population, again recommending a dose cap of 120 mg/day MED. The study
looked at periods before and after disseminating the Washington Opioid Dosing Guideline and
found that a
“yellow flag dosing threshold [greater than 120 mg/day] may be a useful tool for improving the safety of opioid prescribing practices by discouraging further dose escalation. Both chronic and high-dose opioid use rates declined among incident users after the Washington Guideline implementation” (Garg et al., 2013, p. 1620). In Preventing Prescription Opioid Overdose there is a section of the review that
evaluates the avoidance of dangerous drug combinations. In agreement with the Southern
Oregon Opioid Prescribing Guidelines (OPG, 2013), “opioid co-administration with other central
nervous system depressants, particularly alcohol and benzodiazepines significantly increases
morbidity and mortality.” In agreement with methadone dosing guidelines and in response to an
increase in methadone-related adverse events, the U.S. Drug Enforcement Administration (DEA)
and drug manufacturers “agreed to limit the distribution of methadone 40 mg dispersible tablets
to only those facilities authorized for treatment of opioid addiction” (Morgan & Weaver, 2010,
p. 512).
Risk for Overdose from Methadone Use for Pain Relief states that
“interventions such as the use of prescription drug monitoring programs, appropriate screening and monitoring before prescribing opioid pain relievers, regulatory and law enforcement efforts, and state policies aimed at providers and patients involved in diversion of these drugs continue to be essential elements in addressing this public health emergency” (Vital signs: Risk for overdose from methadone used for pain relief - United States, 1999-2010, 2012, p. 496). (Gudin, Mogali, Jones, & Comer, 2013, p. 118) concludes that “there is a defined
increase in rates of adverse events, overdose, and death when these agents,” referring to opioids,
benzodiazepines, and alcohol are used in combination. “Urine testing and Prescription
IMPACT EVALUATION OF A PDMP USE POLICY
14
Monitoring Programs are two indispensable tools that can identify patients who are non-adherent
to treatment, have filled multiple prescriptions at multiple pharmacies, and/or are abusing
prescription drugs and/or illicit drugs.”
Further support for the risks of opioids and benzodiazepines use together is seen in an
editorial by Webster, 2010, p. 801). “The fact that benzodiazepines and opioids have frequent
appearances as co-intoxicants in opioid-related deaths presents yet another risk” which is
compounded by the fact that “In 2006, about half of all U.S. opioid related deaths involved more
than one type of drug with benzodiazepines mentioned most frequently.”
Screening Criteria
The literature addressing opioid prescription screening tools references a screening tool
with a
“prescription abuse checklist of five criteria: overwhelming focus on opiate uses, pattern of early refills, multiple telephone calls or unscheduled visits, episodes of lost or stolen prescriptions, and supplemental sources of opioids. Of these criteria, patterns of early refills and supplemental sources (e.g., from other prescribers) of opioids could be identified in a prescription monitoring program database” (Butler et al., 2004, p. 65).
(Chabal et al., 1997) took the five criteria described in (Butler et al., 2004) and applied
them to patients enrolled in a pain clinic. It was found that “the criteria had good reliability and
can be applied during normal clinical interactions.” This provides additional validation for those
criteria that can be tested in a PMP.
Reinforcing some of the criteria for screening in the Butler study, a study was completed
on the Michigan Medicaid population on patients identified as having three or more prescriptions
from two or more providers in six months. The study looked at evidence of medication
management agreements, early refills, increasing doses, telephone call frequency, reports of lost
or stolen medications, history of abuse, and medical chart documentation. It was found that
IMPACT EVALUATION OF A PDMP USE POLICY
15
concurrent use of non-opioid analgesics, escalating opioid dosage and number of providers best
predicted the number of opioid prescriptions. This study confirms the need for dosage review
and also introduces the concept of multiple prescribers being correlated to opioid use.
Expanding on the concept of multiple providers, a West Virginia study compared live
subjects with persons who died due to drug overdose. The study found that a significantly greater
proportion of deceased subjects were “doctor shoppers” and “pharmacy shoppers” than were the
live subjects. Doctor shopping was defined as having received prescriptions from three or more
clinicians during the six months prior to death. Pharmacy shopping was defined as filling
prescriptions in four or more pharmacies in the same time frame. This study further validates
looking at multiple prescribers and pharmacies in prevention of deaths due to drug overdose,
these being criteria that can be searched in a prescription monitoring database.
Doctor shopping is a conceptual theme that shows up throughout the literature. (Julie &
Hall, 2012) compiled an analysis on the topic and raised a number of issues related to doctor
shopping such as economic costs, ethical issues, legal ramifications from those who benefit, as
well as risk to patients. They point out that a “positive consequence of doctor shopping has
occurred in the professional and legal sectors wherein Prescription Drug Monitoring Programs
have helped to “catch” instances of doctor shopping and would relieve the prescriber of veritably
needing to have a “crystal ball” to determine who may be deceiving them or not.”
Barriers in Implementing Screening Tools
Some barriers exist in regards to the implementation of PDMPs. One rather large
stumbling block exists in the financial impetus required to move forward for many states. Some
administrative experts have noted that
IMPACT EVALUATION OF A PDMP USE POLICY
16
“although prescription drug monitoring programs are designed and maintained at the discretion of each state, federal funds from the Bureau of Justice have assisted in facilitating interstate data exchange and in creating and expanding data collection and analysis systems” (Gugelmann & Perrone, 2011, p. 2258). They also offer more insight by outlining additional barriers in the form of interstate data
communication. They mention how important this aspect of a system is in stating that the
“function and success of drug monitoring programs continue to be limited by variability in data collection systems and the interstate exchange of information. New interstate collaboration initiatives through the National Association of Boards of Pharmacy and the Alliance of States with Prescription Monitoring Programs attempt to address these issues” (Gugelmann & Perrone, 2011, p. 2258).
The cost to payers is another financial burden to add to the list when discussing barriers
to implementing PDMPs. The time spent for healthcare professionals to use such screening tools
is also a cost-related concern. One study that focuses on mitigating risks for payers states that,
“in the future, they could make it easier for time-challenged physicians to use these data by
linking the PMP to their electronic prescribing tools, or to pharmacy computers at the time of
dispensing” (Katz et al., 2013, p. 299).
Others in the field identify lack of knowledge in relation to identifying drug-related
behaviors and risk factors. They conclude that
“What is truly needed (and indeed it is glaring that such a study has not yet been carried out) are large epidemiologic surveys of risk factors evaluated prospectively as patients enter into treatment and are then followed for a meaningful time period (i.e., 1-year minimum, if not longer). The conclusions from such studies would be extremely valuable to practicing clinicians in that they could be employed not to justify excluding patients from treatment with opioids in the face of severe pain, but to plan for and put in place safeguards and structures in the treatment approach that would potentially enable such patients to enjoy good outcomes.” (Passik & Kirsh, 2003, p. 186).
Some of the literature on opioid prescribing limitations suggests that physicians are
concerned that the use of screening tools may limit their ability to offer the kind of treatment
they aspire to provide. One article iterates this idea by stating that “pain physicians are concerned
IMPACT EVALUATION OF A PDMP USE POLICY
17
that there will be a backlash against their work and their patients, resulting in a return to under
treatment of pain” (Wiley, 2007, p. 5).
IMPACT EVALUATION OF A PDMP USE POLICY
18
CHAPTER 3 - METHODOLOGY
Introduction
This chapter will cover the methodology of this study, including the research questions being
answered, the setting of the research, the study design, research methods, how findings will be
reported, and ethical considerations.
This study evaluates impact of a pharmacy PDMP use policy. The researchers compared
a pharmacy that implemented a PDMP use policy with a control pharmacy, given other factors
being similar (i.e., type of prescriptions, number of prescriptions filled annually) to evaluate the
impact of pre- and post-implementation of a pharmacy PDMP use policy.
This is a quantitative study of patient records in the PDMP database utilizing opioids for
pain management by comparing the following during pre- and post-implementation, and between
groups: (1) overutilization by maximum morphine equivalent daily dose or maximum methadone
Due to limitations in the data set (missing days supplies), the number of occurrences of
more than one concurrent long-acting opioid prescription and more than one concurrent short-
acting opioid prescription were not analyzed.
IMPACT EVALUATION OF A PDMP USE POLICY
25
Early Refills
Due to limitations in the data set (missing days supplies), early refill analysis was not
conducted.
Data Security and Handling
Data was stored on an encrypted, password-protected portable hard drive while it
contained any de-identified PDMP flat-file data. After all PDMP de-identified flat-file data
elements were removed, it remained on the encrypted, password protected portable hard drive,
and was permitted to leave OHA premises for data analysis. At the conclusion of the study all
data was destroyed by the researchers by repartitioning and reformatting the hard drive.
Reporting Findings
Results have been presented, in partial fulfillment of the Masters of Healthcare
Administration degree, to the Oregon Board of Pharmacy (OBOP), the sponsoring pharmacy, the
PDMP, and at Pacific University. A bound copy remains in the offices of the Pacific University
Masters of Healthcare Administration Program and may be submitted for further publication.
Ethical Considerations
As we endeavor to portray the value of this PDMP utilization, which aims to improve the
safety of opioid utilizers, it is important that we as researchers, “respect the participants and the
sites for research” (Creswell, 2003, p. 89). The nature of the research that was conducted for
evaluating the PDMP use policy does not include “vulnerable populations such as minors (under
the age of 18), mentally incompetent participants, victims, persons with neurological
impairments, pregnant women or fetuses, prisoners, and individuals with AIDS” (Creswell,
2003, p. 89). Despite the exclusion of these vulnerable populations, there are some ethical
considerations.
IMPACT EVALUATION OF A PDMP USE POLICY
26
One crucial ethical consideration is
“gaining the agreement of individuals in authority (e.g., gatekeepers) to provide access to de-identified data at research sites. This often involves writing a letter that identifies the extent of time, the potential impact, and the outcomes of research” (Creswell, 2003, p. 90).
Pacific University and the sponsoring institution, the “gatekeepers” in this instance, have
taken steps to ensure that both members of the research team had appropriate access to the
patient data at the sponsoring institution. The sponsoring institution has granted this study’s
researchers association via the Affiliation Agreement for student experience under the current
contract with Pacific University. This document lists the expectations of the researchers and their
role as students. The researchers have followed the sponsoring institution’s policies and
guidelines. Additionally, an Oregon Prescription Drug Monitoring Program Data Use Agreement
has been approved and followed by the researchers of this study.
Another ethical consideration “arises when there is not reciprocity between researcher
and the participants” (Creswell, 2003, p. 90). In the instance of this research, the need for
evaluating the PDMP utilization policy as it was addressed in the literature review highlights the
fact that this tool, if it is effective, will serve to benefit individuals who require opioids in their
day-to-day lives by identifying risk factors that can be reviewed by a pharmacist and
communicated to a prescribing physician in an effort to ensure the patient’s safety. The PDMP
use policy is to be used to minimize aberrant behavior and deaths due to overdose. This
motivation is certainly in the best interest of the research subject. Furthermore, the outcome of
this research has ramifications for more patients than those utilized for this study. Validation of
this PDMP use policy could potentially make opioid use safer for all of Oregon if utilized
statewide.
IMPACT EVALUATION OF A PDMP USE POLICY
27
The social stigma associated with pain management medication creates an ethical
consideration that must be taken into consideration. It is important that researchers “anticipate
the possibility of harmful, intimate information being disclosed during the data collection
process” (Creswell, 2003, p. 91). To this end, data that contained identifiers were not provided to
the researchers by the Oregon Health Authority.
This document does not “use language or words that are biased against persons because
of gender, sexual orientation, racial or ethnic group, disability, or age” (Creswell, 2003, p. 92).
The data provided does not contain any information about sexual orientation, racial or ethnic
group, or disability of participants.
Summary
In summary, we have discussed in detail the methodology of this study, which includes
the research questions being answered, the setting of the research, the study design, research
methods, how we have reported findings, and ethical considerations. This methodology has been
use to answer the following question; should the criteria at the point of opioid prescription filling
be used by pharmacists accessing the PDMD to identify unsafe prescribing practices and patient
aberrant behaviors which could indicate abuse, addiction, or increased risk of overdose and
mortality?
In an effort to answer these questions, this research study has adopted a quantitative
approach. This is because there is a body of known information and a specific problem with hard
evidence to use in validating specific criteria. The criteria evaluated are: (1) overutilization by
maximum morphine equivalent daily dose or maximum methadone daily dose, (2) multiple long-
acting or multiple short-acting concurrent opioid prescriptions, (3) multiple concurrent
prescribers, (4) multiple concurrent pharmacies, (5) concurrent prescribing of an opioid and
IMPACT EVALUATION OF A PDMP USE POLICY
28
benzodiazepine or opioid, benzodiazepine and carisoprodol, and (6) early refills from those
filling opioid prescriptions at a retail pharmacy.
The aforementioned techniques have been used for analyzing and presenting the data,
which the literature suggests may indicate aberrant behavior on the patient’s part and put the
patient at risk for overdose or mortality.
IMPACT EVALUATION OF A PDMP USE POLICY
29
CHAPTER 4 - RESULTS
Maximum Morphine Equivalent Daily Dose
Table 1 and Table 2 are each composed of three columns pre-implementation and post-
implementation of the PDMP use policy. The “Cluster ID” column is a numbering the patient
samples collected. We have called it a cluster ID because it represents a cluster of records
pertaining to the pre-implementation or post-implementation time period. The “Max MEq”
column is the maximum morphine equivalent daily dose. The “>120 mg (Y/N)” column denotes
whether the dose in the “Max MEq column” is greater than 120 mg. Table 3 and Table 4
summarize the “>120 mg (Y/N)” data for the intervention and control groups respectively for
statistical analysis. Figure 1 and Figure 2 graphically represent Table 3 and Table 4 respectively.
IMPACT EVALUATION OF A PDMP USE POLICY
30
Table 1: Max Morphine Equivalent Daily Dose (mg) at the Intervention Pharmacy
Pre-Implementation
Post-Implementation Average 116
Average 101
# Yes 21
# Yes 13 # No 29
# No 37
Cluster ID
Max MEq
>120 mg
(Y/N)
Cluster ID
Max MEq
>120 mg
(Y/N) 1 90 N
1 70 N
2 230 Y
2 60 N 3 62 N
3 100 N
4 116 N
4 60 N 5 90 N
5 70 N
6 60 N
6 45 N 7 145 Y
7 52.5 N
8 180 Y
8 180 Y 9 150 Y
9 50 N
10 180 Y
10 180 Y 11 120 N
11 154 Y
12 140 Y
12 50 N 13 154 Y
13 120 N
14 110 N
14 40 N 15 160 Y
15 60 N
16 120 N
16 180 Y 17 30 N
17 60 N
18 200 Y
18 60 N 19 180 Y
19 60 N
20 170 Y
20 70 N 21 90 N
21 60 N
22 92 N
22 66 N 23 145 Y
23 100 N
24 158 Y
24 150 Y 25 100 N
25 75 N
26 150 Y
26 15 N 27 75 N
27 50 N
28 30 N
28 60 N 29 90 N
29 100 N
30 50 N
30 75 N 31 30 N
31 30 N
32 83 N
32 60 N 33 45 N
33 146 Y
34 196 Y
34 165 Y 35 165 Y
35 75 N
36 190 Y
36 90 N 37 60 N
37 50 N
38 30 N
38 50 N 39 30 N
39 240 Y
40 200 Y
40 200 Y 41 50 N
41 105 N
42 135 Y
42 180 Y 43 50 N
43 300 Y
44 50 N
44 57.5 N 45 50 N
45 50 N
46 46 N
46 46 N 47 110 N
47 110 N
48 330 Y
48 330 Y 49 200 Y
49 255 Y
50 60 N
50 45 N
IMPACT EVALUATION OF A PDMP USE POLICY
31
Table 2: Maximum Morphine Equivalent Daily Dose (mg) at the Control Pharmacy
Wiley, J. (2007). As opioid prescriptions increase, so does abuse. Alcoholism & Drug Abuse
Weekly, 19(4), 5. Retrieved March 24, 2014, from
http://www.alcoholismdrugabuseweekly.com
IMPACT EVALUATION OF A PDMP USE POLICY
80
APPENDIX A
Sponsor Support Letter
September 2, 2014 To Whom It May Concern: I am writing in support of the research proposed by Matthew White and Micaiah Fifer titled “Evaluation of Pharmacist Screening Criteria for Intervention in Opioid Prescription Dispensing to Enhance Patient Safety and Reduce Aberrant Behavior”. As a pharmacist with over 20 years of practice experience in Southern Oregon, I have continuously worried about the inappropriate use of opiates and the impact on our communities. As stated in the rational for this research proposal “unintentional prescription drug-related hospitalizations and deaths have risen steadily since 2000 indicating that there is a clear safety issue that corresponds to opioid abuse.” It is clear that my worries are not unwarranted. The Oregon Prescription Drug Monitoring Program (PDMP) has finally given the medical community a resource to identify potentially aberrant behavior. However, each pharmacy and pharmacist is struggling to understand this new data in relationship to the safety of our patients and our communities. With each pharmacist interpreting this data using different processes, patient are treated inconsistently across pharmacies and doctors are receiving mixed messages regarding their patients. The research proposed by Mr. White and Mr. Fifer will provide pharmacists with standard screening criteria and allow for consistent use of the data provided by the Oregon PDMP. I feel confident that consistent evaluation of opiate use across all pharmacies would help the medical community provide safe pain management for our patients while keeping our communities a safer place to live. Sincerely, [Sponsor Omitted]
IMPACT EVALUATION OF A PDMP USE POLICY
81
APPENDIX B
Public Health Division / Multnomah County Health Department IRQ
IMPACT EVALUATION OF A PDMP USE POLICY
82
IMPACT EVALUATION OF A PDMP USE POLICY
83
IMPACT EVALUATION OF A PDMP USE POLICY
84
IMPACT EVALUATION OF A PDMP USE POLICY
85
IMPACT EVALUATION OF A PDMP USE POLICY
86
IMPACT EVALUATION OF A PDMP USE POLICY
87
IMPACT EVALUATION OF A PDMP USE POLICY
88
IMPACT EVALUATION OF A PDMP USE POLICY
89
IMPACT EVALUATION OF A PDMP USE POLICY
90
IMPACT EVALUATION OF A PDMP USE POLICY
91
IMPACT EVALUATION OF A PDMP USE POLICY
92
IMPACT EVALUATION OF A PDMP USE POLICY
93
IMPACT EVALUATION OF A PDMP USE POLICY
94
IMPACT EVALUATION OF A PDMP USE POLICY
95
IMPACT EVALUATION OF A PDMP USE POLICY
96
IMPACT EVALUATION OF A PDMP USE POLICY
97
IMPACT EVALUATION OF A PDMP USE POLICY
98
APPENDIX C
IMPACT EVALUATION OF A PDMP USE POLICY
99
APPENDIX D
Public Health IRB Approval Letter
IMPACT EVALUATION OF A PDMP USE POLICY
100
APPENDIX E
Southern Oregon IRB Approval Letter
IMPACT EVALUATION OF A PDMP USE POLICY
101
APPENDIX F
Pacific University IRB Exempt Application
IMPACT EVALUATION OF A PDMP USE POLICY
102
IMPACT EVALUATION OF A PDMP USE POLICY
103
IMPACT EVALUATION OF A PDMP USE POLICY
104
IMPACT EVALUATION OF A PDMP USE POLICY
105
IMPACT EVALUATION OF A PDMP USE POLICY
106
IMPACT EVALUATION OF A PDMP USE POLICY
107
APPENDIX G
Pacific University IRB De-Identified HIPAA Form
IMPACT EVALUATION OF A PDMP USE POLICY
108
IMPACT EVALUATION OF A PDMP USE POLICY
109
IMPACT EVALUATION OF A PDMP USE POLICY
110
APPENDIX H
Pacific University IRB Approval Letter
IMPACT EVALUATION OF A PDMP USE POLICY
111
APPENDIX I
Timeline
October 2013
• Formulation of partnership / PDMP chosen as subject
Nov-Jan 2013
• PDMP as screening tool discussed and chosen as topic
February-March 2013
• Chapter 1 drafted
March 2014
• Chapter 1 submitted For review to Pacific
• Research committee team personnel selected and formally joined
• IRB project account created for Pacific IRB
• Oregon Board of Pharmacy contacted to discuss committee membership and future
surveying of members
April 2014
• Literature review conducted
• Capstone committee formed
• Ky Fifer signed affiliation / CHP agreement to intern at [Organization Omitted]
May 2014
• Literature review submitted to Pacific for review
• Capstone committee form submitted to Pacific for review
IMPACT EVALUATION OF A PDMP USE POLICY
112
• Methodology submitted to Pacific for review
June 2014
• Initial PDMP Data Use Agreement submitted to OHA
• Revised Chapter 1 submitted to Pacific
• Revised literature review submitted to Pacific
• Draft of chapters 1-3 submitted to Carole Londeree for APA format review
July 2014
• Revised methodology submitted to Pacific
• Project proposal presentation at Pacific to faculty and 2nd year students
• Letter of support requested from OHA for Pacific IRB
• OHA notifies research team that data must be de-identified on OHA site
• Sponsoring pharmacy IRB gives oversight to Pacific IRB
August 2014
• Signed initial draft of Data use Agreement sent to OHA
• OHA reviews and outlines required updates for Data Use Agreement
• Chapters 1-3 reviewed and approved by Dr. Bobby Nijjar from Pacific
University
• Updated Data Use Agreement is sent to Pacific personnel and is signed
• Updated information uploaded to Pacific IRB processing website
• Initial submission to Pacific IRB submitted by Dr. Bobby Nijjar
• OHA IRB submitted via electronic submission process
• Further Data Use Agreement addendums are implemented by OHA personnel
IMPACT EVALUATION OF A PDMP USE POLICY
113
September 2014
• [Sponsor Omitted] letter of support from [sponsor Omitted] provided for IRB submission
• Sponsor letter of support uploaded to Pacific IRB
October 2014
• Data Use agreement signed by Pacific University
• CV and resumes submitted to Oregon Health Authority upon request
• New Oregon Health Authority Data Use Agreement addendums added via email
correspondence from Oregon Health Authority
• On-site meeting with Oregon Health Authority personnel to discuss Data Use Agreement
updates
• Pacific University IRB told to hold submission pending updated Oregon Health Authority
paperwork
• Oregon Health Authority Data Use Agreement, Initial Review Questionnaire, and HIPAA
forms drafted / submitted to Oregon Health Authority
• Asante grants IRB exemption
• Updated Data Use Agreement signed by Pacific research team
• CV’s, support letter, Data Use Agreement, Initial Review Questionnaire, HIPAA De-
identified Form, Southern Oregon IRB Exemption Letter submitted to Oregon Health
Authority
November 2014
• Additional correspondence with Oregon Health Authority for Data Use Agreement
• All pertinent documents submitted to Oregon Health Authority IRB
• OHA IRB determines project does not utilize human subjects and is formally approved
IMPACT EVALUATION OF A PDMP USE POLICY
114
December 2014
• Project formally submitted for review for Pacific IRB
January 2015
• Project formally approved by Pacific IRB
February 2015
• De-identified data is picked up from OHA by researchers