Keywords
risk evaluation and mitigation - health information interoperability - workflow -
REMS - patient safety
Background and Significance
Background and Significance
When safety risks for medications must be mitigated, the Food and Drug Administration
(FDA) can require a Risk Evaluation and Mitigation Strategy (REMS) program.[1] REMS programs are designed to promote patient safety by documenting and sharing
information that can be used to inform practice. Information collected through REMS
programs can provide quality observational data about the safety of medications in
real-world settings.[1]
[2]
[3]
[4] REMS programs include communication and risk management plans for certain drugs
to ensure that the benefits of prescription drugs outweigh their risk in real-world
settings. This program may be in place for a certain period or on an ongoing basis.[1]
[5]
The major components of a REMS program include a medication guide, communication plan,
and elements to assure safe use (ETASU).[6] The medication guide and communication plan are intended to inform patients and
health care providers on potential risks associated with the medication, respectively.
ETASU is used for medications that would otherwise be unavailable due to the serious
safety risks and are unique to each REMS program. Elements may include training and
certification for health care providers, certification of pharmacies or other entities
that dispense the medication, limits on the setting where the medication can be dispensed,
required documentation of safe conditions prior to dispensing the medication, ongoing
monitoring requirements for patients receiving the medication, or a registry for patients
receiving the medication.[7]
[8]
The FDA approved a total 297 REMS programs overall. As of January 2022, the number
of active REMS programs was 62, of which 13 required medication guide and 58 required
ETASU elements. ETASU requirements varied by program with 41 programs requiring dispenser
certification, 39 requiring prescriber certification, 38 requiring prescriber training,
and 29 requiring patient enrollment. Illustrating the complexity of staying current
with the REMS requirements, over the past 7 years, the number of newly approved REMS
programs ranged from 4 to 12 per year, the number of revised programs ranged from
3 to 17 revisions per year, and the number of modifications in these revised programs
ranged from 32 to 63 per year.[9]
REMS programs have significant administrative requirements for tracking and monitoring.
The time and complexity required to meet requirements have increased since program
inception in 2007 as there have been more involved ETASU.[10]
[11] Participation of multiple stakeholders is required for an effective implementation
of REMS programs, including but not limited to pharmaceutical manufacturers, prescribing
providers, pharmacists, support staff, and patients. Providers who prescribe medications
with REMS requirements carry a large burden. In addition to their already demanding
patient care workload, prescribers may need to obtain certification, understand REMS
requirements, enroll patients, ensure monitoring orders are placed, and safe use conditions
are met. Similarly, meeting REMS program requirements increases pharmacist workload.
A pharmacist may need to obtain certification for their pharmacy, ensure that providers
and patients are enrolled, obtain dispense authorizations prior to filling a medication,
provide required patient education, and monitor compliance with REMS requirements
within their pharmacy or health system. Both providers and pharmacists may document
REMS-related activities in their existing health information technology (IT) systems
(e.g., electronic health record [EHR], pharmacy dispensing platform/software, patient
portal) in addition to systems specifically designed for this purpose.[12] Double documentation involves both workflow inefficiencies and the possibility of
inaccurate documentation. In addition, stakeholders have reported confusion about
how best to meet the requirements and the parties responsible for fulfilling and documenting
requirements.[13] Thus, there is a range of ways to accomplish REMS requirements.
Some efforts have been made to address the administrative burden and alleviate the
confusion associated with REMS programs. The FDA convened a group of stakeholders
to obtain feedback on REMS programs, and standardizing REMS information in pharmacy
systems was identified as a high-priority initiative.[4]
[13] In addition, the National Council for Prescription Drug Programs (NCPDP) created
a REMS REFEREnce Manual, which outlines technical standards for allowing prescribing
systems to communicate with REMS administrators.[14] The Office of the National Coordinator for Health IT (ONC-IT) has also developed
guidance, yet adoption remains low.[15] A substantial need exists for standardization and automation of the multi-stakeholder
REMS process to reduce the burden on prescribers and improve medication safety and
access for patients.[4]
[8]
[13] Thus, we sought to characterize the barriers to REMS program implementation and
use, best practices, and reported program outcomes through addressing the following
questions:
-
What are common features of REMS workflows in practice?
-
What are the reported workflow processes and health informatics strategies used by
organizations to meet REMS programs requirements?
-
What are barriers and facilitators to implementation and use of REMS programs?
Methods
Scoping Literature Review
We conducted a scoping literature review following the Preferred Reporting Items for
Systematic Reviews and Meta-analyses guidelines. Literature searches were performed
in PubMed, Embase, and Web of Science. Searches were restricted to articles published
in English only from January 2007 to July 2021. We looked at articles describing workflows,
automation, or electronic data exchange as it relates to REMS programs. The search
was performed on July 30, 2021. The search terms are available in [Supplementary Appendix A] (available in the online version). The review was registered in PROSPERO (identifier:
CRD42021271944).
Eligibility Criteria
Two blinded study authors reviewed articles based on the article's title and abstracts.
A third author was available to resolve any study selection disputes. We included
journal articles that described workflows, automation, or electronic data exchange
related to REMS programs. We excluded articles that did not include a workflow change
and articles that did not measure outcomes from the workflow change. We included peer-reviewed
articles and conference abstracts. The studies included randomized control trials,
cohort studies, pre- andpostobservation studies, and cross-sectional studies.
Extraction and Data Analysis
Two blinded authors independently conducted the quality assessment and data extraction
for all included articles. A third author was available should consensus not be achieved
by the two authors. The authors used the Covidence (
http://www.covidence.org
) screening and data extraction tool. The following data were extracted: study identification,
methods, population, intervention, outcome variables, and p-values.
Results
Studies Included and Excluded
Initial search results included 299 citations. After removing duplicates and articles
not relevant to our aims, seven studies were included in the final review. [Fig. 1] depicts the flow of citation identification and study selection. Of the seven studies,[16]
[17]
[18]
[19]
[20]
[21]
[22] three were full-text articles,[18]
[21]
[22] and four were conference abstracts.[16]
[17]
[19]
[20]
[Tables 1] and [2] depict study design and purpose and implemented workflow process changes and outcomes,
respectively. While most identified studies focused on evaluating REMS program compliance
after a change in workflow processes,[16]
[18]
[19]
[20]
[21] one study assessed time efficiency in the adoption of a pharmacy REMS portal and
another tested the feasibility of a real-time pharmacy claims database to improve
the efficiency of claims adjudication.[15]
[17] The majority of studies were reports of quality improvement projects conducted within
medical centers.
Fig. 1 The flow of study identification and selection.
Table 1
Details and main purpose of included studies
|
Author (year)
|
Title
|
Study design
|
Setting
|
Main purpose
|
|
Adams et al[16] (2010)[a]
|
A Real Time Opioid Class REMS: Technical Feasibility
|
Cross-sectional
|
Mock pharmacy/claims database
|
To test the feasibility of using real-time pharmacy claims to verify patient, prescriber,
and pharmacy enrollment in a training database
|
|
Tsapepas and Morales-Castro[17] (2015)[a]
|
Innovations through Information Technology a Process for Mycophenolate Risk Evaluation
Mitigation Fulfillment
|
Cross-sectional
|
Hospital
|
To evaluate the impact of REMS compliance via multidisciplinary approach
|
|
Chan-Liston et al[18] (2019)
|
REMS Pharmacy Tasks: The Adoption of an Innovative Electronic Support System
|
Quasi-experimental, pretest-posttest
|
REMS-certified pharmacies dispensing lenalidomide, pomalidomide or thalidomide
|
To evaluate the pharmacy adoption rates of an online REMS Pharmacy Portal designed
as an alternative for REMS-certified pharmacies to perform mandatory pharmacy dispense
confirmations and to assess whether Pharmacy Portal uptake was affected by the pharmacy
daily dispense volume
|
|
Do et al[19] (2019)[a]
|
Improving Adherence to Pregnancy Screening in Patients on Teratogenic Medications
Using an Electronic Medical Record Alert System: A Quality Improvement Initiative
|
Quasi-experimental, pretest-posttest
|
Pediatric Rheumatology clinic at a single tertiary care center
|
To improve physician adherence in pregnancy screening for female patients on mycophenolate
|
|
Mahmoudjafari et al[20] (2019)[a]
|
Development of Processes to Ensure Timely Administration of Tocilizumab in the Setting
of Cytokine Release Syndrome after Administration of Tisagenlecleucel or Axicabtagene
Ciloleucel
|
Cross-sectional
|
Large academic medical center
|
To evaluate the implementation of standard policies and treatment algorithm to treat
the acute toxicities seen with tisagenlecleucel and axicabtagene ciloleucel
|
|
Prokes and Root[21] (2019)
|
A Retrospective Analysis of Adherence to Risk Evaluation and Mitigation Strategies
Requirements for Pulmonary Arterial Hypertension Drugs
|
Quasi-experimental, pretest-posttest
|
Large academic medical center
|
To assess Risk Evaluation and Mitigation Strategies (REMS) program compliance for
pulmonary arterial hypertension (PAH) drugs following the initiation of more rigid
protocols and informatics change
|
|
Kostrzewa[22] (2021)
|
Optimization of REMS Program Compliance in a Large Academic Health System
|
Cross-sectional
|
Large academic medical center
|
To evaluate the impact of EHR optimization on REMS program compliance
|
Abbreviations: EHR, electronic health record; REMS, Risk Evaluation and Mitigation
Strategy.
a Study available only as conference/congress abstract. Full-text article not available.
Table 2
Implemented workflow process changes and outcomes of included studies
|
Author (year)
|
Workflow change implemented
|
Groups
|
Outcomes
|
Conclusions
|
|
Adams et al[16] (2010)[a]
|
Utilization of a real-time pharmacy claims to verify patient, prescriber, and pharmacy
enrollment in REMS program at the time of adjudication: mock prescriber (2.2 M records),
pharmacy (700K), and patient (20 M) records were created; mock pharmacy claims (10,000)
were created; five test conditions simulating enrollment or nonenrollment were created
|
N/A
|
• Mock pharmacy claims (10,000) employing the test records and 5 test conditions resulted
in 100% accuracy in rejecting invalid claims and accepting valid claim
• Maximum processing time < 0.5 seconds
|
A system using pharmacy claims can process large volumes quickly and accurately. The
volume of claims processed (60/second) is ∼3-fold higher than would be required to
accommodate an ‘‘all opioid’' REMS (20/second) and ∼20-fold higher to accommodate
an ER opioid/methadone REMS (3/second). Assuming a REMS covering 30 million prescriptions,
the added pharmacy burden would be 4,000 hours versus 1.5 million hours for a web/IVR
system
|
|
Tsapepas and Morales-Castro[17] (2015)[a]
|
Development of a mycophenolate medication use guide to outline REMS requirements to
include:
(1) Prescriber enrollment
(2) Pregnancy testing
(3, 4) Patient education and patient/prescriber acknowledgment form
(5) Mycophenolate pregnancy registry
Guide served as framework to customize CPOE system with order-entry forms with CDS
to guide prescribers through REMS components for females
|
N/A
|
• Since implementation, 119 FCBP have listed for transplantation and 88% have completed
REMS program requirements
• Monthly compliance rates range from 60 to 100%
|
Since implementation, REMS program compliance has increased, including education,
patient-prescriber acknowledgment form, and pregnancy testing
|
|
Chan-Liston et al[18] (2019)
|
A self-service pharmacy portal (PP) was developed (December 2013) to allow REMS-certified
pharmacies to obtain confirmation numbers instead of calling a Customer Care Representative
(CCR) or using the Interactive Voice Response System (IVR) system
|
CCR or IVR vs. PP
|
• Change over time (quarterly) in percentage of pharmacy dispense confirmations obtained
via the PP (2014-Q2 - 67%, 2016-Q2: 98%, 2018-Q1: 98.5%)
• PP adoption rates were not affected by the dispensing volume of the pharmacy
|
REMS-certified pharmacies rapidly adopted PP use to verify patient and provider enrollment
and obtain dispense authorization numbers
|
|
Do et al[19] (2019)[a]
|
BPA alert to prompt ordering of a urine hCG test for female patients age 10 and older
prescribed mycophenolate
|
Before BPA vs. after BPA
|
• Urine hCG at encounter where mycophenolate was prescribed: 11% (8/71 encounters)
vs. 79% (114/144 encounters), p < 0.00001
• No urine hCG within 7 days of prescribing mycophenolate: 72% (51/71) vs. 3% (5/144),
p < 0.0001
|
Implementation of BPA significantly increased urine hCG pregnancy screening at encounters
where mycophenolate was prescribed
|
|
Mahmoudjafari et al[20] (2019)[a]
|
Annual pharmacy competency outside of the REMS requirements for all inpatient pharmacy
staff to ensure awareness for prompt administration of tocilizumab
PRN entry of tocilizumab released concurrently with cell infusion order readily available
on the patient's Medication Administration Record for immediate dispensing
“Prepare and give now” entry for tocilizumab within our toxicity treatment order
set in preparation of potential future admission
Attending required to call the inpatient pharmacy for verbal order to dispense dose
|
N/A
|
• Seven patients received at least one dose of tocilizumab (range 1 to 4 doses). Median
time from dispense to administration: 53.5 minutes
|
Education efforts and EHR modifications resulted in ability to dispense and administer
tocilizumab within 2-hour time frame dictated by REMS requirements
|
|
Prokes and Root[21] (2019)
|
For all four of the medications (macitentan, bosentan, ambrisentan, and riociguat):
1. Therapy restricted to adult pulmonary and cardiology, and pediatric cardiology,
neonatology, and critical care attendings and fellows enrolled in the respective REMS
program
2. All female patients must also be enrolled in the REMS program prior to initiation
(male patients only for bosentan)
3. Complete baseline LFTs (for bosentan)
4. Pregnancy tests for FCBP
5. Ordering processes altered to include order panels with the medication and necessary
laboratory tests
6. Providers must answer required questions before the order can be processed
7. Pharmacists must verify patient and prescriber enrollment in the respective REMS
programs, negative pregnancy tests for FCBP, and LFTs (for bosentan) via CDS alert
which fires upon pharmacist verification to perform this step and this is documented
in a pharmacy intervention within the EHR before dispensing the medication
|
Pre-informatics (n = 50) vs. post-informatics (n = 44)
|
• Primary endpoint: REMS protocol adherence (pregnancy tests performed within 30 days
of medication initiation for FCBP, LFTs ordered within 30 days of bosentan initiation,
and initiation of therapy order documented by an approved attending enrolled in the
REMS program) improved during the post-informatics period (71 vs. 95%; p = 0.07)
• Secondary endpoints:
Pregnancy tests performed within 30 days of medication order (36.4 vs. 100%; p = 0.01)
LFTs within 30 days of drug initiation (91.7 vs. 100%)
• Documentation of initiation of therapy ordered by a REMS-certified prescriber remained
unchanged (87.5 vs. 86%)
|
Initiation of more rigid ordering protocols for the endothelin receptor antagonist
(ERAs) (macitentan, bosentan, or ambrisentan) and riociguat improved pharmacist and
physician compliance with REMS requirements
|
|
Kostrzewa[22] (2021)
|
Six inpatient medications on health system formulary required action for REMS program:
alvimopan, ambrisentan, bosentan, clozapine, macitentan, riociguat. EHR optimization
included:
(1) Banner on EHR patient main page alerting provider patient is on a REMS medication
and prompts pharmacist to complete a note
(2) Note template to walk pharmacist through all required information for each unique
REMS program
|
N/A
|
• Results reported over 1-year period after EHR optimization go-live (August 2018
to July 2019)
• Of 777 inpatient admissions for 545 unique patients during which an inpatient REMS
medication was administered, 318 admissions had at least one note (56 admissions had > 1)
resulting in 40.93% compliance
|
EHR optimization can significantly influence REMS program compliance
|
Abbreviations: BPA, best practice advisory; CCR, customer care representative; CDS,
clinical decision support; CPOE, computerized provider order entry; EHR, electronic
health record; ER, emergency room; FCBP, female of childbearing potential; hCG, human
chorionic gonadotropin; IVR, interactive voice response; LFT, liver function tests;
PP, pharmacy portal; PRN, pro re nata (as needed); REMS, Risk Evaluation and Mitigation
Strategy.
a Study available only as conference/congress abstract. Full-text article not available.
Common Features of REMS Workflows in Practice and Strategies for REMS Program Compliance
The most common features and strategies reported in REMS workflow studies included
EHR optimization/customization,[17]
[19]
[20]
[21]
[22] customized order sets,[20]
[21] clinical decision support (CDS) alerts,[19]
[20]
[21] logging pharmacy interventions, or completing pharmacy notes to verify REMS compliance
with each order.[21]
[22] Other strategies include the development of a pharmacy portal to verify REMS enrollment,[18] requiring prescribers to contact dispensing pharmacies to verify order,[20] educating providers and pharmacists,[18]
[20] and prescriber restriction based on specialty.[21]
One study focused on the process of pharmacy claims adjudication[16] to determine the feasibility of integrating health care IT platforms to reduce claims
processing time. Mock patients, providers, pharmacies, and claims were created. Five
different test conditions simulating differing enrollment statuses were created and
assigned to claims. Four test conditions simulated one of three stakeholders as “not
enrolled (invalid claim),” and one test condition simulated all three stakeholders
as “enrolled (valid claim).” This is the only study identified that reported automation
of data exchange in REMS program processes, albeit in a mock setting. No study reported
integration with third-party vendors, REMS administrators, or pharmaceutical manufacturers
to automate data exchange per NCPDP standards to incorporate REMS processes into provider
and pharmacy workflow seamlessly.
All studies noted an improvement or benefit to the implemented workflow process changes,
though only three studies reported baseline metrics for comparison. None of the studies
mentioned time, resources, or costs incurred to develop and execute the reported workflow
process changes.
Bias and Heterogeneity
Standard analysis for bias was not conducted due to study design and low sample sizes.
Additionally, studies varied significantly in intervention, outcome reporting, duration
of evaluation, and follow-up. Therefore, quantitative comparison or meta-analysis
of included studies could not be performed.
Barriers and Facilitators
Reported facilitators to implementation included staff training and education through
in-person or live interactive training,[18] annual pharmacy staff competency,[20] or staff meetings, and EHR education documentation.[22] Other facilitators included ongoing oversight and maintenance by designated groups
or individuals, soliciting feedback from frontline staff,[22] and deploying an iterative Plan-Do-Study-Act quality improvement design.[19] Implementation barriers included lack of consistency among EHRs in how and where
female reproductive potential was documented, including the criteria used to define
reproductive potential,[21] attempting to tackle a large volume of REMS programs and changes instead of focusing
efforts, and overlooking the need to continually reeducate staff.[22]
Discussion
Our study aimed to review the literature describing the implementation of systems
and workflows to improve the process of REMS, including the design criteria of such
systems, evaluation metrics, reported outcomes, and research gaps. The impetus for
this review originated from the apparent gap in peer-reviewed literature that addresses
key REMS workflow implementation issues. The seven studies included in the final review
employed technology-based workflow interventions and showed numeric improvements in
the process outcomes of rates of adherence to REMS requirements, time to dispense
complex medications in an acute care patient setting, and successful adoption of an
online REMS pharmacy portal.
Two full-length manuscripts and one abstract described improvement in compliance with
REMS elements for specific medications after customizing existing CDS and computerized
physician order entry functionalities in the commercially available EHR (Epic, Verona,
Wisconsin, United States).[21]
[22] One abstract reported a similar result from a previously published intervention[23] that streamlined the ordering process using medication logic modules within the
EHR (Allscripts, Chicago, Illinois, United States). The functionalities used included
system restriction of ordering based on physician specialty, CDS popup alerts or passive
banners within the patient chart (known as Best Practice Advisories or BPAs), order
sets/panels, required hard-stop questions embedded within the medication order, and
links to medication information and medication use guidelines from the electronic
Medication Administration Record (MAR).[17] Another abstract reported the ability to use order sets, help text, and the MAR
within the EHR to ensure adherence to the REMS requirement of the timely administration
of tocilizumab in the setting of cytokine release syndrome after the infusion of chimeric
antigen receptor T-cell therapy.[20]
A study from 2010 reported the technical feasibility of using mock pharmacy claim
adjudication transactions to fulfill REMS patient, pharmacy, and provider enrollment
requirements.[16] An important step to streamlining REMS programs is defining electronic data standards
that support information exchange among the various REMS stakeholders.[24] The REMS program requirements for individual medications must be documented in a
standard, structured format and stored in a central repository. The FDA requires manufacturers
to submit some product information using the Structured Product Labeling (SPL) standard
approved by Health Level 7, but traditionally this has not included REMS information.
Beginning in 2022, manufacturers will also be required to submit REMS information
via the SPL standard.[25] NCPDP added REMS transaction standards to the SCRIPT Standard Implementation Guide
to support incorporating REMS within the existing e-prescribing workflow, including
REMS initiation and query requests and responses.[14] In addition, the NCPDP Telecommunication Implementation Guide includes standards
to facilitate REMS bidirectional communication between dispensing providers (e.g.,
pharmacies), reimbursement adjudicators, and REMS administrators and between these
entities and switches or intermediaries as determined by business needs.[26] Additional transactional standards are necessary for data exchange between EHRs
and REMS administrators for medications given in hospitals or clinics where NCPDP
standards do not apply. Adopting the NCPDP SCRIPT standards, the ONC-IT Interoperability
Standards Advisory identifies the standard calling to “allow prescribers to communicate
with a REMS administrator” as a standard for implementation by the industry but rates
the adoption level of this standard as none or minimal.[15] Our review did not find any studies describing or evaluating the implementation
of the REMS transactions components of the NCPDP SCRIPT or Telecommunications standards
in a clinical environment.
To reduce the administrative burden of dispensing providers, one study reported rapid
adoption of an online REMS portal created by the REMS pharmaceutical sponsor. This
portal allows REMS-certified pharmacies to obtain dispense authorization online instead
of calling a customer care representative or using the interactive voice response
system. This study demonstrates pharmacy's strong preference for more straightforward
mechanisms to interact with REMS.[18] Evaluations of automated REMS programs should measure improvements in workflow and
time savings over nonautomated REMS programs.
One study reported a lack of standardization across EHRs in the representation of
clinical concepts needed to determine if REMS elements are required. Endothelin receptor
antagonists and soluble guanylate cyclase inhibitors (e.g., riociguat) are teratogenic
medications with REMS programs that require monthly pregnancy tests for females of
child-bearing age. The authors found a lack of consistency between EHRs in determining
child-bearing age, with one study using 56 years as the upper cutoff and another used
59 years.[21] Others have reported that EHR vendors do not supply CDS tools to fulfill REMS requirements,
resulting in the need for each institution to allocate resources to customize their
systems.[23] The American Pharmacists Association (APhA) convened a large group of REMS stakeholders
in 2010 who discussed that lack of REMS standardization could lead to reduced patient
access and limited provider participation. The group recommended that all stakeholders
(patients, prescribers, pharmacists, manufacturers, wholesalers, and system vendors)
work collaboratively with the FDA to standardize REMS programs.[27]
In 2015, the FDA launched the “Common REMS Platform” initiative to provide a set of
open, electronic data standards to be used by technology providers and REMS programs
to reduce the burden of REMS compliance. The proposed common REMS platform intends
to integrate REMS requirements into health IT systems within the workflow and support
interoperability and streamlined communication between stakeholders.[24] A group of experts in 2016 who participated in a cooperative FDA workshop identified
opportunities in the identification of data elements, refinement of the REMS process
steps, and facilitating interoperable data exchange.[28] None of the studies that matched our inclusion criteria discussed using the FDA
REMS Platform as a reference for implementation..
Low adoption of the NCPDP SCRIPT REMS data exchange standards recommended by ONC-IT
and slow progress on the development of the Common REMS Platform are due to several
factors. An important factor is the large number of stakeholder entities involved
in the REMS process, including the FDA, pharmaceutical manufacturers, REMS administrators,
prescribing providers, dispensing pharmacies, drug distributors, prescription benefit
management companies, and e-prescribing and health IT vendors, each with their systems
and processes. A second factor is the complexity of the drug distribution system,
as not every medication product follows the traditional prescribing from a provider
to a community pharmacy for the patient to pick up and administer. A growing number
of medications can only be dispensed from certain specialty pharmacies, and other
medications are for in-clinic administration, only bypassing the e-prescribing transactions.
REMS data standards should be flexible enough to accommodate the complexity and variability
of the drug distribution channels. Other barriers include different information systems
used, inconsistent interpretation of requirements, and the lack of a sustainable funding
model to pay for the required costs. The APhA REMS stakeholders discussed multiple
opinions on which entities should financially contribute to the cost of REMS activities,
but they did not reach a consensus.[27]
Our scoping literature review is not without limitations. The small number of studies
that met the criteria of our scoping review revealed the scarcity of peer-reviewed
literature addressing the comprehensive evaluation of the outcomes of REMS implementations.
This scarcity of literature does not imply that provider organizations are not conducting
evaluations of REMS programs as such evaluations might be occurring, and even may
be shared through various forums, but the organizations may lack the resources, expertise,
or the incentive to undertake the peer-review submission process. The included studies
in our review demonstrated a lack of consistency in their outcome metrics, comparator
groups, and methods to reduce research bias. Using a comparator or baseline groups
to study the impact of implementing a REMS improvement and employing techniques such
as cluster randomization and interrupted-time series analysis could minimize bias
in future studies.[29] An ongoing study designed by the FDA and health system collaborators uses a multimodal
approach to evaluating REMS programs.[30] This multiyear study aims to understand patterns of drug use, adherence to safety
requirements, and patient outcomes under REMS programs, and the experience of physicians
and patients will be incorporated into the outcomes. While this study addresses a
gap in the literature, its published protocol does not include outcomes related to
the workflow, time burden, data quality, and does not plan to incorporate the experience
of pharmacists who are essential stakeholders in the REMS process.
One strength of our review is that, to our knowledge, it is the first comprehensive
review of the literature identifying key workflow issues for health system to consider
as they implement and evaluate REMS programs. It also demonstrates the need for more
research assessing the impact of REMS-related workflow process changes and emphasizes
that future research should be of more rigorous methodologies. A collaborative approach
for the adoption of the REMS data exchange standards and development of automated
REMS programs would need to include all stakeholders and government regulators to
ensure that systems requirements and implementation plans are patient-centric, flexible
to accommodate the complexity of drug distribution, can meet clinical safety and regulatory
needs, and have a sustainable funding model to support the development, maintenance,
and operational costs. The adoption of electronic medical records was greatly accelerated
by targeted funds for adoption and meaningful use. Perhaps a similar infrastructure
funding is needed to stimulate the development and adoption of automated REMS systems
with meaningful outcome metrics.
This comprehensive scoping literature review provides an opportunity to review key
issues that should be considered when performing a standardized evaluation to measure
improvements in patient safety for REMS programs. REMS programs can be a burden to
health systems and providers, and effective methods for addressing REMS program requirements
remains unclear. Health systems invest heavily in resources and personnel to implement
REMS programs and remain compliant with the requirements with little or no assessment
of its impact on patient outcomes.
Conclusion
This scoping review found limited high-quality evidence describing REMS workflows,
automation, and electronic data exchange. Strategies to address REMS program requirements
included interventions such as developing order sets, CDS alerts, a centralized pharmacy
portal, and manual interventions or education. The wide variation in reported strategies
reflects the complex and disconnected nature of REMS programs. While there have been
significant efforts by the FDA, standard development organizations, and other stakeholders,
additional efforts are needed to automate REMS programs.
Clinical Relevance Statement
Clinical Relevance Statement
This is the first scoping review summarizing workflows, automation, or electronic
data exchange associated with REMS programs. There is a lack of high-quality evidence
describing methods to automate and streamline REMS program requirements despite efforts
by the FDA's Common REMS Platform and other stakeholders. Additional efforts are needed
to automate REMS programs.
Multiple-Choice Questions
Multiple-Choice Questions
-
What interventions to improve REMS workflows have been reported in recent studies?
-
Developing an API between an EHR and a REMS administrator registry.
-
Incorporating data exchange standards into clinical workflows.
-
Implementing a SMART on FHIR application in the EHR.
-
Creating institution-specific clinical decision support in the EHR.
Correct answer: The correct answer is d. This study found that localized EHR changes are the most
common method for incorporating REMS program requirements into clinical workflows.
No studies evaluating data exchange in a clinical environment were found.
-
Select the best statement about REMS data standards
-
REMS transaction standards need to be created to facilitate data exchange between
e-prescribing systems and REMS administrators.
-
NCPDP provides standards for data exchange between healthcare settings where REMS
medications are administered (e.g., hospitals) and REMS administrators.
-
NCPDP REMS standards were recognized by ONC-IT ISA but adoption in practice has been
slow.
-
REMS program requirements are available in Structured Product Labeling (SPL) format
for all medications with a required REMS program.
Correct answer: The correct statement is c. NCPDP REMS standards were adopted by the ONC-IT ISA
but adoption in practice has been slow. These standards address bidirectional communication
between electronic prescribing systems or retail pharmacies and REMS administrators.
Data exchange between health care settings where medications are administered (e.g.,
hospitals) and REMS administrators is outside the scope of NCPDP standards. The FDA
will require REMS program information in SPL format beginning in 2022.
