https://orcid.org/0000-0001-5043-6206
Figures
Abstract
Purpose
The US Food and Drug Administration (FDA) Amendments Act of 2007 authorized the FDA to require risk evaluation and mitigation strategy (REMS) programs for drugs with important safety concerns. REMS can have elements to assure safe use (ETASU), such as patient registries, dispensing restrictions, and physician training and certification requirements. We aimed to understand physician experiences with and perceptions of a selection of ETASU REMS.
Methods
Physicians prescribing 1 of 4 ETASU REMS-covered drugs: natalizumab, riociguat, sodium oxybate, and vigabatrin.
Data collection/Extraction methods
Qualitative content analysis to summarize physician responses to open-ended questions.
Results
Of 31 physicians (14 female), 6 prescribed riociguat, 6 vigabatrin, 7 sodium oxybate, and 12 natalizumab (5 for Crohn’s disease, 7 for multiple sclerosis), most demonstrated good understanding of the rationale for and requirements of the ETASU REMS but believed that the programs had limited effect on clinical practice. Some physicians reported that the ETASU REMS made them more comfortable with prescribing covered drugs due to heightened oversight, facilitated discussions about treatment, and were likely more beneficial for non-specialists. Concerns were raised about the administrative effort needed to comply with the programs and the potential misuse of patient health information transmitted to manufacturers.
Citation: Sarpatwari A, Brown BL, McGraw SA, Dejene SZ, Abdurrob A, Kesselheim AS (2023) Physician experiences with and perceptions of risk evaluation and mitigation strategy programs with elements to assure safe use. PLoS ONE 18(7): e0288008. https://doi.org/10.1371/journal.pone.0288008
Editor: Rafael Van den Bergh, World Health Organization, BELGIUM
Received: February 10, 2023; Accepted: June 16, 2023; Published: July 6, 2023
Copyright: © 2023 Sarpatwari et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting information files.
Funding: This project was funded by a Greenwall Foundation Making a Difference in Real-World Bioethics Dilemmas Grant. Dr. Sarpatwari and Dr. Kesselheim’s work is also supported by grants from Arnold Ventures and the FDA (Contract Number 75F40120C0044).
Competing interests: The authors declare no conflict of interest.
1. Introduction
The Food and Drug Administration (FDA) approves new prescription drugs if the benefits appear to outweigh the risks. However, some new drugs have severe risks for certain populations, such as pregnant women [1, 2], while other new drugs have risks that can be reduced with better surveillance, such as regular lab tests [3, 4]. To ensure that these drugs can still be FDA-approved and made available for the patients who might benefit from them, Congress gave FDA the authority to require manufacturers to set up risk evaluation and mitigation strategy (REMS) programs [5]. Some REMS programs consist of a medication guide and a communication plan to instruct patients and physicians on safe drug use. Others include additional elements to assure safe use (ETASU), which can include patient registries, dispensing restrictions such as requiring evidence that a patient is taking birth control or has received the appropriate screening blood tests, and training and certification of prescribing physicians and dispensing pharmacies [6]. As of May 24, 2023, 63 FDA-approved drugs or drug classes had REMS programs, of which 59 (94%) had ETASU, while only 2 (3%) had a medication guide and/or a communication guide alone [7].
Despite the prevalence of ETASU REMS, in a 2013 report, the US Department of Health and Human Services Office of the Inspector General cast doubt on their effectiveness [8]. The evidence to date has been mixed. One review of extended-release/long-acting opioids found that between 2012 and 2016 only 28% of prescribers had taken the continuing education tied to the drugs’ ETASU REMS [9]. However, a different study found a lower incidence of opioid overdose among Medicaid beneficiaries following the program’s implementation (which the authors acknowledge could not be disentangled “from many other ongoing initiatives with similar goals”) [10].
Varied findings have also been reported for the impact of ETASU REMS on physician prescribing practices. An investigation of the thrombopoietin-agonist eltrombopag (Promacta) observed decreased off-label use while the drug was covered, although that indication was later FDA-approved [11]. A study of transmucosal immediate-release fentanyl (TIRF) drugs that were supposed to be restricted to opioid-tolerant patients found that by 60-months post ETASU REMS implementation, 35%-55% of patients met this criterion and 34% of surveyed prescribers had prescribed these drugs inappropriately [12]. However, a separate investigation of Medicare Part D beneficiaries reported that between 2010 and 2014, the TIRF ETASU REMS was associated with a sustained decrease in the percent of prescriptions among patients without known opioid tolerance [13]. Finally, a study of the ETASU REMS for erythropoiesis stimulating agents darbepoetin alfa and epoetin alfa did not find a change in non-evidence-based initiation of either drug following initiation of the program [14].
Given the uncertainty over the impact of these programs, we sought to better understand physicians’ experiences with and perceptions of the operation of ETASU REMS in clinical practice.
2. Methods
We conducted a descriptive phenomenological study, using semi-structured guides to interview a cohort of physicians who prescribed ETASU-REMS covered drugs, and qualitatively analyzed their perceptions and experiences to learn how these programs can be better designed and implemented.
To select the programs being examined, we ranked all 40 ETASU REMS existing as of January 2016 on the rigor of their requirements, including: (1) prescriber certification (score 0 if no, 1 if yes), (2) monitoring or testing (0 if no, 1 if recommended, 2 if mandated, 3 if data reporting requirements), (3) administration or dispensing restrictions (0 if no, 1 if certification required, 2 if only limited specialty pharmacies or centers could qualify), and (4) manufacturer access to patient health information (0 if no, 1 if yes). We selected four with scores higher than the median: natalizumab (Tysabri), riociguat (Adempas), sodium oxybate (Xyrem), and vigabatrin (Sabril). Table 1 provides details on the indications and risks of these drugs, and on their monitoring and testing requirements.
ETASU = elements to assure safe use, PML = progressive multifocal leukoencephalopathy, REMS = risk evaluation and mitigation strategy, * = Removed in June 2016.
2.1. Interview guide, physician recruitment, and conduct of interviews
We developed a semi-structured interview guide with three primary topics: utility and burden; privacy; and recommended improvements (S1 Appendix in S1 File). To minimize bias, we posed open-ended questions. We tested the guide in pilot interviews and made minor revisions to it for clarity based on our experiences.
Eligible participants must have prescribed the drug of interest within two years of their interview. To recruit participants, we identified physicians associated with national patient support groups for diseases treated by the drugs. Aiming to assemble a diverse group, we intentionally sought out physicians from states across the country. Invitation emails to these physicians included a fact sheet providing an overview of the study and its goals.
Interviews were conducted between 2016 and 2017 by two research assistants, AA (male, BA) and SZD (female, BA), who were trained through practice interviews led by AS (male, PhD/JD). In each interview, the interviewer noted his or her background and spoke alone with the participant. Interviews were approximately one hour, conducted over the phone, recorded, and transcribed. Field notes were taken as needed. Owing to resource constraints, repeat interviews were not conducted, nor were transcripts returned to participants for comment or correction. Participants completing an interview received a $150 gift card honorarium. A target of at least 5 interviews per drug were sought, with interviews stopped once data saturation was reached.
2.2. Content analysis
Reporting of study methods and results adhered to Consolidated Criteria for Reporting Qualitative Research guidelines [15]. We employed qualitative content analysis to summarize physician responses to the open-ended questions [16, 17]. These responses can be coded using deductive categories defined a priori as well as categories derived inductively, based on concepts newly identified through constant comparison or by line-by-line reading of the interview text [18]. After reading the transcripts, the research team generated a consensus start list of codes. One person (SAM) then coded each transcript using NVivo, meeting with other members of the team to review and refine the list and code definitions (S2 Appendix in S1 File). Finally, the research team reviewed the data to identify themes within and across codes. Again, owing to resource constraints, participant feedback on findings was not sought. This study was approved by the Brigham and Women’s Hospital Institutional Review Board.
3. Results
Among 31 physician participants, 6 prescribed riociguat, 6 vigabatrin, and 7 sodium oxybate (Table 2). Five physicians prescribed natalizumab for Crohn’s disease and 7 for multiple sclerosis. Fourteen were female, 13 practiced in Massachusetts, and 28 worked at academic medical centers.
3.1. Knowledge of ETASU REMS
Participants demonstrated good knowledge of the ETASU REMS. Most understood why the FDA required the programs for the drugs that they prescribed. For example, one participant (physician 13, natalizumab) explained, “My understanding of this program is that it was mandated by the FDA for the rerelease of the drug,” which had been removed owing to the risk of progressive multifocal leukoencephalopathy, “to limit potential, maybe to limit irresponsible or inappropriate use of the agent.” Most participants also knew the requirements of the programs (Table 3). Summarizing the ETASU REMS for vigabatrin, one participant (physician 7) stated, “Most physicians…have to register is one component of it. Then patients…have to fill out recognition or information about the potential risk of the medication. Plus, they have to have regular examinations before treatment by ophthalmologists to exclude one of the significant potential side effects of the renal toxicity.”
3.2. Impact on practice
The dominant perspective was that the ETASU REMS had limited impact on clinical practice (Table 3). Reasons for this perspective included that participants were already familiar with the benefits and risks of the drug and believed that requirements for educating, consenting, and monitoring or testing patients were steps they would have taken anyway. One participant (physician 27, sodium oxybate) estimated the effect on his willingness to prescribe at “zero percent,” while another (physician 3, riociguat) said “that’s very much a standard list of things we go through.” Another participant (physician 26, sodium oxybate) noted, “I have to say there is nothing that has really popped up in the Xyrem REMS Program where I was like, ‘Wow, I would have missed that.’”
Participants highlighted benefits, such as extra regulatory oversight, which made some more willing to prescribe the covered drugs. One participant (physician 5, riociguat) explained that without its monitoring system and central pharmacy, “the onus would be upon me and my team to have a more active role” Another participant (physician 30, sodium oxybate) thought the ETASU REMS encouraged more frequent formalized discussions with patients: “I’m more often having discussions, particularly…about alcohol…I’m doing it in a formal way rather than saying it once and assuming that nothing changes and everybody remembers it.” A second participant (physician 17, natalizumab) commented that this discussion was better “for the patient and the patient-physician relationship.” A few participants felt that the ETASU REMS requirements presented minimal hassle. One participant (physician 24, natalizumab) said, “It virtually takes 60 seconds to do the forms,” while another commented that the requirements, though time-consuming, were reasonable.
Several participants, however, questioned whether the ETASU REMS were worthwhile. Some were skeptical that the programs protected safety. One participant (physician 21, natalizumab) observed that although the ETASU REMS required documentation of the completion of monitoring tests, the program did not ensure that the testing was done correctly or that the results were accurate. Many participants also believed that the ETASU REMS imposed burdens on their practice from increased paperwork, patient monitoring or testing, and communication between the practice, pharmacies, and patients. Some participants were bothered by needing to respond to queries about adverse events from the sponsor. One participant (physician 27, sodium oxybate) called such requests a “nuisance,” because the manufacturer asked the prescriber to fill out “more than a page” of paperwork and “classified everything as an adverse event.” Some participants expressed frustration over the additional work required to submit prescription orders when the ETASU REMS did not link to electronic health record systems.
3.3. Specialists vs. non-specialists
Several participants believed that the ETASU REMS were particularly beneficial for physicians who did not have extensive experience treating the disease for which the covered drug was indicated (Table 4). As one participant (physician 5, riociguat) commented, “Ultimately, I do worry that certain caveats of safety can be missed without a REMS program in a community setting where the practitioner only sees 1 or 2 patients with pulmonary hypertension either per-year or average.”
Yet some participants reported that the ETASU REMS drove patients to specialists. One participant (physician 13, natalizumab) explained: “I think that if you’re an average, practicing gastroenterologist without an interest in ulcerative colitis or Crohn’s disease…the requirement may make this a non-starter and would say, ‘I’m going to send you to the university for evaluation.’” This participant also noted that the majority of his patients were referred by another gastroenterologist. Other participants commented that the additional work associated with the ETASU REMS may be too burdensome for community physicians.
3.4. Data collection
Some participants identified potential benefits of program-related data collection to the field and their practice (Table 4). For example, one participant (physician 23, natalizumab) noted that mandatory data collection meant that the denominator was known. However, many participants expressed some unease about a requirement for patients to grant pharmaceutical manufacturers access to personal health information, which was part of the enrollment forms for natalizumab and vigabatrin. First, some participants argued that pharmaceutical manufacturers were not involved in patient care and that their use of such data would likely not benefit patients. Second, concerns were raised about potential breaches of confidentiality, though one participant (physician 24, natalizumab) noted that since the precedent for data collection by third parties had already been set by insurers, “whether or not the manufacturer gets this information doesn’t really matter.” Finally, some participants recognized the need for limited data collection but were concerned that the patient release statements were unnecessarily broad. Underlying such views was the perception that pharmaceutical companies were using the data for marketing. For example, one participant (physician 6, riociguat) called required data sharing a “business strategy.”
4. Discussion
The rationale for and requirement of the ETASU REMS were well known by physicians prescribing 4 drugs subject to these requirements. Some physicians reported greater comfort prescribing covered drugs on account of this oversight, and some believed the programs promoted physician-patient discussion. However, concerns were raised about the administration time required for compliance and potential improper use of patient health information.
Although their primary purpose is to reduce drug risks, ETASU REMS may have other benefits. By promoting better communication with patients, these programs may help strengthen the physician-patient relationship, increasing patient trust and shared decision making. There is some support for this conclusion in the medical literature. In a survey of Kaiser Permanente Southern California health care providers who had prescribed 1 of 21 REMS-covered drugs between January and June 2013, a majority of oncologists, primary care physicians, and other non-pain management or surgical specialists reported that REMS had a moderate or substantial impact on their interactions with patients [19]. Additionally, in our previous study on patient experiences with the same ETASU REMS as in the present study, we found that the programs made some patients more comfortable following their physicians’ advice [20].
The time and resources needed to meet requirements were a frequent point of discussion in the interviews. The FDA is currently partnering with the MITRE Corporation to develop “an open source proof-of-concept prototype” to better integrate REMS programs into existing health information systems and health care delivery processes, including electronic prescribing [21]. In addition to this initiative, the FDA can require programs to have online data entry portals instead of faxed or mailed submissions, while drug manufacturers can compensate physicians for reporting required clinical information. Such payment would have to be commensurate with effort so as not to create a perverse incentive to prescribe an ETASU REMS-covered drug.
Steps can also be taken to mitigate concerns raised by physicians about data privacy. These concerns mirrored those expressed by patients [14] and can be mitigated by requiring manufacturers to offer clearer explanations about what patient health information can be collected under ETASU REMS, for what purposes, and over what time. Direct FDA administration or required routine third-party audits of these programs might help reassure physicians that the data collected is being properly used for purposes such as pharmacovigilance.
Limitations of our study include physician participants from just 4 ETASU REMS. However, these programs shared core elements with other programs. Second, most physicians interviewed were from Massachusetts and academic medical centers. Third, our use of patient support groups might have resulted in skewed recruitment of specialists. Their views and experiences therefore may not have been representative of all physicians who prescribe these drugs. Work is currently underway to understand a broader range of physician experiences with REMS programs [22].
Our findings suggest that physician knowledge about ETASU REMS was strong. For some physicians, the programs provided reassurance and facilitated discussions about treatment. However, reforms are needed to ensure more efficient implementation and better protection of patient health information.
References
- 1. Tkachenko E, Singer S, Sharma P, Barbieri J, Mostaghimi A. US Food and Drug Administration reports of pregnancy and pregnancy-related events associated with isotretinoin. JAMA Dermatol. 2019;155(10):1175–9.
- 2. Rostas S, Kim M, Gabardi S. Risk evaluation and mitigation strategies: a focus on the mycophenolic acid preparations. Prog Transplant. 2014;24(1):33–6. pmid:24598563
- 3. Mijovic A, Clozapine-induced agranulocytosis. Ann Hematol. 2020;99(11):2477–2482. pmid:32815018
- 4. Parham JS, Goldberg AC. Mipomersen and its use in familial hypercholesterolemia. Expert Opin Pharmacother. 2019;20(2):127–131. pmid:30526168
- 5.
21 USC § 355–1.
- 6. Sarpatwari A, Avorn J, Kesselheim AS. Using a drug-safety tool to prevent competition. N Engl J Med. 2014;370(16):1476–8. pmid:24738666
- 7.
US Food and Drug Administration. Approved risk evaluation and mitigation strategies (REMS). Accessed June 22, 2023. https://www.accessdata.fda.gov/scripts/cder/rems/index.cfm.
- 8.
US Government Accountability Office. FDA lacks comprehensive data to determine whether risk evaluation and mitigation strategies improve drug safety. Published February 12, 2013. Accessed June 22, 2023. https://oig.hhs.gov/oei/reports/OEI-04-11-00510.asp.
- 9. Heyward J, Olson L, Sharfstein JM, Stuart EA, Lurie P, Alexander GC. Evaluation of the extended-release/long-acting opioid prescribing risk evaluation and mitigation strategy program by the US Food and Drug Administration: a review. JAMA Intern Med. 2020;180(2):301–9. pmid:31886822
- 10. Esposito DB, Cepeda MS, Holick CN, Knox C, Desai VC, Liu N et al. Incidence of opioid overdose among patients using ER/LA opioid analgesics before and after implementation of the class-wide opioid risk evaluation and mitigation strategy. J Pain Res. 2020;13:157–9. pmid:32021405
- 11. Sarpatwari A, Franklin JM, Avorn J, Seeger JD, Landon JE, Kesselheim AS. Are risk evaluation and mitigation strategies associated with less off-label use of medications? The case of immune thrombocytopenia. Clin Pharmacol Ther. 2015;97:186–93. pmid:25670524
- 12. Rollman JE, Heyward J, Olsen L, Lurie P, Sharfstein J, Alexander GC. Assessment of the FDA risk evaluation and mitigation strategy for transmucosal immediate-release fentanyl products. JAMA. 2019;321(7):676–85. pmid:30778596
- 13. Fleischman W, Auth D, Shah ND, Agrawal S, Ross JS. Association of a risk evaluation and mitigation strategy program with transmucosal fentanyl prescribing. JAMA Netw Open. 2019;2(3):e191340. pmid:30924899
- 14. Sarpatwari A, He M, Tessema FA, Gagne JJ, Kesselheim AS. Changes in erythropoiesis stimulating agent use under a risk evaluation and mitigation strategy (REMS) program. Drug Saf. 2021;44(3):327–35. pmid:33206339
- 15. Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. Int J Qual Health Care. 2007;19(6):349–57. pmid:17872937
- 16. Hsieh HF, Shannon SE. Three approaches to qualitative content analysis. Qual Health Res. 2005;15(9):1277–88. pmid:16204405
- 17. Cho J, Lee EH. Reducing confusion about grounded theory and qualitative content analysis: similarities and differences. Qualitative Rep. 2014;19(32):1–20.
- 18. Curry LA, Nembhard IM, Bradley EH. Qualitative and mixed methods provide unique contributions to outcomes research. Circulation. 2009;119(10):1442–52. pmid:19289649
- 19. Rashid N, Cheetman TC, Aranda G Jr, Guerrero VN, Bartell H, Coghlan RJ. A descriptive evaluation of prescriber perceptions and experience with risk evaluation and mitigation strategy (REMS) programs in an integrated health care system. Ther Innov Regul Sci. 2015;49(4):503–10. pmid:30222443
- 20. Sarpatwari A et al. Patient and caregiver experiences with and perceptions of risk evaluation and mitigation strategy programs with elements to assure safe use. JAMA Netw Open. 2022;5(1):e2144386. pmid:35050352
- 21.
Confluence. Risk evaluation and mitigation strategies (REMS) integration. Last updated May 17, 2023. Accessed June 22, 2023. https://confluence.hl7.org/display/COD/Risk+Evaluation+and+Mitigation+Strategies+%28REMS%29+Integration.
- 22. Sarpatwari A, Mitra-Majumdar M, Bykov K, et al. A multi-modal approach to evaluate the impact of risk evaluation and mitigation strategy (REMS) programs. Drug Saf. 2021;44(7):743–51. pmid:33904111