In this issue of Anesthesia & Analgesia, Morey et al.1 suggest that regulatory mechanisms are not aligned with patient care in their Special Article entitled “Joint Commission and Regulatory Fatigue/Weakness/Overabundance/Distraction: Clinical Context Matters.” They suggest that overregulation or poor regulation interferes with health care provider activity to the detriment of their patients’ welfare. The authors further recommend that regulatory decisions should use a new stepwise scale representing standards-of-evidence. To address these concerns, I will examine the foundational authority for regulatory agencies, their fundamental limitations, and how regulatory bodies function, paying particular attention to how they do or do not remain linked to the realities of clinical practice. In addition, I will consider the recommendations of Morey et al. and suggest additional measures that may have a more timely impact.
Morey et al. discuss 4 authorities: the US Pharmacopeial Convention (USP), the Food and Drug Administration (FDA), the Joint Commission (JC), and the Centers for Medicare and Medicaid Services (CMS), each with complementary roles in health care regulation. The USP and JC are private, nonprofit, industry standards organizations that define benchmarks of quality and performance. The USP consolidates expert opinions regarding pharmaceutical quality, substance definition, and characterization.a The JC is one of the several accrediting organizations established to define consistent standards of clinical care to ensure quality.b Compliance with industry standards developed by organizations such as JC does not imply delivery of the highest quality of clinical care, but rather serves as a floor, beneath which performance may suffer. The role of industry standards has been recognized by the CMS and FDA.2,3
Congress empowered the FDA to regulate medical products that travel in interstate commerce.4 The FDA was intended to represent patient interests by preventing incidents where legally marketed medical products proved to be unsafe. In the 1962 Kefauver-Harris Amendment to the Food, Drug and Cosmetic Act, Congress determined that a lack of scientifically demonstrated clinical benefit also implied that a drug was inherently unsafe.5 The FDA has the responsibility to conduct unbiased premarket review of medical products using scientific evidence presented by the product sponsors. Methodological changes to the Act are made by Congressional Amendments or through administrative law (“regulations”) written by the staff of the Federal Executive Branch of government.6
The CMS was established in 1961 to conduct an independent review of medical products to determine the value of compensation for legally marketed products paid by Medicare or Medicaid.7
Ideally, clinical aspects of a new medical product are reviewed in detail by FDA physicians with substantial practical experience in the use of the same types of products. However, the FDA does not always have a subject matter expert for every medical specialty. Furthermore, physician reviewers at the FDA are not required to maintain clinical competence in their specialty, although they have the option to do so. Moreover, a regulatory decision maker is typically not the subject matter expert for the product under review and may not even be a clinician. This sometimes results in a different conclusion with respect to safety and clinical benefit than that reached by clinician specialists. In my opinion, this is the main reason regulatory decisions are sometimes out of alignment with the perspective of physicians who use a regulated medical product.
Outside expertise is sometimes enlisted to advise FDA decisions. However, these recommendations are not binding and may not be followed if they conflict with precedent decisions. Therefore, outside experts need to be well informed of FDA legal constraints and understand the regulation of medical products.
For outside experts to participate in regulatory review, they must distinguish between the various levels of evidence needed by the FDA. Approval of drugs is based on a different standard than medical devices because the uncertainty of devices is readily assessed using standard engineering principles, whereas the uncertainty of drugs is far more complex because of the potential for unexpected biological interactions. Therefore, novel devices are inherently more predictable than novel drugs. New drugs are approved when industry sponsors present “substantial evidence” of safety and efficacy.8 Substantial evidence typically requires treatment success from replicated clinical studies that are rigorously controlled and use strong methods to mitigate potential bias.9 The findings that the FDA reviews associate outcome measures with specific patient identifiers so that the data may be independently verified rather than being based solely on summary information reported in medical literature. Verification of each data point is one feature of an FDA review that also distinguishes its analysis from that conducted by private nonprofit review organizations. Other aspects of a drug that may impact use are evaluated, including noncompliance, risk of abuse, and product packaging. Human factors involved in use conditions are routinely considered. Standardized color coding or shape of drug containers within a drug class or drugs used by a medical specialty may be considered. However, regulatory bodies also evaluate unintended consequences such as the potential for inattention to expiry dates if the container label was not used for drug identification. Requiring a change in shape of a drug container may raise costs so much for a generic drug manufacturer that a product is discontinued, leading to shortages in supply. Data demonstrating an improved risk/benefit relationship are generally needed to support a compulsory regulatory action.
In contrast to drug approval, a high-risk medical device may be approved when data support “reasonable assurance of safety and effectiveness.”10 FDA approval usually requires data from one rather than replicated clinical studies. For practical reasons, a clinical device study may not be as well controlled, or effectively mitigate bias, as a clinical drug study. In the case of most medical devices having only moderate risk, the FDA can “clear” the new device if information is presented to demonstrate substantial equivalence to a legally marketed device. These devices must be labeled with an indication statement that does not change its intended use compared with an earlier legally marketed (“predicate”) device with respect to safety or effectiveness.11
Information supporting substantial equivalence need not always include clinical data. Many mechanical ventilators are examples of such devices because the technology and treatment effects are well understood.12 If the indication for the device changes, perhaps when it is used for a new patient population, the standard of evidence will likely increase, including clinical evaluation in the affected patient population. For example, high-frequency oscillatory ventilators were distinguished from earlier mechanical ventilators so that premarket clinical data became necessary for review.13
Morey et al. used the example of glucose monitors, stating “the FDA issued further guidance (nonbinding recommendations) in 2014 for meters used in hospitals that tightened the accuracy requirements and reiterated that these meters should not be used with critically ill patients.”
Why would the FDA issue such a statement for patient population that would seemingly benefit from having their blood glucose measures immediately available? The reason is that the FDA is compelled to make their decisions based on data. Let us consider 2 hypothetical situations for illustration.
In the first scenario, the user is a patient with diabetes mellitus vested in his or her care and able to accept the responsibility of daily monitoring at home. Morey et al. imply that lay users are unsophisticated because they include “elderly patients with little education.” In fact, these patients have been deemed capable of using the device competently by their physician and are trained to do so by a nurse. Most critically, the FDA reviewed clinical data associated with the use of each device in the intended population before allowing it to be marketed. Thus, the public health benefits cited by Morey et al. were scientifically anticipated.
In a second hypothetical case, imagine the home-use device described earlier, but now being used in the intensive care unit (ICU) without prior review of data demonstrating safety and effectiveness. In this case, the sponsor could not sell or hospitals legally purchase the device because it had not been cleared for this indication. A physician, unaware of the limitations of the underlying technology, elects to use the patient’s home-use device to make clinical decisions based on its displayed glucose level. If an elderly patient with little education can use it, surely a trained physician can! In our hypothetical scenario, the untrained physician uses the device and discovers that the patient is hyperglycemic. When small doses of insulin fail to treat the problem, the physician starts an insulin infusion. Unbeknownst to the physician, this home glucose monitor reports falsely elevated glucose levels in patients receiving icodextrin for intraperitoneal dialysis. The patient experiences permanent brain damage from severe iatrogenic hypoglycemia and eventually dies. You may think that this sounds hyperbolic. Sadly, it is an accurate description of several case reports previously reported in Anesthesia & Analgesia.14–17
Morey et al. suggest that a home glucose monitor should be no more dangerous in the hands of a critical care physician than in the hands of the home user. Well, maybe. However, the FDA makes its decisions on data not hunches. The FDA need not have even anticipated the specific risks of falsely elevated glucose readings in ICU patients receiving icodextrin when issuing its guidance. Agency rationale was likely much more simple: the company had not presented relevant data to support safety and efficacy, so the device could not be labeled or legally marketed for use in ICU patients. It is up to the industry sponsor who owns marketing privileges for each product to develop the data, including appropriate user training, to see that a point-of-care glucose monitor can be used in the complex setting of a critical care unit. The fact that the device can be used safely at home is not evidence that it is safe in an altogether different clinical environment.
The various evidentiary standards used for marketing decisions are already described in the Code of Federal Regulations, Title 21 (C.F.R. 21) and familiar to experienced medical product developers. Furthermore, the information used to secure permission to market a product is summarized in product labeling (drugs and high-risk medical devices) and/or in summary documentation published by the FDA on their Web site. Thus, while the suggestions for transparent and evidence-based regulatory review offered by Morey et al. appear reasonable, much of their thinking has already been embedded in the existing regulatory framework for premarket review.
Although I previously reviewed drugs and devices at the FDA, I do not speak for the agency. Moreover, as an anesthesiologist I share the concern of Morey et al. that imposition of a strict 1-hour expiration by a credentialing body on the use of arterial lines, based on sterility limits alone, might compromise trauma care and other emergencies. Doctors have always had to make choices in the face of competing priorities. Sometimes this means acting for individual patients in contrast to a product label or a broad-based determination by a national standards organization. This is art within the practice of medicine. It is why doctors rather than regulators are entrusted by patients with their care. Furthermore, the states, not the FDA or the JC, license doctors to practice. Therefore, when a conclusion by a national regulatory body adversely affects clinical care in a community, doctors may need to enlist help from their state representatives to directly engage these regulators.
Acknowledging that alignment between regulatory bodies and practicing clinicians may be improved, how can this be best implemented? In the case of the FDA, formal systemic changes to regulatory decision making may be accomplished by congressional statute or through executive administrative actions. Both these avenues for change are lengthy, complicated processes subject to legal and clinical interests.
Stakeholders outside regulatory bodies have more expedient opportunities to present opinions that may affect changes for specific product lines. There are 2 main ways that clinicians and other interested parties can give feedback to the FDA.
One opportunity to express opinion directly to the FDA is in response to a new proposed regulation or draft guidance when they are announced on the FDA Web site and in the Federal Register.c The public is invited to critique agency thinking by submitting comments to the docket.d The agency will then revisit their proposals in the light of submitted criticisms and revise the document before it becomes permanent guidance. Critique of a draft guidance document allows practicing clinicians to influence FDA regulatory policy before it is accepted by industry standards organizations like the JC or other regulatory organizations like CMS.
I must emphasize that medical specialists infrequently participate in this activity. I encourage readers to heed the call for action by Morey et al. and to review and provide feedback on draft guidance documents. Getting involved while guidance is under development is far more efficient and effective than complaining after the fact.
Once FDA policy for a drug or device has become established, individual experts and societies of experts can still present a Citizen’s Petition to the agency to revisit a decision or policy they believe to be erroneous.18 A recommendation from external experts should be supported by compelling data to be an effective argument.
Recently, the Office of Chief Scientist at the FDA has begun to engage academic centers to develop “regulatory science,” that is scientific information, technology, and methods to provide new tools that improve regulatory review of medical products. The FDA has sponsored Centers of Excellence in Regulatory Science as an initiative that includes critical path research in collaboration with the FDA to promote innovation, new educational programs, and cosponsorship of scientific meetings.e
The FDA has a well-established process for physician feedback. This process, combined with the legal requirement that FDA decisions be based on evidence, generally results in the agency reaching rationale conclusions on product labeling. If so, this may be an appropriate model for other regulatory agencies that influence the practice of medicine, including the JC, CMS, and USP.
In summary, Morey et al. offered methodological suggestions to improve clarity and conduct of regulatory decision making to better align them with the realities of clinical practice. This editorial responds to their suggestions by identifying aspects of the existing regulatory framework that relates to their suggestions. In addition, various options for scholarly clinicians to engage with the FDA are presented as potentially more expedient options than a sweeping reformation of regulatory law. While I generally concur with Morey et al. that regulatory alignment needs better alignment with clinical practice, I think that physician experts should take more ownership of regulatory sciences using existing channels. These comprise including service as experts to regulatory bodies, advising industry sponsors how to better conduct scientific studies, and timely commentary on FDA opinions and draft guidance documents.
Name: Lex Schultheis, MD, PhD.
Contribution: This author helped write the manuscript.
Attestation: Lex Schultheis approved the final manuscript.
Conflicts of Interest: Lex Schultheis consulted for Flexion Therapeutics, Inc., and consulted for Partners.org, Department of Anesthesia Massachusetts General Hospital and provided pro bono consulting for Zygood, LLC, Otomagnetics, LLC, and Cummins Tri-Star LLC.
This manuscript was handled by: Steven L. Shafer, MD.
a Available at: http://www.usp.org/about-usp. Accessed March 29, 2015.
b Available at: http://www.jointcommission.org/about_us/about_the_joint_commission_main.aspx. Accessed March 29, 2015.
c Available at: https://www.federalregister.gov/agencies/food-and-drug-administration. Accessed March 28, 2015.
d Available at: http://www.fda.gov/RegulatoryInformation/Dockets/Comments/default.htm. Accessed March 28, 2015.
e Available at: http://www.fda.gov/ScienceResearch/SpecialTopics/RegulatoryScience/ucm301667.htm. Accessed March 28, 2015.
1. Morey TE, Sappenfield JW, Gravenstein N, Rice MJ. Joint Commission and regulatory fatigue/weakness/overabundance/distraction: clinical context matters. Anesth Analg. 2015;121:394–6
2. Joint Commission Perspectives. Oak Brook, IL: Joint Commission Resources. 2014;34:1–11
3. Federal Food and Drugs Act of 1906 (The “WILEY ACT”) Public Law Number 59-384 34 STAT. 768 (1906) 21 U.S.C. Sec 1–15. (1934) (Repealed in 1938 by 21 U.S.C. Sec 329 (a))
4. 21 U.S.C. Food Drug and Cosmetic Act § 301(a) ch. 675, §1, 52 Stat. 1040.. 1938 Enacted June 25
5. 21 U.S.C. Food Drug and Cosmetic Act (Drug Amendments October 10, 1962), Public Law 87–781, 76 STAT 780.
6. Administrative Procedure Act (APA), Pub. L. 79–404, 60 Stat. 237. 1946 enacted June 11
7. Social Security Amendments of 1965, Pub. L. 89–97, 79 Stat. 286. 1965 enacted July 30
8. Food and Drug Administration, Substantial evidence 21 C.F.R. § 514.4. (2014)
9. Food and Drug Administration, Adequate and well-controlled studies 21 C.F.R. § 314.126. (2014)
10. Food and Drug Administration, Determination of safety and effectiveness 21 C.F.R. § 860.7. (2014)
11. Food Drug and Cosmetic Act 21 U.S.C., Sections 510(k) (Medical Device Amendments 1976).
12. Food and Drug Administration, Continuous ventilator 21C.F.R. § 868.5895. (2014)
13. Food and Drug Administration. Guidance for Industry and FDA Modifications to Devices Subject to Premarket Approval (PMA) - The PMA Supplement Decision-Making Process. 2008 Rockville, MD Center for Devices and Radiological Health
14. Kroll HR, Maher TR. Significant hypoglycemia secondary to icodextrin peritoneal dialysate in a diabetic patient. Anesth Analg. 2007;104:1473–4
15. Wurz J, Regli B. In one ear and out the other: communication barriers as a risk factor for critical incidents. Anesth Analg. 2007;104:1319–21
16. Disque A, Dhillon A, Gritsch A. Icodextrin causing glucose meter error and severe hypoglycemia after deceased donor renal transplant in a patient receiving continuous ambulatory peritoneal dialysis. A A Case Rep. 2013;1:89–91
17. Pitkin A, Rice MJ. Editorial comment: icodextrin causing glucose meter error and severe hypoglycemia after deceased donor renal transplant in a patient receiving continuous ambulatory peritoneal dialysis. A A Case Rep. 2013;1:92
18. Food and Drug Administration, Citizens Petition 21 C.F.R. § 10.30. (2014)