CDC’s Covid-19 Communication Failures Reforming the Process for Public Health Recommendations

The federal Centers for Disease Control and Prevention (CDC) has been widely criticized for its missteps in responding to the Covid-19 pandemic. CDC’s failures, which veteran public health experts described as “notable,” include knowingly releasing flawed diagnostic tests,[1] neglecting its own best practices in risk communications,[2] and issuing recommendations for temperature screening methods that “do not work.”[3] Its latest August 2022 guidance, regarding isolation and masking for people with or recovering from Covid, continues to sow confusion.[4] CDC guidance for schools was so flawed that its 2021 recommendations drew a public rebuke from the very scientists it cited—and its earlier recommendations for preemptive coordinated school closures likely contributed to the large and discouraging decline in students’ academic performance since 2020, including a growing gap in test scores between low and high poverty elementary schools, as well as declines in kids’ mental health, and adverse effects on parents’ jobs and wages.[5]

In response to these and other criticisms, CDC director Rochelle Walensky has announced her plan for an “ambitious” overhaul of the agency. But the proposed reforms, as described in an August article in STAT, are underwhelming.[6] Under the plan, a reformed CDC would prioritize gathering data to rapidly dispense public health guidance, rather than crafting scientific papers. Walensky also focused on the need to reform the culture of CDC without identifying any actions necessary to do so. While the announcement does suggest a new awareness of poor performance—a first step to any improvement—it does not initiate concrete steps that could not immediately be reversed by the next CDC director.

Effective CDC reform is needed, as new communicable disease threats may be growing, given the continuing challenges posed by Covid-19,[7] as well as rapid growth in the total number of monkeypox cases in the U.S.—already above 20,000—and newly reported evidence of locally transmitted polio in two New York counties. But reform will require action by Congress, which created CDC and appropriated $9.5 billion for CDC programs in fiscal year 2022.[8]

Congress should require changes to CDC’s process for issuing public health advice and guidance so that it is less autocratic, more responsive to public concerns, more explicitly science-based, and more transparent. In particular, CDC needs to adopt and follow good guidance practices, similar to the ones that the Food and Drug Administration has long been required to follow.[9] FDA’s experience demonstrates that it is feasible and worthwhile for CDC to adopt similar practices.

Congressional action is needed because the Biden administration won’t push meaningful reforms to CDC’s guidance programs. In fact, the U.S. Department of Health and Human Services (HHS) recently issued a regulation preventing some of these very reforms.[10] As part of the administration’s implementation of President Biden’s Executive Order 13992,[11] HHS in late July reversed a Trump administration regulation that would have required CDC and most other agencies within HHS to comply with certain good guidance practices.[12]

DOWNLOAD PDF

CDC’s role in issuing guidance is important because incoherent, confusing, or inconsistent advice or guidance can be damaging, especially early in a pandemic, when everyday activities may pose novel and uncertain risks. Examples of bad CDC advice are numerous. Early in 2020, CDC knowingly released a flawed test for health-care providers to assess whether people were sick from the novel coronavirus.[13] Later in 2020, CDC recommended that people in non-health-care settings who have a temperature of 100.4°F (38.0°C) or higher on at least two occasions should practice social distancing with self-quarantine for 14 days. CDC also endorsed the use of a non-contact thermometer applied to the forehead for measuring temperature.[14] Independent researchers concluded that use of such temperature screening programs to identify infected persons was, at best, marginally effective because approximately half of infected persons never develop a fever.[15] Others later found that no-contact temperature screening simply “does not work.”[16]

A CDC advisory panel’s recommendations for preemptive school closures contributed to closures of approximately 100,000 U.S. public schools for more than 8 weeks in 2020, although a large majority offered distance learning.[17] More than 20% of schools also discontinued subsidized meal programs, which caused real harm to many children.[18] A February 2021 CDC report on transmission of the new coronavirus in schools was roundly critiqued by Dr. Joel Zinberg.[19] He noted that CDC’s decision to relax social distancing standards from two meters to one meter was long overdue, coming eight months after the World Health Organization had made the same recommendation.[20] CDC’s guidance on schools to support safe in-person learning drew a public rebuke in 2021 from the very scientists it cited,[21] who claimed that CDC misinterpreted their research and that keeping schools closed or even partly closed was unwarranted.

CDC’s recommendations and guidance on controlling and preventing Covid have substantial influence, even when they are badly flawed. CDC guidelines have been relied on by organizations such as the New York City Department of Education when it announced health and safety measures for the 2021–22 school year,[22] and the National Restaurant Association in its policy for in-person events.[23] Such organizations likely defer to CDC not out of respect for the quality of the underlying scientific analysis but because uniform federal standards are useful in reducing debate about appropriate control measures.

The good guidance practices rejected by the Biden administration’s HHS include commonsense measures necessary for proper management of CDC’s programs. First, they generally require that the agency solicit public comments about new guidance, and they require the agency to publicly respond to major concerns raised by those comments. Soliciting public comment is necessary to collect private information about the benefits and risks of a specific public health guidance, given that federal regulatory agencies are not omniscient. Requiring a public response to major concerns helps to demonstrate a commitment to transparency and to ensure that guidance is communicating what it was intended to communicate.

Second, under the good guidance practices, all guidance documents include a standardized statement making it explicitly clear that its contents do not have the full force and effect of law and are not meant to bind the public unless specifically included in a contract. Such statements prevent agencies from issuing guidance containing new requirements, which should be adopted only through a different and more rigorous rule-making procedure. Third, good guidance practice requires the agency to establish and maintain a standardized, searchable, and comprehensive database of all its guidance documents. Such a database provides clarity about which recommendations arise in what context, which are in effect, and which have been withdrawn or are otherwise no longer in place.

In its 2022 final rule, HHS offered weak explanations for its rejection of the 2020 good guidance practice regulation.[24] It said that the 2020 regulation established “procedures well beyond anything required by applicable law,” which created “unnecessary hurdles that hinder the Department’s ability to issue guidance.” CDC also rejected the requirement for public comment because “any benefit derived from the [public’s] ability to formally comment on guidance and providing [sic] the Department’s responses to comments is outweighed by the Department’s interest in quickly and responsively communicating” its current thinking on its rules and policies. The 2022 final rule provides no convincing evidence in support of this claim. In addition, and somewhat surprisingly, the HHS 2022 final rule does not analyze the experience of FDA, an important agency within HHS that, for two decades, has issued guidance following its own good guidance practice regulation.[25]

FDA’s good guidance practice regulation—which was issued in the fall of 2000, following efforts by major industry associations to have Congress require FDA to do so—should be front and center in discussions of how to improve agencies’ guidance. The regulation has helped prevent FDA from issuing guidance documents that were really regulations and from offering so-called podium policy—in which officials make off-the-cuff recommendations from behind a podium. Instead, FDA has adopted a standardized and regularized process that controls nearly all formal use of statements about what regulated entities “should” or “ought to” do. That process explicitly allows regulated entities not to follow FDA recommendations, provided that they comply with existing statutory and regulatory requirements as stated in relevant statutes and regulations. Importantly, FDA’s regulation allows it not to seek public comment prior to issuing guidance if it determines that doing so is not necessary or appropriate.

FDA’s guidance documents cover a wide variety of topics, from drug development, drug safety, and food safety to the agency’s response to Covid-19. In 2020, FDA issued dozens of guidance documents related to Covid-19, including guidance regarding its decision not to take enforcement action for the marketing of medical products for unapproved uses. Some examples of its recent guidance illustrate their breadth.

• A 2021 guidance document advised about testing for the presence of methanol in pharmaceutical products containing alcohol (ethanol) and isopropyl alcohol that were used during the pandemic. (In spring 2020, FDA became aware of instances in which people with alcohol dependencies drank hand sanitizer products that were labeled as ethanol alcohol but that, in fact, were contaminated with methanol, leading to serious adverse health consequences consistent with methanol poisoning.)[26]
   
• A 2021 guidance sought to clarify the regulatory landscape of face masks, barrier face coverings, face shields, surgical masks, and respirators, as well as help expand the availability of these devices for use by the general public and health-care professionals in health-care settings, as appropriate.[27]
   
• A 2020 guidance offered regulatory flexibility regarding packaging and labeling of shell eggs sold by retail food establishments during the Covid-19 emergency when consumer demand for eggs increased and appropriately labeled retail packaging was not available for all eggs.[28]
   
• A 2022 guidance document described hearing aids, personal sound amplification products, their respective intended uses, and the regulatory requirements that apply to each.[29]
   
• A 2022 guidance for industry described FDA’s enforcement discretion policy for infant formula, with the aim of helping increase the supply of infant formula in the United States.[30]

Public Comments

During the Covid public health emergency, FDA routinely solicited formal public comments on Covid-related guidance documents even when they would take effect without an opportunity for public comment in advance. Such comments occasionally led to revisions to the original guidance, as occurred in the case of the aforementioned 2021 mask guidance. The 2021 guidance includes recommendations about alternatives when FDA-cleared or National Institute for Occupational Safety and Health (NIOSH)–approved N95 respirators are not available and removed FDA’s prior recommendations regarding emergency use authorizations (EUAs) for decontamination of face masks and filtering face-piece respirators.

Volume of Guidance Documents

The total volume of guidance documents issued by FDA is substantial, suggesting that its guidance program is a useful model for CDC. FDA’s database shows that it finalized 114 guidance documents in 2019, 163 in 2020—which included the stress-filled early months of the Covid-19 pandemic—and 91 in 2021. FDA is on pace to issue 124 final guidance documents in 2022, having issued 62 final guidance documents in the first half of 2022.

Similar data for the number and scope of CDC guidance documents are not readily available. CDC provides no downloadable list of guidance documents. CDC’s webpage on its Covid-19 guidance is organized by audience (e.g., “employers, business owners & community leaders,” “individuals & consumers,” “parents & caregivers”) and by topic (e.g., “masks,” “disease information,” “K–12 schools”).[31] As of October 5th, 2022, the page contained 73 total “items.” It is very likely, however, that not all these are unique Covid-related CDC guidance documents. Some of the documents, for example, do not contain recommendations (e.g., “Post-Covid Conditions, CDC Science”).[32]

Other documents appear to simply announce changes in regulatory requirements. For example, the guidance document “Testing Strategies for SARS-CoV-2” states: “Reporting of negative results for non-NAAT tests (rapid or antigen test results) is no longer required. However, testing sites must still report data for all positive diagnostic and screening testing completed for each individual test.”[33] That document links to a separate document that outlines HHS laboratory reporting requirements under the CARES Act.

Transparency and Public Engagement

FDA’s guidance program has several strengths that CDC’s guidance program lacks. In addition to the features discussed earlier (general requirement for public comment, explicit statements as to the nonbinding nature of guidance,[34] and a searchable database of all guidance documents), FDA also provides many opportunities for members of the public to participate with FDA in the development of guidance, such as through advisory committee meetings and public meetings. Different FDA programs publish lists of draft and final guidance documents expected to be released during the current year. For example, the 2022 list from FDA’s Center for Food Safety and Applied Nutrition includes 26 entries, and the 2022 list from its Center for Drug Evaluation and Research contains 116 entries.[35] These lists of pending guidance documents signal opportunities to engage with FDA in the development of such recommendations and can signal pending changes in the agency’s recommendations.

The failure of the HHS 2022 final rule to appropriately acknowledge these strengths in rejecting the 2020 good guidance practices rule could, in theory, make the 2022 rule vulnerable to legal challenge. In general, the Administrative Procedure Act authorizes federal courts to set aside arbitrary and capricious agency actions.[36] A successful challenge to the final rule, however, would require a plaintiff with standing—meaning that the plaintiff has suffered an “injury in fact” that is fairly traceable to the challenged action of the defendant, e.g., CDC, and that a favorable decision by the court would be able to redress that injury.[37] Since the 2022 final rule is essentially procedural, it is not clear that any potential plaintiff would have standing to challenge it.

Individual CDC guidance documents are also generally not subject to judicial review. Courts can and do review regulations promulgated by CDC, as the Supreme Court did when it struck down CDC’s second eviction moratorium in August 2021.[38] But courts can only review “final agency actions”—which, as the Congressional Research Service notes, can “shield certain agency guidance documents from judicial review if such guidance does not legally bind the public.” Guidance and recommendations may have significant impacts on society, but they cannot be reviewed in court if they are not legally binding.

The lack of independent review of CDC’s recommendations and guidance means that the agency has a special responsibility to make them as well reasoned and scientifically supported as possible. While the agency has revealed some awareness of the criticisms of its recommendations and guidance, it has not recognized the full scope of necessary reforms. For example, an internal CDC review, completed in March 2021, addressed only whether its Covid guidance was “evidence-based and free of politics.”[39] Missing from that review was a thoughtful inquiry into how to improve public confidence in the scientific basis of CDC’s recommendations and guidance.

CDC guidance documents are typically not supported by a scientific or technical document explaining the scientific basis for the agencies’ recommendations, relative to alternative recommendations. They occasionally contain a section titled “rationale,” which may offer points to consider, but without explicit reasoning to defend a specific recommendation to control the spread of coronavirus over a plausible alternative. CDC also offers “Science Briefs” on many topics related to Covid,[40] but these often read like summaries of scientific literature and not interpretations of scientific findings that point to, or help to understand, specific recommendations to control and prevent disease.

Even CDC documents published in support of its guidance can do so only awkwardly. For example, on August 11, 2022, CDC issued a press release with 10 bullet points containing new Covid-19 guidance relaxing certain isolation and masking recommendations.[41] The 800-word announcement does not link to a separate guidance document explaining the change in CDC’s assessment of scientific evidence of benefits and risks of isolation and masking. It does reference an article coauthored by Dr. Greta Massetti and 11 other members of CDC’s Covid-19 response team, “Summary of Guidance for Minimizing the Impact of Covid-19” [emphasis added], which provides a summary of all current guidance, rather than a justification of specific changes in recommendations for isolation and masking.[42]

CDC would do well to adopt a good guidance practice regulation modeled on FDA’s, but in addition, the regulation should require that each guidance contain a reasoned scientific basis for its policy recommendations. This would include a systematic review of relevant studies, an effort to distinguish between mere associations and causal relationships, and an assessment of how specific recommendations would lead to greater public health benefits or potentially higher risks. In the case of school closings, for example, the requisite scientific basis for a recommendation to close schools would note the reduction in transmission among students and school personnel that closures might be expected to provide. But it would also have to acknowledge the unintended risks to public health—those that arise from missed learning and child development opportunities, as well as from the loss of school-based safety nets, such as free lunch programs for children. It would also note how these risks might vary with the age of the students, class size, or time each week spent learning in person.

A requirement to provide such a scientific basis, if combined with an open docket, might help in instances where CDC’s own staff work points to risk-reducing strategies that are not incorporated into CDC’s recommendations. For example, a February 2021 paper published in CDC’s Mortality and Morbidity Weekly Report (MMWR) demonstrated that “knotting” the ear loops on a surgical mask could improve fit (and therefore protection against the virus).[43] This “knotting” recommendation appears to be inexplicably missing from CDC guidance.

Administrative reform of CDC’s process of advising others how to control disease would not be a panacea—the agency’s weak response to Covid-19 has several origins. In seeking lasting improvements to CDC’s programs, however, Congress would do well to consider more than just the usual topics—congressional funding of CDC operations and authorities granted to CDC to require actions by private parties. Brian Miller and others, for example, have convincingly argued that CDC’s role in fighting chronic non-communicable disease should be scaled back, and it should refocus its energies on communicable disease.[44]

Even the agency’s flagship public health communications tool, MMWR, sometimes reflects CDC’s problematic culture. The MMWR report on the detection of poliovirus in wastewater in New York appeared on August 16—days after it was reported in the national press.[45] One can’t help but wonder how much this publication was delayed by negotiations over who would get credited in the MMWR report, which lists 27 coauthors individually, plus the “U.S. Poliovirus Response Team,” which has scores of CDC and state and local agency staff named as members. CDC, which is rightly proud of MMWR’s high impact factor and rank among scientific and technical journals focused on public health,[46] has also taken no steps to make its data and code open access, as Science, Nature, and the American Economic Association have done.[47] Several steps may thus be necessary and appropriate for meaningful and robust reform of CDC’s programs.

To ensure consistent, systematic opportunities for public input into CDC guidance development and to buttress CDC’s use of scientific reasoning, Congress should require that CDC adopt a good guidance practice regulation modeled closely on FDA’s regulation of its own guidance program. It would require open public dockets for all CDC guidance and solicitation of public comment in advance, except for public health emergencies or similar cases where such public comment is inappropriate. It would require a database of all CDC guidance both past and current. And it would require CDC to respond to major questions raised in public comments on draft guidance. Importantly, it would also direct CDC to provide a reasoned scientific basis for its recommendations.

Please see Endnotes in PDF