According to at least one source [1], it takes an average of 17 years for a proven intervention to be fully implemented. By that metric, we would have no treatments for COVID-19 besides those identified anecdotally. How did the boots on the ground of our healthcare systems actually respond?
I’ve written lately about the several thousand clinical treatments and studies our first-responding physicians tried, but a new study seeks to understand how our institutions tried to identify the signal from the noise and create guidelines. Guidelines typically change slowly; they are relatively fixed. But since the pandemic required a more nimble structure, guidelines morphed into “living” documents – able to change to “follow the science.”
Researchers identified 105 of these on-the-fly, “living," as opposed to static, protocols from 123 US healthcare systems, 52 academic centers, 31 VA system hospitals, and 12 community hospitals, along with 10 reference documents from “guideline-issuing organizations,” like the WHO or CDC. They characterize:
- how the treatments were applied – to the suspected, confirmed, or higher risk disease
- whether disease severity or risk was used to guide treatment, as well as the basis for those criteria
- use of laboratory results in guiding care
- caveats to guide physicians in how limited the recommendations might really be
96.8% of the protocols clearly and strongly recommended that active intervention by physicians on behalf of their patients – no one could watch people dying of COVID-19’s without trying something. Those at higher risk or that were more severely affected were singled out deserving special attention. There was also a nod to the belief in “Do no harm,” as many of the guidelines detailed appropriate safety monitoring for the immunocompromised or those receiving repurposed drugs. The following image captures much of the paper’s findings:
The severity of illness (SOI) and risk factors (RF) were often used in combination to direct therapy. To a lesser degree, SOI – how ill the patient was – was considered a more reliable guide than just their risk. Many of the risk factors were based on the conditions “on the ground,” the local risks. Despite the various co-morbidities widely discussed in the media, the predominant risk factors were age and being immunocompromised either due to a specific disease, or treatment as part of organ transplantation, or cancer. Risk factors also included the standard sources of complications, heart, lung, kidney disease, diabetes, and its fellow traveler obesity.
Because COVID-19 is primarily a respiratory disease, oxygenation and respiratory rate were both used as markers of disease severity. Controversy still follows the best treatment for oxygen desaturation. In less harried times, the oral tradition of medicine suggests that we not “treat the findings of an X-ray, but the patient’s symptoms.” That rule of thumb went by the wayside as a third of the protocols identified imaging as a means of determining severity.
Despite media controversy, 90% of the protocols recommended hydroxychloroquine until it became clear that it had little if any, therapeutic value. Steroids and anti-inflammatory agents were mentioned 60% of the time, mostly to be condemned and not recommended. That too changed when the data showed a marked benefit. Many protocols recommended that infectious disease specialists be consulted in treatment management, using them as gatekeepers, and requiring them to approve using therapeutic agents like antiviral drugs.
Caveats
“Many protocols highlighted a sense of urgency regarding the development and operationalization of treatment protocols and included statements about “frequent updates” and “do not print” given the “rapidly changing” nature of recommendations … Other notable emerging themes involved the inclusion of “warning” or cautionary statements that therapeutics recommended were not FDA-approved or evidence-based.”
The urgency to do something and not stand idly by comes through loud and clear in this setting. That has been the long-time approach of medicine’s doers – the surgeons and proceduralists; the futility of doing nothing emboldened medicine’s “thinkers,” those who have the usual luxury of mindful observation, to join with the doers, at least this time. And again, you can see "Do no harm" creeping in those cautionary statements.
There is literature that seeks to identify the factors necessary in formulating guidelines for innovative care. These include organizational structure, the characteristics, e.g., invasiveness and complications, of the treatment, the strength of its supporting evidence, and external environmental factors – most often whether the hospital or physician down the block will be the first to the treatment and the subsequent "prestige."
“…during the early pandemic, we found that the 2 constructs that most heavily influenced institutional recommendations were the external environment and provider and organizational factors. Strength of the underlying evidence played a relatively minor role in driving adoption of practices.” [Emphasis added]
Given a choice between doing something or nothing, physicians acted. The environment, in this instance, the ever-growing queue of severely ill entering the emergency department, trumped evidence. Physicians moved their organizations as much as they could to continue to save lives. There were no flip-flops, no politically inspired therapeutic agendas. As this study shows, medicine acted empirically, using only observational data in the absence of randomized controlled studies. As the pandemic loses its urgency, we have to ask how we will better incorporate living documents, observational data into evidence-based medicine.
Source: How Are Emerging Data Translated Into Clinical Practice? A Mixed Methods Investigation of Coronavirus Disease 2019 Institutional Treatment Protocols Open Forum Infectious Diseases DOI: 10.1093/ofid/ofab072
