Skip to Main Content

Medical schools teach students a four-part “virtuous cycle” in which one step positively reinforces the next: Assess the patient. Implement a therapeutic plan. Assess the patient’s response. Revise the therapeutic plan as needed.

In an emergency department, this cycle can be completed in minutes. In the cancer clinic, it can take months.

Mastering the virtuous cycle is understood to be a central measure of medical competence. Yet when the patient is not one person but an entire society, this cycle is fractured and ad hoc in ways that would make any patient demand a new doctor.


We’ve all been witness to — and victims of — this failure in the pandemic. The superb accomplishments of therapeutic medicine cannot address the population-based issues that Covid-19 has raised. But we can use the virtuous cycle as a way to switch gears to employ approaches drawn from disciplines like public health.

For the first step, assessment, doctors were unable to define the most basic clinical course of severe Covid-19, despite billions of dollars invested to achieve interoperable electronic health records over the past 30 years. It took clinicians and researchers months to identify the interplay of inflammation, coagulopathy, and cardiac dysfunction, and then only through a jury-rigged combination of conference calls and small studies shared through disparate nuggets of preprints.


In retrospect, the clinical course of Covid-19 would have been plain to see from any systematic study of the experiences within dozens, then hundreds, of hospitals treating patients with the disease in China, Europe, and then in the U.S. early in 2020. This failure of assessment was multiscale: Medical and public health establishments had neither the mechanisms, governance, nor institutional will share these clinical data in real time at the national level, let alone internationally. The just-unveiled platform may offer some promise. Funded by corporate and nonprofit foundations, it features data on Covid-19 cases from multiple health research institutions and hospitals. It has grown from a small volunteer network into a comparatively substantial effort.

For step 2, implement a therapeutic plan, we have witnessed the staggeringly rapid and welcome development of new classes of vaccines that are effective against Covid-19, and their being vetted and delivered an unprecedented nine months from when the pathogen was identified. This achievement is all the more notable because it sets the stage for even more timely vaccine development for future threats.

Yet now, with the delivery of these vaccines, authorities are stumbling badly. Millions of available doses have yet to be injected, and each delayed injection adds to continued and — with new variants emerging — increased transmission of Covid-19.

Independent of the disorganization and lack of planning at the national level (and until recently the unapologetic forswearing of any responsibility there) some of our most feted academic health centers had multiple false starts with articulating and implementing vaccine distribution. The IT infrastructure required to track vaccinations and any adverse effects has had to be developed in extremis because there is no pre-existing software available at the right scale.

The assessment of step 2’s therapeutic intervention in step 3 is, unfortunately and unsurprisingly, no better than the initial assessment called for in step 1. There’s simply no sophisticated dashboard by which policymakers and others seeking a bird’s-eye view can understand the impact of the interventions that have fitfully been attempted.

The last part of the cycle, revising the therapeutic plan based on how the patient is responding, simply hasn’t been possible given the neglect of the earlier parts of the cycle.

Moreover, viewing the population as merely a sum of individual patients rather than a whole can miss the forest for the trees. Karthik Dinakar of The Chelsea Project, a nonprofit program to reduce the spread of Covid-19 in the working-class city of Chelsea, Mass., points out that there can and should be a conception of community risk distinct from individual risk. In Chelsea, the life expectancy is 73 years for men and 75 year for women, lower than the national average of 76 years for men and 81 years for women. The highest-risk population, according to the state’s vaccine prioritization, amounts to only about 1,500 people over age 75. “The vast majority of deaths in the city have been in the 40-55 range because of occupational risk and overcrowded housing,” Dinaker told us by email. The authoritative status conferred on the individual clinician’s lens prioritizes risk scores based on an “average” individual, without regard for the transmission and infection dynamics that are specific to a community, such as occupational risks and overcrowded housing.

There is a fundamental gap in the way we relate individual health care to broader public health that inhibits our ability to respond to emerging health threats. Our analogy to the virtuous cycle of therapy shows that we’re missing people in crucial but unrecognized roles — an assemblage of clinical teams writ large, whose primary responsibilities are not to individual patients but to the patients of their health care systems. These teams would be tightly linked with regional and national teams to assess what is happening with a novel infectious disease, and then strategically implement responses, measure the results, and revise accordingly. In that way, insights would be shared with the thousands of health care providers suffering under the weight of the pandemic — and demonstrate a path for amelioration of other community-wide public health challenges.

We need experienced, coordinated teams of health professionals and analysts across local, regional, and national care systems whose primary job is to evaluate data rapidly and with full understanding of the limitations of data and idiosyncrasies required for reliable interpretation.

Public health systems elsewhere in the world may offer instruction. In early 2020, for example, critical care doctors were surrounded in their ICUs by Covid-19 patients — all sick, many dying. These intensivists were learning that their standard therapies for respiratory illnesses usually weren’t effective. Many small trials were launched with the best of intentions, but systematic, large-scale trials to see beyond the individual details of each case and assess the effectiveness of treatments more broadly were scarce.

In this context, jarring in its unexpectedness, appeared a June 2020 preprint from researchers in the United Kingdom reporting results from a well-designed clinical trial called RECOVERY, which at the time included more than 6,000 Covid-19 patients. This trial showed that treating patients with 6 milligrams daily of dexamethasone for up to 10 days sharply decreased deaths from Covid-19.

The success of the RECOVERY trial reflected its approach: Investigators dismantled bureaucratic obstacles and streamlined ponderous processes, including those of data sharing and aggregation. As the RECOVERY team’s Martin Landray told the BMJ, “What we are seeing is what happens when people are empowered and motivated.” Implicit in this comment is that this performance could not happen routinely.

New health teams aimed at implementing a virtuous therapeutic cycle at a regional and national scale will need infrastructure for continual monitoring of health and treatment that is readily within our reach technologically but has so far been thwarted by sociopolitical and cultural considerations. Even if this infrastructure was present, the governance, expertise, and familiarity with the required planning and analyses to ensure that these teams would be successful are missing.

They could take a page from emergency room resuscitation, for which clinicians have recognized the importance of defining responsible teams and rehearsing again and again in so-called mock codes the full virtuous cycle required for this potentially lifesaving process.

Even the most talented and committed doctors cannot properly respond to a pandemic by ministering to one patient at a time. Individual health requires public health, and those engaged with the latter have been underfunded and untended for too long. The good news is that there are some clear, simple, and, in the larger scheme, inexpensive steps that can be taken to begin shoring up public health in the U.S. The bad news is that we have time and again failed to heed even the most basic lessons to establish the virtuous therapeutic cycle in this area, instead counting on raw luck to avoid the ravages of a pandemic.

It’s time to break an emerging pandemic cycle with a collective, therapeutic one.

Isaac Kohane is professor and founding chair of the Department of Biomedical Informatics at Harvard Medical School. Jonathan Zittrain is professor of international law and professor of computer science at Harvard University and co-founder of its Berkman Klein Center for Internet & Society.

  • An integrated national record system with sophisticated data management and analysis tools would have delivered results about effective therapies MUCH sooner. This is an utterly worthwhile goal if done with proper privacy protection, and accessible only by doctors and medical scientists. Insurance providers must use their own databases as they are not relevant to actual medical progress.

Comments are closed.