By Govind Persad, Monica Peek, and Seema Shah
Vaccines are no longer our only medical intervention for preventing severe COVID-19. Over the past few months, we have seen the arrival and wider availability of treatments such as monoclonal antibodies (mAbs), and more recently, of novel oral antiviral drugs like Paxlovid and molnupiravir.
The recent Delta and Omicron surges have made these therapies scarce. The Delta variant led the federal government to resume control over mAb supply and promulgate allocation guidelines. The Omicron variant exacerbated scarcity because only one of the currently available mAbs, sotrovimab, appears to be effective against it. While Paxlovid and molnupiravir are effective against Omicron, both will likely be in short supply for many months. Paxlovid is currently constrained by a lengthy manufacturing process. Molnupiravir — which is substantially less effective — is contraindicated for use in patients under 18 and not recommended for use during pregnancy.
To allocate COVID-19 vaccines, the CDC’s Advisory Committee on Immunization Practices, the National Academies of Sciences, Engineering and Medicine (NASEM), and the World Health Organization (WHO) identified ethical goals for prioritization, such as maximizing benefit and minimizing harm, mitigating health inequities, and reciprocity. These committees, particularly the NASEM and WHO committees, included ethics experts as well as experts in social science, biology, and medicine. Current federal guidelines for therapy allocation, in contrast, do not identify ethical objectives or involve ethics expertise.
In an open-access Viewpoint in Clinical Infectious Diseases, we identify ethical goals for the allocation of scarce therapies. We argue that the same ethical goals identified for vaccine allocation–in particular maximizing benefit, minimizing harm, and mitigating health inequities — are also relevant for therapy allocation. Because many people have now taken steps to mitigate pandemic scarcity, for instance by protecting themselves through vaccination, we argue that reciprocity is also relevant.
Because therapies work differently from vaccines, prioritization strategies to achieve these ethical goals will differ as well. For most therapies, unlike vaccines, people should not be prioritized based on a heightened risk of exposure because these therapies are designed to treat people who have already been infected. And while vaccines and boosters reduce transmission, it is not clear that all therapies do. Prioritizing individuals who may transmit illness to others, such as high contact workers, may therefore be less relevant.
The NIH COVID-19 Treatment Guidelines Panel asserts that the highest priority group, Tier 1, comprises both “Immunocompromised individuals not expected to mount an adequate immune response to COVID-19 vaccination or SARS-CoV-2 infection” and “Unvaccinated individuals at the highest risk of severe disease.” Those at highest risk are further defined as anyone older than 75, and everyone older than 65 with clinical risk factors such as cancer, cardiovascular disease, chronic lung disease, diabetes, obesity, pregnancy, and sickle cell disease. The second tier is unvaccinated people over 65 or with any clinical risk factor. Tier 3 is vaccinated people over 75 or over 65 with clinical risk factors, and the final priority group is vaccinated people over 65 years or under 65 with clinical risk factors (Tier 4).
Even though allocating scarce treatments necessarily involves both clinical and ethical considerations, the NIH guidance identifies no ethical foundations. This leads to several problems. First, the NIH’s placement of unvaccinated older adults in Tier 1 alongside people with severe immunocompromise is inconsistent with both maximizing benefits/minimizing harms and reciprocity. An unvaccinated person over 75 with no clinical risk factors, or a person over 65 with a BMI of 30, is likely at far lower risk than a person of the same age who is immunocompromised, and even at lower risk compared to a younger immunocompromised person — yet the NIH places both in Tier 1. Similarly, an unvaccinated 20-year-old with diabetes (Tier 2) should not be ranked ahead of a 70-year-old who received 1 dose of the J&J vaccine but no booster and has chronic kidney disease (Tier 3).
Immunocompromise also often produces and correlates with health inequities. Further, unlike immunocompromised people, unvaccinated people could have readily protected themselves without requiring these very scarce therapies. This makes reciprocity relevant — particularly now that a recent Kaiser Family Foundation poll reported that Americans who are racial minorities or have serious health conditions are now more likely to have received vaccines than others, and less than 2% of unvaccinated adults report that it is “hard for them to get a COVID-19 vaccine.” Older and medically vulnerable people who continue to face high risk of COVID-19 illness after vaccination should not be asked to wait in line behind adults who refused vaccines. Accordingly, we argue that immunocompromised people should be prioritized over those who are eligible for, but declined, vaccine. (Notably, prophylactic monoclonal antibodies are prioritized for immunocompromised and vaccine-ineligible people rather than for those who have declined to be vaccinated.)
The NIH guidance’s reliance on one-size-fits-all age cutoffs at 75 and 65 is similarly inconsistent with benefiting people and preventing harm, and is also inconsistent with mitigating health inequities. Prior to the advent of vaccines, people who faced various forms of disadvantage, including racism, faced higher risk of severe illness and death at earlier ages. Age-75 cutoffs for vaccine eligibility were therefore criticized for exacerbating inequities and failing to track actual risk. This differentiation in risk continues among unvaccinated adults, and may continue even among people who have been vaccinated. We also know that people who live to 75 have been more economically advantaged, less likely to have experienced life-shortening medical conditions, and less racially diverse than the broader American public. Reflecting this, recent research found that considering social vulnerability (using place-based metrics like the Social Vulnerability Index) alongside age in vaccine allocation would have both saved more lives and narrowed disparities compared to allocating by age alone. Using social vulnerability as a factor would also help differentiate those who are unvaccinated due to access barriers from those who are unvaccinated due to ideological commitments, consistent with the reciprocity principle.
Given staffing shortages due to the Omicron variant, prioritizing health workers who are providing needed services — an option the NIH guidelines ignore — is also consistent with benefiting people and preventing harm, mitigating health inequities, and reciprocity. Health workers are needed to benefit people and prevent harm, merit reciprocity in view of their contribution to stemming the harms of the pandemic, and provide benefits that are especially needed by people on the short end of health disparities. Notably, nursing workforces are aging, and older physicians are especially likely to serve in settings — such as rural areas, where 20% of physicians are 60 or over — where their services are particularly scarce.
Last, the NIH guidance asserts that “Providers should use their clinical judgment when prioritizing the use of anti-SARS-CoV-2 mAbs for treatment or PEP in a specific situation.” But allocation among patients under scarcity is a public health and public policy decision, not a clinical one: we should not place clinicians facing patient care pressures in the position of spending a “weekend poring over the charts of Covid patients to figure out who should get scarce treatments.” And the listed clinical criteria for severe COVID-19 risk are based on 2020 research and not validated for the current Omicron variant. Additionally, people from more disadvantaged communities are likelier to have undiagnosed illness, which suggests that priority criteria based on diagnosed illness may exacerbate inequity.
Instead of the NIH’s tiers based on age, unvaccinated status, and diagnosed illness, we suggest using a categorized priority system, sometimes also termed a “reserve system,” that prioritizes portions of the pool of available therapies for specific groups. Massachusetts has already successfully used such a system. For instance, 20% of the available supply could be prioritized for people in the most socially vulnerable geographic areas. Another 20% of supply could be prioritized for scarce health workers. The remaining supply could be open to all who are clinically eligible, stratified by criteria including diagnosed illness, age, and unvaccinated status—but with attention to how some of these criteria can stand in tension with ethical goals like reciprocity and mitigating health inequities. A categorized priority system allows states and hospitals to realize multiple ethical values in a way that is operationalizable under time pressure, ensures that no one group (such as unvaccinated adults) consumes all available therapies, and avoids the need for time-consuming, morally distressing, and potentially biased allocation judgments by individual clinicians.