Rationing Health Care in the Time of COVID


As health care spending in the United States climbed to seemingly unsustainable heights 40 years ago, some observers predicted that rationing of health care would inevitably dominate American medicine. Since then, there have been periodic debates in academic circles regarding whether and how rationing would become standard practice in American medicine, as well as research into priority-setting methods. Most Americans, however, have been either unaware of or bewildered by the notion of rationing. Except for a brief political moment in 2009 when opponents of the Affordable Care Act falsely claimed that the reform law would engender rationing of care, the “R” word rarely has appeared in public discourse.

However, as thousands of COVID-19 patients flooded hospitals across the US in 2020, clinicians suddenly found themselves forced to make difficult life-and-death decisions regarding which patients should receive intensive care. With the demand for care quickly outstripping the supply of beds and ventilators, providers faced enormous pressure to triage patients ethically and humanely. For many Americans, this was their first encounter with medical triage and rationing of care on a broad scale. Triage, which originated during the Napoleonic wars, is an accepted medical practice that allocates scarce resources to those most in need of immediate treatment and most likely to benefit from it. Organ transplant centers in the United States long have used triage based on principles of clinical need, potential benefits and risks, tissue compatibility, and predicted survival. However, COVID-19 changed the calculus slightly by imparting greater priority to the predicted survival of potential recipients. This motivated many transplant centers to mandate vaccination and has spurred broader discussion about rationing methods and their ethical use.

The recent surge in Delta variant cases has led some states to invoke crisis standards of care. For example, the Idaho Department of Health and Wellness declared a statewide hospital resource crisis, enabling medical facilities to triage patients and ration care. In Montana and Alaska, where intensive care units (ICU) are operating well above capacity, physicians are being forced to make gut-wrenching choices in assigning beds to patients, and in Oregon, hospitals have been compelled to postpone treatments and surgeries, placing patients at risk of poor outcomes from non-COVID-related conditions. These same dire circumstances prevail in other states where low vaccination rates and political resistance to mask mandates have given rise to the worst outbreak of COVID cases since the earliest days of the pandemic.

Could the United States have avoided such rationing of care? Quite possibly, but we lacked the imagination to envision and prepare for such a scenario, we failed to act decisively and quickly to abate the virus’ spread when it first appeared, and we have not effectively countered the misinformation and distrust of vaccines that have hampered vaccination efforts. Consequently, the COVID-19 death toll recently surpassed 730,000, with thousands more hospitalized in intensive care. A Kaiser Family Foundation analysis estimates that over 90,000 COVID-19 deaths among unvaccinated adults between June and September 2021 could have been prevented with vaccinations. Yet, even with vaccines, crisis standards of care are critically important tools for clinicians to use in such extreme situations.

Crisis standards of care (CSC) are frameworks employed by health care providers for the allocation of resources, such as medical equipment or staff, in times of extreme resource scarcity. Crisis standards utilize triage to prioritize patients based on predicted outcome. The algorithms used to “score” patients generally consider exclusion criteria (i.e., conditions that suggest a low likelihood of survival despite maximal delivery of care), as well as measures of organ function that reflect the severity of the patient’s illness. The Sequential Organ Function Assessment (SOFA) Score, for example, numerically quantifies the number and severity of failed organs. However, the National Academy of Medicine has cautioned against using such scores or similar calculations as predictors of outcome because they may not accurately reflect the condition of patients with respiratory illness (e.g., COVID-19), and may have “limited or no utility” when allocating elastic resources that can be substituted or given in reduced dosages (e.g., medications or dialysis). Although methods for estimating the likelihood of survival in patients with COVID exist, states have not replaced SOFA scoring with COVID-specific scoring in their CSCs, and it is unclear whether they will, given that some CSCs were explicitly designed to be non-specific.

Among 29 states with published crisis standards of care, a recent study found wide variation in the rationing methods used and in the degree of discretion afforded providers, with some states delegating decision making entirely to individual hospitals, and others specifying the exact health conditions that warrant the transfer of patients from the ICU to palliative care. In contrast, the 21 states without CSCs leave triage decisions to individual organizations and clinicians. For many providers, this is an unwanted and stressful burden during public health crises.

Critics of CSCs claim that these frameworks introduce racial and ethnic inequities into decision making, citing “color-blind” triage frameworks that ignore documented racial and ethnic disparities, thus leading to higher rates of comorbidities in communities of color and placing those populations at higher risk of not receiving care. In addition, some CSCs appear to discriminate against older and disabled individuals despite a lack of evidence demonstrating a clear link with survival likelihood.

How, then, should scarce health care resources be allocated in times of crisis? First, in the absence of clear public health guidelines, the responsibility for setting triage rules should not fall to individual hospitals and physicians. At the start of the COVID-19 pandemic, only Arizona and New Mexico implemented CSCs, while physicians in New York and elsewhere struggled mightily without them. Ideally, a single national CSC framework may be most desirable. However, the diversity of beliefs, values, and opinions across the nation would make consensus difficult, if not impossible, to achieve. Nevertheless, the United States would benefit from a national commission—composed of clinicians, ethicists, and members of the public—and charged with developing a uniform CSC that adheres to values of equity and transparency. A potential model for the commission might be the Washington State committee that, many years ago, successfully made tough decisions to allocate scarce dialysis machines.

In the absence of national standards, we offer the following suggested principles for incremental improvement:

  • All states should adopt CSCs, with the 21 states currently lacking CSC frameworks remedying this situation as quickly as possible.
  • Equity should be a priority, with all state CSCs emphasizing equitable access to treatment for all.
  • Objective, standardized triage methods, based on disease severity and likelihood of survival, should be integral to all CSCs.
  • Yet, CSCs should be flexible regarding the use of disease-specific scoring methods for conditions that are less amenable to a one-size-fits-all scoring system.
  • Among patients with identical severity and survival scores, the use of lotteries may offer the most equitable means of allocating resources.
  • Public health professionals should manage CSCs without political interference from public officials.
  • To the extent possible, CSCs should be uniform within states, so that rationing decisions are consistent among hospitals and clinicians in the same state.

The United States remains within the steely grip of COVID-19, and cannot afford to repeat the mistakes of the past. Crisis standards of care—if implemented in timely, ethical fashion—offer opportunities for mitigating public health threats effectively. However, much work is needed to improve their design and equitable application in all states and territories of the United States both now and in future public health crises.

Cohen AB, Katchmar A. Rationing Health Care in the Time of COVID. Milbank Quarterly Opinion. October 26, 2021. https://doi.org/10.1599/mqop.2021.1026.

About the Authors

Alan B. Cohen became editor of The Milbank Quarterly in August 2018. He currently is a research professor in the Markets, Public Policy, and Law Department at the Boston University Questrom School of Business, and professor of health law, policy and management at the Boston University School of Public Health. He previously directed the Scholars in Health Policy Research Program and the Investigator Awards in Health Policy Research for the Robert Wood Johnson Foundation. Earlier in his career, he held faculty positions at Johns Hopkins University and Brandeis University, and spent 8 years at the Robert Wood Johnson Foundation. He is a member of the National Academy of Social Insurance. He received his BA in psychology from the University of Rochester, and his MS and ScD in health policy and management from the Harvard School of Public Health.

See Full Bio

Amanda Katchmar is a Kilachand Honors College student and MPH candidate at Boston University.

See Full Bio