Bioethics Forum Essay
When Endemic Disparities Catch the Pandemic Flu: Echoes of Kubler-Ross and Rawls
Over the past several years, I have served on a couple of panels considering the clinical and ethical challenges posed by pandemic flu. Our concern was the threat posed by the avian variety brewing in China; not once did we discuss an alternative viral vector. Nonetheless, in light of the current fears that pandemic swine flu might be emerging in Mexico, the panels’ deliberations offer useful insight into how groups warm to the threat of pandemic illness and turn almost primal fears about survival and existence into cogent and rational policies. Understanding how this dialogue evolves and generates public health policies can help us assess those policies.
Let me provide one perspective. The pattern of discourse at these panel sessions is reminiscent of Elizabeth Kubler-Ross’s account of the stages of grief. Denial, anger, bargaining, depression and acceptance progressively mark the deliberations. And after completing this sequence, we typically come to discussions of aftermath and bereavement. We ask how we will persevere and reconstitute civic society after a pandemic. Ultimately, we conclude that our resilience will depend on the justness of our decisions and attention to endemic and emergent health care disparities.
The meetings all start with the somber realization that a pandemic flu will forever change one’s life and existence. It is akin to walking into a doctor’s office and learning about a malignant biopsy result. We enter from the sunshine into the darkness of a PowerPoint presentation laying out facts about death and devastation on the scale of warfare. I am reminded of a Village Voice piece by the writer Paul Cowan as he battled the leukemia that would take his life. “The world is composed of the sick and the not-yet-sick,” wrote Cowan: crossing some lines forever changes one’s perceptions and priorities.
So it is when listening to one of these flu presentations. ’All the conventions we take for granted might come to an end if a pandemic came to our shores. This realization takes hold as the group hunkers down to listen to an epidemiologist who explains how a pandemic could occur. First there might be mutations of a bird flu in China that enables transmission from human to human. A cluster of illness develops in a rural province. A tourist makes a visit and boards a jumbo jet home to New York. On the way he – and everyone he bumps into – incubates and spreads contagion. And there you have it: a pandemic leaps 10,000 miles from the hamlet where it all began.
As the morning progresses, data is shared about the toll a pandemic would take. Models depict progressively worse scenarios. For example, the U.S. Health and Human Services Pandemic Influenza Plan (11/2005) envisions that a “moderate” pandemic akin to the 1958 or 1968 flu season would result in 90 million Americans becoming ill, or 30% of the population. It would result in 45 million outpatient visits, 865,000 hospitalizations, and 128,750 admissions to intensive care, with 64,875 of those requiring ventilatory support. There would be 209,000 deaths.
A “severe” pandemic, like the 1918 Spanish flu, would afflict the same number of people but extract a catastrophic toll. There would be 9.9 million hospitalizations, 1,485,000 million ICU admissions, 742,000 needing a ventilator, and 1,903,000 deaths.
At this point, denial kicks in and the group rallies around a hopeful comment. Someone suggests that we really don’t know how this will play out and it hasn’t happened yet. Fears of a pandemic flu remain hypothetical. We can’t compare death rates from a modern epidemic to the carnage of 1918, before there were antibiotics or ventilators and when troop movements just after World War I did more for viral spread than nature never could. Someone else observes that virulent viral outbreaks often burn themselves out. A virus’s lethality is often paired with frailty. A virus whose effects are horrific but brief and contained tends not to travel well. The Ebola virus can have fatality rates of 50% to 80%, although its geographic range has been limited and outbreaks have been infrequent. Similarly, Severe Acute Respiratory Syndrome was devastating to those who contracted it, with a death rate estimated at upwards of 15 to 20 percent in China and 6.5 percent in Toronto, but there were just over 8,000 cases worldwide according to a WHO report.
Reassured – and collectively gasping for alternative scenarios – our group dynamics grow more hopeful. No one wants to cause a panic about hypotheticals. Okay, the group inclines, we’ll grant you this might happen, but it won’t be so bad. Let’s think about rather minor inconveniences, let’s plan for what we might handle: how would hospitals cancel elective surgeries and ensure that nurses and doctors report to work as usual before things got back to normal?
There is some reassurance in this level of denial. But it lasts only as long as the psychological defense is sustained. Once the true gravity of the threat is exposed, avoidance is viewed as increasingly irresponsible; an abdication of a collective duty to use our superior knowledge of the threat to act responsibly and engage in contingency planning.
At this juncture, though, we overcome our defense mechanisms and entertain the inconvenient truths about what might need to be done to avert disaster. For example, using the antiviral drug oseltamivir phosphate (Tamiflu) as prophylaxis against infection would require 56 days of medication, the period during which an Avian flu surge would put individuals at risk. A five-day course of 75 mg pills taken twice a day for conventional flu can be had on the Internet (with a prescription, I hope) for about $100. Even with a sizeable bulk discount, treating tens of millions of Americans would be hugely costly.
And now Kubler-Ross’s anger. If we can bail out insurance giants, why can’t we make proper provisions for universal prophylaxis, itself a sort of insurance policy? Sometimes anger expresses proper indignation. Increasing the availability of antivirals from current levels, designed to protect essential personnel, to a broader swath of society makes good sense.
Now we are into the bargaining phase. What would it cost to prevent this pandemic pharmacologically? One thing we’ve learned from the public discussion of the swine flu threat is that the federal government has stockpiled 50 million doses of Tamiflu. Would that be enough to counter an Avian threat? Could we ramp up production? And here is the bargaining: Could we make a deal with Big Pharma? Buy in bulk? Get a bargain? And why is the cost of these drugs off the table anyway? Isn’t the provision of a common defense the basic role of government anyway? In fact, isn’t a federal role appropriate, especially given the unprecedented stimulus spending? If nothing else, wouldn’t it put a lot of people to work and be economically valuable in its own right?
But even as we speak of averting the whole dire scenario through simply taking a couple pills for a couple of months, depression begins to seep back into the room. Oseltamivir phosphate is effective in the treatment of swine flu and some strains of avian flu, but there is no guarantee it will work against a new variant, either as a prophylactic or a therapeutic agent. And then consider ventilator availability. In 2005, Erich Giebelhaus and Lewis Soloff surveyed New York City hospitals to gain an “accurate number of ventilators and ‘surge ventilators’ … by type of ventilator and population served.” Every hospital responded. As of December 2005, they found there were 1,857 full-featured ventilators that could be used by adults or kids, an additional 561 suitable only for adults and 270 more for neonates, for a total of 2,688 full-featured ventilators. There were also another 614 automatic resuscitators, 790 portable anesthesia machines (which could deliver oxygen), and 186 portable ventilators.
The New York City Department of Health and Mental Hygiene used these data to further extrapolate ventilator availability in the case of a pandemic. Employing CDC FluSurge 2.0 projections for avian flu, they made their estimates based on overall population attack rates of 15%, 25%, and 35% and the percentage of these patients who would require ICU care. Of patients needing ICU care, they further risk-stratified what percentage would need a ventilator. They calculated data for ICU admission rates of 15%-25% and ventilatory requirements for admitted patients ranging from 50%-100%, assuming that 60% of ventilators are already in use (for patients with other illnesses and those undergoing surgeries), a 25% attack rate with 15% and 7.5% of ill patients needing an ICU bed and ventilatory support, respectively. This combination of factors left 514 unused ventilators. However, when truly pandemic attack rates occur, things become more dire. If the attack rate were 35% with 25% of these patients needing ICU care and 100% of such patients needing a ventilator, there would be a deficit of 1578 ventilators in the system, assuming that 60% of ventilators were already in use. The deficit jumps to 2,662 ventilators if all ventilators were already in use for other purposes. This deficit is roughly equivalent to all the full-featured ventilators in New York City. Although this is bad, it would be worse yet in adjoining areas where the medical infrastructure is less well provisioned.
When these constraints become clear, we realize that none of our defense mechanisms can assure us that a dire pandemic would not have a devastating impact. With this realization comes the first inkling of acceptance: if the “big one” struck, there would be no way around some sort of allocation scheme to make use of available resources. And when we accept this new reality, we begin to think differently about our charge. We start to reconcile ourselves to the hard choices that will have us progressively abandon an individualistic approach to patient care and adopt a harder, more utilitarian strategy.
In the sphere of hospital-based clinical ethics, this might mean the rescinding of normal policies and procedures designed to protect patient and family choice. For example, under New York law, when there is an objection to a do-not-resuscitate order, the dispute goes to the ethics committee for mediation and a statutorily mandated 72-hour hold. And if a family objects to a brain death determination on religious or moral grounds, then practitioners are obliged to make a “reasonable accommodation.”
We need to ask if policies like these, which protect patients and families under normal circumstances, still make sense when catastrophe strikes. Although they would logically improve population-based outcomes, changing these policies would alter hard-won achievements prizing patient and family choice, if only on a temporary basis. Moreover, they would likely engender suspicion from groups that already feel disenfranchised because of pre-existing inequities of care that have been well documented nationally and locally.
What happens when these endemic disparities catch the pandemic flu? We already see discordant mortality rates from swine flu when comparing Mexico to the United States. Within the context of U.S. healthcare, will disparities mean that some individuals or groups will be put at greater risk? Will lack of medical infrastructure lead to especially hard triage decisions in some hospitals and less onerous ones in others? And will the allocation of ventilators make inequities worse?
On one of the avian flu panels I served, we became very concerned that scarce ventilators might be directed preferentially to health care professionals who became ill in the performance of their duties. Although this might make sense if these essential personnel could recover in time to be redeployed to provide care, the time course to recovery makes the rationale for preferential access suspect. We were concerned that privileged access to ventilators might mean that the only people to secure a ventilator in some communities would be health care professionals. We viewed this as patently unfair because it would mean that others, including children, might be denied ventilators, leading “to the appearance of favoritism, in which those who devised the rationing system appeared to reserve special access for themselves.” We concluded that allocation of ventilators should only be based on clinical indications.
The problem of inequity is further compounded if we address the fair distribution of ventilators at a community level. If we return to the prevalence and distribution data collected by Giebelhaus and Soloff, we see that the distribution of full-featured ventilators across the City was quite variable across the boroughs. The average number of ventilators per 100,000 people was 23.2 and ranged from 14.1 in Queens to 39.2 in Manhattan.
These disparities would only get worse in a pandemic, when triage decisions would be institutionally based and where ventilators would become fixed assets. Not only would institutions not want to share their limited resources, but no one would want to accept a ventilator that could be a source of contagion. Moreover, it would be unwise to transfer patients from overburdened facilities to facilities not yet overrun by the pandemic. It would harm the patient and could compromise another institution, thereby increasing scarcity of limited resources.
Disparities would only become more critical during a pandemic flu. Imagine an outbreak equally affecting two neighborhoods in Manhattan and Queens. Rationing ventilators would be especially harsh in Queens, where the ventilator density is just over a third of what it is in Manhattan. This poses an immediate challenge to fairness and equity, if triage decisions are to be made locally. When appeals processes are taken out of the mix and opportunities for third party advocacy are decreased, some communities might rightly feel that they have lost voice. Scarcity combined with the loss of some regulatory protections could lead to hidden rationing: decisions might appear fair when viewed from within the perspective of one hospital, but grossly inequitable when compared to choices made at facilities with more ample resources.
I fear that if we fail to rectify these disparities, it will lead to a moral haunting – a complicated bereavement, if you will – when the pandemic ends and some of us emerge on the other side. Imagining the aftermath of such biologic terror, when we are compelled to ask ourselves about legacy and the consequences of our actions. One colleague – I believe he was a pediatrician – raised the question of aftermath when he questioned a utilitarian ventilator-rationing scheme that would have favored adults over children. He asked what sort of society would we become if we didn’t take care of the children. What would be our future after the pandemic when we tried to come back together as a community? How could we heal after making any decision that sacrificed one segment of the community for another?
While inaction in the face of disproportionate death is not equivalent to the intentional infliction of harm, neither is inaction permissible. To stand by in the face of this unequal burden is to be complicit in the outcome. Ideally, we should strive for a generic decrease in health disparities. But short of that aspirational goal, when confronting the more pragmatic threat of pandemic flu, it seems only just to level the playing field before disease strikes.
A couple of years ago in the Hastings Center Report, Lori Uscher-Pines and colleagues made a similar argument [registration required] for social justice on a global scale in the face of the pandemic threat. On a more local scale, if a pandemic were to come to New York City, equity could be achieved, in part, by using federal ventilators held in a reserve stockpile to normalize ventilator densities in areas at risk. The prepositioning of emergency supplies is more than a polemical call for action. With global surveillance, public health authorities will have advance knowledge to prepare for an approaching pandemic.
In the case of New York City, if all the boroughs were at a hypothetical equal threat of a pandemic avian flu outbreak, the first of reserve ventilators should go to Queens and then to Brooklyn and Staten Island so that their ventilator density would equal the prepandemic status quo in Manhattan. Once equity had been achieved with Manhattan, each county would receive additional ventilators according to their need. This will help ease the inequities of triage decisions faced by hospitals with different levels of infrastructure. Moreover, a promise of prepositioning ventilators ahead of a wave of illness will help build trust with communities that need to be part of the process of planning for a pandemic event and that may harbor concerns about fairness. Ultimately, surviving any plague is really like John Rawls’ veil of ignorance.
As we look to the future, all we know is how we are now situated, privileged or not, rich or poor, blessed with family or not. But after a pandemic flu we cannot know what we might have lost in treasure or good fortune. How might it be after the dreaded loss of a child from a random infection? What will become of one’s workplace or neighborhood if the flu is tsunami-like in its effects? Writing in the Journal of Health Care for the Poor and Underserved of Hurricane Katrina’s effect on his hometown of New Orleans, Wayne Riley, the newly appointed president of Meharry Medical College, recounts wholly unanticipated devastation and pain for him and his extended family upon his return as an aid worker. The point of Riley’s compelling narrative is that none of us can predict the consequences of disaster. We might try to make prospective choices based on our current sense of self-interest, but we don’t entirely know what those interests will be in the aftermath of crisis, when the old status quo and perks of privilege are pierced.
There is nothing more akin to the proverbial commons than an epidemic, which looks past individuals and takes aim at the herd. Protection comes from group preservation. For this reason, we need to be truly liberal in our planning. We need to make every effort to be fair, not knowing whether an equitable approach will be a tax or a boon for each of us individually. Of that we will be ignorant. What is certain is that history will have recorded the justness of our decisions and actions, whether it be for a current or a future pandemic threat.
Acknowledgements: I want to thank my fellow members of the NYS DOH/ NYS Task Force on Life & the Law, NYS Workgroup on Ventilator Allocation in an Influenza Pandemic and The Greater New York Hospital Association’s Ethics Working Group on Avian Flu for educating me about the pandemic threat and the privilege of rich conversation about these difficult issues. I am especially grateful to Dr. Debra Berg of the New York City Department of Health and Mental Hygiene and Dr. Nathaniel Hupert of Weill Cornell Medical College for their expert advice and their ongoing efforts to protect the public health. The opinions expressed herein are my own and do not necessarily represent the views of others or any of the aforementioned agencies or entities.
Joseph J. Fins is a Professor of Medicine and Public Health and Chief of the Division of Medical Ethics at Weill Cornell Medical College. He is a Fellow of The Hastings Center.