SEVERAL weeks ago, I went to a case presentation of a patient with influenza A who had been ventilated and spent 8 days in the intensive care unit (ICU). The patient was part of this year’s unexpected and unseasonal surge in cases across Australia and, given the uncertainty about this atypical event, concern was expressed about what might play out over the course of the rest of the year.
When I asked what the ICU team would do if the demand for ventilators outstripped the supply, the reply was that patients would be transferred to other centres.
What would happen if, during a pandemic, other centres did not have spare capacity and could not accept transfers? The immediate reply from a senior ICU consultant was that they would have to decide which patient, or patients, would not be offered ventilation. This was neither a hypothetical question, nor was the response unpremeditated.
So how can clinicians prepare to deal with the challenges of an influenza pandemic? The first thing to recognise is that, unlike the challenges of unexpected crashes, explosions and natural calamities, the spectrum of individual clinical consequences of an influenza outbreak is generally understood and can be anticipated. However, in a pandemic the number and severity of cases is not predictable, and it is this challenge that I will explore first.
During the 2009 H1N1 influenza pandemic, approximately 23% of the residents of New South Wales experienced an influenza-like illness. Fortunately, the disease was mild for most people, and less severe in the elderly. Overall, 131.6 fewer died per 100 000 of people aged older than 80 years than is usual most winters, with the result that in 2009 NSW’s all-age all-cause mortality was lower than average. While about 25% of the influenza cases occurred in people older than 70 years of age, unusually most deaths occurred in persons younger than 60 years.
According to a study of the impact of the 2009 pandemic on ICUs in Australia and New Zealand published in the New England Journal of Medicine, the demand for ventilation and extracorporeal membrane oxygenation (ECMO) stretched, but did not overwhelm, ICU capacity. During the 3-month study period, 722 patients with confirmed H1N1 influenza were admitted to an ICU in Australia or New Zealand. The majority were under the age of 65 years, with pregnant women, people with obesity (body mass index > 35) and indigenous people at greatest individual case-based risk.
However, only 51 people died from influenza in NSW between May and September 2009. So, it seems that particular strain of H1N1 influenza did not have a serious impact on most of those infected while the few patients who were seriously ill were well served by high quality, high-tech health care, including admission to ICU and the use of ECMO.
It is probably fair to say that in 2009 we got lucky.
Once again, I turned to our state and national management plans hoping to find information and support for health care leaders, clinicians and the community who will have to deal with painful resource allocation decisions in the event of a severe influenza pandemic.
Neither our Queensland Health Pandemic Influenza Plan, May 2018 nor the Australian Health Management Plan for Pandemic Influenza provide much help, as they do not make any quantitative comment on the possible scale of a pandemic disaster.
Fortunately, the American Pandemic Influenza Plan: 2017 Update includes a table (on p 44) that quantifies the impact for a variety of scenarios. At this point, I suggest that you also read the concise review of pandemic influenza by Taubenberger and Morens (2006) and, for a much broader historical and social perspective, Laura Spinney’s 2017 book Pale rider: the Spanish flu of 1918 and how it changed the world.
As you will see from these documents, the scope of a serious pandemic is another kettle of fish.
“An estimated one third of the world’s population (or ≈ 500 million persons) were infected and had clinically apparent illnesses during the 1918–1919 influenza pandemic. The disease was exceptionally severe. Case-fatality rates were > 2.5%, compared to < 0.1% in other influenza pandemics. Total deaths were estimated at ≈ 50 million and were arguably as high as 100 million”.
In 1918, the elderly generally experienced a mild influenza. The vast bulk of deaths were among the young, both children and young adults. ICUs with ventilators and ECMO had not been invented; there were no antivirals or influenza vaccines, nor any antibiotics to treat secondary pneumonia. Nor were there the countless patients whose lives are now sustained by the ongoing application of high-tech, high intensity medicine. In 1918, nobody had any expectation that they could be cured if they became very sick. At best, patients received nursing care, tender loving care, and perhaps a little morphine.
While we may be fairly confident about the proportion of the community infected and the range of influenza severity in individuals, nobody can predict the distribution of severity nor the impact of pandemic we will experience in our community. Perhaps, as a starting point for discussion, the 1918 pandemic should be considered a worst-case scenario.
If a pandemic with a similar impact occurred in Australia today, with a population of around 25 000 000 people, about 8 million (one-third) would develop influenza. Of these, 200 000 (2.5%) would die over several months, in addition to the routine baseline of about 160 000 (in 2017) deaths annually. The latter may increase if health services were overwhelmed and had difficulty sustaining modern therapies for those who are dependent on them.
In 1918, many died very quickly and it is unclear how many of those who died from fulminant disease could be saved using modern technology.
Even if, for example, only half as virulent as in 1918, the pandemic would still be a disaster of cataclysmic proportions (this is within the range suggested in the US Pandemic Influenza Plan). ICUs in NSW were stretched when, in 2009, 51 people died. In NSW, with a population of around 7 million, if one-third of the population were infected with influenza with a case-based mortality of about 1.25%, about 30 000 people could die.
About 30 people currently die each week in my community of about 200 000. In this scenario, an additional 100 would die every week for 2–3 months.
It is clear that, given average length of stay in the ICU and the inevitable absenteeism of skilled staff due to illness, the sheer volume of demand would rapidly overwhelm our capacity to provide 21st century highly intensive treatment to match the needs and expectations of all patients.
How should we respond as health workers in the event that our health system has more critically ill patients than it has capacity to treat?
Back to Australia’s pandemic plans. The Queensland Plan does not address how we should deal with clinical demand that exceeds capacity, and refers the reader to the national document for high-level decision making. The national Plan candidly states:
“An influenza pandemic represents a significant risk to Australia. It has the potential to cause high levels of morbidity and mortality and to disrupt our community socially and economically.”
And follows on with the optimistic:
“Like any other hazard, Australia will approach this risk by undertaking activities to:
Subsequently it states:
“The Australian Government and state and territory governments will work together to develop new models of care to manage patients and agree on influenza triage criteria (if required).”
This reactive approach seems inappropriate in the light of warnings from other jurisdictions:
“As the demand for critical care resources during a pandemic will likely be sudden and occur over a short time frame it will be too late at that point for a hospital and ICU to create a plan to respond to the surge in volume of critically ill patients.”
Which brings me back to the question: How can clinicians prepare to deal with the challenges of an influenza pandemic?
The consideration of this question immediately raises supplementary questions that are far blunter:
- If there are more sick people than we can possibly treat, how should we decide who to treat?
- How should our community be prepared for the realities of a pandemic and how might we reach consensus in prospectively deciding appropriate responses?
- What ethical framework would guide decision making, and would the law protect clinicians and others who are forced to make choices between patients?
- Which parts of health care would be wound back (and treatments curtailed) when staffing was cut by illness? For example, might it be more effective to close ICUs so that the large numbers of staff required to treat small numbers of people could care for much larger numbers of people and prevent a larger number of deaths?
- What is the role of palliative care, how would it be made available for all the people who were dying, and who would deliver it? I believe that palliative care has a big part to play in disaster management.
No doubt you can think of additional important questions.
Clearly, senior ICU consultants have already been thinking about what they will have to do; however, there is little evidence that planners have moved beyond organisational matters. My review of the large number of Australian documents that emerge on a routine internet search for pandemic disaster planning in Australia did not find anything that addressed the practical problems of resource allocation – if such documents exist, they should be highly visible.
In fact, I found more discussion of how to deal with a surfeit of dead bodies than guidance for the ethical clinical care of those still alive.
For a far broader consideration of these issues I suggest that you read two documents from the United States: Crisis standards of care: a systems framework for catastrophic disaster response (2012) and the Louisiana Department of Health and Hospital pandemic influenza plan 2011. These documents (among others) were created in the aftermath of Hurricane Katrina and other disasters that exposed a variety of difficult challenges across the US community. They openly discuss urgent issues such as community engagement in the ethically challenging task of prioritisation in resource allocation, the need for legal protections for clinicians and volunteers who must act outside their formal roles, and community norms during extreme events, and, interestingly, the role of palliative care as a means of demonstrating that, even when life cannot be prolonged, care will be continued.
Our Australian plans do not address such issues.
Australian society has metamorphosed since 1919. We no longer accept instruction without explanation, and expect to be treated when we are seriously ill. These two factors alone make public engagement and effective communication essential before the onset of an influenza pandemic. Without prior understanding of the principles and details of our pandemic responses, the community may be very skeptical when, as clinicians and community leaders, we have no choice but to implement painful choices in the interests of the community as a whole.
Our community also needs to know before the event that, in times of calamity such as an influenza pandemic, we will still care for them, even when we cannot provide the treatment they have been encouraged to expect. A big part of this response is well supported front-line palliative care services.
Dr Will Cairns OAM is on the verge of retirement from his medical career, first as a GP, and subsequently as a specialist in palliative medicine.
The statements or opinions expressed in this article reflect the views of the authors and do not represent the official policy of the AMA, the MJA or InSight+ unless so stated.