Study says 1918-like pandemic could kill 51-81 million

The Lancet, the prestigious British medical journal, has published new research which attempts to estimate the mortality if a 1918-category pandemic occurred today.

According to the report, this study was unlike earlier theoretical models which made strong assumptions about attack rates and case-fatality rates among influenza cases. Instead, these researchers sought to conduct quantitative analyses based on actual vital registration data gathered during the 1918-20 pandemic.

51-81 million deaths, most in developing nations; income correlation

Their bottom line: Should a strain of influenza similar to what caused the 1918-20 pandemic to emerge in our day, it could kill 51-81 million individuals (median estimate = 62 million). This estimate is based strictly on recorded patterns of mortality in countries which had nearly complete vital registration systems for the study period of 1915-23.

According to this calculation, most deaths in a modern pandemic would occur in 15-29 year-old individuals, followed by those aged 0-14 years and 30-44 year-olds. One of the most striking estimates: 96% of the pandemic-related deaths would take place in the developing world.

The registration data from 1918-20 showed:

• The 1918-20 mortality rates were extremely high, reaching nearly 8% in one province of India.


• There was huge variation among the mortality rates. Among the locations evaluated, the pandemic death rate varied 31-fold - from a low of 0.25% to a high of 7.8%. (By comparison, global mortality from all causes of death in 2000 was 0.92%.)


• About half of the variation in mortality can be explained by per-head income alone.
  Pandemic mortality was strongly negatively related to this variable. For higher income households, pandemic mortality was lower.


• The data confirms the well-known observation that, unlike the 1957-58 and 1968-70 pandemics, mortality was concentrated in young adults, not elderly individuals.

Reasons for/against 1918 being the "worst case"

The researchers also commented about whether 1918 was a "worst case" pandemic:

In most discussions of influenza, the 1918-20 pandemic sets the upper limit, in terms of mortality, on what might occur in future pandemics. However, there is no logical or biological reason why that pandemic - albeit very severe - should represent the maximum possible mortality in a future pandemic. Random genetic mutation could, in principle, produce a more lethal virus... In addition to this uncertainty about what is genetically possible, future mortality could be larger if the 1918-20 pattern of low older adult mortality were in fact due to some acquired immunity from the pandemics of the mid-19th century.

Even as they suggest reasons why 1918 might not have been a worst case (and others have stated the same), the researchers believe their analysis "provides a plausible upper bound on pandemic mortality." I wonder about that. I don't claim to be a scientist or mathematician, so I'm hesitant to question the work of those who are. But I wonder about that claim. If nothing else, this research effort doesn't appear to factor in people today whose health is frail and whose lives are made possible (or whose health is stabilized) by taking daily medicines or treatments which weren't available in 1918. Think about how many people need:

• outpatient services such as dialysis


• in-home oxygen supplies


• prescription medicines: heart medicine, diabetes medicine, high blood pressure medicine, insulin for diabetes, the HIV cocktail, painkillers, etc.

These people may succumb to a virulent influenza at a much higher rate than the total population for two reasons. First, given their more delicate health, their bodies simply might not endure the flu as well. Secondly, if medical supply chains are interrupted, these people may lose access to the medicines which keep them alive. So even if they don't contract the flu, they could die from lack of needed medicines.

In coming days, I'll talk more about such factors which complicate our situation today versus in 1918. [UPDATE 12-26-06: Post now available.]

Lower mortality today?

The researchers also suggest reasons why a modern pandemic caused by a 1918-category virus could have lower mortality than what they estimate (51-81 million people). But these reasons strike me as not fully convincing.

First, symptomatic medical management is better now than in 1918-20. However, although individuals with access to health care in high-income and middle-income countries might benefit, health-care systems could become overwhelmed, which would attenuate this effect.

My thoughts: The hope in the first sentence is dashed in the second sentence. If we can make a blanket statement: it doesn't greatly matter what medical tools we have today vs. 1918 if we can't produce and deliver enough of them when and where needed. Overwhelmed hospitals that have no pandemic flu vaccine, limited antivirals, not enough ventilators, and perhaps shortages of other critical medicines and supplies (and staff) certainly "would attenuate" the benefit of today's "better-than-1918" medical management.

Second, antivirals such as zanamivir [Relenza] and oseltamivir phosphate [Tamiflu] might have a positive effect on the reduction of transmission and case-fatality rates. Because we have not yet seen the next pandemic virus, the magnitude of this effect cannot be quantified.

My thoughts: A broad reliance on antivirals continues to be hampered by two or three issues. First is availability. As has been discussed widely before, Roche and its licensees have a lot of work ahead of them in meeting the demand for Tamiflu. Second is speed of delivery. As I've reported, the bird flu strain of H5N1 causes such rapid sickness in humans that it's difficult to get Tamiflu to the victims quickly enough. Some experts, including HHS Secretary Michael Leavitt, have expressed concern about this. Third, while the World Health Organization and others expect antivirals to be effective on any pandemic flu strain, these newer drugs remain relatively untested on H5N1 (because there have been so few human cases of H5N1 infection).

Third, vaccination with a lag of 4-6 months from the onset of a pandemic could reach a large fraction of the high-income populations. The speed of the epidemic, perhaps affected by various efforts at quarantine, will determine the potential benefit of vaccination. [snip] In view of the restricted vaccine production capacity and the reality of health system coverage, vaccination would have little or no effect on the poorest populations.

My thoughts: I remain mystified why a vaccine continues to be dangled in front of us as a supposed resource. Of course, having a vaccine would be the best defense against a flu virus. But:

• A well-matched vaccine cannot be defined until after the pandemic begins (and the exact flu strain is known) - thus, the 4-6 month lag mentioned above, and


• The big problem: We remain hamstrung by limited vaccine manufacturing capacity (mentioned above). At current capacity, it would take about 30 years to make enough vaccine to innoculate the world population against a novel (new) virus. If new research efforts yield a universal vaccine (protects from all viruses), perhaps it will use faster manufacturing approaches. But such is not on the near-term horizon.

The last reason why a 1918-category pandemic today might have a lower mortality than estimated:

Fourth, in 1918-20, a large proportion of deaths was due to secondary bacterial pneumonia after primary viral pneumonitis. Antibiotics for pneumonia could have a substantial effect on case-fatality rates. In middle-income and low-income settings, prompt access to antibiotics could be the most affordable strategy that has the largest effect on mortality.

My thoughts: Antibiotics for secondary bacterial pneumonia certainly would help. Yet, the severe damage often done by primary viral pneumonia in 1918 and among many current H5N1 victims is striking (and untreatable by antibiotics, of course). As the World Health Organization wrote in Avian Influenza - Assessing the Pandemic Threat:

As expected, many of the deaths in 1918 were from pneumonia caused by secondary bacterial infections. But Spanish flu also caused a form of primary viral pneumonia, with extensive haemorrhaging of the lungs, that could kill the perfectly fit within 48 hours or less. The disease was so severe and its clinical course so unfamiliar that influenza was not even considered when the first cases appeared.

Study methodology

Some specifics about the study's methodology:

• Because an influenza pandemic might increase mortality for a year or two after the pandemic's peak year, the researchers assigned a three-year window (1918-1920) to the pandemic. They calculated the average mortality rate in the surrounding three-year periods of 1915-17 and 1921-23, and subtracted this from mortality in 1918-20. The residual was attributed to the pandemic.


• Although the pandemic itself didn't last three years, the three-year time-frame captures the effect of influenza on other causes of death. (The pandemic began in Spring 1918.)


• The mortality estimates of a 1918-category pandemic on the 2004 world population take into account: population size, age composition of populations, and changes in per head income for the world, regions, and major countries.


• Pandemic mortality is a function of both (a) the influenza attack rate and (b) the influenza case fatality rate. Available mortality data did not allow the researchers to determine how much of the variation in outcome was due to transmission factors or case-fatality factors.

The researchers were from Harvard University, the University of Queensland, and the Johns Hopkins Bloomberg School of Public Health. Their report - Estimation of potential global pandemic influenza mortality on the basis of vital registry data from the 1918-20 pandemic: a quantitative analysis - appears in the December 23/30, 2006 issue (Vol 368) of The Lancet (free registration).