A new study published on the preprint server medRxiv* demonstrates the influence of prior infectious outbreaks in a large population on succeeding pandemic outcomes. This paper demonstrates that places in the United States that suffered higher numbers of deaths in the 1968 influenza pandemic have lower rates of death from coronavirus disease 2019 (COVID-19) among older people as compared to younger populations. This finding deserves detailed study to elucidate the underlying protective mechanism operating in these cases.
Study: The 1968 Influenza Pandemic and COVID-19 Outcomes. Image Credit: Koy_Hipster / Shutterstock.com
The COVID-19 pandemic led to the implementation of many different social and economic restrictions in an attempt to contain the transmission of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), as well as spurring research into vaccines to arrest the outbreak.
However, many researchers are seeking to determine the best way to prepare for future pandemics, as well as understand what factors underlie population-level susceptibility to such pathogens. Earlier work has shown that racial, demographic, social, and policy factors contribute to the spread and clinical severity of the current pandemic, while the responses to the crisis have varied widely as well.
Host factors such as biology, genes, and immune function also play a major role in COVID-19 outcomes. At the same time, it is important to understand how past exposures to crisis situations, health-related or otherwise, affect current responses, both individual and public, to the COVID-19 pandemic. The mechanism linking these situations could be behavioral, which can include readiness to comply with public health measures like wearing masks, institutional, or social, to name a few possibilities.
Some interesting postulates have emerged in previous studies, such as the drastic decline in tuberculosis after the great 1918 Spanish flu pandemic; or the part played by plague and smallpox exposures in enhancing the expression of a human immunodeficiency virus (HIV)-resistance gene.
The current study looks at the 1968 H3N2 novel influenza virus pandemic, often called the Mao Flu or Hong Kong Flu. The spread of this flu caused about 100,000 deaths in the U.S., out of about 200 million living at that time and 1-4 million deaths globally. About a third of the current population of the U.S. was living at the time.
About 40% of the deaths from the 1968 pandemic were among those under 65 years, unlike the COVID-19 outbreak. However, no nationwide lockdowns occurred, and vaccine development was slow. Schools and universities were closed in 23 states.
The current study looks for evidence of a link between severe disease in the 1968 flu pandemic and the outcomes from the COVID-19 pandemic, up to the end of 2020. The researchers tried to uncover individual, social, and policy-related factors that could account for such a link, if present.
The findings of the current study show that counties with a high mortality rate in the 1968 flu outbreak showed lower deaths among those aged 60-69 years, but not among the 40-59 years age group. People who lived in the 10% of counties with the highest ’68 flu deaths rates had 1-2% lower COVID-19 death rates compared to the average county rates.
The COVID-19 mortality rates were found to be inversely related to the density of population, while they were higher in the elderly, Black, and Hispanic populations. These factors are not related to the 1968 flu outcomes and are therefore not likely to be confounding factors. However, the ’68 flu death rates were linked to the proportion of elderly in the population and, to a smaller extent, with the Black population.
In locations where the ’68 flu-related death rates were high, the current COVID-19-related deaths are lower in those age groups that were alive at the time of the earlier pandemic. The greatest and most significant differences in death rates are among those who are 50 years or older as compared to those younger than 50, with slightly lower but still highly significant differences among those who are 60+ compared to controls aged 59 years or less.
For those over 60 years of age, hospital admissions due to COVID-19 are also lower as compared to the overall rate, while the difference between under-60 and over-60 groups in terms of hospitalization rates are also greater. Importantly, this trend is observable in nursing home data as well, where the case fatality data for COVID-19 is much lower among those who had higher mortality during the ’68 pandemic.
Using 48,000 healthcare records, linking a medical procedure, including hospitalization, to a COVID-19 diagnosis made within 30 days before the procedure, the same trend was detected, favoring older groups born before 1968 in places where the flu pandemic had a significant impact.
The researchers found no evidence that this reduction in mortality was due to an increase in hospital beds at the time of the earlier pandemic, while it appears that the places worst affected by that outbreak had lower mask-use rates than other locations. Finally, compliance with public health containment directives was low in these counties.
The researchers indicate that the outcomes of the 1968 flu pandemic are linked to those of the current outbreak. The 1-2% reduction in death rates due to COVID-19 seen in the top 10% of counties in terms of 1968 flu mortality is not due to confounding factors and is robust across populations including hospitals and nursing homes.
The locations with adverse outcomes in 1968 had lower COVID-19 mortality rates, ruling out a missed variable that would make these places vulnerable to both outbreaks. The highest significance is seen in the fall of 2020, declining with the rapid and extensive spread of COVID-19 and vaccine rollouts.
This may be affected by the rising numbers of deaths in the U.S. that broadens the county sample size. Alternatively, nursing homes may have adopted better norms of care over time as the current pandemic evolved, leading to a reduction in mortality that is independent of other factors.
The primary factor operating here appears to be individual, rather than not behavioral, social, or related to public health policy. The reduced case fatality ratio in nursing home residents is especially striking, as these individuals have less control over the risk mitigation practices followed around them.
The evidence presented here suggests a biological phenomenon, perhaps trained innate immunity, that exerts a general antiviral effect. This could be strengthened by genetic and lifestyle factors that were selected by the 1968 pandemic and later. Earlier studies have demonstrated such a link between 1918 flu exposure and cardiovascular disease risk.
Much more work will be needed to validate these findings and explore putative mechanisms. Secondly, the public health interventions implemented in the current pandemic may perhaps affect future generations in terms of their vulnerability to new outbreaks, which deserves careful consideration as well.
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.