Originally published 2020-05-03.
In mid-March 2020, as the coronavirus spread across the United States, The New York Times obtained an internal US government response plan warning that the pandemic could last 18 months or longer and might arrive in multiple waves of illness
. The document, about 100 pages long and dated 13 March 2020, was marked unclassified but "for official use only" and not for public release.
The warning turned out to be realistic, and in some ways too optimistic. COVID-19 came in repeated waves for well over 18 months, the first vaccines arrived in December 2020, and the World Health Organization did not end its global health emergency until 5 May 2023. Here is what the plan said, the scientific modelling that shaped decisions that same week, and how those early projections compare with what actually happened.
What the federal plan warned
The plan was a contingency document, written to help agencies prepare for a long emergency rather than to predict an exact course. Its main expectations were:
- A long timeline. The outbreak could continue for a year and a half or more, with cases rising and falling in waves.
- Pressure on hospitals. Growing numbers of infections among older and high-risk people would push the health care system towards its limits.
- Shortages. Supply problems could affect health care, emergency services and other parts of critical infrastructure.
- Strain on government. State and local agencies, essential services and communications networks could be stretched and at times unreliable.
The document also reviewed tools the government could use, including school closures and the Defense Production Act.
The Defense Production Act
The Defense Production Act took effect on 8 September 1950, early in the Korean War. It allows the president to require companies to prioritise contracts for goods considered vital to national defence and to fund expansion of domestic production.
On 18 March 2020, President Donald Trump signed an executive order invoking the act and declaring ventilators and personal protective equipment essential to national defence. A further order on 23 March targeted hoarding and price gouging of medical supplies. In early April the administration used the act to direct several companies, including General Electric and Medtronic, to secure materials for ventilator production, while carmakers such as General Motors and Ford were enlisted to build ventilators.
Imperial College's model and the 18-month estimate
The same week, on 16 March 2020, a team from Imperial College London led by epidemiologist Neil Ferguson published its now-famous Report 9 on non-pharmaceutical interventions. It estimated that, if the virus spread unchecked, there could be around 510,000 deaths in Great Britain and 2.2 million in the United States. Those figures described a scenario with no response at all, not a forecast of what would happen.
The report compared two broad strategies:
- Mitigation aimed to slow the spread, protecting the most vulnerable and flattening the peak of hospital demand without stopping transmission. The authors concluded this would still likely lead to hundreds of thousands of deaths.
- Suppression aimed to push case numbers down and keep them low, using population-wide social distancing combined with home isolation of cases, quarantine of households and school closures.
The modellers judged suppression the preferred option where it was possible, but warned that measures would need to stay in place, or be switched on and off as cases rose, until a vaccine was available, potentially 18 months or more
. They also described COVID-19 as the most serious threat from a respiratory virus since the 1918 H1N1 influenza pandemic, the so-called Spanish flu.
The report helped shift policy in both the UK and the US towards lockdowns and stay-at-home guidance. It also drew criticism: some scientists questioned its assumptions about how many people would follow quarantine rules, and the computer code behind it was attacked as unreliable, although later independent efforts reproduced its results.
How the predictions compared with reality
Several of the early warnings proved accurate. Shortages of masks, gowns, testing materials and ventilators were real and severe in spring 2020, especially in hard-hit cities. The virus did come in waves, driven by seasonal patterns, relaxed restrictions and new variants such as Alpha, Delta and Omicron.
The vaccine timeline was faster than the 18-month estimate. The UK was among the first countries to approve a COVID-19 vaccine, from Pfizer and BioNTech, on 2 December 2020, and the US began emergency use later that month, roughly nine months after the plan was written. However, vaccines did not end the pandemic on their own. Uneven global access, waning immunity and new variants kept the virus circulating, and the WHO's emergency declaration stayed in place until May 2023, more than three years after it was issued on 30 January 2020.
The human cost was enormous. More than a million people in the United States died of COVID-19, and the global count of confirmed deaths passed seven million. The WHO estimated that the true toll, measured as excess deaths, was around 14.9 million in 2020 and 2021 alone.
Planning scenarios versus predictions
A common misunderstanding at the time was to treat documents like the federal plan or the Imperial report as forecasts with fixed end dates. In reality they were scenarios, tools for asking "what if?" so that governments could prepare. Their numbers depended on assumptions about how the virus spread, how people behaved and how quickly science could deliver treatments and vaccines, all of which were uncertain in March 2020.
Hindsight can also be unfair to early warnings. When a model's warnings lead governments to act, the worst case does not happen, which can make the warning look exaggerated afterwards.
Lessons for the next pandemic
- Stockpiles matter. The supply shortages the plan predicted showed how much depends on having protective equipment and medical supplies ready before a crisis.
- Clear communication builds trust. Explaining that projections are scenarios, not promises, helps people understand why advice changes as evidence grows.
- Science can move fast. Years of earlier research on mRNA and coronavirus vaccines made the record-breaking vaccine timeline possible.
- Health advice is personal. For questions about vaccines, boosters or your own risk, talk to a doctor or pharmacist.
Read more on how the outbreak unfolded in our earlier coverage of doctors on the front line of COVID-19 and one Italian's account of the stages his country went through.
Sources: Live Science (Yasemin Saplakoglu), 21 March 2020, the original report this article was based on; Imperial College London: Report 9; Wikipedia: Defense Production Act of 1950; Wikipedia: COVID-19 pandemic; Wikipedia: Neil Ferguson.
COMMENTS