Ideas to Action:

Independent research for global prosperity


Global Health Policy Blog


Pandemic risk is largely underestimated and actions to prepare for outbreaks are grossly underinvested. The spread and severity of COVID-19 felt like a surprise to many policymakers – a clear case in point. But careful analyses are suggesting that future pandemic risks are significant.

Last month, CGD hosted an event on predicting the frequency and scale of future pandemics to discuss what’s to come after the COVID-19 pandemic and the critical task of preparing for future outbreak probabilities. We were joined by Ben Oppenheim and Nicole Stephenson, who are modeling future pandemic risk at Metabiota; Robert Agyarko and Cristina Stefan, who are examining risk with an eye to ensuring rapid outbreak response at African Risk Capacity; as well as Dean Jamison, who has long argued that pandemic risk is underappreciated.

Here are three key takeaways from the event:

1. The next pandemic could be much sooner and more severe than we think

COVID-19 is broadly viewed as being a “once in a lifetime” or “once in a century” pandemic. Modeling work based on historical data shows that this is not necessarily the case.

Ben Oppenheim and Nicole Stephenson reminded us that COVID-19 is not the only pandemic of this decade; we’ve seen a drumbeat of epidemiological events. Their analysis, based on historical data on epidemic frequency and geographic distribution, shows that the frequency and severity of spillover infectious disease – directly from wildlife host to humans – is steadily increasing. Figure 1 illustrates this trend. (To the right of the red dashed line is a projection, and the dashed blue line shows the 95 percent prediction interval.)

Figure 1. Trend: frequency and severity of spillover events

Source: A Global Deal for our Pandemic Age: Report of the G20 High Level Independent Panel on Financing the Global Commons for Pandemic Preparedness and Response, 2021

Using their model, based on simulated epidemic and pandemic catalogs, the Metabiota team also estimated the probability of another event as damaging as COVID-19. Metabiota’s approach is significant, Dean Jamison emphasized, because it applies techniques from both epidemiological and catastrophe modeling to estimate the expected impact of future outbreaks, and assesses long-term pandemic risk in a way that previous epidemiological literature did not.

Their results are worrying and put risk much higher than many seem to expect. They estimate the annual probability of a pandemic on the scale of COVID-19 in any given year to be between 2.5-3.3 percent, which means a 47-57 percent chance of another global pandemic as deadly as COVID in the next 25 years. These numbers illustrate the importance of preparing immediately for future outbreaks. Outbreaks will occur – it’s just a matter of when and where.   

2. The most vulnerable countries are at a higher risk for spillover events and less equipped to respond

We have some insight into where outbreaks occur, based on that same historical data. Figure 2 from Metabiota illustrates the geographical distribution of spillover events related to coronaviruses, filoviruses, and other non-influenza risks and shows a high incidence across Sub-Saharan Africa and South Asia – the size of the red point is proportional to the number of spillover events. In other words, low-income countries are disproportionately affected by infectious disease spillover events, and they are also the first line of defense against pandemic threats.

Figure 2. Geographic distribution of 88 reported spillover events since 1960

Source: Metabiota, Human Epidemic Database, 2021. (For a description of the approach to epidemiological data collection and structuring, see here.)

Recognizing both this heightened risk and the need to respond, African Risk Capacity (ARC), a specialized agency of the African Union, is working on financing options for outbreak response through risk profiling, contingency planning, parametric insurance and risk transfer, and outbreak modelling. ARC’s previous modeling work shows that the annual exceedance probability (the probability of a given magnitude event or greater to occur over the next year) of experiencing an outbreak having 6+ Ebola cases is 10 percent each year in Uganda, for example (Figure 3). Robert Agyarko from ARC pointed to the need for rapid containment to check spread and reduce impact.

Figure 3. Example exceedance probability/recurrence interval results for Ebola cases, Uganda (as of September 2019)

Source: African Risk Capacity, presentation at CGD’s event, July 20, 2021

And first and foremost, Robert Agyarko reminded us, we need short-term access to vaccines now, in addition to building long-term capacity for all regions to respond to COVID-19 and anticipate future pandemic threats. The African continent is still acutely experiencing the COVID-19 pandemic, in a fast-moving third wave – with only just over 1 percent of Africans fully vaccinated. The most important thing we as a global community can do now is ensure equity in global vaccine access.

3. What we spend and do now to contain spillovers can mitigate risk

While these numbers may be alarming, it’s better to have a clearer picture of what’s to come to best prepare. To rapidly contain outbreaks, individual countries and global health institutions need resilient health systems and both outbreak preparedness and response mechanisms in place. And there is no substitute for documenting pandemic events with complete data to ensure that policymakers see a clear and present risk.

Early action and effective preparedness are absolutely essential to mitigating risk. And mitigation works. Metabiota’s analysis, based on simulated event catalogs, estimates that preventing just 1 percent of spillover events could avert a significant number of deaths in the future – 100 times fewer deaths over a 10 year period (Table 1).

Table 1. Impact of spillover reduction – estimates based on simulated event catalogs

Pathogen group Expected deaths over next decade Deaths averted over 10 years from 1% reduction in spillover events
Pandemic Flu (excluding seasonal flu) 2,200,000 22,000
Coronaviruses 1,500,000 15,000
Viral Hemorrhagic Fevers (e.g. ebola virus, nipah, etc.) 300,000 3,000
Total 4,000,000 40,000

Source: Oppenheim B, Stephenson N., Bertozzi S., Gadsden P., Guerrero, J., Madhav N, Jamison DT, The Value of Viral Surveillance. (2021) Working paper.

But how do we get there? Our speakers all offered ways forward to mitigate risk:

  • To be better prepared and mitigate pandemic risk, Dean Jamison says three distinct types of financing are needed: prevention, preparedness and surveillance; response and countermeasures; and then reconstruction

  • Ben Oppenheim and Nicole Stephenson from Metabiota laid out a clear set of priorities to mitigating pandemic risk:

    1. Develop a clearer picture of the drivers of disease spillover, and how they’re changing over time

    2. Shape political incentives to improve outbreak detection and reportin

    3. Build strong incentives for early action to contain outbreaks before they spread

    4. Invest in preparedness, both within the health sector and beyond it

  • And Robert Agyarko and Cristina Stefan shared African Risk Capacity’s Epidemic Preparedness Index (EPI), a comprehensive metric based on simulated behavior of diseases and response capacity to determine outbreak preparedness levels. The EPI assesses the ability to detect, communicate, and respond to an epidemic, tailored to national and subnational formats, and therefore what countries and regional bodies can do to prepare and mitigate outbreak risk.


Modeling shows that the next pandemic could be much closer than we think, and early action and preparedness – and financing for those mitigation strategies – are key to ensuring we do not have a repeat of the current pandemic. COVID-19 is the upper bound but part of broader picture of increasing spillover, increasing mortality, increasing burden, increasing death, which will require increasing efforts to detect, respond and contain.  

CGD supported scaling up those efforts via its contributions to the report of the G20 High Level Independent Panel on Financing the Global Commons for Pandemic Preparedness and Response. More and smarter money is key to ensuring that the world can prevent, detect, and respond to future pandemic threats, and clearer governance is urgently needed to make the global system work better to save lives and economies. It is long past time for new global funding to keep the world safer from pandemics.

There are many ways of interpreting and simulating trends in pandemic risk, and other groups working on this modeling. Check back this fall for the next event in this series, where a team from Imperial College London will share their work on modeling the return on investment in pandemic preparedness, and costs of life years lost due to the pandemic. And see CGD’s other work on financing for outbreak preparedness and response.

Predicting the future requires assumptions and caveats; our collective track record on predicting waves of the current pandemic illustrates the difficulties of capturing all relevant variables driving transmission. Ironically, longer-term estimates of the probabilities of spillover events can be more accurate than the near-term dynamics of COVID-19 spread. And better pandemic risk science will enable better predictions in the future; the world could use a scientific body to systematically assess and build consensus on what drives pandemic risk and how it changes over time, or an intergovernmental panel on pandemic risk, similar to the Intergovernmental Panel on Climate Change (IPCC).

Risk is constantly evolving. The next pandemic could hit us tomorrow, and the best way to prepare is to build up public health infrastructure, surveillance, reliable data, and medical countermeasure capacity for everyday, especially in the most vulnerable countries.


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.