How many lives have we lost to COVID-19? 

It’s surprisingly hard to answer this question. During the pandemic, countries have struggled to track this number. Counting deaths has always been a challenge for public health systems. Now, even when so many lives are at stake, and everyone is paying attention, we still fall short. 

We estimate that the total number of COVID-19 deaths worldwide is at least double what has been reported—6.9 million total deaths since January 2020. Nearly every country has had a hard time recording COVID-19 deaths accurately. But the share of missing deaths varies widely from country to country. 

Never has it been more critical to improve our data systems than now. Knowing COVID-19’s toll is critical for leaders when it comes to decision-making, resource allocation, and preventing the next pandemic. As individuals, understanding the toll COVID-19 has had on lives can influence the steps we take to protect ourselves. 

How does the Institute for Health Metrics and Evaluation count the number of COVID-19 deaths that have been missed? We start by applying the lessons that we’ve learned from leading the Global Burden of Disease study. We looked at the 56 countries that reported total deaths from all causes before—and more importantly—during the COVID-19 pandemic. Next, we evaluated how many excess deaths had occurred in these countries; that is, the number of deaths that occurred over and above what would have been expected in previous years. In our analysis, we factored in information about patterns of mortality that are unique to each country, including seasonal changes in mortality.   To evaluate how many of these excess deaths were likely due to COVID-19 versus other drivers, such as delays in accessing lifesaving health care, and increases in mental health disorders, we reviewed studies published over the course of the pandemic. 

Once we estimated how many COVID-19 deaths were missing from the official counts in our original subset of countries, we used this information to estimate the number of COVID-deaths that had likely gone unreported in other countries around the globe. In generating our estimates, we also factored in data that was relevant for tracking COVID-19 deaths, such as the estimated rate of detection of new infections. We will update these numbers regularly because they are likely to change over time as we gain new insights. Our estimates likely undercount the total number of deaths from COVID-19. 

Deaths From COVID-19 Have Gone Uncounted in Most Countries 

On average, in high-income countries, we estimate that around one out of three deaths from COVID-19 may be missing from official tallies. As a whole, the percentage of unreported COVID-19 deaths is higher in low-and middle-income countries than it is in high-income countries. For example, our research indicates that around two out of three deaths may be missing from government statistics in sub-Saharan Africa and South Asia.  

Certainly, wealthy countries have more resources to track deaths from COVID-19. High-income countries were hit hard early in the pandemic when COVID-19 testing rates were very low, meaning that many COVID deaths were probably missed. Many deaths from COVID-19 among older adults in long-term care facilities were not recorded. Reporting in high-income countries has improved substantially over time, however, as testing rates have increased. 

Around two out of three deaths may be missing from government statistics in sub-Saharan Africa and South Asia

While financial and administrative resources are helpful for recording deaths, commitment from national leaders is also essential for counting deaths accurately. Alongside high-income countries, some upper-middle-income countries stand out for tracking most of their COVID-19 deaths, including Argentina, Thailand, Costa Rica, and Colombia. In these countries, we estimate that around 3 percent or less of COVID-19 deaths are missing from government statistics.  

There are other countries—some in Eastern Europe and Central Asia—where most deaths have gone uncounted. For example, in Azerbaijan and Belarus, more than 90 percent of COVID-19 deaths appear to be missing from official statistics. 

There are also some countries in Latin America and the Caribbean region where it appears that a high numbers of deaths are missing. In Ecuador, we estimate that 64 percent of COVID-19 deaths are unaccounted for. Around 65 percent of COVID-19 deaths may be absent from government tallies in Mexico. Our findings are similar to the Mexican government’s conclusions: In March 2021, the government published a report estimating that the true number of COVID-19 deaths was around 60 percent higher than official statistics indicated. 

In the United States, we estimate that 37 percent of COVID-19 deaths are missing from official counts. There is wide variation across the country. Massachusetts; Washington, DC; North Dakota; Hawaii; and Rhode Island have tracked all or nearly all of their COVID-19 deaths. In these locations, we estimate that less than 1 percent of deaths went unrecorded. In contrast, Kentucky and Maine may have missed as many as 62 and 59 percent, respectively, of their COVID-19 deaths in official counts—the highest percentages in the United States.  

Similar to the trends we’re seeing in vaccination globally, resources make it easier for countries to track deaths from COVID-19. At the same time, we see that a commitment to counting deaths by national leaders is essential. Accurate tracking of COVID-19 indicators—not just deaths, but also cases, testing, and immunization—is essential for making lifesaving decisions. More accurate data collection can help a government implement social distancing measures in a timely manner, or help a hospital system make sure it has enough beds, staff, and supplies to handle a surge. Now is the time to strengthen our data systems and invest in tracking important metrics in real time. Our health depends on it. 
 

ACKNOWLEDGEMENTS: The authors would like to thank Professors Haidong Wang and Mohsen Naghavi, Dr. Darrah McCracken, Tahiya Alam, William Dangel, Ruri Syailendrawati, and Maja Pašović for their feedback on this post. 

EDITOR’S NOTE: The authors are employed by the University of Washington’s Institute for Health Metrics and Evaluation (IHME), which produced the COVID-19 forecasts described in this article. IHME collaborates with the Council on Foreign Relations on Think Global Health. All statements and views expressed in this article are solely those of the individual author and are not necessarily shared by their institution.