The COVID-19 pandemic did not have a single peak. Depending on whether you measure confirmed infections, deaths, hospital strain, or broader societal damage, the answer shifts dramatically, and it shifts again depending on which part of the world you are talking about. Global excess deaths climbed steeply through late 2020, dropped briefly, then surged to their highest point around mid-2021, driven largely by catastrophic waves in South and Southeast Asia. Meanwhile, the sheer volume of confirmed infections hit its zenith much later, during the Omicron waves of early 2022, even though those infections were generally less deadly per case. Understanding which “peak” matters requires looking at several overlapping crises that struck different regions at different times.
Excess Deaths Peaked at Different Times Across Regions
Confirmed case counts are a poor stand-in for how bad things actually were at any given moment, because testing capacity varied wildly between countries and over time. Excess mortality, which compares the number of people who actually died against how many would have been expected to die in a normal year, is a more reliable gauge. A WHO-led analysis of excess mortality found that the global toll rose steadily through the end of 2020, dipped briefly, then spiked sharply in the middle of 2021 before beginning a sustained decline. But that global curve masks enormous regional variation. The Americas saw their excess death peak around January 2021, Europe’s worst period came at the end of 2020, and South-East Asia experienced a devastating spike in mid-2021 when more than twice the expected number of deaths occurred in a single period.1Nature. The WHO estimates of excess mortality associated with the COVID-19 pandemic
These regional differences were not random. They tracked the timing of each major variant’s arrival, the availability of vaccines, and the stringency of public health measures. A country that locked down early and effectively might have delayed its worst wave by a year or more, only to face a different variant under different conditions. This means that asking “when was the peak” without specifying a region produces an answer so averaged that it is nearly meaningless for any individual country’s experience.
The First Wave and Europe’s Early Crisis
The earliest severe wave hit parts of East Asia and then Europe in early 2020. Italy became the first Western country to face a full-blown outbreak, and its experience became a template for what other nations would soon endure. Researchers estimated that Italy’s first wave peaked around March 14, 2020, with roughly 26,575 estimated daily cases and about 672 deaths on that single day. After a summer lull, a second, broader wave struck, peaking on November 12, 2020, with an estimated 60,425 daily cases and 695 deaths. The second wave was wider, lasting longer and producing a secondary hump in December before finally declining.2PubMed Central. The first year of COVID-19 in Italy: Incidence, lethality, and health policies
Italy’s trajectory, two distinct waves within the first year, became common across much of Europe and the Americas. The first wave tended to be smaller but caught health systems completely off guard. The second was often larger in raw numbers but met with somewhat better clinical knowledge and expanded ICU capacity. Both of these early waves were caused by the original strain and early variants, before Delta and Omicron changed the game entirely.
Delta and the Global Death Peak
If you had to pick one period that represented the pandemic’s deadliest chapter globally, mid-2021 has the strongest claim. The Delta variant, first identified in India, was substantially more transmissible than earlier strains and caused more severe disease. India’s experience was staggering: infections resurged dramatically in late March 2021, and the country reported about 19 million confirmed cases between late March and June of that year, roughly twice the total from the entire preceding 12 months. Researchers estimated that around a third of India’s population was infected during that three-month stretch, including reinfections.3PubMed Central. COVID-19 pandemic dynamics in India, the SARS-CoV-2 Delta variant and implications for vaccination
Delta was not just more contagious. It was more clinically dangerous. A prospective study at one Indian hospital found that patients hospitalized during the Delta-driven second wave had significantly worse hypoxia, required more intensive care and ventilatory support, and had in-hospital mortality roughly double that of first-wave patients.4medRxiv. Greater Covid-19 Severity and Mortality in Hospitalized Patients in Second (Delta Variant) Wave Compared to the First: Single Centre Prospective Study in India Images of overwhelmed cremation grounds and oxygen shortages in Indian cities during April and May 2021 became some of the pandemic’s most indelible scenes. Delta subsequently spread worldwide, driving severe waves across Southeast Asia, Africa, and parts of Europe and the Americas through the second half of 2021, and this period aligns with the global excess mortality peak identified in the WHO analysis.
Omicron and the Largest Infection Waves
By late 2021, Omicron had arrived and rapidly displaced Delta. Omicron was extraordinarily transmissible. Household transmission rates for Omicron were estimated at around 51%, compared with about 36% for Delta, meaning that once one person in a home caught it, the odds of others being infected were substantially higher.5Nature Communications. Increased household transmission and immune escape of the SARS-CoV-2 Omicron compared to Delta variants Omicron also had a pronounced ability to dodge immune protection from both previous infection and vaccination, and its sub-lineages, including BA.1, BA.2, BA.4, and BA.5, continued to evolve for greater immune escape throughout 2022.6PubMed Central. A Detailed Overview of SARS-CoV-2 Omicron: Its Sub-Variants, Mutations and Pathophysiology, Clinical Characteristics, Immunological Landscape, Immune Escape, and Therapies
The result was that global confirmed case counts during the Omicron waves of January and February 2022 dwarfed anything seen before. Many countries recorded daily case numbers five to ten times higher than their previous records. Yet the relationship between infections and deaths had fundamentally changed. In South Africa, where the Omicron wave was among the first to be studied closely, researchers found that the incidence of infection became decoupled from hospitalization and death rates during the Omicron wave, compared with the ratios seen in previous waves.7PubMed Central. Population Immunity and Covid-19 Severity with Omicron Variant in South Africa A large share of the population had some degree of immunity from prior infection, vaccination, or both, and Omicron appears to have been intrinsically less severe in the lower respiratory tract.
So if you define the peak by raw infections, early 2022 was it for much of the world. If you define it by the proportion of infections that killed people, that peak was earlier, in 2020 and the Delta period of 2021.
When Hospitals Were Most Overwhelmed
For the people working inside health systems, the peak was whenever their local ICU beds ran out. A systematic review of hospital surge during the pandemic found that mortality among hospitalized COVID patients climbed steeply as occupancy rose. Patients admitted during the highest-burden periods faced up to four times the risk-adjusted odds of dying compared with those admitted when hospitals had more capacity. One study showed a roughly 59% increase in the odds of death as ventilator-capable bed occupancy went from empty to full.8PubMed Central. Measures and Impact of Caseload Surge During the COVID-19 Pandemic: A Systematic Review
In the United States, ICU occupancy during weeks of high COVID admissions jumped by about 39 percentage points on average, a roughly 68% increase relative to normal levels. That surge was most extreme early in the pandemic, when ICU occupancy spiked by over 53 percentage points during peak admission weeks, before decreasing somewhat in later waves as hospitals adapted.9JAMA Health Forum. COVID-19 Admission Rates and Changes in US Hospital Inpatient and Intensive Care Unit Occupancy Large metropolitan hospitals bore the heaviest burden. The practical consequence was that people who needed emergency care for any reason, not just COVID, faced worse outcomes during surge periods because staff and equipment were stretched thin.
China’s Delayed and Compressed Peak
Not every country’s pandemic timeline followed the global pattern. China’s strict zero-COVID policy suppressed large-scale outbreaks for nearly three years but ultimately produced one of the sharpest and most compressed epidemic peaks anywhere. When containment measures were lifted in December 2022, Omicron spread at an extraordinary rate of about 0.42 per day, translating to a doubling time of roughly 1.6 days. Researchers estimated that about 97% of China’s population was infected during December 2022 alone, with the nationwide epidemic peaking around December 23.10PubMed Central. Swift and extensive Omicron outbreak in China after sudden exit from ‘zero-COVID’ policy
Death counts in Chinese cities also peaked during the fourth week of December 2022.11JAMA Network Open. Excess All-Cause Mortality in China After Ending the Zero COVID Policy China’s experience illustrates how policy choices could shift the timing of a country’s peak by years, but not necessarily eliminate the toll. The population had comparatively little prior natural immunity, and the wave, when it came, was the fastest mass infection event documented during the entire pandemic.
Japan’s Unusual Late Mortality Pattern
Japan presents another outlier. The country’s excess mortality was actually negative in 2020, meaning fewer people died than expected, likely because masking, social distancing, and reduced influenza transmission offset early COVID deaths. Excess mortality then gradually rose, crossing above expected levels in early 2021 and climbing through mid-2021 before plateauing. A sharp second increase then carried excess mortality to its highest point in late 2022, consistent across all 47 prefectures.12BMJ. Excess mortality during and after the COVID-19 emergency in Japan: a two-stage interrupted time-series design Japan’s mortality peak thus lagged behind most Western countries by roughly two years, a reflection of its initially successful containment followed by large Omicron-driven waves in a population with a high proportion of elderly residents.
Why Reported Case Numbers Were Always Imprecise
Any discussion of when the pandemic peaked has to reckon with a basic problem: reported case counts were never an accurate mirror of reality, and the degree to which they were wrong changed over time. Early in 2020, when testing capacity was severely limited, many infections went entirely undetected. Research demonstrated that changes in the daily number of tests conducted and the percent testing positive could bias estimates of how fast the virus was spreading, sometimes making it look like transmission was slowing when in fact it was just being tested less.13PubMed Central. The impact of changes in diagnostic testing practices on estimates of COVID-19 transmission in the United States Separate work reinforced that epidemic curves of reported cases did not always reflect the true growth rate because testing rates themselves were changing.14PubMed Central. Changes in testing rates could mask the novel coronavirus disease (COVID-19) growth rate
Wastewater surveillance offered a partial corrective. In communities where researchers tracked viral RNA in sewage, wastewater signals sometimes spiked when clinical case counts stayed flat, especially after events like college students returning to campus or phased reopenings.15PubMed Central. Metrics to relate COVID-19 wastewater data to clinical testing dynamics By the Omicron era, widespread use of at-home rapid tests, which were rarely reported to public health authorities, meant that official case counts understated reality even more than before. The true infection peak in many countries was almost certainly higher and possibly earlier than what the confirmed-case dashboards showed.
Damage That Did Not Follow the Infection Curve
Some of the pandemic’s worst consequences peaked on timelines only loosely connected to infection waves. Mental health deteriorated sharply and early. A study of anxiety, depression, and worry in the United States found two major peaks during 2020, likely aligned with the initial lockdowns and the fall/winter surge, followed by gradual improvement through 2021 and 2022.16PubMed Central. Covid and mental health in America For many people, the psychological low point came well before the largest infection waves.
Educational losses followed yet another trajectory. A global analysis of PISA test scores found that pandemic-era school closures were associated with an average decline of 12 points in mathematics, equivalent to more than seven months of lost learning. The effect scaled with the duration of closures: each additional week of school shutdowns was associated with a further decline in achievement.17PubMed Central. COVID-19, school closures, and student learning outcomes: New global evidence from PISA This damage accumulated gradually and was not tied to any single infection peak; it was a product of cumulative disruption spread across 2020 and 2021.
Long COVID added yet another lagging peak. In a two-year longitudinal study, about a third of participants experienced symptoms lasting more than 90 days after infection. Those symptoms, including fatigue, cognitive difficulties, and activity limitations, peaked in severity around six months after the initial infection. Eight percent of participants still had not returned to their pre-COVID health status two full years later.18PubMed Central. Two-Year Longitudinal Study Reveals That Long COVID Symptoms Peak and Quality of Life Nadirs at 6–12 Months Postinfection Because long COVID is a downstream consequence of infection, the peak burden of long COVID trailed the peak of each infection wave by months, creating rolling waves of chronic illness that continued well after case counts had subsided.
The Virus in Animal Populations
One dimension of the pandemic that did not peak and subside in a neat curve was the spread of SARS-CoV-2 into animal populations. Mink and white-tailed deer, in particular, showed sustained animal-to-animal transmission, raising concerns about the virus establishing permanent reservoirs outside humans.19PubMed Central. SARS-CoV-2 as a Zooanthroponotic Infection: Spillbacks, Secondary Spillovers, and Their Importance In one documented case in the United States, a mink-adapted version of the Delta variant spread from a mink farm to free-ranging white-tailed deer in nearby counties during late 2022 and early 2023, well after Delta had been displaced by Omicron in human populations.20Nature Communications. SARS-CoV-2 Delta variant re-emerges in US farmed mink and free-ranging white-tailed deer in 2022–2023 Old variants can persist and mutate in animal hosts, potentially spilling back into humans in altered form. From an evolutionary standpoint, the pandemic’s “peak” in animal reservoirs may not have arrived yet, and it represents one of the least predictable aspects of the virus’s long-term trajectory.