A reinvigorated multilateralism in health: lessons and innovations from the COVID-19 pandemic

Executive summary As of May 31, 2022, there were 6·9 million reported deaths and 17·2 million estimated deaths from COVID-19, as reported by the Institute for Health Metrics and Evaluation (IHME; throughout the report, we rely on IHME estimates of infections and deaths; note that the IHME gives an estimated range, and we refer to the mean estimate). This staggering death toll is both a profound tragedy and a massive global failure at multiple levels. Too many governments have failed to adhere to basic norms of institutional rationality and transparency, too many people—often influenced by misinformation—have disrespected and protested against basic public health precautions, and the world's major powers have failed to collaborate to control the pandemic. The multiple failures of international cooperation include (1) the lack of timely notification of the initial outbreak of COVID-19; (2) costly delays in acknowledging the crucial airborne exposure pathway of SARS-CoV-2, the virus that causes COVID-19, and in implementing appropriate measures at national and global levels to slow the spread of the virus; (3) the lack of coordination among countries regarding suppression strategies; (4) the failure of governments to examine evidence and adopt best practices for controlling the pandemic and managing economic and social spillovers from other countries; (5) the shortfall of global funding for low-income and middle-income countries (LMICs), as classified by the World Bank; (6) the failure to ensure adequate global supplies and equitable distribution of key commodities—including protective gear, diagnostics, medicines, medical devices, and vaccines—especially for LMICs; (7) the lack of timely, accurate, and systematic data on infections, deaths, viral variants, health system responses, and indirect health consequences; (8) the poor enforcement of appropriate levels of biosafety regulations in the lead-up to the pandemic, raising the possibility of a laboratory-related outbreak; (9) the failure to combat systematic disinformation; and (10) the lack of global and national safety nets to protect populations experiencing vulnerability. This Commission report aims to contribute to a new era of multilateral cooperation based on strong UN institutions to reduce the dangers of COVID-19, forestall the next pandemic, and enable the world to achieve the agreed goals of sustainable development, human rights, and peace that governments are committed to pursue as members of the UN. We address this Commission report to the UN member states, the UN agencies and multilateral institutions, and multilateral processes such as the G20 and the G7. Our aim is to propose guideposts for strengthening the multilateral system to address global emergencies and to achieve sustainable development. In issuing this report, we commend the excellent work of many important international studies that have preceded our own, most notably those from the Independent Panel for Pandemic Preparedness and Response and the G20 High-Level Independent Panel on Financing the Global Commons on Pandemic Preparedness and Response. Section 1 of this Commission report provides a conceptual framework for understanding pandemics. Section 2 provides an annotated chronology of the COVID-19 pandemic and thematic findings regarding several issues. Section 3 presents our policy recommendations, particularly around multilateral cooperation centred at WHO to address global health crises, and around investments in preparedness for future health crises through strong national health systems and international financing and technology cooperation with the world's lower-income regions. Methodology The Lancet COVID-19 Commission was established in July, 2020, with four main themes: developing recommendations on how to best suppress the epidemic; addressing the humanitarian crises arising from the pandemic; addressing the financial and economic crises resulting from the pandemic; and rebuilding an inclusive, fair, and sustainable world. 1 The 28 Commissioners are global experts in public policy, international cooperation, epidemiology and vaccinology, economics and financial systems, sustainability sciences, and mental health. The Commissioners oversaw the work of 12 thematic Task Forces, which met on an ongoing basis (once every 2 weeks or once per month) to support the work of the Commission. These Task Forces included a total of 173 experts. The Commission Secretariat acted as liaison among the Task Forces. The Task Forces published short pieces on their respective areas of focus on the Commission website and in peer-reviewed journals, contributing to the efforts of the overall Commission. Key findings • The proximal origin of SARS-CoV-2 remains unknown. There are two leading hypotheses: that the virus emerged as a zoonotic spillover from wildlife or a farm animal, possibly through a wet market, in a location that is still undetermined; or that the virus emerged from a research-related incident, during the field collection of viruses or through a laboratory-associated escape. Commissioners held diverse views about the relative probabilities of the two explanations, and both possibilities require further scientific investigation. Identification of the origin of the virus will help to prevent future pandemics and strengthen public trust in science and public authorities. • WHO acted too cautiously and too slowly on several important matters: to warn about the human transmissibility of the virus, to declare a Public Health Emergency of International Concern, to support international travel protocols designed to slow the spread of the virus, to endorse the public use of face masks as protective gear, and to recognise the airborne transmission of the virus. • As the outbreak became known globally in early January, 2020, most governments around the world were too slow to acknowledge its importance and act with urgency in response. It was mainly the countries in WHO's Western Pacific region, primed by their experience with severe acute respiratory syndrome, that reacted with urgency to the outbreak, and that generally pursued a suppression strategy that led to low cumulative mortality, although the omicron variant (B.1.1.529) has been undoing some of these gains. • Coordination among governments was inadequate on policies to contain the pandemic, including travel protocols to slow the global transmission of the virus, testing strategies, public health and social measures, commodity supply chains, data standards and reporting systems, and advice to the public, despite the very high interdependence among countries. • Epidemic control was seriously hindered by substantial public opposition to routine public health and social measures, such as the wearing of properly fitting face masks and getting vaccinated. This opposition reflects a lack of social trust, low confidence in government advice, inconsistency of government advice, low health literacy, lack of sufficient behavioural-change interventions, and extensive misinformation and disinformation campaigns on social media. Public policies have also failed to draw upon the behavioural and social sciences; doing so would have led to more successful implementation of public health interventions and helped to increase social trust, prosociality, equity, and wellbeing. In many cases, policies and decision making have not been informed by robust and continuously updated evidence syntheses. • Public policies did not properly address the profoundly unequal effects of the pandemic. Heavily burdened groups include essential workers, who are already disproportionately concentrated in more vulnerable minority and low-income communities; children; women, who face employment, safety, and income losses, exacerbated by the adverse consequences of school closures; people living in congregate settings, such as prisons or care homes, especially for older populations; people living with chronic conditions and disability; Indigenous Peoples; migrants, refugees, and displaced populations; people without access to quality and affordable health care; and people who face the burdens of long COVID. • Among high-income countries, those with strong and resilient national health systems—including public health systems that complement clinical health care—have generally fared better at addressing COVID-19 and maintaining non-pandemic-related health services. In low-income and middle-income countries (LMICs), where health systems tend to be under-resourced and fragmented, better outcomes were seen when previous experiences with outbreaks and epidemics were built upon, and when community-based resources—notably community health workers—were used to support screening and contact-tracing capacity and trust-building within communities. • Rapid development of multiple vaccines has been a triumph of the research and development system and the result of long-standing public and private investment and cooperation. However, the lack of a multilateral and coordinated approach by governments to manage intellectual property rights, technology transfer, international financing, the allocation of vaccines from multinational pharmaceutical companies, and the support for vaccine production in LMICs for use in those countries, has come at a great cost in terms of inequitable access to vaccines. • Economic recovery depends on sustaining high rates of vaccination coverage and low rates of new, clinically significant COVID-19 infections, and on fiscal and monetary policies to mitigate the socioeconomic effects of the pandemic and prevent a financial crisis. Emergency global financing from the International Monetary Fund, the World Bank, and regional development banks had a salutary role, although much larger financial flows from high-income to low-income regions were warranted. • The sustainable development process has been set back by several years, with a deep underfinancing of investments needed to achieve the Sustainable Development Goals (SDGs) and the aims of the Paris Climate Agreement. In most countries, the pandemic diverted resources and policy attention away from longer-term goals, thereby reversing progress towards the SDGs in many countries. Key recommendations • The world requires globally coordinated efforts to bring an end to the COVID-19 pandemic on a rapid and equitable basis. Countries should maintain a vaccination-plus strategy that combines mass vaccination, availability and affordability of testing, treatment for new infections and long COVID (test and treat), complementary public health and social measures (including the wearing of face masks in some contexts), promotion of safe workplaces, and economic and social support for self-isolation. A vaccination-plus strategy with the goal of protecting populations should be implemented on a sustainable basis, rather than as a reactive policy that is abruptly turned on and off. • WHO, governments, and the scientific community should intensify the search for the origins of SARS-CoV-2, investigating both a possible zoonotic origin and a possible research-associated origin. The search for origins requires unbiased, independent, transparent, and rigorous work by international teams in virology, epidemiology, bioinformatics, and other related fields. • WHO should expand the WHO Science Council to apply urgent scientific evidence for global health priorities, including future emerging infectious diseases. This Council should include experts from diverse fields and from all six WHO regions, and should include younger people and have gender parity. Establishing an understanding of exposure routes and the highest-risk environments for transmission should always be among the first essential steps for scientists in response to future disease threats, because this knowledge should determine effective control strategies for reducing risk. • Governments, represented at the World Health Assembly (WHA) by their national health ministers, should establish stronger means of cooperation and coordination in the response to emerging infectious diseases. Strengthened cooperation should be incorporated in a new pandemic agreement and in updated International Health Regulations (IHR), as were adopted in 2005 after the outbreak of severe acute respiratory syndrome and which now need updating. • WHO should be strengthened. The WHA should create a WHO Global Health Board composed of the six WHO regions, represented by heads of state on a rotating basis, and selected by the governments of each region. Reforms of WHO should include a substantial increase of its core budget. The world community should not establish new centres of global health policy and finance that would compete with, or even undermine, the central role of WHO. • We call for a dual track to prevent future emerging infectious diseases. To prevent natural spillovers, governments should coordinate on the global surveillance and regulation of domestic animal and wild animal trade, and take stronger measures against dangerous practices. To prevent research-related spillovers, WHO should be given new oversight authority regarding the biosafety, biosecurity, and bio-risk management of national and international research programmes that are engaged in the collection, testing, and genetic manipulation of potentially dangerous pathogens. • The WHA, in conjunction with the G20 countries, should adopt a 10-year global strategy to bolster research and development capacity and commodity production capacity—including for vaccines—for every WHO region, including in the low-income regions of the world. WHO should help several low-income and middle-income countries (LMICs) to achieve WHO's stringent regulatory authority status. • Countries should strengthen national health systems on the foundations of public health and universal health coverage, grounded in human rights and gender equality. Strong public health systems should include strong relationships with local communities and community organisations; surveillance and reporting systems; robust medical supply chains; health-promoting building design and operation strategies; investments in research in behavioural and social sciences to develop and implement more effective interventions; promotion of prosocial behaviours; strong health education for health promotion, disease prevention, and emergency preparedness; effective health communication strategies; active efforts to address public health disinformation on social media; and continuously updated evidence syntheses. The health-care system should include universal health coverage that is centred around primary health care and ensures that patients have access to quality care for pandemic-related and non-pandemic-related health issues, including mental health. Community health workers and community-based organisations should be well trained and supported. • In addition to strengthening health systems, each country should determine and expand national pandemic preparedness plans to prevent and respond to newly emerging infectious diseases. Preparedness plans should include improved surveillance and monitoring; definition and protection of vulnerable groups; international notifications; cooperation within WHO regional groups; emergency financing; guidelines on behavioural, social, and environmental interventions, travel protocols, and safe schools and workplaces; robust health-commodity supply chains (eg, personal protective equipment, diagnostics, therapeutics, and vaccines); effective risk communication and active opposition to misinformation and disinformation; training of public health professionals; and provision of adequate staffing. • A new Global Health Fund should be created that is closely aligned with WHO. This Fund should combine and expand the operations of several existing health funds and add new funding for three windows of financing: commodities for disease control, pandemic preparedness and response, and primary health system strengthening in LMICs. We propose that the Global Health Fund should have its headquarters in (Continues on next page)(Key recommendations continued from previous page)Geneva, Switzerland, but have strong regional offices in each of the six WHO regions. The Fund would thereby have centralised overall funding but decentralised programme design and implementation, to foster strong ownership by the countries of each region and to reflect regional needs and priorities, rather than being under top-down control from Geneva or from a few donor countries. • The UN member states, with particular responsibility of the G20 countries, should adopt a new financial architecture to scale up financing for LMICs to meet the urgent challenges of pandemic preparedness, the Paris Climate Agreement, and the Sustainable Development Goals. The new financial architecture should include increased sustainable development funding from all sectors: official institutions, the private sector, foundations, and civil society. The Commission issued its first statement in The Lancet 2 on the occasion of the UN General Assembly on Sept 14, 2020. The second statement of the Commission was published in The Lancet 3 on Feb 12, 2021, around the time of the launch of the global vaccination effort. In October, 2021, the Commission issued its third statement, directed towards the G20. 4 For the final report of the Commission, each Task Force prepared a report drawing on original analysis and synthesis of evidence to generate recommendations relevant to their specific Task Force. Recommendations from these reports provide the basis for the Commission's final report. Additionally, the Secretariat oversaw a detailed analysis of the key policy recommendations of other COVID-19 commissions, panels, and working groups. The Secretariat and members of the Task Forces then examined the best available syntheses of evidence to inform and test the key claims made in the report. The Commission focuses on the public policy of pandemic preparedness, response, and recovery, specifically in the areas of public health, virology, social policy, macroeconomics, international finance, and geopolitics. The Commission is not an investigative group, nor a body of biomedical specialists in key fields such as virology, vaccine development, and medicine. The Commission's focus is on science-based policy, global cooperation, and international finance. The Commission uses UN nomenclature for all countries and locations mentioned in the report. Section 1: conceptual framework for understanding pandemics Five pillars of the successful fight against emerging infectious diseases There are five basic pillars of a successful fight against emerging infectious diseases. The first is prevention: to stop an outbreak before it occurs by taking effective measures to prevent the emergence of a new and dangerous pathogen. The second is containment: to eliminate the transmission of disease from infected individuals to susceptible individuals after a disease has emerged. The third is health services: to save the lives of people with the disease and ensure the continuity of other health services, including those for mental health. The fourth is equity: to ensure that economic and social burdens are shared among the population and that the most vulnerable groups and individuals are protected. The fifth is global innovation and diffusion: to develop, produce, and distribute new therapeutics and vaccines in an equitable and efficient manner. To accomplish these five pillars requires an ethical framework of prosociality—the orientation of individuals and government regulations to the needs of society as a whole, rather than to narrow individual interests.5, 6, 7, 8 In the 14th century, authorities in Venice, Italy, battled plague outbreaks by requiring ships to remain at anchor for 40 days before landing (the word quarantine derives from the Italian quaranta giorni, or forty days), as an early and incipient form of prosocial regulation. 9 Prosociality nowadays includes voluntary behaviours by individuals, such as the proper use of face masks, in addition to government regulations, such as the enforcement of workplace safety standards, to prevent the transmission of disease. Challenges of prosociality arise especially in circumstances of strategic dilemmas, in which the pursuit of narrow self-interest by each member of the society ends up weakening the society as a whole. By turning from the pursuit of narrow self-interest to the pursuit of shared interests, members of society can increase the wellbeing of all. Prosociality generally requires some form of the Golden Rule (doing to others what you would have done to you) or the Kantian Imperative (acting according to maxims that can be universal laws). Pandemics have many strategic dilemmas, and therefore require cooperative responses rather than selfish—and self-defeating—behaviours. Prosociality was at a low ebb in many societies during the past 2 years. In many countries, social trust in government and other authorities among citizens has declined markedly in the past two decades (and over the past four decades in the United States) 10 —related, at least in part, to the persistent increase in socioeconomic inequalities. In places of low social trust, prosocial behaviours are rejected by many groups within society. Additionally, at the national level, many governments showed themselves to be untrustworthy and ineffective. 11 At the global level, cooperation among governments was undermined by rancour among the major powers. This hostility gravely weakened the capacity of international institutions such as WHO to conduct their assigned roles in the pandemic response. Success also requires preparedness. Building these five pillars after an outbreak has started is far too late, as the world has learned the hard way with COVID-19. This pandemic broke out at a moment of weak global preparedness. Despite ample previous warnings of increasing pandemic risks, at least since the outbreak of severe acute respiratory syndrome in 2003, most of the world was not prepared for COVID-19. Rapid response to a new outbreak to control community transmission When an outbreak occurs, time is of the essence. A core characteristic of emerging infectious diseases such as COVID-19 is the exponential growth of new infections in the initial stages of mass transmission. According to the basic model of the spread of an emerging infectious disease, the number of new infections per day is proportional to the number of infectious people in the population multiplied by the share of the population that is susceptible to infection (ie, the share of the population that lacks immunity). With the emergence of a new pathogen, most or all of the population is susceptible to infection, so the number of new cases is proportional to the number of current infectious cases, which implies an exponential growth of new infections. A single new case of COVID-19 at the start of the pandemic became hundreds, or in some situations thousands, of cases within a month. The original SARS-CoV-2 variant that was first identified in Wuhan, China, had a doubling time of approximately 3 days, meaning that, over a 30-day month, a new (index) case would lead to roughly ten doublings, or 1024 (210) new infections on the tenth doubling. 12 The basic lesson of this rapid growth of infections is the need to act on a new outbreak as soon as possible. If the public health system can quickly identify the index case, public health authorities can then trace the contacts of that person, so they can all be quarantined during the period of potential infectiousness to others.13, 14, 15 Case identification followed by contact tracing and isolation or quarantine can slow and reduce transmission.16, 17, 18 A major challenge to the test and quarantine approach for COVID-19 was the high proportion of asymptomatic infections, especially in young people. The problem arises when community transmission (transmission among individuals beyond the index case) is already well underway, as public health workers might not be able to trace the contacts of hundreds of infected individuals. The ability of the public health system to identify cases, trace contacts, and isolate infected individuals can be overwhelmed in just a few weeks of uncontrolled community transmission. Even with community transmission, all is not necessarily lost. Identifying and isolating a high proportion of infected individuals early in their infectious period, so that on average each infected person gives rise to less than one new infected person, is sufficient. With widespread access to quality and affordable community testing, infected individuals can learn quickly of their own SARS-CoV-2 infection and of their potential infectiousness to others, and can use face masks, adopt physical distancing, and isolate as soon as possible after infection. By doing so, these people are more likely to give rise to less than one new infection. This decline in infections can occur even without extensive contact tracing, but requires supportive personal behaviours and access to testing early in the epidemic, as was achieved in Republic of Korea. 19 In short, pandemic control is based heavily on prosocial actions by individuals (eg, getting tested, keeping a physical distance, and isolating when infective), and these individual measures rely heavily on public policies (eg, trusted information for the community, access to testing sites, and an economic framework such as guaranteed paid leave) to support self-isolation. Four COVID-19 control regimes COVID-19 presents special challenges in terms of control, as transmission occurs from presymptomatic (before symptoms) and asymptomatic (without symptoms) individuals as well as from those with symptoms. Such transmission makes COVID-19 control especially difficult, because people who are infected are often not aware of their infectivity. In technical terms, the reproduction number, R, denotes the number of infections caused by each infectious person. At the start of an outbreak, when no control measures have been put in place, R is denoted as R 0, called the basic reproduction rate. In the uncontrolled first wave of the COVID-19 pandemic in Wuhan, R 0 was around 2·4. 20 With intensive contact tracing and the isolation of infected individuals, in addition to the implementation of a range of public health and social measures, 21 the effective reproduction rate can be reduced to less than 1, and the epidemic will decline. Epidemiologists therefore distinguish between four kinds of COVID-19 control regime. First is the uncontrolled scenario, R 0=2·4. In this case, the epidemic eventually ends when most of the population has been infected and therefore has eventually acquired immunity (at least temporarily). This kind of mass infection that eventually results in mass acquired immunity through natural infection is sometimes called the herd-immunity strategy. 22 This strategy is highly problematic for COVID-19, because many infected people will die of the disease, and many who survive have what is known as long COVID. 23 Second is the limited control case in which R is reduced to less than R 0=2·4 but remains greater than 1. In this case, the epidemic still grows exponentially, but less rapidly than with no control measures. The peak in the number of new infections per day is lower and occurs later than in the uncontrolled case. We have classified this control scenario as a mitigation strategy. During the COVID-19 pandemic, this strategy was often referred to as flattening the curve of the epidemic.24, 25 The eventual (cumulative) number of cases of infection during the epidemic is nearly the same as in the uncontrolled epidemic, but the infections are spread out over a longer time, and therefore put less pressure on the capacity of hospital and health-care systems at any given time. Third, through the implementation of combinations of layered control measures—including widespread testing, contact tracing, and isolating; proper use of face masks; physical distancing; limitations on mass gatherings; and improved ventilation systems at workplaces—R decreases to less than 1, so the epidemic declines.16, 26 Early implementation of public health and social measures—including closing businesses and venues, banning public events, launching public information campaigns, and requiring the use of face masks—is more effective at keeping cumulative cases and deaths low than implementation at a later stage. 27 We call this a suppression strategy. However, if circulation of the virus is not brought to zero, and infected people continue to arrive from other areas, this strategy needs to be implemented on an ongoing basis to contain each new outbreak in the community, and becomes progressively more difficult as highly transmissible variants emerge. Fourth, by means of aggressive testing, contact tracing, and isolating, R is kept near 0. This strategy, which is sometimes called a containment strategy and was known in China as a zero-COVID strategy, can be viewed as an intensive application of the suppression strategy. 28 In principle, deaths can be kept to near zero, and infections can be held to a very small portion of the population. A successful long-term exit from a containment strategy (ie, lifting of the containment measures) depends on the successful containment of transmission in the rest of the world, or on sufficient protection from vaccinations and highly effective therapeutics to tolerate a subsequent spread of the virus without incurring high death rates and serious disease. Some countries in the WHO Western Pacific region that adopted a containment strategy during the first 2 years of the pandemic later abandoned the strategy during the period in which the omicron variant (B.1.1.529) was dominant (known as the omicron wave), after a sufficiently high proportion of the population had been vaccinated. What is the basis for choosing between these strategies? The herd-immunity strategy might be adopted if policy measures to reduce transmission are believed to be too onerous, too costly, or too ineffective to justify any steps to reduce R, or if the burden of infection in terms of deaths and illness is viewed as too small to justify any control measures. The herd-immunity strategy was originally advocated by some pundits on the grounds that acquired immunity would protect from COVID-19 reinfection for an extended period; however, it has since been found that immunity tends to wane over time, and new SARS-CoV-2 variants arise that evade acquired immunity. The herd-immunity strategy downplays not only the deaths but also the serious disease burden of long COVID. 29 The flatten-the-curve strategy is likely to be adopted when the over-riding concern of policy makers is the surge of patients into the hospital system, and stronger control measures are viewed as too costly, unnecessary, or infeasible. The epidemic response of many governments has been led by political considerations and hospital-system administrators rather than by public health considerations and specialists. The suppression strategy is likely to be adopted if suppression of the pandemic is deemed to be feasible and at sufficiently low cost to justify the stronger actions needed. Some opponents of the suppression strategy have argued that suppressing infection is futile, because in the future the virus will inevitably evade controls until herd immunity is reached. Yet this argument of futility misses an essential point, which is that a major purpose of suppression strategies is to buy time until the arrival of better tools—such as vaccines and therapeutics—at which time the response can be re-evaluated and perhaps eased. Even if a full-scale epidemic eventually arises, the temporary suppression of the epidemic by a year or two can buy time for mass vaccinations or the arrival of effective therapeutics, thereby saving lives and avoiding long-term health effects. The final tally of costs and benefits of a suppression strategy will necessarily be provisional until there is an exit policy, either through worldwide containment or through a high level of vaccination and access to effective medicines. When the number of cases has already reached very high levels, even stronger measures—notably national lockdowns of the population, causing substantial dislocations of daily life and economic activity—might have to be invoked to regain some measure of control over the pandemic. Lockdowns with high rates of adherence can cause pronounced, albeit temporary, reductions of R, thereby causing a rapid decline in the number of new infections and the total number of infected people. If the lockdown is simply followed by a relaxation of controls to the pre-lockdown status quo, then R quickly returns to the pre-lockdown level, and within a short period of time the exponential growth of new infections restores the pre-lockdown rate of daily infections. The proper use of lockdowns is to serve as a temporary expedient to provide time for national health systems to build up and enable a more comprehensive set of public health and social measures, so that less disruptive measures—such as testing, contact tracing, and isolation—are in place after the lockdown is lifted. 30 Prosociality for pandemic control Suppression of the virus requires a range of public health and social measures, which are also known as prosocial behaviours. A person who tests positive for COVID-19 should not partake in activities that pose a risk of infection to others out of concern for others, not only for themselves, and they should reasonably be able to expect that others will behave in the same way. Additionally, encouraging prosocial behaviour during pandemics and beyond is valuable for mental and physical health, which could help to address the pervasive adverse effects of the pandemic on mental health, particularly during lockdown.31, 32, 33, 34 Prosocial behaviours include those that reduce transmission of the virus to others, support others to keep safe, promote health and social care within and outside health-care facilities, and promote social cohesiveness and mutual aid. Such prosocial actions include testing for infections, including frequent use of rapid diagnostic tests; isolating in the event of a positive test; precautionary quarantining after exposure, before receiving test results; wearing well fitting face masks in public indoor settings; maintaining physical distancing in public and other indoor spaces; meeting outdoors rather than indoors; working online from home where feasible; maximising outdoor air ventilation, upgrading filtration to minimum efficiency reporting value (MERV) 13 filters, and using portable air cleaners with high-efficiency particulate air filtration and other evidence-based air cleaning approaches—such as germicidal ultraviolet light—in high-risk settings, particularly when ventilation and filtration are not possible; and getting immunised as soon as vaccines are available. National and local governments need to provide support so that people can make these necessary behaviour changes. Such support includes prompt deployment of high-quality testing, with widespread accessibility and affordability; government provision of public isolation and quarantine facilities for people living in conditions that prevent isolation at home; provision of financial and social support for people in isolation or quarantine; provision and deployment of high-quality and timely public information to support healthful and prosocial behaviours; increasing indoor air ventilation above minimum standards and enhancing air filtration efficiency (MERV 13 or higher) in mechanically ventilated buildings; and provision of free and easily accessible vaccination. Crucially, prosociality applies between governments as well as between individuals. If two neighbouring countries have a shared open border, the maintenance of R near 0 is feasible only if both countries pursue a suppression (R 100 μm in diameter) that fall quickly to the ground (usually within 2 m of the source)—land directly on the mucous membranes of a susceptible person in the near-field. The third method of transmission occurs through touch, or indirect contact via a contaminated object known as a fomite, in which pathogens are transferred—usually by hand—to the mucous membranes of a susceptible person. A paradigm shift in how we view and address the transmission of respiratory infectious diseases is underway. 118 Airborne transmission in both the near-fields and the far-fields is a crucial, if not dominant, exposure pathway for SARS-CoV-2 and other respiratory viruses. Laboratory, field, modelling, and case studies have shown that airborne transmission through the inhalation of a virus-laden aerosol is important, if not dominant, for COVID-19.119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132 Although transmission can occur through touch, it is rare for respiratory viruses, and touch and spray transmission are not likely to contribute to widespread transmission or superspreading events. As nearly all transmission occurs indoors, the way in which we design and operate building ventilation and filtration systems can reduce transmission. Long-standing erroneous thinking about airborne transmission led WHO to discount the role of this transmission route at the start of the pandemic. The downplaying of airborne transmission can be traced to the misinterpretation of observations and experimental results from around 100 years ago. Because most transmission occurs when people are in close contact, it was wrongly assumed that transmission was through spray rather than through airborne aerosols. In fact, much of the close transmission is through aerosols, because people release considerable quantities of aerosol in addition to large droplets, especially when talking and coughing, 133 and also because aerosol is most concentrated close to the source, like cigarette smoke particles near a smoker.134, 135 Therefore, although transmission via the airborne route by virus-laden aerosol can occur both in the near-field and in the far-field, the risk of near-field transmission for a single person in proximity to an infected person is generally greater than the risk of far-field transmission. 134 Nonetheless, the greater frequency of transmission by close contact, combined with a desire of scientists to refute miasma theory—the prevailing theory of the transmission of respiratory infectious disease from the mid-to-late 19th century, in which vague explanations for the causes of disease, such as so-called bad air, were perpetuated with little to no causative basis 136 —led to the desire to promote infection-control recommendations centring on hygiene and sanitation. Such recommendations contributed to the perpetuation of the erroneous idea that spray transmission was the dominant mode of spread of respiratory infectious diseases, including COVID-19. 136 Numerous publications have attempted to overturn mistaken ideas about transmission routes for respiratory infectious diseases,117, 137, 138, 139, 140, 141, 142, 143 and have initiated a paradigm shift towards more accurate definitions. 117 Alas, WHO was slow to acknowledge the airborne transmission of SARS-CoV-2, and was therefore slow to emphasise the range of measures needed to limit indoor transmission. Failures and successes of international cooperation The world has paid a high price for the combination of poor preparedness and failures of cooperation to address COVID-19. The multiple failures of international cooperation include (1) the lack of timely notification of the initial outbreak of COVID-19; (2) costly delays in acknowledging the crucial airborne exposure pathway of SARS-CoV-2 and in implementing appropriate measures at the national and global levels to slow the spread of the virus; (3) the lack of coordination among countries regarding containment strategies; (4) the failure of governments to examine and adopt best evidence for controlling the pandemic and managing economic and social repercussions from other countries; (5) the shortfall of global funding for LMICs; (6) the failure to ensure adequate global supplies and equitable distribution of key commodities, including protective gear, diagnostics, medicines, medical devices, and vaccines, especially for LMICs; (7) the lack of timely, accurate, and systematic data on infections, deaths, variants, and health system responses; (8) the poor enforcement of appropriate levels of biosafety regulations in the lead-up to the pandemic, raising the possibility of a laboratory-related outbreak; (9) the inability or unwillingness to combat systematic disinformation; and (10) the lack of global and national safety nets to protect populations experiencing vulnerability. Nonetheless, there have been some important bright spots in the national and global responses to COVID-19. The most important has been the public–private partnerships for the rapid development of vaccines. Also notable were the actions of higher-income countries to support households, businesses, and employers through fiscal and labour market measures to mitigate the adverse effects of the pandemic, and to inject funds into the health-care sector. We also highlight the positive role of the multilateral financial institutions. The World Bank provided nearly US$14 billion in fast-track support for COVID-19-related relief efforts and approved $12 billion in 2020 for countries to buy and deliver vaccines. 144 The International Monetary Fund (IMF) also provided urgent support of approximately $170 billion for around 90 countries. There were also important regional responses, such as the Caribbean Association of Doctors sharing epidemiological data on COVID-19 and the Caribbean Community ensuring coordinated action between countries, the EU's Inclusive Vaccine Alliance and ambitious recovery plan, and the African Vaccine Acquisition Task Team, a new initiative for surveillance enhancement and vaccine purchasing and sharing. Additionally, established health organisations such as the Global Fund to Fight AIDS, Tuberculosis, and Malaria (known as the Global Fund) partnered with countries to reprogramme up to 5% of their current grants towards supporting COVID-19 responses. 145 There has also been rapid acceleration in the digital transformation of health systems and innovation in health-system delivery. Regional differences in mortality rates The official cumulative death toll from COVID-19, reported by the Institute for Health Metrics and Evaluation (IHME), was 6·9 million as of May 31, 2022. However, the IHME estimates the true death toll at more than twice this number, 17·2 million, 146 and even higher values have been estimated by others (The Economist 147 estimates a death toll of 19·4 million as of February, 2022). Major differences in COVID-19 mortality are seen between WHO regions (for the countries in each WHO region, see appendix p 1). For each WHO region, we show the reported cumulative deaths and total cumulative deaths estimated by IHME until May 31, 2022, both in absolute numbers (figure 2A ) and per million population (figure 2B). Figure 2 Estimated and reported cumulative deaths from COVID-19, globally and by WHO region, as of May 31, 2022 Estimated and reported cumulative number of deaths (A) and cumulative number of deaths per million population (B), globally and by WHO region, as of May 31, 2022. All data are from the Institute for Health Metrics and Evaluation (IHME), accessed May 31, 2022. 146 Note that the IHME reference scenario provides a range of cumulative and daily estimated infections and cumulative and daily estimated deaths, and we refer to the mean estimate in all figures. Reported cumulative and daily deaths from May 2 to May 31, 2022, were modelled on the basis of past data. The WHO Western Pacific region, which includes east Asia and Oceania, stands out for its very low average mortality rate, both in terms of reported deaths (125 per million population) and estimated total deaths (300 per million population) attributed to COVID-19. The region with the next lowest number of reported deaths is the WHO African region, for which the reported death rate is 165 per million population; however, the total death rate estimated by IHME is more than ten times higher, at 1774 per million population. The undercounting of deaths is consistent with serosurveys (blood testing for previous infection) that suggest high infection rates in Africa and a massive undercounting of actual infections. 148 The WHO South-East Asia region similarly has a relatively low reported mortality rate (366 per million population) and a far higher estimated total death rate (2549 per million population). It is well established that deaths from COVID-19 in India were vastly undercounted during the delta (B.1.617.2) wave of April–June, 2021. 149 The WHO Eastern Mediterranean region has a mid-range reported death rate (734 per million population) and a high estimated total death rate (2779 per million population). The remaining two WHO regions have the highest estimated total death rates: 4144 per million population for the European region and 4051 per million population for the region of the Americas. Particular attention should be paid to Latin America, which has the highest number of excess deaths relative to population. Although the region represents 8·4% of the global population, as of the middle of July, 2022, more than 111 614 cumulative cases of COVID-19 per million population and nearly 2603 deaths per million population from COVID-19 have been reported, according to Reuters. 150 The international distribution of COVID-19 death rates is almost the opposite of what might have been expected before the pandemic. On the basis of the 2019 publication of the Global Health Security Index, which ranked the United States first and the UK second in the world in terms of preparedness for pandemics and epidemics, it was widely assumed that the United States and Europe had the strongest pandemic response capacities and would fare best in a pandemic. 151 In general, the report gave high preparedness scores to countries in the European region and the region of the Americas, and generally much lower scores to the countries of the Western Pacific region. For example, China was ranked 30th and New Zealand was ranked 32nd. The 2019 assessment failed to predict the poor quality of the public policy response to the pandemic in the European region and the region of the Americas, and the much higher quality of response in the Western Pacific region. Other than in the Western Pacific region, national pandemic control systems were very disappointing compared with expectations in 2019. The broad pattern of SARS-CoV-2 transmission across regions can be seen in figure 3 . Only the Western Pacific region held the rate of new infections per day to very low levels. The other regions all experienced several waves, and none used the phases in which infection rates were low to move to a suppression strategy. The biggest wave in terms of deaths, claiming millions of lives in just a few weeks, was the delta wave from April to June, 2021. This wave resulted in astounding mortality, particularly in India, with an estimated 2 million deaths worldwide in these two months. The omicron variant brought another enormous global wave of infections, starting in December, 2021. Figure 3 Estimated number of infections per 100 000 population by WHO region, Feb 4, 2020–May 31, 2022 Data from the Institute for Health Metrics and Evaluation, accessed May 31, 2022. 146 Other than regional cooperation among the countries of the Western Pacific region, there was little early effort across governments to coordinate approaches to limit transmission of the virus during the pandemic. National governments have failed to perceive, or to articulate, the core logic of a weakest-link game: to successfully control the transmission of the virus, each country is dependent upon the actions of other countries, so a cooperative approach is necessary to achieve the desired outcome. Instead, national governments generally took actions on their own with disregard for any effects on, or from, other countries. Regional policy choices in confronting the pandemic Suppression strategies in the Western Pacific region The countries of the Western Pacific region generally adopted suppression strategies, and were broadly successful in their implementation. These strategies comprised two phases: from the outbreak in late 2019 until early 2022, when the omicron variant first emerged; and then from early 2022 onwards in the context of the omicron wave. During the first 2 years, the region generally suppressed transmission of the virus, during which they implemented a vaccination campaign in 2021. With the emergence of the omicron variant, most countries of this region then shifted from suppression to mitigation as they adopted a policy known as living with the virus, counting on high vaccination coverage to keep mortality rates relatively low. Only China maintained a suppression strategy during the omicron phase, resorting to new lockdowns of the population in major urban areas. Following the initial outbreak in Wuhan at the end of December, 2019, around 3000 confirmed new cases of COVID-19 per day were reported in China during February, 2020. Thereafter, China succeeded in suppressing the pandemic to fewer than 100 new cases daily by early March, 2020, and to fewer than ten new cases per day by late April, 2020. The methods of suppressing transmission included a rigorous temporary lockdown of Hubei Province, combined with aggressive case-seeking, large-scale testing, tracing of contacts, use of QR codes to track the movement of individuals, and the isolation of all infectious people and their close contacts. 152 China closed its international borders and required new arrivals to quarantine for an extended period. Notably, China was intent to contain the pandemic from Jan 23, 2020, adopting a policy that became known in China as the zero-COVID strategy.153, 154 From early 2020 onwards, Australia, Cambodia, China, Hong Kong SAR, Lao PDR, New Zealand, Singapore, Viet Nam, and several small Pacific island states attempted to pursue a suppression strategy based on the Asia-Pacific Strategy for Emerging Diseases and Public Health Emergencies, which was adopted by the WHO Western Pacific region in the aftermath of the epidemic of severe acute respiratory syndrome in 2003. 155 Singapore was the only country of the Western Pacific region to have a substantial outbreak during the first half of 2020, which occurred when COVID-19 spread through the hostels of migrant workers. 156 The other countries of the region maintained fewer than ten cases per million people per day for most of 2020. 146 In 2020, transmission was suppressed without the benefit of vaccines. In 2021, countries of this region implemented successful vaccination campaigns alongside their suppression strategies. By the end of 2021, Australia, Cambodia, China, Hong Kong SAR, New Zealand, Singapore, and Viet Nam had all achieved full vaccination rates of adult populations of more than 60%. 157 At the end of 2021, the omicron variant reached the region, and both the feasibility and desirability of continuing the suppression strategy were called into question. The R 0 of the omicron variant is several times higher than that of the previous variants, making the feasibility of a suppression policy exceedingly difficult (figure 4 ). The social costs of transmission of the virus were considerably reduced, but certainly not eliminated, by the widespread—though incomplete—vaccination coverage achieved during 2021. Figure 4 Estimated daily infections and estimated cumulative infections in the Western Pacific region, Feb 4, 2020–May 31, 2022 Data from the Institute for Health Metrics and Evaluation, accessed May 31, 2022. 146 Owing to the difficulty in continuing the suppression strategy and the widespread vaccination coverage, most of the countries of the Western Pacific region relaxed their suppression strategies in early 2022, reverting to a more limited mitigation strategy. Cases of COVID-19 rapidly increased in Australia, Hong Kong SAR, New Zealand, Singapore, and Viet Nam in the first months of 2022, and deaths rose too, yet the death rates remained low because of the vaccination coverage (Figure 5, Figure 6 ). As a result, cumulative deaths per 100 000 population as of May, 2022, in the Western Pacific region were far lower than in any other region of the world. The suppression strategy during 2020 and 2021 therefore had a lasting benefit, as it gave time for high rates of vaccination coverage. Figure 5 Estimated daily infections and estimated daily deaths in the Western Pacific region during the omicron (B.1.1.529) wave, Jan 1–May 31, 2022 Data from the Institute for Health Metrics and Evaluation, accessed May 31, 2022. 146 Figure 6 Estimated daily infections and estimated daily deaths in Australia, China, Hong Kong SAR, New Zealand, and Singapore during the omicron (B.1.1.529) wave, Jan 1–May 31, 2022 Data from the Institute for Health Metrics and Evaluation, accessed May 31, 2022. 146 China, in contrast to most of the other countries in the region, chose to maintain a strict suppression strategy even during the omicron wave. Given the remarkable infectivity of the omicron variant, suppression during this period has required new and stringent lockdowns of the populations in major urban centres that have omicron outbreaks, including Shanghai and Beijing. There has been debate within China about the balance of costs and benefits of the zero-COVID policy with such an infectious variant. China is using the period of lockdowns to increase vaccine coverage in the population that remains unvaccinated. Flatten-the-curve policies in the Americas The contrast between the policy response of the Western Pacific region with that of the region of the Americas has been stark from the outset. Few if any countries in the Americas were prepared for the pandemic. In the United States, the initial test kits provided by the US CDC were faulty, 158 and during the first few months of 2020, the rate of testing was very low and tests were only available sporadically.159, 160, 161 Local public health departments capable of systematically testing, tracing, and isolating infected people were scarce, and US federal officials repeatedly downplayed the importance and risks of the outbreak. The United States, like many member countries of the Organisation for Economic Co-operation and Development, chronically underinvested in public health before the pandemic, devoting only 2·5–3% of the total health sector budget to public health.162, 163 The high mortality rate of the region of the Americas reflects the failures of this region to take concrete measures to suppress the epidemic, and high vulnerability to deaths from COVID-19 due to structural characteristics. Such characteristics include older populations;163, 164, 165 a high prevalence of comorbidities (eg, diabetes, chronic obstructive pulmonary disease, cardiovascular disease, and kidney disease) within the population;166, 167 the high proportion of the population living in congregate settings such as care homes,168, 169 prisons, 170 worker hostels, and homeless centres; 171 and social determinants of health including racial and ethnic disparities, income inequalities, and inequitable access to high-quality health care.172, 173 Another factor was the increase in anti-vaccine propaganda in the Americas that caused tens of millions of people to refuse vaccines, and hundreds of thousands to needlessly lose their lives.16, 174 Evidence suggests that the case ascertainment rate (defined as the number of confirmed new infections relative to the number of actual new infections) was much less than 20% in the first half of 2020 175 —meaning that viral transmission increased rapidly during February and March, 2020, with little awareness by authorities and the public until mid-to-late March. By early April, 2020, the United States was reporting around 30 000 cases per day (90 per million population), with the actual number of cases estimated to be more than five times higher. Between Feb 4, 2020, and May 31, 2022, there were an estimated 1·2 million deaths attributed to COVID-19 in the United States (figure 7 ). Figure 7 Estimated cumulative infections and estimated cumulative deaths in the United States, Feb 4, 2020–May 31, 2022 Data from the Institute for Health Metrics and Evaluation, accessed May 31, 2022. 146 The situation in the rest of the Americas was similar (figure 8 ). The pandemic arrived several weeks later in much of Latin America, but it then raged with little respite or sustained control. In South America as a whole, confirmed new cases reached 60 per million population on May 22, 2020, and then never decreased below that rate until the end of 2021. Canada also sustained major waves of infection, although generally with fewer cases per day per million people than in the United States. Figure 8 Estimated cumulative infections and estimated daily infections in the region of the Americas, Feb 4, 2020–May 31, 2022 Data from the Institute for Health Metrics and Evaluation, accessed May 31, 2022. 146 Many countries in the Americas imposed temporary lockdowns from roughly mid-March to the end of April, 2020. Yet, unlike in the Western Pacific region, these lockdowns were not meant to be a prelude to comprehensive suppression through testing, tracing, and isolating, and complementary policies on travel, public hygiene, and other regulations. The lockdowns in the Americas were generally designed to slow the transmission of the virus rather than to keep R below 1. These lockdowns were described by most countries as flattening the curve to avoid an overflow of patients with COVID-19 in hospitals. No countries in the Americas pursued a suppression strategy. Flatten-the-curve policies in Europe Governments in the European region did not aim to suppress the pandemic, only to slow the transmission of the virus. Several countries in western Europe saw large numbers of infections early in the pandemic—notably Belgium, France, Germany, Italy, Spain, and the UK. International travel to and from all these countries is extensive, and many infected people arrived from China during January and early February, 2020.176, 177 There was little testing in the first weeks of the outbreak, and a massive surge of cases occurred in March, 2020. To lessen the pressure on hospitals, European countries adopted tough lockdown measures, 178 although—as in the region of the Americas—the focus was on flattening the curve rather than on ultimately suppressing the pandemic. By June, 2020, case transmission in Europe declined to low levels (figure 9 ), yet European governments failed to implement continued rigorous measures of physical distancing (including the wearing of face masks and bans on large gatherings). On the contrary, the decline in the number of cases by June, 2020 led governments in Europe to rescind control measures and to encourage a return to life as usual, especially in the context of Europe's upcoming summer holidays; this relaxing of control measures led to another wave of infections in countries across Europe in September, 2020.179, 180 Figure 9 Estimated cumulative infections and estimated daily infections in the European region, Feb 4, 2020–May 31, 2022 Data from the Institute for Health Metrics and Evaluation, accessed May 31, 2022. 146 This second wave gave rise to renewed restrictions and partial lockdowns throughout Europe, which again led to a decline in cases by June, 2021. Once again, policies were eased in time for vacations in July, 2021, setting the basis for a third wave in October, 2021—this time due to the new delta variant, which was first identified in India. A moderate decline in cases of the delta variant by late October and November, 2021, was soon followed by a fourth wave of infections due to the omicron variant, beginning in December, 2021. Compared with other WHO regions, Europe was relatively more successful at vaccinating a substantial portion of its population rapidly in 2021, which contributed to limiting the number of severe cases and deaths in subsequent outbreaks later in 2021 and in 2022. 181 Sub-Saharan Africa's hidden epidemic At the outset of the pandemic, conditions of poverty, urban crowding, and fragmented health-care systems gave rise to fears that sub-Saharan Africa would suffer the most extreme outcomes in the world as the virus spread through the continent. On the surface, at least, that was not the case, as the numbers of reported cases and deaths remained low. Estimated infections are about 100 times higher than reported infections in the region, but still remain lower than in all other WHO regions (figure 10 ). Although some scientists speculated that pre-existing immunity from intensive exposure to other pathogens—including other coronaviruses—could have contributed to the African region's low reported case load, 182 a simpler explanation now seems most likely. Several serosurveys suggest that Africa experienced large waves of infection but that most cases went unreported, in part because they did not lead to severe disease. A serosurvey in six districts of Zambia showed that the case ascertainment rate was only one reported case per 92 actual infections, or 1·1%. 183 Similar findings apply to other countries—such as the Republic of the Congo, 184 the Democratic Republic of the Congo, 185 Malawi, 186 and Kenya 187 —suggesting that up to 80% of the population had already been infected by COVID-19 before the arrival of the omicron variant. Figure 10 Estimated cumulative infections and estimated daily infections in the African region, Feb 4, 2020–May 31, 2022 Data from the Institute for Health Metrics and Evaluation, accessed May 31, 2022. 146 Although partly attributable to low testing capacity, the under-reporting of cases is also driven to an important extent by Africa's demography: the population is young, giving rise to a relatively high proportion of asymptomatic and mild cases. The median age in Africa is around 18 years, compared with between 30 years and 35 years across Asia, the Americas, and Oceania, and around 42 years in Europe. 188 Because the risk of severe disease and death from COVID-19 increases exponentially with age, 189 it is not surprising that Africa—and sub-Saharan Africa in particular—has recorded relatively fewer deaths, despite very limited containment policies and generally inadequate health systems. Even estimates of the total number of deaths from COVID-19 in the African region, which are up to ten times higher than the official numbers, place it as the region with the second-lowest number of deaths per million people. South-East Asia and the delta surge The South-East Asia region saw a wide disparity in health-system capacities and consequent outcomes for COVID-19 management, spread, and mortality. Several countries (eg, Sri Lanka, 190 Thailand, 191 Lao PDR, 192 and Cambodia 193 ) were able to contain the spread of COVID-19 in the first year of the pandemic as a result of effective contact tracing, wearing of face masks, and physical distancing. Other countries (eg, Pakistan, Bangladesh, Malaysia, and India) saw sharp rises in infections, but death rates remained relatively low. 146 The delta wave caused a spike in cases in most countries in the region, with much higher mortality, although actual deaths have been substantially under-reported in several countries (figure 11 ). This under-reporting was perhaps most notable in India during the delta wave from March to July, 2021. India was among the first countries to impose travel restrictions, suspend international flights, and impose a strict lockdown early in the pandemic. 194 In March, 2020, India had 654 cumulative confirmed cases of COVID-19, 146 and the adopted restrictions aimed to prevent community transmission and give the health system time to ramp up. Figure 11 Estimated cumulative infections and estimated daily infections in the South-East Asia region, Feb 4, 2020–May 31, 2022 Data from the Institute for Health Metrics and Evaluation, accessed May 31, 2022. 146 After lockdown was relaxed at the end of May, 2020, India saw a surge of cases during June and July; however, because case counts had decreased by the end of the year, the Indian government relaxed most controls early in 2021. 195 In March, 2021, two factors combined to markedly increase the number of infections. First, the highly infectious delta variant emerged, and other variants of concern arrived from other countries. 196 Second, several events in the country—such as elections, public protests, and religious festivals—brought large aggregations of people together, and most participants in these events did not wear face masks.197, 198 The combination of new variants and crowds proved devastating in India. All levels of government—central, state, and local—were unprepared for the speed and the scale of the delta surge as it gathered momentum in March, 2021. Hospitals were overwhelmed within weeks, and breakdowns in the medical supply chain contributed to a severe shortage of oxygen, hospital beds, and pharmaceuticals. 199 In smaller towns and rural areas, large numbers of patients went untreated. 200 India reported roughly 20 million COVID-19 infections and 250 000 deaths attributed to the disease between Jan 1 and June 30, 2021, but the actual numbers are estimated to be vastly higher. Seroprevalence of COVID-19 IgG antibodies in non-vaccinated individuals older than 6 years increased from 24% in December, 2020 and January, 2021, to 62% in June and July, 2021, confirming that hundreds of millions of people were infected during the delta wave. 201 The IHME estimates that there were around 417 million infections and 1·6 million deaths from COVID-19 in India between April 1 and July 1, 2021, compared with just 18 million reported cases and 252 997 reported deaths (figure 12 ). Another study estimates between 3·1 million and 3·4 million deaths from COVID-19 in India between the start of the pandemic and Sept 1, 2021, compared with 440 000 reported deaths; these figures suggest that the reported death count was roughly 14% of the actual value. 149 Figure 12 Estimated daily infections and estimated cumulative deaths in India during the delta (B.1.617.2) wave, April–June, 2021 Data from the Institute for Health Metrics and Evaluation, accessed Feb 19, 2022. 146 After the delta surge, cases decreased markedly and vaccination efforts accelerated; as of Sept 1, 2022, more than 76% of the eligible population of India had been vaccinated with a single dose and more than 70% were fully vaccinated. 202 Despite the omicron surge in January, 2022, hospitalisation rates and deaths remained low. Across the South-East Asia region, the omicron wave saw infections spike, even in countries that had managed to contain the spread of COVID-19 in the first year of the pandemic (figure 11). However, hospitalisation rates and deaths remained low in most countries. By June, 2022, at least 60% of the population in all countries except Myanmar had received a full course of vaccination. 181 The Eastern Mediterranean region The Eastern Mediterranean region has reported relatively low death rates. However, as in the African and the South-East Asia regions, the number of deaths has been substantially under-reported (figure 13 ). With ongoing conflict in Palestine, Yemen, Syria, and Somalia, in addition to the takeover of Afghanistan by the Taliban in 2021, surveillance and health systems have limited capacity. Additionally, with the prevalence of refugees and displaced people congregating in emergency shelters and crowded settings, disease outbreaks are highly likely. Figure 13 Estimated cumulative infections and estimated daily infections in the Eastern Mediterranean region, Feb 4, 2020–May 31, 2022 Data from the Institute for Health Metrics and Evaluation, accessed May 31, 2022. 146 This region has great extremes in terms of COVID-19 vaccination coverage. For example, in the United Arab Emirates, about 90% of the eligible population is fully vaccinated and 95% is vaccinated with at least a single dose as of June, 2022. 203 However, in Yemen, only 2% of the population is fully vaccinated and 5% is vaccinated with at least one dose. 204 In Afghanistan, about 13% of the eligible population is fully vaccinated, with about 10% projected to be vaccinated with at least a single dose. 205 In addition to the destabilising effects of political instability and violence, the low vaccination rates are also related to low literacy rates and to vaccine hesitancy among the population. Premature lifting of public health and social measures In March, 2022, governments around the world removed many of the remaining public health and social measures regarding face masks, indoor gatherings, large events, physical distancing, and testing, especially in countries with high vaccination rates. This removal of measures coincided with the emergence and prevalence of the highly infectious omicron variant, with a reproduction rate estimated to be three to four times that of the delta variant. 206 The omicron variant led to the greatest surge yet in the number of confirmed cases per day, estimated by the IHME to be roughly 45 million per day at the peak in January, 2022, compared with the previous worldwide peak of around 15 million per day in April, 2021. IHME estimates that 4·3 billion people, or 54% of the world's population, were infected with SARS-CoV-2 between Dec 1, 2021 and May 31, 2022. 146 There are many costs associated with the premature lifting of public health and social measures. First, a considerable portion of the population is not immunised, and many vulnerable groups—such as people who are immunocompromised and who cannot mount an effective immune response to vaccination—will never be able to achieve effective immunity. 207 Second, vaccination coverage provides only partial and waning protection against infection, although more protection against hospitalisation and death. 208 Additionally, coverage of a population with a third booster dose is in general far lower than coverage with two vaccines. Third, further variants and returning waves of infection can be expected as current immunity wanes, so dismantling basic controls and even advertising the end of the pandemic is premature, and likely to be harmful when restrictions become necessary again. Fourth, high infection rates mean high rates of long COVID, which disrupts education and jobs, brings physical and mental suffering to individuals and their families, creates a considerable burden on health services, and undermines economic recovery. One widespread view among epidemiologists is that the virus is likely to become endemic, recurring seasonally (notably during the cold weather in each hemisphere, because of the much greater risk of indoor transmission) yet with lower mortality, similar to the annual mortality burden of seasonal influenza, 209 and with annual vaccines targeting the dominant variant, as for influenza.210, 211 However, there is little assurance that we have achieved such predictability, and influenza too can give rise to devastating new variants and pandemics, such as the outbreaks in 1957–58 and 1968 that are each estimated to have claimed between 1 million and 4 million lives.212, 213, 214 Although the omicron variant of SARS-CoV-2 has been less lethal than the preceding delta variant, future variants are as likely to be more virulent or less virulent, and existing vaccines are likely to become less protective against serious illness and death. Three major risks should be kept in mind at this stage. First, future variants might be highly infectious and deadly. Second, people who are currently unvaccinated (ie, have not received any doses of a COVID-19 vaccine)—roughly 2·5 billion people or 33% of the global population as of Aug 19, 2022 146 —remain vulnerable to morbidity and mortality and provide opportunities for the virus to circulate freely, increasing the risk of dangerous mutations occurring. Third, the omicron wave has shown that serious economic disruptions can occur even in highly vaccinated populations, 215 as it led to shortages of health and other essential workers, school closures, disruptions to business operations, and prevented the restoration of tourism and international travel. Widespread political ineffectiveness One of the most striking occurrences of the pandemic has been the irresponsibility of several influential political leaders. National and local politicians act on various motives, including political timelines and electoral cycles, fears of public backlash, narrow economic interests, and lack of knowledge and expertise. Because the pandemic was a novel experience for most leaders, humanity was vulnerable to their learning curves, ignorance, and mixed motives. National responses were often improvisational, occasionally bordering on the absurd. 216 Several national leaders made highly irresponsible statements in the first few months of the outbreak, neglecting scientific evidence and needlessly risking lives with a view to keeping the economy open.217, 218, 219, 220, 221, 222, 223 Leadership styles and public messaging have been instrumental in building trust and leveraging effective responses, and we note that the gender of political leaders seems to have had a role in the rhetoric, with women leaders more often concerned than male leaders about individual-scale effects and social welfare.224, 225 A rather small group of governments, many led by women, kept death rates much lower than in other countries. The Western Pacific region—including Australia, China, New Zealand, Republic of Korea, and Taiwan, Province of China—was largely successful at maintaining relatively low death rates. 226 In Europe, two Nordic countries stand out as having lower death rates than the rest of the continent—Iceland (about 210 deaths per million population 227 ) and Norway (about 280 deaths per million population 228 ). In contrast with its neighbour, Norway, Sweden's response was lax, 229 resulting in a much higher death rate (about 1500 deaths per million population). 230 National policy choices made a vast difference in health outcomes. Disinformation in media False news, propaganda, and demagoguery are nothing new. Yet the means of deploying these malicious forces change over time with each new medium of communication. In the past two centuries, newspapers, radio, and television have each served as purveyors of misinformation and disinformation. Now we are experiencing new forms of disinformation through digital media, 231 which are especially powerful in disseminating ideas because of their reach (eg, Facebook has nearly 2·4 billion subscribers), immediacy, relative anonymity, and the ease of creating echo chambers of tightly clustered groups receiving utterly different impressions of the facts from other online communities.232, 233, 234, 235 The COVID-19 pandemic is not the first crisis in which misinformation has been shared, 236 and it is not the only issue that is permeated by lies and distrust. The spread of misinformation and disinformation is common among climate change deniers, the tobacco industry's continued fight against warning labels, and parents who refuse or delay routine childhood vaccinations. 237 The rapid speed at which news is shared, the deliberate spread of misinformation and disinformation by political leaders, and the lack of truth-telling and adequate oversight creates an overwhelming environment that fosters distrust in health officials and promotes the idea that individual opinions have equal weight to the best available scientific evidence.237, 238 This distrust has real-world consequences. One study in Italy found that poor public understanding of the risk of COVID-19 infection, even during the peak of COVID-19 incidence in 2020, was related to 11 411 people breaking lockdown regulations. 238 Some media outlets erroneously promoted dangerous or experimental treatments, such as hydroxychloroquine and ivermectin, resulting in unnecessary visits to hospital emergency departments and shortages of such medications for people with legitimate needs.239, 240, 241 All countries proved to be highly vulnerable to disinformation and misinformation regarding the pandemic, with one study documenting that 46% of people in the UK and 48% of people in the United States were exposed to false information. 242 A 2020 study of the most-viewed videos about COVID-19 on YouTube found that more than 43% of these videos contained misleading information. 243 In the United States, the political far-right has promoted anti-science rhetoric, as shown by their opposition to COVID-19 vaccines and anti-COVID-19 prevention measures. Anti-science rhetoric and disinformation about COVID-19 are now promoted by several elected members of the US Congress, 244 and are often disseminated by cable television news channels and podcasts. By some estimates, between 100 000 and 200 000 Americans lost their lives because they refused COVID-19 vaccinations and were open in their defiance. 245 This anti-science movement has globalised with tragic consequences. 246 Pope Francis raised his powerful voice against the dangerous infodemic regarding vaccines, declaring: “To be properly informed, to be helped to understand situations based on scientific data and not fake news, is a human right.” 247 WHO is the leading international authority on health and should be supported by its member states and by other UN agencies to lead the global response to a health crisis. However, various political leaders publicly undermined WHO, disseminated campaigns against it and its advice, and even tried to halt its funding.234, 248, 249, 250, 251 Public attitudes towards pandemic control and investments in behavioural sciences Public attitudes towards the pandemic differed markedly around the world. Public attitudes are shaped by various factors, including national culture, education level, peer groups, social media, and the actions and statements of national leaders. Other important aspects include emotions, past habits, capability, convenience, and social norms. 252 Human behaviour is central to transmission of the disease and suppression of the pandemic. Behavioural science provides evidence about the effectiveness of interventions and policies across contexts and the processes required to bring about change; for example, evidence-based optimal behaviours for pandemic control include getting vaccinated and tested, isolating when infectious, wearing face masks, and physical distancing. 253 At a minimum, behavioural change requires appropriate motivation, capability, and opportunity; the absence of any one of these will undermine public health efforts. The COM-B model 252 is a synthesis of frameworks of behaviour change has identified nine types of intervention strategy and seven policy options to underpin them. These strategies are most effective when selected in combination and according to a scientific understanding of behaviour in the political and socioeconomic context. Motivations for behaviour change include self-interest, belief in the effectiveness and feasibility of behaviours, and reducing aversive emotions. Capability includes knowing exactly what to do, when, and why, and is closely linked to health literacy. Desired behaviour will not occur with motivation and capability if people do not have the opportunity—both physical (eg, easy access to vaccines and financial resources to self-isolate) and social (eg, normative influences to wear face masks). Health and safety cultures relating to infectious diseases should be built into societies, embedding risk assessment and management into everyday life to maintain behaviours. 254 There is evidence that trust in government helps to promote prosocial behaviours within the population. An empirical study published in 2021 concluded that greater trust in government regarding COVID-19 control was cultivated for governments that were perceived as well organised, fair, and to be disseminating clear messages and knowledge on COVID-19. This greater trust in government was also significantly associated with a greater adoption of health and prosocial behaviours and lower rates of decline in these behaviours over time. 255 Differences in national cultures also shape national responses. Cultural psychologists, for example, have distinguished between so-called tight and loose cultures, with tight cultures characterised by strong public adherence to social norms and loose cultures by lax adherence. 256 Societies also differ along a spectrum of individualism versus collectivism in terms of social norms, in which individualism champions the freedom of individual choice whereas collectivism champions conformity with the group. Tight and collective cultures were more attuned than loose and individualistic cultures to respect the social norms of physical distancing and to adhere to isolation when potentially infected, leading to greater success in controlling the pandemic. Social media also had a crucial role in how public health and social measures were perceived by the public. Scientific understanding of the virus and its characteristics have continued to evolve throughout the pandemic, and as more information has been gathered about the genetic make-up, reproduction rate, and methods of transmission of the virus, and about new variants of concern, public health and social measures have had to be adjusted. However, the reliance of social media on so-called clickbait headlines and short commentaries has had a polarising effect on many communities. Social media left little room for the deep discussion of issues and enabled rapid orthodoxies to develop, which in turn often delegitimised necessary adjustments to advice and to public health and social measures in response to the evolving understanding of the virus. The preparedness of populations also differed as a result of their different exposures to and experiences with previous pandemics. Residents of countries in the Western Pacific region, remembering the 2003 epidemic of severe acute respiratory syndrome, strongly adopted public health and social measures such as wearing face masks, increasing hygiene, and avoiding large public gatherings. Figure 14 shows survey data from YouGov on the use of face masks in public places. Notably, survey participants in countries of the Western Pacific region and the South-East Asia region began wearing face masks early in the pandemic and maintained the highest reported use throughout 2020–21. Surveyed populations in the Eastern Mediterranean region, the European region, and the region of the Americas were later and slower to start wearing face masks, missing an early opportunity to suppress the pandemic. Additionally, after reaching a maximum in late 2020, the use of face masks declined in these regions at higher rates than in the Western Pacific region. YouGov did not conduct this survey in any countries in the African region. 257 Figure 14 Proportion of the surveyed population who wear face masks in public places by WHO region, March 15, 2020–Dec 15, 2021 Data from YouGov. 257 Another indicator of public attitudes to prosocial public health measures is public protests over COVID-19 restrictions. The Carnegie Endowment for International Peace tracks global protest actions and identifies protests that are related to COVID-19. 258 Such protests have overwhelmingly taken place in the region of the Americas and the European region, as shown in figure 15 . Emphasis on the value of liberty, in particular in high-income countries within these regions, has been widely invoked in these protests to reject mandates on face masks and vaccines, alleging that such mandates infringe on individual freedom. Figure 15 Countries in which coronavirus-related protests were held, March, 2020–November, 2021 Countries in which protests were held, coloured by WHO region. Protests were most heavily concentrated in the European region (dark grey; 22 protests) and the region of the Americas (dark blue; 16 protests), with only a few protests in the Western Pacific region (light blue; four protests), the South-East Asia region (yellow; two protests), the Eastern Mediterranean region (orange; five protests), and the African region (purple; one protest). For countries in light grey, no protests were recorded. Data from the Carnegie Endowment for International Peace. 258 An additional and alarming aspect of the public response in many countries has been widespread vaccine hesitancy, encouraged by a hostile and coordinated anti-vaccine movement that has spread dangerous and false information about the health risks of vaccines and has campaigned against vaccine mandates. Anti-vaccine disinformation in the United States has been traced to well organised right-wing groups that use targeted social media. 259 Global evidence suggests that vaccine hesitancy is highest among people with lower levels of formal education and with lower incomes, emphasising the need for more evidence-based health education and communication on preventive care and health maintenance in public schools and from government, trusted community leaders, and medical professionals, in addition to the regulation of anti-science messages on social media.174, 260, 261 Unequal health effects of the pandemic The effects of the pandemic have been hugely unequal across the world, both within and between countries. Disease effects can be distinguished from economic and social effects, although they are deeply intertwined. Case fatality rates (defined as the number of confirmed deaths per confirmed case) have been high in older people, with mortality rates at least ten times higher in people older than 65 years than in those younger than 40 years. Other groups that have been disproportionately affected include people with comorbid medical conditions, particularly those with immunocompromised status; people living with physical disabilities or mental disorders; residents of congregant settings; and essential workers.262, 263, 264, 265, 266 First responders, service-sector workers, and other essential workers often receive low wages and, in some countries, low access to high-quality health care, and therefore disproportionately experience health, economic, and social effects. The majority of health-care workers, in particular, are women. Ensuring mental health and wellbeing is a crucial part of preparedness and response to pandemics and other similar threats. 267 There does not need to be a trade-off between saving lives and safeguarding mental health.268, 269 A study conducted across 15 countries found a small negative association between COVID-19 policy stringency and mental health, consistent with previous work, and a negative association between the intensity of the pandemic (measured by the average number of deaths per day per 100 000 people) and mental health. However, because early and targeted suppression policies not only minimise deaths but also lead to lower average levels of stringency through better transmission control, the mental health of people in countries that aimed to suppress the virus was found to be equivalent to or better than the mental health of people in countries with a flatten-the-curve strategy. 270 It is clear that various aspects of mental health declined during the COVID-19 pandemic. 271 Numerous sources documented a clear and consistent pattern of increased anxiety and depression in the early months of the pandemic, which declined during later months in some countries but not in others. 272 A systematic review estimates more than 53 million additional cases of major depressive disorders and 76 million additional cases of anxiety disorders globally in 2020. 273 Data on short-term emotions, life evaluations, loneliness, and self-harming behaviour were more mixed. Daily emotions were more negative (fewer positive emotions and more negative emotions), whereas reports of loneliness and self-harm initially increased but later returned to their pre-pandemic levels. Population-level trends conceal the extreme strain felt by some individuals. Pre-existing inequalities in mental health have remained during the pandemic. For instance, people living in marginalised communities with lower socioeconomic status and mental health conditions reported poorer mental health than the general population.30, 274 The pandemic has also introduced new profiles of risk, with younger people (aged 18–34 years), women, and people with children under the age of 5 years showing the largest increase in psychological distress. 275 Stark socioeconomic effects Socioeconomic divides have widened since the start of the pandemic. First, essential workers are disproportionately concentrated among vulnerable minority and low-income communities. Since the start of the pandemic, many people whose work is computer-based or telephone-based have been encouraged to work from home.276, 277 People who could work online, such as office workers, fared better in terms of health and socioeconomic wellbeing than goods-producing workers (eg, those in agriculture, mining, manufacturing, or construction) and workers in industries categorised as essential (eg, health-care workers and grocery store workers) who have been required to work in-person. Onsite work often entailed substantial risk of infection (such as at meatpacking plants) or temporary closures of worksites. Workers of colour in the United States, who are over-represented in high-risk, essential industries, are also more likely than other populations to have lower incomes, live in multigenerational households, and use public transport to travel to work. These groups of workers have been at the greatest risk of serious illness and death from COVID-19. 278 Inadequate efforts were made to prioritise the health and safety of these workers, first in requiring precautions in their workplaces, and later in reducing structural barriers to accessing vaccines when they became available.262, 263, 279 Workers in the retail trade, personal services, and leisure and accommodation (eg, restaurants, hotels, tourist locations, entertainment, and sports) were also especially affected by closures and lockdowns. Of the 22·1 million job losses in the United States between January and April, 2020, 8·3 million (38%) occurred in the leisure and hospitality sector, 280 although this sector accounted for only 11% of pre-pandemic employment. 281 In general, these differences strongly exacerbated income inequalities, as the online workforce is considerably higher paid on average. Additionally, informal employment accounts for 60% of global employment; the International Labour Organization projected in early 2020 that 1·6 billion informal workers would be the most severely affected and would lose 60% of their earnings in the first month of the pandemic. 282 In India, nearly 1 million front-line community health workers went on strike in December, 2021, owing to a lack of protective equipment and wages of $40 a day, as they were asked to be the soldiers of the pandemic. 283 Additionally, health workers worldwide are facing increased violence, initially because they were viewed as vectors of disease and then because of polarised politics surrounding vaccination. 284 Second, women have borne a disproportionate socioeconomic burden, 285 and existing widespread gender inequalities in terms of labour, income, personal safety, education, and food security intensified over the course of the pandemic, with important regional, national, and local variations and disparities.286, 287, 288 Although the overall death toll of COVID-19 is higher among men than women, 289 survey results confirm a disproportionate effect of the pandemic on the livelihoods of female workers.290, 291, 292 Worldwide, women comprise 70% of front-line health and social workers. 293 The poorest women subsidise health care globally; it is estimated that half of the contributions by women to health care, which totals approximately $1·5 trillion annually, are in underpaid and unpaid work. 294 Women faced an enormous increase in the demands on their time during the pandemic.295, 296 When parents are charged with supervising a child's remote learning, or when they cannot rely on schools consistently being open (perhaps due to hybrid schooling situations or quarantines after COVID-19 exposures), they might not be able to do their own work. These situations had a disproportionate effect on women, who dropped out of the labour force at higher rates than men.297, 298 In Latin America, the share of women in the workforce declined from 51·4% in 2019 to 46·9% in 2020. In the United States, the number of women in the workforce who were mothers decreased by 21·1% between March and April, 2020, whereas the number of working fathers decreased by only 14·7%. 299 Women are also disproportionately represented within the population of care centres for older people, many of which experienced outbreaks of COVID-19 and many deaths from the disease. In the United States, for example, women account for 70% of residents in nursing homes. 285 Women faced an epidemic of domestic violence alongside the COVID-19 pandemic, as confirmed in several meta-analyses.300, 301 Additionally, a WHO survey showed that across 105 countries, 68% of women faced disruption in family planning services early in the pandemic. 302 Third, children younger than 18 years are considerably affected by the indirect repercussions of the pandemic. 303 Nearly 80 million children are at risk of vaccine-preventable diseases as a result of fragile health systems that have been disrupted by the pandemic. 2 Children also face threats to their immediate and long-term wellbeing from interruption in other essential services, particularly in-person schooling. 304 Resources permitting, schools provide food, safety, nurturance, sociality, cognitive development, and education. 305 However, 195 countries closed schools during the pandemic, affecting more than 1·5 billion children and young people and posing enormous long-term and unrecoverable costs to them, their parents, and the economy. 306 School closures have had devastating effects on student learning, mental health, socioemotional outcomes, and lifetime earning potential, such as education backslides, increasing drop-out rates, and increased abuse and neglect. In-person schooling was deprioritised even as other non-essential or less essential community and economic activities continued.307, 308, 309 School closures also affected the safety of children, increasing their exposure to abuse. 310 Lack of economic support at home and the lack of protection offered by schools affected girls in particular, as it put them at increased risk of mental health problems, violence, child marriage, pregnancy, female genital mutilation, and HIV infection, with limited or no access to services.304, 310 11·2 million girls and young women globally are now at risk of not returning to care centres, schools, or universities. 304 Across the United States and Mexico, at least 62 million elementary-age and secondary-school-age children (aged 5–18 years) were physically out of school for at least 13 months continuously. 311 In Latin America, more than 165 million young people stopped attending classes in person. 312 To deal with these school closures, the education sector pivoted to online learning and digital resources, immediately redefining access to learning by access to the internet. 310 As of December, 2020, 64% of low-income students in the United States were attending school solely via computer, compared with 48% of high-income students. Black (66%) and Hispanic (64%) students were almost twice as likely as White (34%) students to be learning fully remotely, and were also twice as likely as White students who were also remote to have no live access to a teacher. 311 UNICEF estimates that a third of the world's schoolchildren were unable to access digital learning. 3 School closures have also affected children's physical health, food security, and nutritional status. Since the beginning of the pandemic, UNESCO estimates that nearly 370 million children across 50 countries have missed meals since school closures began, and that globally an estimated 39 billion in-school meals have been missed as a result of pandemic-induced school closures. 313 An average of four in ten in-school meals have been missed by children around the world, and this number increases to as many as nine in ten in some countries. 310 The absence of a structured, school-led routine and peer interactions has disrupted the lives of children; amplified the anxiety caused by isolation and their fears of the disease; and led to the loss of physical, intellectual, and social engagement. Motivation levels in children have declined because of the inability to play outdoors (which also affects their physical health), meet friends, and be in the classroom. Globally, a median of one in five young people aged 15–24 years said that they often felt depressed or had little interest in doing things. 314 One estimate from the World Bank suggests that this generation of children could potentially lose an estimated $10 trillion globally in their life earnings. 310 The loss of learning could also potentially increase learning poverty levels to 63% and drive countries even further off-track from achieving their learning poverty goals.315, 316 The estimated economic loss for south Asia due to school closure is projected to be between $622 billion (best-case scenario) and $880 billion (worst-case scenario). 310 For 2020 alone, the IMF estimated learning losses from mandated school closures at 20–25% of the school year in advanced economies and twice that in emerging market and developing economies (figure 16 ). 318 Figure 16 Duration of school closures worldwide between March, 2020 and October, 2021 School closure is defined as the government-mandated closures of educational institutions affecting most or all (≥80%) of the student population enrolled from pre-primary to upper-secondary levels. Reproduced from Agarwal R, 317 by permission of the International Monetary Fund. As the pandemic entered its third year, schools continued to impose stringent restrictions. Although efforts should be made to reduce undue risk of infection to children, it should further be recognised that it is possible to keep schools open without putting students and adults at excessive risk.319, 320 Because of the lower risk of severe COVID-19 among children, and considering the harms of school closures, in-person learning should be prioritised. Fourth, wealthy households experienced an increase in net worth even as hundreds of millions of workers worldwide lost their jobs. When the pandemic began, the major central banks (the Federal Reserve, the European Central Bank, the People's Bank of China, the Bank of Japan, and the Bank of England) engaged in massive monetary expansion, commencing around March 20, 2020, to forestall a credit squeeze or financial panic. This expansion resulted in an enormous increase in financial asset prices, especially of equities but also including real estate and wholly speculative assets such as cryptocurrencies and non-fungible tokens. The net worth of the richest ten people in the world increased to $1·5 trillion, 321 and these ten people possess six times more wealth than the poorest 3·1 billion people. Severe lockdowns also had important social, gender, and economic consequences. In Latin America, approximately 26 million people lost their jobs in 2020; this region had the largest losses in terms of hours worked worldwide and reduction of labour incomes. 312 Lockdowns also resulted in increased gender-based violence. Although the United Nations Population Fund projected in April, 2020 that if the lockdown continued for 6 months or more, 31 million additional cases of gender-based violence could be expected to occur, lockdown policies remained gender-blind. 322 Stringent lockdowns in India helped to slow the spread of COVID-19 but brought about severe economic and social hardships: gross domestic product (GDP) contracted by 24% for the second quarter of 2020 and 7·3% for the year 2020. An estimated 120 million jobs were lost for the duration of the lockdown, and millions of migrant workers returned to their home villages under severe distress, some trekking extremely long distances on foot, with little money, food, or water.323, 324, 325 Fifth, all regions have seen incidents of discrimination, xenophobia, racism, and attacks against people scapegoated for spreading the virus. 326 For example, political leaders and media outlets referred to SARS-CoV-2 as the Chinese virus, and racially motivated crimes against members of the Asian American and Pacific Islander communities in the United States have surged.327, 328 Additionally, when scientists in South Africa detected a new COVID-19 variant (which was later named the omicron variant) and informed WHO, some countries immediately imposed travel bans on African countries that had not yet recorded the variant.329, 330, 331 Indigenous Peoples have a higher risk of infection with severe socioeconomic, cultural, and health consequences, due to poorer baseline health status and lower access to health care and sanitation services than the general population. 332 Indigenous women and girls have often been disproportionately affected by the pandemic, as many live in extreme poverty with diminished incomes and limited access to mainstream and high-quality health care. 333 Internally displaced people, including people in areas controlled by armed groups, are especially vulnerable to infectious diseases and have limited access to safe housing, sanitation, and health care. 334 Similarly, refugees and asylum seekers who are forced to flee from their homes to escape war, persecution, or natural disaster often have no place to live other than refugee camps, which are often overcrowded and provide inadequate access to sanitary facilities. In such situations it is almost impossible to maintain physical distance, practise safe hand hygiene, and stay healthy. 335 Without lasting and durable solutions for internally displaced people, refugees, asylum seekers, undocumented migrants, and people who are stateless, they will continue to face limited or even non-existent access to essential services, including social services and health care, even if the pandemic abates. The burdens of long COVID Long COVID, chronic COVID syndrome, long-haulers, post acute sequelae of COVID, or post-COVID condition are some of the terms used to denote the persistence of symptoms or emergence of new symptoms at least 3 months after SARS-CoV-2 infection, irrespective of viral status, 336 with long COVID the most recognised term in scientific literature. 337 WHO provides the clinical case definition of post-COVID condition as symptoms 3 months from the onset of COVID-19 that last for at least 2 months and cannot be explained by an alternative diagnosis. 338 This syndrome and its underlying mechanisms are still poorly understood, as it affects people regardless of the severity of their COVID-19 infection, including younger adults, children, and those who were not hospitalised.23, 339, 340 Reported symptoms of long COVID include fatigue, cough, chest tightness, chest pain, shortness of breath, sore throat, abnormalities of taste and smell, palpitations, myalgia, joint pain, numbness or tingling, skin rashes, hair loss, diarrhoea, urological problems, headache, neurocognitive issues including altered memory and concentration, cognitive blunting (known as brain fog), insomnia, impaired balance and gait, and mental health problems such as anxiety, post-traumatic stress disorder, and depressive symptoms, and these symptoms can be continuous or relapsing and remitting in nature. People with long COVID often report impaired quality of life and issues with education and employment. The social and financial effects of COVID-19 also contribute to the issues surrounding long COVID, including mental ill health.336, 341, 342, 343, 344, 345 The accurate reporting of long COVID is complicated. Disparities in epidemiological data are probably the result of differences in the base population, the accuracy of diagnosis, the reporting systems, and the capability of health-care systems. 337 There are several challenges in the diagnosis of long COVID, including the length of the follow-up period, the accuracy of self-reporting, the symptoms examined, the reliance on parent-reported symptoms for children, negative PCR tests (false-negative results) in some patients, an absence of antibodies in patients who do not seroconvert, and difficulties in establishing a direct link between symptoms and the infection.336, 337, 346, 347 Studies worldwide have reported various rates of long COVID and a wide range of symptoms; however, it is suggested that up to 35% of patients who were treated for COVID-19 on an outpatient basis and up to 87% of patients who were hospitalised with COVID-19 have continuing symptoms.348, 349 Underlying chronic conditions such as diabetes, hypertension, and cardiovascular diseases might also worsen after COVID-19 infection, requiring closer monitoring. 336 Some risk factors for long COVID are older age, female sex, having more than five symptoms during the acute stage of the infection, comorbidities including the presence of autoantibodies, and previous psychiatric disorders.341, 350, 351, 352 Conversely, vaccination against COVID-19 reduces the odds of developing long COVID after infection, and some people who had been infected with COVID-19 before vaccination and had long COVID reported improved symptoms after vaccination.353, 354, 355 Women, who are over-represented among populations in understaffed, forward-facing patient care roles and often lack access to adequate personal protective equipment (PPE, which is often poorly fitting for female bodies), may have an increased risk of long COVID. 356 UN Women reported in 2020 that around 70% of health-care workers infected in Spain, Italy, and the United States in the early months of the pandemic were women. 357 Studies show that long COVID could be related to organ damage, persistence of the virus in the body, post-viral syndrome, persistence of chronic inflammation or immune response (autoantibody generation), post-critical-care syndrome, complications related to comorbidities, reactivation of the Epstein-Barr virus induced by COVID-19-related inflammation, and adverse effects of medications, among other factors.336, 358, 359, 360 Long COVID has substantial physical, mental, social, and economic effects, and long COVID might itself be an emerging pandemic. 361 As COVID-19 is likely to become endemic, many people could have long-term health care and social care needs, which could overburden the systems of countries and of school and work environments. These people will require multidisciplinary and stigma-free care, which is not readily available in many settings.342, 362 Although the recording of exact epidemiological data for long COVID is difficult with existing knowledge and capacity of health-care systems, continuous research and regular surveillance—in addition to international scientific collaboration—are needed to inform health-care systems, social-care systems, and governments when developing treatment, rehabilitation, support algorithms, and policies and practices related to disability, education, occupation, and finance.337, 340, 363 The need for strong health systems An effective response to COVID-19 and any future pandemic requires not only a checklist of policy actions but also health systems that can deliver needed outcomes and can provide support to individuals to perform prosocial behaviours. The United States provides an example: although the government was able to develop and distribute vaccines, only 64% of the population was fully vaccinated as of Jan 31, 2022. This relatively low level of vaccine coverage compared with that of other countries is a consequence of widespread public distrust in the government and public health departments in the United States. Public health officials have little community presence, except in emergencies, and therefore little opportunity to build or rebuild trust. As in many other countries, the United States has invested little in community-based public health services, with the preponderance of health outlays directed at health care rather than public health. Within the health-care sector, the focus has been on secondary and tertiary health care rather than on universal access to primary health care. In 2019, US public health outlays of $104·5 billion accounted for a mere 2·8% of the total health outlays of $3·759 trillion.364, 365 Many LMICs—particularly those that have faced other epidemics, such as HIV, Ebola, and Zika—were able to integrate their responses to COVID-19 with well established community screening and contact-tracing capacity, and to rapidly deploy community health workers. 366 As members of the community they serve, community health workers are instrumental in improving population health,367, 368, 369 and are crucial in linking people to preventive and clinical services, dispelling misinformation, and fostering trust.370, 371 For example, in Bhutan, Brazil, and Kenya, community health workers had an essential role in pandemic responses, including encouraging the adoption of public health and social measures and vaccination uptake.372, 373, 374 However, despite their pivotal role, these workers continue to be undersupported and overworked.370, 375 Strong national health systems and universal health coverage have been priorities in the Sustainable Development Goal (SDG) era to prevent and treat communicable and non-communicable diseases and reduce catastrophic health expenditure. Countries with universal health coverage and health systems centred around primary health care, as called for by SDG Target 3.8, were able to scale up emergency services for patients with COVID-19 while continuing to ensure quality health care for health needs unrelated to the pandemic. There have been vast differences between the effects of COVID-19 on populations in countries with resilient national health systems and universal health coverage and in countries without such systems.376, 377, 378, 379, 380 In low-resource settings in particular, the COVID-19 pandemic has had a marked effect on immunisation programmes, maternal mortality, malaria control efforts, and prevention and treatment programmes for tuberculosis and HIV. Routine immunisation programmes for children have been disrupted and maternal deaths, stillbirths, ruptured ectopic pregnancies, and maternal depression have increased.381, 382 Estimates suggest that in 2020 there were 680 000 deaths from HIV, 383 and an excess of 1·4 million tuberculosis deaths, 14 million malaria cases, and 69 000 malaria deaths globally.384, 385 Approximately two-thirds of these additional malaria deaths were linked to disruptions in the provision of malaria prevention, diagnosis, and treatment, such as anti-malaria drug coverage, the distribution of insecticide-treated nets, and diagnostics. 385 High levels of income inequality breed a lack of social trust and solidarity,386, 387 which in turn erodes support for public services. In the absence of public services, inequalities persist or increase, creating a feedback loop of distrust and inequality that sociologist Bo Rothstein has called a “social trap”. 388 The lack of social solidarity is especially acute regarding marginalised groups in society, such as people experiencing homelessness, people who are incarcerated, people who use drugs, people experiencing domestic abuse, people from racial and ethnic minorities, and people with low incomes. These highly vulnerable groups should be prioritised by the public health system. Public health services also address another dimension of vulnerability that is largely overlooked by the curative health system: the high and increasing prevalence of chronic, non-communicable diseases. COVID-19 has imposed especially high morbidity and mortality burdens on people with comorbid conditions, such as obesity, diabetes, kidney disease, and heart disease.264, 389, 390, 391 These chronic, non-communicable diseases can be mitigated through public health and social measures, through preventive and rehabilitative efforts to promote healthy behaviours such as nutritious diets and more active lifestyles, and by addressing the environmental determinants of non-communicable diseases, including air pollution and the built environment. 392 Increasing evidence suggests that exposure to air pollution increases the risks associated with COVID-19 infection, including the mortality risk.393, 394, 395 The curative health industry pays little attention to the long-term health benefits of lifestyle and environmental changes to reduce risks. The COVID-19 pandemic has also shown that insufficient attention was previously paid to the design and management of ventilation and filtration systems for healthy indoor environments, including safe workplaces, safe schools, and safe public transport. SARS-CoV-2 transmission occurs primarily indoors because of the high concentration of aerosols in indoor spaces with poor ventilation or filtration.396, 397, 398, 399 Outdoors, there is much greater dilution and dispersion of airborne particles, which considerably reduces the concentration of aerosols in the near-field and far-field. Building-related interventions can reduce the spread of COVID-19. Inadequate ventilation and unfiltered, recirculated air have been associated with SARS-CoV-2 transmission in several studies that examined large outbreaks across various indoor environments, 400 showing that building-level strategies are key to reducing the spread of airborne infectious diseases.40, 127, 401, 402, 403 Examples of effective building-level control measures include increasing outdoor air ventilation in buildings through mechanical heating, ventilation, and air conditioning systems and opening windows, and upper-room ultraviolet germicidal irradiation in high-risk settings.402, 404, 405, 406 Existing ventilation standards are not effective against the transmission of respiratory diseases.407, 408 It is imperative, for example, that the American Society of Heating, Refrigerating and Air-Conditioning Engineers, the US CDC, and WHO develop ventilation targets for respiratory infectious diseases, including for non-health-care settings. Strategies to improve ventilation include upgrading MERV filters in heating, ventilation, and air conditioning systems to MERV 13 and using portable air cleaners with high-efficiency particulate air filters. Setting ventilation rates on the basis of reducing risks to health (as opposed to only minimising energy consumption) and requiring upgraded filtration could help to reduce the risk of disease transmission within buildings.409, 410 Despite the relevance of buildings to disease transmission, case cluster investigations routinely—and mistakenly—fail to evaluate building performance or collect data on ventilation and filtration strategies. Such information could enable an understanding of the combination of environmental factors that lead to the highest risks and the most effective control strategies. 411 Safe public transport Relatively little scientific evidence is available to indicate how public transport vehicles, such as buses and trains, might facilitate viral transmission. Additional research in this area is important to ensure that public transport can operate as safely as possible, particularly as an affordable option for essential workers in urban areas. Research suggests that the risk of SARS-CoV-2 transmission on public transport can be substantially reduced by asking people who are unwell to refrain from its use, requiring or encouraging vaccination among passengers, changing transport schedules to reduce crowding, and improving ventilation or filtration. Buses can substantially improve ventilation by opening as many windows, doors, or roof hatches as feasible; 412 deploying recirculating-air cleaning systems that use mechanical filtration or ultraviolet germicidal irradiation; and requiring passengers to wear face masks. Although air travel can exacerbate the spread of infectious disease, and spread within aeroplanes has been reported, the risk of transmission of respiratory viruses among passengers in flight is low.413, 414, 415, 416, 417, 418 This low risk is attributed to onboard systems that deliver a 50:50 mix of outdoor air and recirculated air that passes through high-efficiency particulate air filters in most aeroplanes. 419 In addition, the air is delivered and returned within each row, which renders ventilation highly effective. However, the ventilation systems are not always operating when aeroplanes are parked at the gate, and this represents a time of higher risk. A report published by the National Academies in 2013 highlighted this issue and recommended that gate-based ventilation be operational, especially during pandemics and periods of high risk. 419 Rapid vaccine development yet unequal vaccine distribution and uptake The bright spot of the pandemic has been the rapid deployment of scientific knowledge and evidence, most importantly in the development and introduction of effective vaccines. Putting aside the flawed and unequal distribution of vaccines, and the refusal to assist vaccine producers in LMICs to scale up and distribute vaccines that they themselves had developed, the rapid development of vaccines and scale-up of manufacturing capacities is an impressive example of private–public partnership and transnational scientific collaboration. The US Government has had a long-term and pivotal role in funding the scientific development of several vaccine platforms, most notably the mRNA vaccine technologies that underpin the Pfizer-BioNTech and Moderna COVID-19 vaccines. The US Government was also deeply involved in 2020 in funding the rapid development, clinical trials, and production of Moderna's COVID-19 mRNA vaccine, whereas BioNTech received €100 million in financing from the European Investment Bank 420 and a €375 million grant from the German government. 421 US Government funding has come through multiple channels, including the NIH, the Department of Defense, the US Agency for International Development, and the US Department of Health and Human Services. Moderna's first grant from the Department of Defense (through the Defense Advanced Research Projects Agency) was in 2013, 422 US Department of Health and Human Services funding dates to 2016, 423 and $2·5 billion of funding in 2020 came from the US Department of Health and Human Services and the NIH. The scale-up of vaccine manufacturing capacities was equally impressive. Despite setbacks early on, by September, 2022, more than 12·6 billion doses of a COVID-19 vaccine had been administered worldwide, with roughly 8·8 million doses administered every day. 424 Although demand for boosters is growing, the main constraints against universal vaccination in 2022 are likely to be related to logistics and vaccine hesitancy rather than supply. At least since the epidemic of severe acute respiratory syndrome, the US Government has been actively involved in researching betacoronaviruses and potential vaccines against the diseases they cause. A 2021 study estimates that the NIH spent $17·171 billion between 2000 and 2019 on vaccine platforms, of which an estimated $943 million was spent on mRNA vaccines and another $757 million was spent on vaccines targeting diseases caused by betacoronaviruses. 425 US Government funding for clinical trials of the Moderna vaccine totalled an additional $4·9 billion in 2020. 426 The NIH has long funded crucial bench research at the University of Pennsylvania that led to the use of a modified amino acid (pseudouridine) in the mRNA formulation of both the Pfizer-BioNTech and the Moderna vaccines. The great success of vaccine development led to massive profits for the producers of mRNA vaccines, despite their support from public funds. The market capitalisation of Moderna increased from around $8 billion at the end of 2019 to $65 billion on Jan 24, 2022. 427 Similarly, the market capitalisation of BioNTech increased from around $7 billion at the end of 2019 to $36 billion in January, 2022. 428 Despite having had a decisive role in funding the development of these vaccines, the US Government did not share in the market returns, and indeed purchased the vaccines from these companies on a commercial basis. 429 The US Government's long-term support of vaccine development research illustrates the crucial role of government funding for high-priority public goods such as vaccine technologies. Although private capital markets funded some research, funding and scientific support from the US Government was indispensable. Yet the US Government has not designed appropriate ways to manage the intellectual property that it co-funds and co-generates. Typically, intellectual property that has been funded by the US Government ends up entirely in non-governmental hands, whether at universities or companies. This discrepancy leads to an inequitable outcome (the privatisation of publicly funded wealth generation) and to an inefficient outcome (the underuse of intellectual property because of monopoly pricing by patent holders). The result is the near-complete privatisation of the intellectual property that results from government funding. Universities and companies then exercise monopoly rights to the technologies under patent, and charge prices for vaccine doses that are much greater than the actual costs of manufacturing. In theory, mark-ups compensate companies and universities for their investments in research and development; however, their outlays in this area were partly funded by the government in the first place. Ultimately, taxpayers pay monopoly prices for vaccines that tax dollars helped to fund. Some governments in poorer countries lacked the means altogether to pay monopoly prices, and had to wait in the back of the queue for donations or discounts. By contrast, the AstraZeneca vaccine was developed by a not-for-profit partnership between AstraZeneca and the University of Oxford, UK. 2 billion doses of the vaccine were supplied to more than 170 countries (250 million doses supplied at-cost through the COVAX initiative) between January and November, 2021. However, AstraZeneca announced in November, 2021 that it planned to begin stepping down the provision of doses on a not-for-profit basis. 430 As another example, the non-profit Texas Children's Hospital Center for Vaccine Development at Baylor College of Medicine, Texas, United States, worked with Biological E with no patents to produce and distribute a novel paediatric COVID-19 vaccine for emergency use in India, which was administered to more than 10 million adolescents in its first 10 days.431, 432 Alternative technologies, including recombinant protein COVID-19 vaccines, are not protected by patents and could have been made available early in the pandemic at very low cost or no cost to LMICs, alongside the patent-protected vaccines that use newer technologies. Some of the off-patent vaccines were produced at scale in LMICs by the Developing Country Vaccine Manufacturing Network (DCVMN). However, without early support from governments of high-income countries, complementary efforts to deploy vaccines based on more traditional, off-patent technologies lagged behind efforts by the major multinational pharmaceutical companies to bring newer vaccines to market. This neglect of existing technologies as a complementary strategy hampered the ability to make COVID-19 vaccines on a sufficient scale for the world. In addition, the national regulatory authorities of countries that host a DCVMN member are not designated as stringent, and therefore do not have the freedom to provide vaccines to COVAX or other LMICs. The failures to globalise the vaccine ecosystem to include production in a greater number of LMICs amplified global inequities in vaccine access. Even without government funding, the patent system for vaccines is problematic. Large multinational pharmaceutical companies that develop vaccines are permitted monopolies on the production of these life-saving technologies during the terms of the patent and for the duration of applicable regulatory exclusivity, thereby leaving low-income countries with no effective access to vaccines other than through mechanisms such as the Global Alliance for Vaccines and Immunizations and official development assistance. The public also faces very high prices, mainly as taxpayers through government-funded vaccine-purchasing programmes. A range of policy options incentivise research and development while also ensuring greater fairness in both the pricing of and access to vaccines. For example, governments could share in the profits in proportion to their investment share in research and development outlays. Governments could negotiate prices to balance the research and development incentives and fair access. Governments could force companies to license their intellectual property in return for reasonable fees, a process termed compulsory licensing. The US Government in fact proposed new World Trade Organization rules to facilitate compulsory licensing of the intellectual property of COVID-19 vaccines, 433 but was met with initial strong resistance from European governments. 434 The travails of COVAX The COVAX (COVID-19 Vaccines Global Access) initiative was formed to accelerate the development and manufacturing of COVID-19 vaccines and to provide fair and equitable access to these vaccines for all countries. The launch of this initiative in April, 2020 was a timely and meritorious achievement, well before COVID-19 vaccines were proven and available. However, in practice, COVAX failed to deliver on its targets and timelines because vaccine-producing companies made contracts directly with the governments that paid the highest prices, rather than with COVAX, which insisted on lower prices for low-income countries. Moreover, vaccine-producing countries, such as India, imposed export bans on vaccines rather than delivering them to COVAX as promised. As a result, COVAX was chronically short of vaccine supplies for LMICs, and was further unable to set dates and plans for vaccine delivery and use. 435 Low-income countries in Africa were persistently last in line. As of January, 2022, the percentage of the population that was fully vaccinated was 71% in the EU and 63% in the United States, but only 10% in Africa and even lower in many countries, such as 2% in Nigeria (figure 17 ). 181 In October, 2021, WHO called for every country to fully vaccinate at least 40% of their population by the end of the year; 437 however, this goal was missed, 438 almost without a word of remorse from high-income countries. Bottlenecks eased towards the end of 2021. By the end of the year, COVAX had delivered nearly 1 billion doses to 144 countries, and monthly deliveries were increasing and becoming more predictable, reaching 347 million doses in December, 2021. 439 The EU, China, the United States, the Russian Federation, and India were all stepping up production and exports of vaccines, but not in a manner that was adequately coordinated among the vaccine-producing countries. Figure 17 Percentage of the world's population that is fully vaccinated as of Jan 6, 2022 Reproduced, with permission, from the International Monetary Fund. 436 The COVAX Facility—a mechanism through which to procure and equitably distribute COVID-19 vaccines—was a worthy innovation in principle, but upstream science policy failures and downstream implementation failures prevented the widespread availability of effective and safe COVID-19 vaccines in low-income countries. Such failures included reserving financial incentives largely for multinational pharmaceutical companies to develop vaccines based on new technology—such as mRNA, adenovirus, or nanoparticle vaccines—without commensurate support for vaccines that used more traditional technology and could be made locally by vaccine producers in LMICs. Compounding this issue was a failure to make adjuvant technologies available to vaccine producers in LMICs. US Operation Warp Speed, the Coalition for Epidemic Preparedness Innovations, and other vaccine-funding initiatives operated on the premise that only multinational pharmaceutical companies had the ability to scale up and make effective vaccines, because of the mistaken belief that only new vaccine technologies could address the virus. The ability to grant emergency-use listing of COVID-19 vaccines was restricted to either WHO or one of the six stringent regulatory authorities in the United States, Canada, UK, EU, Australia, and Japan—once again excluding any national regulatory authority based in an LMIC. In addition, despite the COVAX framework, there were few instances in which the vaccine-producing countries (the United States, UK, China, India, the Russian Federation, and countries of the EU) worked operationally together to accelerate the equitable global uptake of vaccines. Research governance Despite many examples of research excellence during the pandemic, there has also been poor-quality research, subject to biases and with misleading results.440, 441 One such example is the controversy around the effectiveness of hydroxychloroquine, of which an initial study suggested effectiveness but several subsequent, well designed randomised controlled trials (RCTs) showed no benefit. Independent assessments of trial data and the potential for biases to influence findings provide safeguards against the reporting of misleading results. Registration of trials in a publicly accessible database is an essential step in assuring quality and reducing the risks of publication bias, whereby the results of negative studies are under-reported. The WHO International Clinical Trials Registry Platform provides an invaluable source of information on current trials evaluating a range of interventions. 442 COVID-NMA is an example of an international initiative, led by a team of researchers from Cochrane and other institutions, that works in conjunction with WHO to generate up-to-date mapping of evidence from trials regarding COVID-19 drug treatments. 443 COVID-NMA is one of many groups producing living evidence syntheses and improving future research by assessing the methodology and transparency of trials. Clinicians, policy makers, and people wanting to understand the best available evidence should consult robust sources such as this, and should not be swayed by the results of individual trials that have not been subject to such appraisal. 444 Analysis of COVID-19 trials shows substantial imbalances between the numbers of drug trials (2465), vaccine trials (109), and public health trials (16), with public health trials in particular being under-represented. 445 Many of the individual drug treatment trials for which results are available recruited small numbers of patients, such that the trials were underpowered and were not able to detect useful beneficial effects. Notable exceptions include the DisCoVeRy trial in patients who were hospitalised, 446 and the PRINCIPLE trial, which recruited participants recovering from COVID-19 at home. 447 Only three of the 16 trials of preventive interventions involved public health and social approaches: two RCTs on the effectiveness of wearing face masks and one RCT on physical distancing.448, 449 The failure to undertake large-scale RCTs of potentially useful interventions, particularly of social and behavioural interventions and of ventilation and filtration regimes, is a missed opportunity to document the effectiveness of affordable preventive measures. However, the study of behavioural interventions using observational data has also been helpful to control the pandemic. 450 Research funders should ensure an appropriate balance between the evaluation of treatments in hospital and in community settings and preventive (including behavioural and social) and therapeutic interventions. WHO has an important role in monitoring the funding, recruitment, and analysis of intervention trials and in reporting progress regularly. The scarcity of support for the development of new drugs, vaccines, and diagnostics by and for biotechnology companies and large-scale producers based in LMICs shows a marked research imbalance. Existing funds prioritised incentive financing for large and multinational pharmaceutical companies, with the expectation that discoveries would somehow quickly reach Africa, southeast Asia, and Latin America. This did not happen, and glaring vaccine inequities emerged as a result. A new financing mechanism is therefore needed for research and development in LMICs. Pandemic economics The COVID-19 pandemic led to a deep global recession in February–April, 2020. In March, 2020, uncertainty was so high that it led to a dash-for-cash and general malfunctioning in financial markets. Central banks and many governments acted promptly to ensure orderly conditions in financial markets and to protect vulnerable households and companies. Governments imposed severe mobility restrictions on all continents, resulting in what was termed the Great Lockdown. Global output in 2020 contracted by 3·1%, with losses particularly severe among advanced economies, which implemented the strictest lockdowns. 451 The United States experienced a most unusual recession: it was by far the deepest but also the shortest on record. The National Bureau of Economic Research Dating Committee asserts that the contraction was limited to March and April, 2020. 452 By Q3 of 2020, global output was already beginning to recover as people, companies, and governments learned more about the virus, uncertainty declined, and mobility restrictions were eased in many places. Like the initial contraction, the recovery was led by advanced economies. The recovery in the United States followed the same pattern, but was even more pronounced. In the table , we compare the actual annualised growth in GDP between 2019 and 2022 with the IMF's pre-pandemic projections from the October 2019 World Economic Outlook.453, 454 In most countries, and in all WHO regions, actual growth was lower than had been predicted in the months before the pandemic. In a few cases, such as Taiwan, Province of China, actual growth (4·3%) was higher than had been predicted (2·0%). Note that actual growth here refers to the growth for 2019–22 as of the World Economic Outlook April, 2022. GDP refers to GDP measured at constant international prices. Table Comparison between the International Monetary Fund's pre-pandemic projection for annualised growth from 2019 to 2022 and actual growth in gross domestic product from 2019 to 2022 World Economic Outlook October 2019 projected growth for 2019–22 World Economic Outlook April 2022 actual growth for 2019–22 Actual minus expected growth rate 2019–22 Estimated cumulative deaths per million population (as of May 31, 2022) WHO region African region 3·5% 1·9% −1·5% 1575·6 Region of the Americas 2·0% 1·5% −0·5% 3995·5 Region of the Americas (excluding United States) 2·3% 0·8% −1·5% 4078·9 Eastern Mediterranean region 2·9% 2·7% −0·2% 2803·4 European region 1·8% 0·3% −1·5% 4289·5 South-East Asia region 6·3% 2·6% −3·7% 2545·6 Western Pacific region 4·6% 3·5% −1·1% 300·6 Western Pacific region (excluding China) 2·4% 1·2% −1·2% 1064·7 World 3·5% 2·0% −1·5% 2228·6 Country Brazil 2·3% 0·5% −1·8% 3648·0 China 5·8% 4·9% −0·9% 15·4 France 1·3% 0·4% −0·9% 2659·4 Germany 1·3% 0·1% −1·3% 2749·6 Italy 0·7% −0·3% −0·9% 4891·5 Mexico 1·8% −0·6% −2·4% 5723·5 New Zealand 2·6% 2·0% −0·6% 203·9 Republic of Korea 2·6% 1·9% −0·7% 500·7 Russian Federation 2·0% −2·3% −4·3% 5360·0 Taiwan, Province of China 2·0% 4·3% 2·2% 165·0 United Kingdom 1·5% 0·4% −1·1% 2910·4 United States 1·8% 1·9% 0·1% 3823·5 Data from the World Economic Outlook October 2019, 453 the World Economic Outlook April 2022, 454 and the Institute of Health Metrics and Evaluation. 146 Actual growth signifies the International Monetary Fund's estimates as of April, 2022. Gross domestic product is measured in international dollars at constant prices. Notably, although all WHO regions experienced slower growth than had been predicted in October, 2019, the growth shortfalls in the Western Pacific region, the Eastern Mediterranean region, and the region of the Americas were lower than in other regions. In general, economies that followed a suppression approach—China, New Zealand, Taiwan, Province of China, and Republic of Korea—had smaller growth shortfalls than the rest of the world. The United States experienced no growth slowdown (at least based on the World Economic Outlook as of April, 2022), but also had a high level of cumulative deaths per million population. The recessions induced by COVID-19 in many countries in 2020 were accompanied by substantial employment losses (figure 18 ). These losses were particularly large in emerging markets and differed in terms of the age and education level of the workers affected. Workers who had received tertiary education were generally the least affected by employment losses. Although employment levels recovered in many advanced economies in 2021 (except among workers with lower levels of education), emerging markets had more persistent employment losses as of March, 2022. 455 Figure 18 Change in employment-to-population ratio in G20 countries Mean percentage point change in employment-to-population ratio between latest data (Q3 2021; except Argentina, Q2 2021) and Q3 2019. Prime: age 25–54 years; youth: age 15–24 years. For all other categories, ages 15–64 years are considered. Basic: primary and lower-secondary education; intermediate: upper-secondary and post-secondary, non-tertiary education; advanced: above post-secondary, non-tertiary education. Data are aggregates for G20 advanced economies (Canada, France, Italy, Republic of Korea, Spain, and United States) and emerging market economies (Argentina, Brazil, Mexico, and South Africa). Reproduced, with permission, from the International Monetary Fund. 455 The changes in GDP measure the direct and indirect effects of the pandemic on output. However, beyond the losses of output, GDP does not measure the massive losses in societal wellbeing that are associated with illness and deaths as a result of COVID-19. To account for societal losses from excess deaths and disease, economists often estimate the monetary value of mortality and morbidity by assigning an economic value to a death or a year of illness. In the United States, for example, US federal agencies assign a value of a statistical life (VSL) to identify the societal benefits of measures to save lives or the costs of failures to save lives. Although there is no single authoritative estimate of VSL, US agencies typically place it between $5 million and $10 million per life lost. One study from early in the pandemic put the VSL in the United States at $4·5 million per life lost in 2020, considering age-specific mortality rates. 456 Because the US GDP per capita was $63 500 in 2020, the VSL was approximately 70 times the GDP per capita. Using such an estimate, the economic value of the loss of life is far larger than the loss of market output. As of the end of May, 2022, there had been approximately 3900 deaths attributed to COVID-19 per million people in the United States, or 0·39% of the population. Using a VSL equal to 70 times the GDP per capita, we see that the economic loss due to morality is therefore 70 × 0·39% of GDP, or 27% of annual GDP—a staggering sum, and far larger than the loss in market output. The global mortality at the end of May, 2022 was approximately 0·2% of the world's population. Using a global VSL equal to 70 times the world GDP per capita, the economic loss is 14% of annual world output. These are very rough illustrations, yet they make the point that the societal losses of life, put in monetary terms, have been enormous, and the societal benefits of reduced mortality rates are similarly enormous. Note that these very simple calculations also do not include any estimates of the costs of morbidity, including those of long COVID, which are also substantial. The long-term economic scarring and legacies of the pandemic are as yet unknown. As of mid-2022, the United States had returned to the pre-pandemic level of economic activity, but was also experiencing inflation caused by the combination of pandemic-related expansion in demand with adverse supply shocks, the war in Ukraine, and the sanctions regime against the Russian Federation. Monetary policy tightening is ongoing. The tightening cycle is shared by many other advanced and emerging market economies. China's economic recovery is facing difficulties, resulting from continued sporadic lockdowns, supply chain disruptions, and financial stresses, including in the real estate sector. COVID-19 will lead to sustained and sizeable output losses in low-income and emerging economies, whereas advanced economies are expected to return closer to their pre-pandemic paths over the next 3 years. 451 Uncertainty remains elevated. Countries that are heavily dependent on tourism and travel suffered extreme losses of national output, and their economies have not yet recovered. Island countries in the Caribbean are a case in point, 457 with countries in this region experiencing some of the largest GDP declines in the world in 2020: Aruba, −22%; Saint Lucia, −20%; Antigua and Barbuda, −20%; and Barbados, −18%, compared with declines of less than 5% in most high-income countries. Caribbean island economies did not bounce back in 2021, as travel and tourism remained depressed. Several oil-exporting countries also experienced a steep decline in GDP in 2020, but that decline was followed by a V-shaped recovery in 2021 as the global demand for petroleum recovered. High-income countries were able to draw from domestic and international capital markets to finance increased government outlays at historically low interest rates in response to the pandemic, whereas low-income countries generally lacked access to international capital markets on comparably favourable terms. According to IMF data, the advanced economies increased their budget deficits relative to GDP by an average of 6·8 percentage points during 2020–21 compared with 2019. LMICs were much more constrained in their borrowing. Countries of sub-Saharan Africa, with the least access to capital markets, increased their deficit-to-GDP ratio by a mere 1·7 percentage points, with most of the increase driven by lower revenue. In essence, high-income countries used deficit financing to stabilise incomes and protect businesses from bankruptcy, whereas low-income countries had to reduce their expenditure. In 2021, high-income countries bounced back to growth with far more vigour than did low-income countries, and long-term prospects suggest a divergent recovery, with advanced economies returning to their pre-pandemic output paths while emerging markets and low-income countries are likely to face a permanent negative output shock. 458 For the lowest-income countries, financial support should come in the form of grants and highly concessional long-term loans. It is clear that economic prospects are intimately tied to pandemic developments. Suppression of the virus has the best outcomes for economics, health, and civil liberties. 459 As hospitalisations and mortality increase, economic activity declines in response to both greater restrictions on economic activity imposed by governments trying to regain control of the situation, and labour shortages as people leave the workforce because of illness, school closures, and voluntary physical distancing. As we have noted throughout this Commission report, the key to escaping both the pandemic and the economic fallout is widespread vaccination and basic public health precautions (a so-called vaccination-plus strategy), combined with the development and distribution of increasingly effective treatments for COVID-19 to lessen its health effects once contracted. The faster the world can act to vaccinate its population, the better the prospects for exiting the pandemic and achieving long-lasting economic recovery. In the words of the IMF Managing Director Kristalina Georgieva: “Vaccine policy is economic policy.” 460 Global finance and the pandemic Central banks around the world reacted swiftly to unprecedented financial disturbances in March, 2020. Through the aggressive easing of monetary policy, extension of various credit facilities, and liquidity swap lines, a vicious circle of negative macrofinancial feedback loops was stopped. Initially, emerging markets experienced large capital outflow, but these patterns reversed quickly as orderly market conditions were restored. The World Bank approved a $14 billion fast-track facility for health emergency response on March 2, 2020, and this facility supported more than 100 countries with credits and grants within the subsequent 3-month period through the global health multiphase programmatic approach. These resources helped countries to source public health and social measure commodities, mobilise health workers, and expand care capacity through procurement of equipment and goods such as oxygen, as part of a $150 billion package announced in April, 2020 for health, social, and economic response. The World Bank also approved $12 billion for financing vaccine procurement and delivery in early October, 2020, even though these resources were limited by their requirement to be allocated on an individual-country basis. The IMF provided two kinds of liquidity assistance for LMICs. The first was emergency credit lines and other IMF financing facilities in 2020–21, provided to around 90 countries and totalling roughly $170 billion. The second was the issuance of special drawing rights in 2021, amounting to $650 billion. However, this important infusion of liquidity was made in proportion to the existing IMF quotas, meaning that most of the special drawing rights allocation went to high-income countries, and only $30 billion to countries in Africa. Nonetheless, this liquidity infusion was still substantial relative to the GDP of many low-income countries—up to 11% of recipient country GDP in some cases. In 2022, the IMF established a new Resilience and Sustainability Trust—partly financed by high-income countries lending some of their special drawing rights to the Trust—which in turn will provide long-term financing to small and low-income countries to help them to tackle structural challenges of global relevance, such as pandemic preparedness and climate change, and thereby strengthen their resilience to shocks. G20 countries also offered some relief via the Debt Service Suspension Initiative, which provided almost $13 billion in relief to 48 eligible countries before it expired at the end of 2021. 461 Some lower-income countries decided not to participate in the programme given its potential effects on their future access to credit. Unfortunately, the G20 did not do as much as it could have done to facilitate a coordinated response to COVID-19. As a powerful complement to the UN, comprising 19 national governments and the EU, the G20 represents 63% of the world's population and 87% of its economic output. The G20 is well poised to adopt a plan to support lower-income countries through foreign direct investment, trade agreements, official development assistance, and loans. Such financing could support achievement of the SDGs and the Paris Climate Agreement, including addressing many of the health, education, environmental, and infrastructure challenges that we have described and that are exacerbated by the COVID-19 pandemic. G20 countries account for 90% of research expenditure, researchers, publications, and patents. 80% of countries worldwide invest less than 1% of GDP in research and development. 462 Special attention should be paid to scaling up investment in research and development, especially through partnerships between governments, universities, and companies in G20 countries, and their counterparts in LMICs. There is a great need for long-term commitments to building capacity in LMICs, ensuring equity and maintenance of a worldwide network of collaboration. Such a plan could facilitate cooperation among scientists, sectors, and institutions to develop and produce tests, vaccines, therapeutics, and other goods, and therefore build capacity in the long term, not only for producing drugs and medical equipment but also to build a 21st-century knowledge economy. However, going forward, this initiative could be leveraged to create innovative, implementable, and coordinated programmes to support global research and development and to scale up investment in the production capacities of LMICs. Global health financing Health financing nowadays is primarily covered by national (domestic) resources, which includes government and private health expenditure in addition to out-of-pocket spending, with only very modest levels of international support by donors for LMICs.463, 464 Total official development assistance for health, as measured by the OECD Development Assistance Committee, came to a mere $23 billion as of 2019, or less than 0·05% of the GDP of donor governments. 465 This assistance amounted to roughly 1·5% of the health financing of LMICs, a very modest sum. 463 The pandemic did not alter this picture in any decisive way. Even these modest amounts of official development assistance have proven decisive in scaling up disease control in many areas. The Global Fund and the Global Alliance for Vaccines and Immunizations have each saved tens of millions of lives with modest outlays of a few billion dollars per year.466, 467, 468 These programmes prove that international financing for health can result in practical solutions, despite being highly underfunded considering their missions. The creation of the Access to COVID-19 Tools (ACT) Accelerator and COVAX in April, 2020 aimed to accomplish the same results for COVID-19, but both fell far short. 469 Raising even a few billion dollars of international support for ACT Accelerator and COVAX proved extremely difficult, despite the urgency of the pandemic. The modest funding was insufficient to obtain crucial PPE supplies, testing supplies, and especially vaccines. Bilateral deals between countries and manufacturers for the purchase of these commodities were prioritised at the expense of COVAX. Many producing countries also implemented temporary export bans on essential products (eg, PPE, pharmaceuticals, and hand sanitiser), with wider effects on access for countries in need. The multilateral development banks similarly provided too little health financing, and too slowly, to make up the pandemic financing gap. 470 The current proliferation of special health funds needs to be rationalised within a single global fund that is closely linked with WHO. The spread of small vertical health funds has become unmanageable, with many key functions slipping through the cracks. In addition to the Global Fund and the Global Alliance for Vaccines and Immunizations, the World Bank Group hosts the Global Financing Facility in support of Every Woman Every Child and the Pandemic Emergency Financing Facility for pandemic response. WHO has a Contingency Fund for Emergencies. The Pandemic Emergency Financing Facility and the Contingency Fund for Emergencies were established during the 2014–15 Ebola outbreak in west Africa. The COVAX Facility is part of the ACT Accelerator, which is co-managed by WHO, the Global Alliance for Vaccines and Immunizations, UNICEF, and a few partners. These multiple funds weaken the international response capacity in health emergencies. Long-term economic and sustainable development recovery from the pandemic The extended period of economic disruption is creating long-term economic and social scarring, including long-lasting difficulties for children whose schooling has been repeatedly disrupted; new debt crises among low-income countries that are facing sustained output and revenue losses amid the tightening of global financial conditions, and in some cases facing years of reduced numbers of tourists; 471 long-term disabilities arising from long COVID and from the mental distress of the pandemic;271, 472, 473 long-term gender inequalities, as women have been on the front line of the pandemic and are disproportionately affected by poverty, with many being forced to reduce their participation in the workforce; 474 and long-term social and political instability arising from increased income inequality and business disruptions as a result of COVID-19 that will take years to overcome. The highly expansionary macroeconomic policies of 2020–21, in particular the enormous increases in the money supply in both the United States and the eurozone, have resulted in high inflation in 2022, with the longer-term consequences still to be seen. That the first 2 years of the pandemic were followed by the war in Ukraine and a stringent sanctions regime against the Russian Federation has further exacerbated the supply shocks due to COVID-19, and has given rise to the fear of an extended period of stagflation, in which the combination of monetary expansions and powerful supply shocks could lead to a combination of economic downturns and high inflation, an event not seen in the world economy since the 1970s. We must also address the ongoing climate and ecosystem crises that threaten major global destabilisation. Earth is now 1·2°C hotter than the pre-industrial temperature, and the 8 years since 2014 have been the eight warmest years since records began in 1870. The next El Niño event, perhaps occurring as early as 2023, is likely to cause a large upward shift in worldwide temperatures, and the world could very soon exceed the 1·5°C threshold that was adopted as a guardrail by the Paris Climate Agreement. 475 The COVID-19 pandemic threatens to divert global attention, financing, and scarce political capital away from the urgent agenda of sustainable development, including the SDGs and the Paris Climate Agreement. We must ensure that the focus on pandemic preparedness and response enhances rather than detracts from the broader sustainable development agenda. Europe's success in launching the European Green Deal while simultaneously confronting the COVID-19 pandemic shows the political feasibility of fighting the pandemic while also intensifying the policy commitments for sustainable development. The Next Generation EU fund commits €750 billion towards recovery and resilience with a focus on a green and digital transformation. In using the funds, EU member states must devote a minimum of 37% of the funding to climate change and a minimum of 20% to the digital transition. The challenges of sustainable development in LMICs are far more acute. Even if the COVID-19 pandemic recedes soon, economic scarring will be long-lasting. Many lower-income countries lack the financial means to promote long-term recovery and to finance sustainable development and commitments under the Paris Climate Agreement. The IMF has established that what are classified as low-income developing countries—a group of 57 low-income and lower-middle-income developing countries—face a financing gap of $300 billion–$500 billion per year to achieve the SDGs, and this gap has increased as a result of the pandemic.476, 477 These funds could be mobilised by a combination of means: increased official development assistance, increased lending by the public development banks (both multilateral development banks and national development banks), increased market borrowing through sustainability-themed bonds (eg, green, social, SDG, or sustainability-linked bonds) and other means, and global coordinated taxation, for example on ultrahigh net worth, greenhouse gas emissions, and financial transactions. There is an opportunity to integrate health, environmental sustainability, and equitable economic recovery into post-COVID stimulus packages, but the potential for doing so has not been realised. Many such packages are likely to result in numerous adverse environmental impacts, particularly because of increases in fossil fuel combustion, 478 which not only accelerates climate change but also results in higher levels of air pollution. Long-term exposure to air pollution is thought to increase the risk of adverse outcomes of COVID-19 infection.395, 479 Despite widespread empirical evidence of the opportunities offered by green recovery strategies, current practice shows that most industrialised countries have only partly implemented sufficiently ambitious green stimulus packages.480, 481, 482, 483, 484, 485 A 2022 analysis shows that the pandemic economic recovery packages of the G20 countries, totalling $14 trillion, allocated only 6% of total stimulus spending to areas that reduced emissions, and 3% was allocated to activities that will increase emissions. 486 Moreover, governments have focused their efforts on measures that could bring immediate benefits to employment and economic growth, such as investments in green energy and transportation, and renovations to increase the energy efficiency of buildings. Although very important, such interventions address only a part of the sustainability challenges faced by high-income countries: an analysis of recovery packages of European countries found that issues such as the agrifood system, circular economy, and nature-based solutions have received little attention so far, although many of these countries have relatively poor performance on such challenges. 487 As the SDGs provide a holistic and inclusive framework that can be universally applied, their integration into public policies is crucial for an effective green transition. For this purpose, the appropriateness of investments and reforms can be analysed with the support of artificial intelligence methods such as machine learning, 488 enabling a rapid assessment of national plans and corrective actions when needed. Section 3: recommendations for ending the COVID-19 pandemic, preparing for the next, and long-term sustainable development All ongoing risks highlight the continued urgency of global cooperation to end the COVID-19 pandemic and to prepare for future pandemics. Although we have largely failed in terms of global cooperation during the first 2 years of this pandemic, putting such cooperation into place is still urgent. We remain far from preparedness for future pandemics, and logic tells us that achieving safety will require the reinforcement of basic tools of pandemic control on a truly global basis: universal vaccine coverage, physical distancing, the use of face masks as appropriate, prudential controls on potential superspreader events, safe workplaces, surveillance for new variants, global protocols for safe international travel, and the scale-up of test-trace-and-isolate regimens to be put in place when community transmission is low to ensure that it is kept low. Global and national strategies to end the COVID-19 pandemic Governments must be vigilant about new variants of SARS-CoV-2 (especially for unvaccinated populations) and waning protection from vaccinations and previous infections. Strong monitoring and coordinated surveillance systems around the world need to be established to assess the risks of new waves of COVID-19. WHO, working with the main vaccine-producing companies and countries, needs to intensify its efforts to ensure high levels of immunisation coverage in all countries, especially in the low-income countries where vaccine coverage remains dangerously low. Countries must then implement a vaccination-plus control strategy that includes mass immunisation; the availability and affordability of testing; treatment for new infections (test and treat); rehabilitation and social support for people with long COVID; and complementary public health and social measures such as the use of face masks, the promotion of safe workplaces, and economic and social support for self-isolation. These vaccination-plus policies should be implemented with the goal of protecting populations on a sustainable basis, rather than implementing reactionary policies that either require the use of face masks and proof of vaccination or relax these requirements on the basis of the volatile reproduction rate of the virus at a given time. In calling for this vaccination-plus strategy, we strengthen the calls of many other commissions for the implementation of evidence-based public health and social measures to control transmission and protect the most vulnerable populations.96, 489, 490, 491 China, the United States, the EU, India, the Russian Federation, and other major regional and global powers must put aside their geopolitical rivalries to work together to end this pandemic and to prepare for the next one and for other global crises. Military conflicts and social turmoil can increase cases of COVID-19 and outbreaks of other infectious diseases and can affect surveillance systems, therefore amplifying humanitarian crises. In this global coordination and cooperation, UN institutions should be pre-eminent. The G20 should have a special role as the point of coordination among the world's major economies and financial centres, given the central role of global finance in pandemic preparedness, response, and recovery. Intensified investigation into origins As a Commission, we strongly support the call for an objective, open, data-driven, transparent, and independent scientific debate about the origin of SARS-CoV-2.44, 45 WHO, governments, and the scientific community should intensify the search for the origin of SARS-CoV-2, including a possible natural spillover or a possible research-related spillover. In the absence of an unbiased, independent, and rigorous search for a natural origin by a multidisciplinary team of experts alongside an unbiased, independent, and rigorous investigation of the research-related hypotheses, the public's trust in science will be imperilled, with potentially grave long-term repercussions. It is therefore crucial to investigate all hypotheses fully, not only to ascertain the source of the pandemic and to protect against future emerging infectious diseases, but also to ensure the integrity of science itself. The perceived lack of transparency to date by leading scientific agencies and laboratories is troubling and needs to be addressed. Maintain WHO as the lead institution for response to emerging infectious diseases WHO should remain at the centre of the multilateral response to emerging infectious diseases. Yet WHO needs strengthening. Such strengthening should include new regulatory authority, more backing by national political leaders, more contact with the global scientific community, and a larger core budget to carry out its many crucial responsibilities. The capacity of WHO should also be complemented by far greater international finance to support the health systems, research and development, and biomedical production capacity of LMICs. Establish a global pandemic agreement and strengthen the IHR The weaknesses and shortfalls of the IHR (2005) in protecting the world against the COVID-19 pandemic have led to the decision by WHO member states to start the process to draft and negotiate new agreements on pandemic preparedness and response, including a possible new Pandemic Treaty. We echo many other reports on COVID-19 in calling for a global pandemic agreement and for a reassessment of and update to the IHR (2005).96, 489, 490, 492 We recommend the following as some of the core constituents of the new pandemic arrangements. First, bolstered WHO authority, with high-level political support from heads of government in a new Global Health Board. Second, the right of WHO to investigate in situ any events that could constitute a new global public health emergency. Third, the creation of a global surveillance and monitoring system for infectious disease outbreaks and transmission, building on existing initiatives 493 —both for disease forecasting and for making informed, data-driven decisions about workplace restrictions, resource allocations, and effective interventions to avoid untargeted, one-size-fits-all restrictions. Fourth, the approval by WHO, the International Civil Aviation Organization, and the International Maritime Organization of standing regulations regarding the processing and control of international travellers and international freight and shipping under global pandemic conditions. Finally, the publication of an annual WHO Report on Global Pandemic Preparedness and Response, which should include findings from the Global Preparedness Monitoring Board for review and adoption by the WHA. The WHA should then establish global health policies to address the gaps and weaknesses identified in the report by drawing on the successes of national and regional health strategies and policies, and adapting them to fit various contexts and country needs. Reform of WHO governance The effectiveness of WHO depends on backing by its member states and, most importantly, the political support of the major powers. We echo some of the recommendations made by other groups to bolster the role and financing of WHO in preventing and responding to future pandemics and strengthening public health systems globally. 492 We strongly urge the creation of a new WHO Global Health Board to support WHO in its decision making and actions, especially on urgent and controversial matters. We support similar calls by other panels for a Global Health Threats Council or Global Health Threats Board.96, 490, 492, 494 This WHO Global Health Board should be composed of heads of government representing each of the six WHO regions, and elected by the member states of those regions. Efforts should be made to ensure inclusive decision making, including gender parity. Additionally, a substantial increase to the core budget of WHO is required to increase its effectiveness at its headquarters in Geneva, at regional offices, and in countries around the world. Moreover, WHO needs the capacity to draw upon large-scale emergency financing in the case of a global public health emergency. WHO should continue to use the infrastructure it has built up so that this infrastructure is in place when another pandemic emerges. An emergency credit line at an international financial institution, designed as a key tool of the new Global Health Fund, could provide the necessary emergency financing channel. Regulations for the prevention of pandemics from natural spillovers and research-related activities and for investigating their origins There is a need for the primary prevention of pandemics through the implementation of preventive measures against both natural spillovers and spillovers from research-related activities. The prevention of natural spillovers would require a One Health approach—an integrated, unifying approach that aims to sustainably balance and optimise the health of people, animals and ecosystems 495 —including the strengthening of veterinary services, the regulation of trade in domestic and wild animals and of wildlife and livestock rearing, the prevention of deforestation, and the enhancement of pathogen-surveillance systems in domestic animals and in humans. 496 Such measures are likely to be highly cost effective and would complement investments in the strengthening of national health systems, the development of vaccines and therapeutics, and other outbreak response strategies described in this report. Strategies to prevent research-related releases should include stronger international and national oversight of biosafety, biosecurity, and biorisk management, including the strict regulation of gain of function research of concern. It is certain that future pandemics will arise from interactions between humans and animals, and that research on viruses will continue with the potential for accidents. It is therefore imperative that society enacts measures to reduce the possibility of both natural spillovers and spillovers from research-related activities, and acts rapidly if these scenarios do occur. Countries must adopt rigorous surveillance over the trade of domestic and wild animals and over research. Calls for One Health approaches to address the risks of the emergence and transmission of zoonotic diseases are common among reports on pandemic prevention and preparedness,489, 490, 492, 494 and many reports have called for greater global surveillance and monitoring of disease risks, with WHO as the coordinating power. The WHA should also adopt new global regulations on biosafety to regulate pathogen-related fieldwork and laboratory work. We must be aware that the collection of viruses and other pathogens from nature can give rise to research-related spillovers. Such activities have been largely unregulated in the past. Moreover, we urgently need global oversight regarding potentially hazardous laboratory experimentation, including gain of function research of concern. The new global regulations should include inspection by international teams of facilities that are involved in the manipulation of dangerous pathogens. Moreover, scientists need to develop much safer means of research on the pathogenicity and infectivity of viruses than the gain of function research of concern methods that have spread with little oversight during the past 20 years. Even today, there is little understanding and clarity about the research on SARS-like viruses that was underway just before the COVID-19 pandemic. Additionally, because disease-causing pathogens are stored and studied in laboratories across the world, countries should maintain peace and address conflict with diplomacy rather than warfare, as armed conflict and the destruction of research institutions risk the escape of pathogens and subsequent outbreaks of disease. When investigating the origins of any novel pathogen, potential hypotheses should not be prematurely rejected to ensure that time-sensitive data—such as early case information and laboratory records—are collected. Countries should encourage the examination of the originally submitted papers on the origins of emerging infectious diseases, the availability of raw scientific data, and the systematic review of grants and documents on planned research. G20 support for finance, research and development, and the production capacities of LMICs The G20 should plan and implement a 10-year effort, with accompanying financing, to ensure that all WHO regions—including the world's lower-income regions—have the capacity to research, develop, produce, and distribute all of the essential tools for pandemic control, including tests, diagnostics, vaccines, therapeutics, PPE, and human resources for health, among others. 494 Specifically, this initiative should include capacity building and innovations for vaccine producers in LMICs and should extend stringent regulatory authority capacity to these countries. Strengthen national health systems and increase investments in primary and public health Governments should allocate an increased share of national income to their health systems, and in the case of low-income and lower-middle-income countries, development aid should complement domestic financing to ensure that all countries—including those with the lowest income—have strong public health systems and health-care systems that are centred around primary health care and can achieve universal health coverage, as called for by SDG 3. Along with increased investments in health systems and medical supply chains, there is also a need to increase support for research and development in the behavioural, social, and implementation sciences to ensure that public and primary health interventions and systems effectively serve their populations within their respective sociocultural and economic environments. The strengthening of health systems should address inequalities for health-care workers and communities in terms of gender, ethnicity and race, income, and accessibility. All reports that have been written on COVID-19 make strong recommendations for national health system strengthening for emergency preparedness and for the maintenance of primary and mental health care during emergencies.96, 490, 491, 492, 493, 494 Alongside the strengthening of health systems and the provision of universal health coverage, more investment is needed to ensure a standing public health capability that can operate effective surveillance systems in each country, adequately trained and resourced outbreak investigation, and response capacity and communications expertise. Quality health education must also become widely accessible. There should also be increased financing for programmes that address other major global health concerns, including HIV/AIDS, tuberculosis, malaria, childhood immunisations, maternal mortality, and neglected tropical diseases. Existing programmes should be enhanced with resiliency and long-term sustainability, and new programmes should be designed with resiliency and sustainability in mind. Health workers, including community health workers, are essential to the function of resilient health systems, and these workers must be well trained, well paid, well supported, and provided with appropriate PPE. Sustainable investments in education and training programmes for community health workers, nurses, mid-level practitioners, and physicians in LMICs are necessary for health professionals in all countries to participate in patient-centred and population-centred health systems that are locally responsive and globally connected. Communities, civil society organisations, and local faith-based groups must be engaged in health-system strengthening, and these investments will not only improve pandemic responses but also improve the health of the populations during non-pandemic times. National pandemic preparedness plans The overarching lesson of the COVID-19 pandemic is the need for national preparedness along with global cooperation and concerted action. Most countries lack meaningful pandemic preparedness plans. The countries that made such plans, such as countries of the Western Pacific region after the outbreak of severe acute respiratory syndrome, have fared best in the COVID-19 pandemic. The enormous vulnerabilities that COVID-19 has exposed in high-income and low-income countries alike shows that plans should be comprehensive, cross-sectoral, and supported by global coordination. We support other commissions, including the Independent Panel for Pandemic Preparedness and Response, 96 in recommending that each WHO member state should adopt, by law, national pandemic preparedness plans that meet international standards laid out in the Pandemic Treaty, and in supporting the WHO IHR regulations and the review of these regulations. Essential components of national pandemic preparedness plans are listed in the panel . Panel Essential components of national pandemic preparedness plans • Improved primary health care and universal access to health care for all as a precondition for effective health systems and public health • Scale-up of community-based public health systems, integrated with primary health care, for surveillance, testing, tracing, monitoring, public education, and social support, and investment in a skilled workforce • Investment in public health and scientific literacy, including through educational programmes in schools and universities, to protect the public against misinformation and disinformation • Inclusive, diverse leadership at all levels of pandemic response by ensuring gender parity and including health workers, civil society, human rights experts, gender experts, and the expertise of social, political, and behavioural scientists • Investment in supporting scientists to present evidence more effectively to decision makers, and in supporting decision makers in their understanding of how to interpret evidence and implement science-based recommendations and policies • Investment in behavioural and social sciences research to develop and implement more effective interventions and policies to change behaviours, and therefore reduce the likelihood of pandemics and viral transmission and increase resilience • Protection of vulnerable groups, including older people, women, children, disadvantaged communities, refugees, Indigenous Peoples, people with disabilities, and people with comorbid medical conditions • Establishment of safe schools, workplaces, and public spaces, including investments in ventilation and filtration systems as tools to fight disease and promote long-term resilience, health, and wellbeing • Permission for schools to implement stringent and evidence-supported control strategies early in an outbreak situation, to prioritise in-person learning • Universal access to digital services, so that digital applications (such as government alerts, transfer payments, emergency services, and contract tracing) reach the entire population • Social support services that address gender-based violence, mental health disorders, loneliness, indigency, and other social needs • National and global supply chains for personal protective equipment, testing, diagnostics, vaccines, medicines, and other essential commodities, and management of medical consumable waste • Global coordination and standardisation of data collection, data quality, monitoring, and reporting, to ensure the disaggregation of data by relevant factors (eg, gender, age, race, and ethnicity) • Application of a gender-responsive approach to pandemics—including in terms of testing, treatment, and prevention—and equal pay for women workers, with special attention given to unpaid and underpaid workers in the health sector, especially community health workers • Universal adoption of updated medical, scientific, and public health training and airborne infectious disease exposure standards that reflect the paradigm shift towards more accurate definitions of transmission routes for respiratory infectious disease, and the most effective mitigation strategies for each one, including enhanced building ventilation and air filtration The Global Health Fund We call for the creation of an integrated and flexible Global Health Fund with three core financing windows: commodities for disease control, pandemic preparedness and response, and primary health system scale-up in LMICs. This Fund should have representation and leadership from LMICs. Funding for the first window would merge the efforts of the existing Global Fund and the Global Alliance for Vaccines and Immunizations. The Global Health Fund would require annual disbursements of the order of $60 billion per year (around 0·1% of the GDP of the high-income countries, which is estimated by the IMF to be about $60 trillion in 2022). 497 This recommended annual funding of $60 billion would be allocated roughly as follows: commodities, $20 billion per year; pandemic preparedness, $15 billion per year; and support for primary health systems, $25 billion per year. Although many reports on COVID-19 call for collective financing mechanisms for sustainable, predictable, flexible, and scalable financing for pandemic preparedness with the support of WHO, the G20, and the international financing institutions, here we are calling for increased support not only for pandemic preparedness but also for health systems and disease control more generally. We also highlight the design and political advantages of consolidating the target funding into a single Global Health Fund, which will ensure consistency, coordination with WHO, and a holistic vision of health that places the health system at the core. The Global Health Fund should be closely aligned with the work of WHO. We propose that the Fund has its headquarters in Geneva, but has strong regional offices in each of the six WHO regions. As such, the Fund's overall funding would be centralised but programme design and implementation would be decentralised, fostering strong ownership by the countries of each region and reflecting regional needs and priorities, rather than operating under top-down control from Geneva or from a few donor countries. The Global Health Fund should be supplemented by an emergency financing mechanism to enable a surge of funding in the face of a global health emergency. Such an emergency mechanism could rely on the borrowing authority of an international financial institution, such as the World Bank, to provide the ability to mobilise at least $10 billion immediately, and possibly far more, in the event of another major global public health emergency. Sustainable development and a green recovery We call on the G20 to conduct an urgent review of SDG financing options and to adopt a package of financing policies to expand SDG financing for LMICs in line with needs. As emphasised by the UN Secretary-General in his remarks to the General Assembly in January, 2022: “We must rescue the 2030 Agenda.” 498 The COVID-19 pandemic is a setback for sustainable development globally, but building forward is a necessity. Bolstering SDG financing—especially for key physical and digital infrastructure, services, and social protection in low-income countries and LMICs—and the adoption and implementation of ambitious sustainable development policies can help to prevent future global shocks, including pandemics, cybersecurity events, or climate events, and to promote resilience. Many options are available to scale up SDG financing: donor countries could scale up official development assistance, expand the financing of the multilateral development banks, and promote large flows from capital markets to sovereign borrowers. SDG-based lending should be offered at lower interest rates and longer maturities. Global coordination on tax enforcement and on new taxes (eg, a globally coordinated tax on ultrahigh net worth) could supplement current tax revenues. Other financing channels could include the issuance of SDG bonds by national, provincial, or municipal governments; the launch of an SDG Fund based on large-scale private philanthropy; and global coordination and management of levies on CO2 emissions. Conclusion: protecting and promoting multilateralism The COVID-19 crisis has exposed major weaknesses in the UN-based multilateral system, resulting from excessive nationalism, tensions among the major powers, chronic underfinancing of global public goods including the UN system itself, lack of flexibility of intellectual property regimes to ensure that global public goods are available to all, lack of adequate sustainable development financing for LMICs, and the erosion of political support for multilateral solutions by the major powers. Despite major efforts to stimulate recovery and a just transition to sustainable development, the lack of ambition in the global response to COVID-19 is like that of other pressing global challenges, such as the climate emergency; the loss of global biodiversity; the pollution of air, land, and water; the persistence of extreme poverty in the midst of plenty; and the large-scale displacement of people as a result of conflicts, poverty, and environmental stress. In this light, our most basic recommendation is the strengthening of multilateralism in all crucial dimensions: political, cultural, institutional, and financial. We call for all countries, especially the richest and most powerful, to support, sustain, and bolster the work of the UN system. We call for awareness of the benefits of multilateralism, solidarity, cooperation, and the shared commitment to sustainable development, whether facing pandemics, ending poverty, keeping the peace, or meeting global environmental challenges. We strongly support the call of the UN Secretary-General for a new Common Agenda, 499 and urge member states to engage in its implementation constructively and swiftly. We encourage member states to enrich their deliberations and decisions with the voices of civil society, the private sector, local governments, parliaments, academia, and young people, among others. We note the timeliness of recommitting to the Universal Declaration of Human Rights, the UN's moral charter, as we celebrate its 75th anniversary in 2023. Declaration of interests SSAK declares grants paid to the Centre for the AIDS Programme of Research in South Africa from the National Institutes of Health (National Institute of Allergy and Infectious Diseases), the South African Medical Research Council, the National Research Foundation, and the European and Developing Countries Clinical Trials Partnership. He is a vice-president of the International Science Council; a commissioner of the African Commission on COVID-19; and a member of the WHO Science Council, the Advisory Council of the Physicians for Human Rights, the Global Medical Advisory Board on COVID-19 for Sanofi Pasteur, and the Scientific Advisory Committee of the Bill & Melinda Gates Foundation. CBa declares consulting fees from and is a member of the advisory board of the Sustainable Development Solutions Network. MEB and PJH are co-directors of a team of scientists at Texas Children's Hospital Center for Vaccine Development, who are co-inventors of a COVID-19 recombinant protein vaccine technology owned by Baylor College of Medicine (BCM) that was licensed by BCM, non-exclusively and with no patent restrictions, to several companies committed to advancing vaccines for low-income and middle-income countries. The co-inventors have no involvement in license negotiations conducted by BCM. Similar to other research universities, a long-standing BCM policy provides its faculty and staff, who make discoveries that result in a commercial license, a share of any royalty income. To date, BCM has not distributed any royalty income to the co-inventors of the COVID-19 recombinant protein vaccine technology. Any such distribution will be undertaken in accordance with BCM policy. CBu declares grants from the Health Research Council of New Zealand, the New Zealand Ministry of Health, and the New Zealand Ministry of Foreign Affairs and Trade; consulting fees from Johnson & Johnson Japan; payment from University of Malaya; and is the president of the Society for Research on Nicotine and Tobacco—Oceania. GCB is co-chair of the UHC2030 International Alliance. AH declares grants and consulting fees from the Wellcome Trust and the Oak Foundation. NJ declares support from the UN Sustainable Development Solutions Network and consulting fees from the University of California, San Francisco Department of Humanities and Social Sciences and the Foundation for Global Governance and Sustainability. ET declares consulting fees from the United Nations Development Programme. VV-F and IS are co-chairs of the Nizami Ganjavi International Centre. MAP declares grants from the Rockefeller Foundation to his institution and is a member of the Board of Directors for Management Sciences for Health, and non-executive Governing Board member of Sidra Medicine. JT declares funds from the Victorian Government to BehaviourWorks Australia for surveys and research on COVID-19 related behaviours, from Monash University and from Melbourne Water. AG was a member of the Open Society Global Drug Policy Program until February, 2022. MH declares funds from the Victorian Government, the New South Wales Government, the Australian Government, and the Macquarie Foundation to the Burnet Institute, Melbourne, VIC, Australia for research and modelling work related to COVID-19, and grants from AbbVie and Gilead Sciences that are unrelated to this work. JVL declares grants from AbbVie, Gilead Sciences, MSD, and Roche Diagnostics, and consulting fees from NovVax, all unrelated to this work. All other authors declare no conflicts of interest. This report does not represent the institutions nor the organisations for which the authors work.

Summary Background An adequate amount of prepaid resources for health is important to ensure access to health services and for the pursuit of universal health coverage. Previous studies on global health financing have described the relationship between economic development and health financing. In this study, we further explore global health financing trends and examine how the sources of funds used, types of services purchased, and development assistance for health disbursed change with economic development. We also identify countries that deviate from the trends. Methods We estimated national health spending by type of care and by source, including development assistance for health, based on a diverse set of data including programme reports, budget data, national estimates, and 964 National Health Accounts. These data represent health spending for 184 countries from 1995 through 2014. We converted these data into a common inflation-adjusted and purchasing power-adjusted currency, and used non-linear regression methods to model the relationship between health financing, time, and economic development. Findings Between 1995 and 2014, economic development was positively associated with total health spending and a shift away from a reliance on development assistance and out-of-pocket (OOP) towards government spending. The largest absolute increase in spending was in high-income countries, which increased to purchasing power-adjusted $5221 per capita based on an annual growth rate of 3·0%. The largest health spending growth rates were in upper-middle-income (5·9) and lower-middle-income groups (5·0), which both increased spending at more than 5% per year, and spent $914 and $267 per capita in 2014, respectively. Spending in low-income countries grew nearly as fast, at 4·6%, and health spending increased from $51 to $120 per capita. In 2014, 59·2% of all health spending was financed by the government, although in low-income and lower-middle-income countries, 29·1% and 58·0% of spending was OOP spending and 35·7% and 3·0% of spending was development assistance. Recent growth in development assistance for health has been tepid; between 2010 and 2016, it grew annually at 1·8%, and reached US$37·6 billion in 2016. Nonetheless, there is a great deal of variation revolving around these averages. 29 countries spend at least 50% more than expected per capita, based on their level of economic development alone, whereas 11 countries spend less than 50% their expected amount. Interpretation Health spending remains disparate, with low-income and lower-middle-income countries increasing spending in absolute terms the least, and relying heavily on OOP spending and development assistance. Moreover, tremendous variation shows that neither time nor economic development guarantee adequate prepaid health resources, which are vital for the pursuit of universal health coverage. Funding The Bill & Melinda Gates Foundation.