PT  - JOURNAL ARTICLE
AU  - J. Paul Callan
AU  - Carlijn J.A. Nouwen
AU  - Axel S. Lexmond
AU  - Othmane Fourtassi
TI  - Categorizing the Status of COVID-19 Outbreaks Around the World
AID  - 10.1101/2021.03.08.21252586
DP  - 2021 Jan 01
TA  - medRxiv
PG  - 2021.03.08.21252586
4099  - http://medrxiv.org/content/early/2021/03/23/2021.03.08.21252586.short
4100  - http://medrxiv.org/content/early/2021/03/23/2021.03.08.21252586.full
AB  - Although the SARS-CoV-19 virus spread rapidly around in world in early 2020, disease epidemics in different places evolved differently as the year progressed – and the state of the COVID-19 pandemic now varies significantly across different countries and territories. We have created a taxonomy of possible categories of disease dynamics, and used the evolution of reported COVID-19 cases relative to changes in disease control measures, together with total reported cases and deaths, to allocate most countries and territories among the possible categories. As of 31 January 2021, we find that the disease was (1) kept out or suppressed quickly through quarantines and testing & tracing in 39 countries with 29 million people, (2) suppressed on one or more occasions through control measures in 74 countries with 2.49 billion people, (3) spread slowly but not suppressed, with cases still increasing or just past a peak, in 31 countries with 1.45 billion people, (4) spread through the population, but slowed a result of control measures, leading to a “flattened curve” and fewer infections than if the epidemic were unmitigated, in 32 countries with 2.24 billion people, and (5) spread through the population with some but limited mitigation in 5 countries with 168 million people. In addition, several countries have experienced increases in cases after disease appeared to have finished spreading due to declining numbers of susceptible people. For some of these countries – for example Kenya, Pakistan and Afghanistan – the resurgences can be explained by the relaxation of control measures (and may have been enhanced by disease spread in population segments that experienced lower infection levels during the first waves). For other countries, the resurgences point to the effects of new virus variants with higher transmissibility or immunity resistance – including most countries in Southern Africa (where the B.1.351 variant has been identified) and several countries in West Africa (potentially due to the B.1.1.7 or other variants). These findings are consistent with mounting evidence of high infection rates in several low- and middle-income countries, both from seroprevalence studies and estimates of actual deaths from COVID-19 combined with estimates of expected mortality rates. We estimate that 1.3–3.0 billion people, or 17–39% of the global population, have been infected by SARS-CoV-2 to date, and that at least 4.5 million people have died from COVID-19 – much higher than reported cases and deaths. Disease control policies and vaccination strategies should be designed based on the state of the COVID-19 epidemic in the population – and consequently may need to be different in different countries.Key Points✥ The state of the COVID-19 pandemic varies significantly in different countries and territories around the world – and policies for disease control and vaccination will need to be tailored accordingly.✥ In any epidemic, there are several possibilities for how the disease will spread over time – and our analysis finds that, in fact, as of 31 January 2021, there were many countries and territories in each of the main categories of COVID-19 epidemic dynamics that might have been expected:Kept out or suppressed quickly through quarantines and testing & tracing – in 39 countries with 29 million people (0.4% of the global population), mostly small island states and a few countries in Southeast Asia. [Category H in the following map and table]Suppressed through control measures (social distancing, hygiene and testing & tracing) –in 74 countries with 2.49 billion people (31.9% of global population), mostly in Europe, East Asia and the Pacific. [Categories F and G]Spread slowly but not suppressed, with cases still increasing or just past a peak – in 31 countries with 1.45 billion people (18.6% of global population), including many countries in Latin America, Eastern Europe and the Middle East, as well as the United States and Russia. [Categories D and E]Spread through the population, but slowed as a result of control measures, leading to a “flattened curve” and fewer infections than if the epidemic were unmitigated – in 32 countries with 2.24 billion people (28.8% of global population), mostly in South and Southeast Asia (including India) and Africa. [Category B]Spread through the population with some but limited mitigation or “flattening the curve” –in 5 countries with 168 million people (2.2% of global population). [Category A]Experienced increases in cases after disease appeared to have finished spreading, which in some countries might have been solely due to relaxation of control measures (especially in wealthier population segments which experienced low infection levels during the first wave) – for example in Kenya and in Pakistan and some Central Asian countries – but which in some countries is likely to be due to new virus variants with higher transmissibility or immunity resistance – for example in most countries in Southern Africa and several in West Africa, and possibly also in parts of South and Central America. [Category J and many countries in Category K] ✥ These findings are backed up by mounting evidence of high infection rates in several low- and middle-income countries. Seroprevalence studies in Kenya, Nigeria, Pakistan and South Africa have reported finding antibodies for SARS-CoV-2 in large percentages of the studied populations – and suggest that current infection levels are likely above 50% in each country. Studies of actual deaths due to COVID-19, combined with estimates of expected mortality rates, similarly suggest that SARS-CoV-2 has, by now, infected more than half of the populations in Bolivia, Ecuador, Peru, Mexico, South Africa, Sudan, Syria, Yemen and Zambia.✥ Countries of all income levels, and from all regions, appear in each of the main disease dynamics categories; however, there are clear income and geographical patterns in states of COVID-19 epidemics around the world. Most high-income countries have controlled the spread of SARS-CoV-2 through measures. Middle-income countries are spread across all categories, and account for 45 of the 63 countries which have slowed the disease significantly but not fully suppressed it. Some low-income have experienced largely unmitigated susceptibility-driven dynamics, while others have “flattened the curve” to varying degrees.✥ We estimate that between one and two out of every five people globally has been infected by SARS-CoV-2 to date, and that at least 4.5 million people have died from COVID-19. Our estimate of total infections – 1.3–3.0 billion people, or 17–39% of the global population – is between 13 and 30 times the number of confirmed cases, and twice to four times as much as previous estimates of total infection numbers. We estimate that 4.6–10.0 million people have died from COVID-19, between 2.1 and 4.5 times the number of deaths attributed to COVID-19.✥ An estimated 8.9-12.5 million lives remained at risk from COVID-19 as of the end of January 2021, prior to vaccination efforts – mainly in high-income countries (2.4–2.9 million), China (2.1 million) and India (1.7–2.9 million). Vaccinations, of course, have already started to reduce these numbers substantially.✥ Our analytical approach is simple but useful – providing insight into the epidemic status even in many low-income countries with limited disease monitoring, and with potential to provide early warnings of significant new variants. We compare the evolution of reported cases with changes in stringency of disease control measures, and check that infection levels are plausible given total reported cases and deaths and the income level of the country. Anomalies in which changes in the evolution of reported cases cannot be explained by changes in the stringency index provide indications of possible significant variants of the virus. Up to the end of January, the data provide indications of the presence of significant new variants in:Most countries in Southern Africa (where the B.1.351 variant, with higher transmissibility and some resistance to immunity, was first identified in South Africa)Several countries in West Africa (likely with higher transmissibility and resistance to immunity, possibly the B.1.1.7 variant, which was first identified in the UK and has been found in Ghana and Nigeria, or possibly a different variant).They also suggest, with less certainty, that the disease dynamics may be affected by new variants in several countries in South and Central America (perhaps the P.1 variant descended from the B.1.1.28 variant which was first identified as coming from the Brazilian Amazon).✥ Different countries should adopt different disease control policies, according to the state of the COVID-19 epidemic in the population.For countries that have kept the disease out or suppressed outbreaks through control measures, their measures need to be kept in place – and potentially strengthened especially in the face of higher-transmissibility variants – until vaccines have been widely administered.For countries in which the disease is spreading slowly, full control measures should be maintained at least until new case numbers fully decline from the peak; later, it may be possible to relax some measures, but if measures are relaxed too soon or too much after cases peak, then significant further outbreaks can be expected (as has already happened in several such countries).For countries in which cases have declined following a flattened curve, there may be room to relax control measures that have the greatest negative health, economic and social consequences – but the most effective control measures will need to be maintained (even when cases remain low for extended periods), and measures may need to be strengthened to tackle variants which higher transmissibility or ability to evade immune responses.For countries in which the disease spread was largely unmitigated, many control measures could be relaxed for most people – although there may be risks if sizeable population segments have much lower infection levels than the general population or from variants with a high degree of immunity resistance.✥ These findings may have implications for the optimal distribution of early batches of vaccines within countries.For countries that have kept the disease out or suppressed outbreaks through control measures, vaccinations should be given first to frontline healthcare and essential workers and to elderly and vulnerable groups (starting with the oldest and most vulnerable).For countries in which the disease is spreading slowly, detailed modelling should be done to determine whether the optimal strategy is to vaccinate at-risk groups first or to vaccinate key transmitters to halt the outbreak and “crush the curve” while waiting for further vaccine supplies to arrive. For any country choosing the key transmitter strategy – as Indonesia has done and has been suggested for the United States – it will be essential to maintain control measures, and to keep higher transmissibility variants out, or otherwise the benefits of a key transmitter vaccination strategy could be lost.For countries in which the cases declined following a flattened curve, vaccinations should probably be given first to elderly and vulnerable groups, but the optimal strategy may switch to vaccinating key transmitters if there are resurgences in cases due to higher-transmissibility or immunity-resistant variants.For countries in which the disease spread was largely unmitigated, vaccination should concentrate on elderly and vulnerable people, because population-level immunity already exists, and the greatest danger lies in vulnerable people becoming infected due to endemic SARS-CoV-2 from variants that will likely circulate over many years.✥ These findings may also have implications for the optimal distribution of the first vaccines across countries. For most countries, the optimal allocation of vaccines doses is likely still to be according to population size – as current recommendations suggest. However, the global optimal allocation strategy might include providing somewhat greater supplies, during the next few months, to countries where using the vaccine to halt spread of the disease might be possible (provided that disease control measures are maintained in those countries).✥ Vaccination strategies will need to account for current and potential future virus variants as well as the likelihood that immunity from vaccination will wane over time. Higher-transmissibility variants of the SARS-CoV-2 virus increase the urgency of distributing vaccines in countries which have controlled the disease to date, and may alter the optimal strategy for countries deciding between vaccination first of elderly and vulnerable people or of key transmitters. Immunity-resistant variants of the virus may reduce the effectiveness of current vaccines, but are not likely to negate fully the protection they offer. Immunity acquired through vaccination is likely to wane over time – like immunity acquired through infection. In many, perhaps most, countries, the time to vaccinate the whole population will exceed the timeframe in which immunity from vaccination wanes or new immunity-resistant variants emerge. Looking to the longer term, therefore, new virus variants and waning immunity are likely to necessitate re-vaccination (with vaccines tailored to the latest variants) on a regular basis – and the optimal long-term strategies for ongoing vaccination will vary widely across countries and will depend on many factors.Summary The state of the COVID-19 pandemic varies significantly in different countries and territories around the world – and policies for disease control and vaccination will need to be tailored accordingly. Although the SARS-CoV-19 virus spread rapidly around in world in early 2020, the state of disease epidemics in different countries diverged rapidly as the year progressed. Many high-income countries have had second or third waves; other countries have seen cases continue to increase gradually; still others have experienced declines in cases to low levels after peaks in mid-2020. Facing different situations, different countries might need to adopt different policies in the coming months, including different disease control measures and vaccination strategies.Each country needs to know its COVID-19 status. There are several possible courses that a disease epidemic can take in a population. The disease can spread rapidly until its runs out of people remaining to infect; the disease can be slowed with control measures but still spread until large numbers of people are infected and immune; the disease can be suppressed or “crushed” by control measures; or the disease can be kept out completely. More complex disease dynamics will occur when a virus mutates, if new variants evade immune responses in people already infected or spread faster than before, or if the disease spreads differently in different segments of a population. Our research suggests that different countries have experienced outbreaks in each of the main possible categories:Susceptibility-Driven Dynamics with Limited Mitigation [Category A] – in which the disease spread until it infect most people and declined due to low susceptibility levels (i.e., low share of the population still able to be infected).Susceptibility-Driven Dynamics Mitigated by Measures [Categories B and C] – in which the disease curve was “flattened” by control measures, but the disease still spread and declined after infecting large numbers of people (but fewer than if there were no mitigation).Susceptibility-plus-Measures-Driven Dynamics [Categories D and E] – in which the disease was slowed significantly but not suppressed, i.e., the curve was “flattened” but not “crushed”, and the disease is still spreading in the population.Measures-Driven Dynamics [Categories F and G] – in which the disease has been constrained to date mainly through control measures (social distancing, hygiene and testing & tracing), but, of course, could spread again if measures are relaxed because only a minority of the population has been infected.Index-Case-Control Dynamics [Category H] – in which the disease has been kept out or suppressed to date through strict control measures (especially quarantines and testing & tracing).Complex Disease Dynamics due to Differences Across Population Segments and/or New Variants [Categories I and J, and many countries in Category K] – in which the disease experienced an apparently susceptibility-driven curve but with low overall infection levels (i.e., share of population infected) because some segments of the population have not had many infections, or in which the disease later shows an unexpected resurgence, due to spreading within the previously less-affected population segments or due to the emergence of immunity-evading strains of the virus.From reported data on COVID-19 cases and disease control measures, we can categorize, for most countries and territories, the dynamics of the disease to date. First, we compare the timing of increases and/or decreases in reported new cases with the timing of changes in the “Stringency Index” of control measures compiled by the Oxford COVID-19 Government Response Tracker – and select the appropriate disease dynamics category. Second, we check if the infection levels expected for the category or categories indicated in the first step, are consistent with predicted ranges from the total numbers of reported cases and deaths using plausible ranges for the case detection rate, death detection rate and infection fatality ratio (IFR) given the country’s income level. This approach yields definitive categories for most countries and territories. COVID-19 disease dynamics are complicated in many countries, due to changes in control measures, seasonal patterns, geographical differences within countries, variability in case testing over time and emergence of new variants; such effects can be seen in the reported cases and deaths for many countries, but they do not obscure the basic drivers of disease dynamics – in other words, which of the categories applies – for most countries.The results suggest that there is a wide variation in the state of the COVID-19 epidemic around the world – as of 31 January 2021 – as illustrated in the map and the table below. COVID-19 has been suppressed through control measures – Categories F, G and H – in 113 countries and territories with 2.52 billion people or about 32.3% of the global population. However, the rest of the world are in different situations. A total of 31 countries, with populations of 1.45 billion people (18.6% of global population), fall into Categories D and E, meaning that the disease spread has been slowed but not suppressed and cases are currently still increasing or just past their peak. In the 23 countries of Category B, home to 2.05 billion people (26.3%), the disease was slowed but not suppressed, and cases have declined fully from the peak. In a further 9 countries with 0.19 billion people (2.5%), the disease spread through the population after initial waves were suppressed. COVID-19 outbreaks in 5 countries with 0.17 billion people (2.2%) were only somewhat mitigated by control measures and the virus has likely infected most of the population, falling into Category A.For several countries, which have apparent anomalies and fall outside the “basic” categories, the methodology provides important insights into epidemic status – pointing to situations where significant differences may exist across population segments or providing early warnings of new variants with higher transmissibility or resistance to immunity. For 5 Arabian Peninsula countries (3 of which are in Category I) and Singapore, it is likely that the virus has spread widely among migrant worker communities but has been controlled in the rest of the population. Categories J and K include 28 countries in which reported cases have surged after first waves which were likely or possibly susceptibility-driven, with curves flattened to various extents as a result of control measures which mitigated the epidemics. For some countries – including (1) Kenya, (2) Pakistan, Afghanistan, Kyrgyzstan and Kazakhstan, and (3) Egypt and Sudan – the second peaks are likely due to relaxation of measures but larger than might be expected due to disproportionate effects of the second waves on population segments (likely more affluent groups) which had lower infection levels during the first waves. For other countries – including (1) most countries in Southern Africa and (2) many countries in West Africa – the data suggest the presence of new virus variants with higher transmissibility and possible resistance to immunity, because resurgences or accelerations in cases happened in several neighbouring countries around the same time, and often without changes in control measures, and the second surges in cases usually involved faster increases than the first waves. The B.1.351 variant, with higher transmissibility and some resistance to immunity, was first identified in South Africa and is known to have caused most cases in the country’s second wave; the B.1.1.7 variant, which has higher transmissibility, has been found in Ghana and Nigeria. Several countries in South and Central America have experienced second waves or surges in cases: Suriname’s might be due to a higher-transmissibility variant (perhaps the P.1 variant that was first identified as coming from the Brazilian Amazon); Bolivia’s was large but could be explained by a significant decline in control measures; increases in Brazil and several other countries across South and Central America might simply be due to relaxation of social distancing behaviours over the Christmas and New Year holiday season although a role for virus variants cannot be discounted.Countries of all income levels appear in each of the main disease dynamics categories; however there are clear correlations between income groups and COVID-19 status categories. Most high-income countries have controlled the spread of SARS-CoV-2 through measures (and thus fall in Categories F, G and H). Middle-income countries are spread across all categories, and account for 45 of the 63 countries which have slowed the disease significantly but not fully suppressed it (Categories B, C, D and E). Some low-income countries have experienced largely unmitigated susceptibility-driven dynamics (Category A), while others have “flattened the curve” to varying degrees (Categories B, C, D and E). A mix of low- and middle-income countries are among the 34 countries in Categories J and K.Clear geographical patterns have emerged in the states of COVID-19 epidemics. There was more diversity in the state of the epidemic within regions earlier in the pandemic, but regional patterns had become clear by the end of January 2021.In the Americas, the disease has spread slowly but has not been suppressed (Categories D and E) in most countries, including those with the largest populations, while many (but not all) of the Caribbean islands have kept SARS-CoV-2 out or under control (Category H).Western and Northern European countries have, for the most part, controlled the disease through social distancing and hygiene measures, through two or three waves, and fall in Categories F and G.Across Eastern Europe, the Levant, the Caucuses and Iran, all countries have constrained growth of the disease significantly, but infection levels in most have grown to moderate levels: different countries in these regions are included in Categories C, D/E and F/G, although their infection levels may all be in the moderate range.In South and Central Asia, the virus has spread widely in most countries and cases have declined. In India, Bangladesh, Nepal and Uzbekistan, the case curve was flattened considerably, and current infection levels are likely moderate (Category B). In Pakistan, Afghanistan, Kyrgyzstan, and Kazakhstan (all in Category J), there have been two peaks in cases. Bhutan has contained the outbreaks of the virus to date (Category F).Many countries in East and South-East Asia have largely kept the disease under control or kept it out (Categories F, G and H). However, Malaysia, Mongolia and Myanmar experienced widespread outbreaks in the second half of 2020, the Philippines appears to be past the peak of its epidemic (Category B), and Indonesia has had a continuous but very slow rise in cases since the start of the pandemic (Category E).In Australia, New Zealand and most Pacific Island States, SARS-CoV-2 has been excluded through quarantines, together with testing and tracing and lockdowns when the virus has spread beyond quarantined individuals (Categories F and H).African countries appear to have differed greatly in how the disease has spread. Many countries appear to have experienced widespread epidemics followed by declines in case numbers, with varying degrees of “curve flattening” due to control measures (Categories A and B). In some countries – Tunisia, Libya, Togo, Botswana and Mozambique – cases spread very slowly (Categories D and E). A few countries appear to have kept the disease out, and a few others appear to have experienced full outbreaks after having previously kept the virus largely out. As described earlier, most countries in Southern Africa and many in West Africa experienced rapid growth in case numbers in December and January (putting many in Categories J and K) – suggestive of the presence of one or more new variants with higher transmissibility and possible resistance to immunity.We estimate that 1.3–3.0 billion people have been infected by SARS-CoV-2 to date, or about 17–39% of the global population. This estimate is between 13 and 30 times the number of confirmed cases, and perhaps twice to four times as much as previous estimates of total infection numbers. We estimate that 4.6–10.0 million people have died from COVID-19, between 2.1 and 4.5 times the number of deaths attributed to COVID-19.An estimated 8.9-12.5 million lives remain at risk from COVID-19, which can be saved through appropriate disease control measures and effective deployment of vaccines. Of these estimated extra deaths, if 90% of the population were to contract SARS-CoV-2, high-income countries account for about 2.4–2.9 million, China for about 2.1 million, and India for about 1.7– 2.9 million. Vaccinations, of course, have already started to reduce these numbers substantially.The findings of this report are backed up by mounting evidence of high infection rates in several low- and middle-income countries. Immunity testing provides direct evidence of the current state of the COVID-19 epidemic. Serological studies in several cities and regions in Brazil, India, Kenya, Pakistan, Qatar and South Africa have already reported finding antibodies for SARS-CoV-2 in large percentages of the studied populations. Note, however, that serological testing will underestimate the number of people who have been infected, due to waning of SARS-CoV-2 antibodies which affects significant numbers of people at about 4-6 months after infection. Consequently, serological testing might understate the actual degree of immunity in a population, because some people may have antibodies at levels below the detection threshold of the serology tests or may have memory B cell or T cell responses, either or both of which will likely reduce the severity of their illness if reinfected, and may reduce their vulnerability to reinfection and their likelihood to pass on the virus to other people if reinfected. In some places, reliable estimates of actual deaths due to COVID-19 may be a substitute for immunity testing to determine the share of population infected to date, at least approximately. Estimates, using a variety of methodologies, in Bolivia, Ecuador, Mexico, Peru, South Africa, Sudan, Syria, Yemen and Zambia all indicate that moderate to high shares of their populations have already been infected.Different countries should adopt different disease control policies, according to the state of the COVID-19 epidemic in the population. The following recommendations for countries in different categories take into account their current infection levels and the potential for additional infections if measures are relaxed or if new variants become common in a country.➢ Category A: Control measures should be relaxed for most people; such relaxation is not likely to lead to many more cases and deaths. In some low- and middle-income countries, wealthier population segments may have implemented greater degrees of social distancing during the epidemic to date, and have much lower infection levels than in the overall population; these segments should maintain social distancing, until vaccines arrive, because otherwise they could experience substantial outbreaks (which may have generated “second waves” in some countries). If and when new virus strains with higher transmissibility and/or resistance to immunity arrive, control measures should be strengthened again to avoid new outbreaks, if the new variants cause high mortality levels and if it seems likely that control measures will be more effective at controlling the new outbreaks than they were during the initial outbreaks.➢ Categories B and C: Control measures currently in place that have the greatest negative health, economic and social consequences could be relaxed. However, many control measures, especially the most effective in limiting virus spread, will need to be maintained, even though current case numbers are low; otherwise, significant resurgences can take place (as has happened, for instance, in Kenya and Bolivia). Population segments that may have maintained lower infection levels during the outbreak to date will need to maintain social distancing. If and when new virus strains with higher transmissibility and/or resistance to immunity arrive, control measures will likely have to be strengthened again to avoid new outbreaks.➢ Categories D and E: Control measures should be maintained at least until new case numbers fully decline from the peak; if measures are relaxed too soon after cases peak, then significant further outbreaks can be expected (as has happened, for instance, in Brazil, Colombia and Paraguay). Once cases fully decline from the peak – through further infections or as a result of vaccination programmes – then, and only then, some of the disease control measures with the greatest negative health, economic and social consequences could be relaxed. For some countries in Categories D and E, it may be possible to push R0_e below 1 and hence “crush the curve” by introducing some additional control measures or improving compliance with existing measures. New virus strains, especially with higher transmissibility, can generate resurgences or accelerations in growth of cases (as seen, for example, in Mozambique and Togo).➢ Categories F and G: COVID-19 control measures, put in place by governments and implemented by citizens, have saved perhaps 13.1–14.2 million lives. To continue to protect these lives, control measures need to be maintained until vaccines become widely available – and strengthened, if necessary, to compensate for new virus variants with higher transmissibility.➢ Category H: Measures to keep the disease out – mainly strict quarantines for new arrivals and testing & tracing of suspected cases – should be maintained until vaccines become widely available.The findings of this report may have implications for the optimal distribution of early batches of vaccines within countries. Current policies in several countries call for deployment of vaccines first to healthcare workers and then by age cohort, starting with the oldest. These plans are aligned with the results of modelling (by Imperial College London and others) which suggest that, when the supply of vaccines is limited, the optimal strategy is to target the elderly and other high-risk groups. However, the models indicate that, if the supply is sufficient to stop transmission of the virus, the optimal strategy switches to targeting key transmitters (e.g., working age people and potentially children) to indirectly protect the elderly and vulnerable. Consequently, the optimal strategy may vary according to the disease status category for each country:➢ Category A: Vaccination should concentrate on elderly and vulnerable people, starting with the oldest and most vulnerable. There is no alternative strategy to consider because population-level immunity already exists, and the greatest danger lies in vulnerable people becoming infected due to endemic SARS-CoV-2.➢ Categories B and C: Vaccinations should probably be given first to elderly and vulnerable groups, and to frontline healthcare and other essential workers. However, if there are resurgences in cases across the population due to higher-transmissibility or immunity-resistant variants, then the optimal strategy may switch to targeting key transmitters, similar to some countries in Categories D and E.➢ Categories D and E: In some of these countries, the optimal strategy may to be vaccinate key transmitters – while maintaining current disease control measures – because it may be possible to halt the outbreak and “crush the curve”, while waiting for further vaccine supplies to arrive (after which disease control measures could be released). This strategy is being pursued by Indonesia and was suggested for the United States of America in a recent paper. However, careful modelling and planning would be necessary, for any country considering such an approach, to determine if a key transmitter strategy would in fact be optimal and if it would be feasible to implement. Further, for such a strategy to work, it will be necessary to keep control measures in place and to keep high-transmissibility variants of the virus out, until enough people have been vaccinated.➢ Categories F, G and H: Vaccinations should be given first to frontline healthcare and other essential workers and to elderly and vulnerable groups (starting with the oldest and most vulnerable).These findings may also have implications for the optimal distribution of the first vaccines across countries. Modelling by the Imperial College London COVID-19 Response Team suggests that the optimal allocation of vaccine doses among countries “is sensitive to many assumptions and will vary depending both on the vaccine characteristics and the stage of the epidemic in each country at vaccine introduction,” and concluded that, “[g]iven this uncertainty, allocating vaccine doses according to population size appears to be the next most efficient approach.” Our findings reinforce the uncertainty strongly: it is very likely that that stage of the epidemic varies greatly across countries. For most countries, the optimal allocation of vaccines doses is likely still to be according to population size – and then for those countries to give doses first to elderly and vulnerable people. However, the global optimal allocation strategy might include providing somewhat greater supplies, during the next few months, to Category D and E countries where using the vaccine to halt spread of the disease might be possible (provided that disease control measures are maintained in those countries). It is clear, in any case, that further modelling of vaccine allocation strategies is essential, taking into account the actual vaccine efficacies and projected available doses by month, as well as allowing for disease stage categories in different countries.Vaccination strategies will need to account for current and potential future virus variants as well as the likelihood that immunity from vaccination will wane over time. Higher-transmissibility variants of the SARS-CoV-2 virus increase the urgency of distributing vaccines, especially in Category F and G countries which may struggle to keep the disease suppressed, and might cause vaccination of key transmitters to be a less effective strategy for Category D and E countries if higher-transmissibility variants mean that they can’t suppress the disease fully with limited vaccinations. Immunity-resistant variants of the virus may reduce the effectiveness of current vaccines, but are not likely to negate fully the protection offered by existing vaccines. Immunity acquired through vaccination is likely to wane over time – like immunity acquired through infection. Looking to the longer term, new virus variants and waning immunity are likely to necessitate re-vaccination (with vaccines effective against the latest variants) on a regular basis. In many, perhaps most, countries, the time to vaccinate the whole population will exceed the timeframe in which immunity from vaccination wanes or new immunity-resistant variants emerge. In making long-term plans, therefore, countries may face a wide range of options for who to vaccinate (elderly and vulnerable populations, key transmitters or entire populations) and for frequency of vaccination (every 6 months, annual, or once if residual benefits are sufficient). Optimal strategies for each country will be complicated to determine, as the choice will depend on many factors, including vaccine effectiveness in reducing mortality and in reducing transmission, how effectiveness wanes over time, mortality rates and transmissibility of new variants (in general and in previously infected or vaccinated people), and, once the risks to life and health from “endemic COVID” decrease to the point where COVID-19 is not an overriding issue, comparison with other health and budgetary priorities.Competing Interest StatementJ.P.C., C.J.A.N. and O.F. work at Dalberg Advisors, a management consultancy whose clients include multilateral agencies, foundations, international development agencies, governments, companies and NGOs. They have prepared this article in a personal capacity, and the work was not funded by any client of Dalberg Advisors.Funding StatementNo funding was received to support this work.Author DeclarationsI confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.YesThe details of the IRB/oversight body that provided approval or exemption for the research described are given below:The research used only publicly available and anonymized data.All necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived.YesI understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).Yes I have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.YesAll data used for this research presented in this report can be freely downloaded from the cited sources. Much of the relevant data used is presented in Annexes A and B. https://covid19.who.int https://ourworldindata.org/coronavirus-data-explorer https://www.bsg.ox.ac.uk/covidtracker https://www.economist.com/graphic-detail/2020/07/15/tracking-covid-19-excess-deaths-across-countries https://doi.org/10.1101/2020.07.23.20160895