Improving modelling for epidemic responses: reflections from members of the UK infectious disease modelling community on their experiences during the COVID-19 pandemic (preprint)

The COVID-19 pandemic both relied and placed significant burdens on the experts involved from research and public health sectors. The sustained high pressure of a pandemic on responders, such as healthcare workers, can lead to lasting psychological impacts including acute stress disorder, post-traumatic stress disorder, burnout, and moral injury, which can impact individual wellbeing and productivity. As members of the infectious disease modelling community, we convened a reflective workshop to understand the professional and personal impacts of response work on our community and to propose recommendations for future epidemic responses. The attendees represented a range of career stages, institutions, and disciplines. This piece was collectively produced by those present at the session based on our collective experiences. Key issues we identified at the workshop were lack of institutional support, insecure contracts, unequal credit and recognition, and mental health impacts. Our recommendations include rewarding impactful work, fostering academia-public health collaboration, decreasing dependence on key individuals by developing teams, increasing transparency in decision-making, and implementing sustainable work practices. Despite limitations in representation, this workshop provided valuable insights into the UK COVID-19 modelling experience and guidance for future public health crises. Recognising and addressing the issues highlighted here is crucial, in our view, for ensuring the effectiveness of epidemic response work in the future.

The interaction of transmission intensity, mortality, and the economy: a retrospective analysis of the COVID-19 pandemic (preprint)

The COVID-19 pandemic has caused over 6.4 million registered deaths to date, and has had a profound impact on economic activity. Here, we study the interaction of transmission, mortality, and the economy during the SARS-CoV-2 pandemic from January 2020 to December 2022 across 25 European countries. We adopt a Bayesian vector autoregressive model with both fixed and random effects. We find that increases in disease transmission intensity decreases Gross domestic product (GDP) and increases daily excess deaths, with a longer lasting impact on excess deaths in comparison to GDP, which recovers more rapidly. Broadly, our results reinforce the intuitive phenomenon that significant economic activity arises from diverse person-to-person interactions. We report on the effectiveness of non-pharmaceutical interventions (NPIs) on transmission intensity, excess deaths and changes in GDP, and resulting implications for policy makers. Our results highlight a complex cost-benefit trade off from individual NPIs. For example, banning international travel increases GDP however reduces excess deaths. We consider country random effects and their associations with excess changes in GDP and excess deaths. For example, more developed countries in Europe typically had more cautious approaches to the COVID-19 pandemic, prioritising healthcare and excess deaths over economic performance. Long term economic impairments are not fully captured by our model, as well as long term disease effects (Long Covid). Our results highlight that the impact of disease on a country is complex and multifaceted, and simple heuristic conclusions to extract the best outcome from the economy and disease burden are challenging.

Prophylactic and reactive vaccination strategies for healthcare workers against MERS-CoV (preprint)

Several vaccines candidates are in development against Middle East respiratory syndrome–related coronavirus (MERS-CoV), which remains a major public health concern. Using individual-level data on the 2013-2014 Kingdom of Saudi Arabia epidemic, we employ counterfactual analysis on inferred transmission trees (“who-infected-whom”) to assess potential vaccine impact. We investigate the conditions under which prophylactic “proactive” campaigns would outperform “reactive” campaigns (i.e. vaccinating either before or in response to the next outbreak), focussing on healthcare workers. Spatial scale is crucial: if vaccinating healthcare workers in response to outbreaks at their hospital only, proactive campaigns perform better, unless efficacy has waned significantly. However, campaigns that react at regional or national level consistently outperform proactive campaigns. Measures targeting the animal reservoir reduce transmission linearly, albeit with wide uncertainty. Substantial reduction of MERS-CoV morbidity and mortality is possible when vaccinating healthcare workers, underlining the need for at-risk countries to stockpile vaccines when available.

Vaccine Equity in Low and Middle Income Countries: A Systematic Review and Meta-analysis (preprint)

Background Evidence to date has shown that inequality in health, and vaccine coverage in particular, can have ramifications to wider society. However, whilst individual studies have sought to characterise these heterogeneities in immunisation coverage at national level, few have taken a broad and quantitative view of the contributing factors to heterogeneity in vaccine coverage and impact. This systematic review aims to highlight these geographic, demographic, and sociodemographic characteristics through a qualitative and quantitative approach, vital to prioritise and optimise vaccination policies. Methods A systematic review of two databases (PubMed and Web of Science) was undertaken using Medical Subject Headings (MeSH) and keywords to identify studies examining factors on vaccine inequality and heterogeneity in vaccine coverage. Inclusion criteria were applied independently by two researchers. Studies including data on key characteristics of interest were further analysed through a meta-analysis to produce a pooled estimate of the risk ratio using a random effects model for that characteristic. Results One hundred and eight studies were included in this review. We found that inequalities in wealth, education, and geographic access can affect vaccine impact and vaccine dropout. We estimated those living in rural areas were not significantly different in terms of full vaccination status compared to urban areas but noted considerable heterogeneity between countries. We found that females were 3% (95%CI[1%, 5%]) less likely to be fully vaccinated than males. Additionally, we estimated that children whose mothers had no formal education were 28% (95%CI[18%,47%]) less likely to be fully vaccinated than those whose mother had primary level, or above, education. Finally, we found that individuals in the poorest wealth quintile were 27% (95%CI [16%,37%]) less likely to be fully vaccinated than those in the richest. Conclusions We found a nuanced picture of inequality in vaccine coverage and access with wealth disparity dominating, and likely driving, other disparities. This review highlights the complex landscape of inequity and further need to design vaccination strategies targeting missed subgroups to improve and recover vaccination coverage following the COVID-19 pandemic. Registration Prospero CRD42021261927

Effectiveness of COVID-19 vaccines against the Omicron (B.1.1.529) variant of concern (preprint)

Abstract Background A rapid increase in cases due to the SARS-CoV-2 Omicron (B.1.1.529) variant in highly vaccinated populations has raised concerns about the effectiveness of current vaccines. Methods We used a test-negative case-control design to estimate vaccine effectiveness (VE) against symptomatic disease caused by the Omicron and Delta variants in England. VE was calculated after primary immunisation with two BNT162b2 or ChAdOx1 doses, and at 2+ weeks following a BNT162b2 booster. Results Between 27 November and 06 December 2021, 581 and 56,439 eligible Omicron and Delta cases respectively were identified. There were 130,867 eligible test-negative controls. There was no effect against Omicron from 15 weeks after two ChAdOx1 doses, while VE after two BNT162b2 doses was 88.0% (95%CI: 65.9 to 95.8%) 2-9 weeks after dose 2, dropping to between 34 and 37% from 15 weeks post dose 2.From two weeks after a BNT162b2 booster, VE increased to 71.4% (95%CI: 41.8 to 86.0%) for ChAdOx1 primary course recipients and 75.5% (95%CI: 56.1 to 86.3%) for BNT162b2 primary course recipients. For cases with Delta, VE was 41.8% (95%CI: 39.4-44.1%) at 25+ weeks after two ChAdOx1 doses, increasing to 93.8% (95%CI: 93.2-94.3%) after a BNT162b2 booster. With a BNT162b2 primary course, VE was 63.5% (95%CI: 61.4 to 65.5%) 25+ weeks after dose 2, increasing to 92.6% (95%CI: 92.0-93.1%) two weeks after the booster. Conclusions Primary immunisation with two BNT162b2 or ChAdOx1 doses provided no or limited protection against symptomatic disease with the Omicron variant. Boosting with BNT162b2 following either primary course significantly increased protection.

Safety, tolerability and viral kinetics during SARS-CoV-2 human challenge (preprint)

To establish a novel SARS-CoV-2 human challenge model, 36 volunteers aged 18-29 years without evidence of previous infection or vaccination were inoculated with 10 TCID50 of a wild-type virus (SARS-CoV-2/human/GBR/484861/2020) intranasally. Two participants were excluded from per protocol analysis due to seroconversion between screening and inoculation. Eighteen (~53%) became infected, with viral load (VL) rising steeply and peaking at ~5 days post-inoculation. Virus was first detected in the throat but rose to significantly higher levels in the nose, peaking at ~8.87 log10 copies/ml (median, 95% CI [8.41,9.53). Viable virus was recoverable from the nose up to ~10 days post-inoculation, on average. There were no serious adverse events. Mild-to-moderate symptoms were reported by 16 (89%) infected individuals, beginning 2-4 days post-inoculation. Anosmia/dysosmia developed more gradually in 12 (67%) participants. No quantitative correlation was noted between VL and symptoms, with high VLs even in asymptomatic infection, followed by the development of serum spike-specific and neutralising antibodies. However, lateral flow results were strongly associated with viable virus and modelling showed that twice-weekly rapid tests could diagnose infection before 70-80% of viable virus had been generated. Thus, in this first SARS-CoV-2 human challenge study, no serious safety signals were detected and the detailed characteristics of early infection and their public health implications were shown. ClinicalTrials.gov identifier: NCT04865237.

Global predictions of short- to medium-term COVID-19 transmission trends : a retrospective assessment (preprint)

From 8th March to 29th November 2020, we produced weekly estimates of SARS-CoV-2 transmissibility and forecasts of deaths due to COVID-19 for 81 countries with evidence of sustained transmission. We also developed a novel heuristic to combine weekly estimates of transmissibility to produce forecasts over a 4-week horizon. We evaluated the robustness of the forecasts using relative error, coverage probability, and comparisons with null models. During the 39-week period covered by this study, both the short- and medium-term forecasts captured well the epidemic trajectory across different waves of COVID-19 infections with small relative errors over the forecast horizon. The model was well calibrated with 56.3\% and 45.6\% of the observations lying in the 50\% Credible Interval in 1-week and 4-week ahead forecasts respectively. We could accurately characterise the overall phase of the epidemic up to 4-weeks ahead in 84.9\% of country-days. The medium-term forecasts can be used in conjunction with the short-term forecasts of COVID-19 mortality as a useful planning tool as countries continue to relax public health measures.

Global predictions of short- to medium-term COVID-19 transmission trends : a retrospective assessment (preprint)

BackgroundAs of July 2021, more than 180,000,000 cases of COVID-19 have been reported across the world, with more than 4 million deaths. Mathematical modelling and forecasting efforts have been widely used to inform policy-making and to create situational awareness. Methods and FindingsFrom 8th March to 29th November 2020, we produced weekly estimates of SARS-CoV-2 transmissibility and forecasts of deaths due to COVID-19 for countries with evidence of sustained transmission. The estimates and forecasts were based on an ensemble model comprising of three models that were calibrated using only the reported number of COVID-19 cases and deaths in each country. We also developed a novel heuristic to combine weekly estimates of transmissibility and potential changes in population immunity due to infection to produce forecasts over a 4-week horizon. We evaluated the robustness of the forecasts using relative error, coverage probability, and comparisons with null models. ConclusionsDuring the 39-week period covered by this study, we produced short- and medium-term forecasts for 81 countries. Both the short- and medium-term forecasts captured well the epidemic trajectory across different waves of COVID-19 infections with small relative errors over the forecast horizon. The model was well calibrated with 56.3% and 45.6% of the observations lying in the 50% Credible Interval in 1-week and 4-week ahead forecasts respectively. We could accurately characterise the overall phase of the epidemic up to 4-weeks ahead in 84.9% of country-days. The medium-term forecasts can be used in conjunction with the short-term forecasts of COVID-19 mortality as a useful planning tool as countries continue to relax stringent public health measures that were implemented to contain the pandemic.

Assessment of Mortality and Hospital Admissions Associated with Confirmed Infection with SARS-CoV-2 Variant of Concern VOC-202012/01 (B.1.1.7) a Matched Cohort and Time-to-Event Analysis (preprint)

Background: The emergence of VOC202012/01 in England, known as B.1.1.7 or informally as the ‘UK variant’, has coincided with rapid increases in the number of PCR-confirmed positive cases in areas where the variant has been concentrated. Methods: To assess whether infection with SARS-CoV-2 variant VOC202012/01 is associated with more severe clinical outcomes compared to wild-type infection, genomically sequenced and confirmed variant and wild-type cases were linked to routine healthcare and surveillance datasets. Two statistical analyses were conducted to compare the risk of hospital admission and death within 28 days of test between variant and wild-type cases: a case-control study and an adjusted Cox proportional hazards model. Differences in severity of disease were assessed by comparing hospital admission and mortality, including length of hospitalisation and time to death.Results: Of 63,609 genomically sequenced COVID-19 cases tested in England between October and December 2020 6,038 were variant cases. In the matched cohort analysis 2,821 variant cases were matched to 2,821 to wild-type cases. In the time to event analysis we observed a 34% increased risk in hospitalisation associated with the variant compared to wild-type cases, however, no significant difference in the risk of mortality was observed. Conclusion: We found evidence of increased risk of hospitalisation after adjusting for key confounders, suggesting increase infection severity associated with this variant. Follow-up studies are needed to assess potential longer-term differences in the clinical outcomes of people infected with the VOC-202012/01 variant.

Genomics and epidemiology of a novel SARS-CoV-2 lineage in Manaus, Brazil (preprint)

Cases of SARS-CoV-2 infection in Manaus, Brazil, resurged in late 2020, despite high levels of previous infection there. Through genome sequencing of viruses sampled in Manaus between November 2020 and January 2021, we identified the emergence and circulation of a novel SARS-CoV-2 variant of concern, lineage P.1, that acquired 17 mutations, including a trio in the spike protein (K417T, E484K and N501Y) associated with increased binding to the human ACE2 receptor. Molecular clock analysis shows that P.1 emergence occurred around early November 2020 and was preceded by a period of faster molecular evolution. Using a two-category dynamical model that integrates genomic and mortality data, we estimate that P.1 may be 1.4–2.2 times more transmissible and 25-61% more likely to evade protective immunity elicited by previous infection with non-P.1 lineages. Enhanced global genomic surveillance of variants of concern, which may exhibit increased transmissibility and/or immune evasion, is critical to accelerate pandemic responsiveness. One-Sentence Summary We report the evolution and emergence of a SARS-CoV-2 lineage of concern associated with rapid transmission in Manaus.

The costs of keeping schools open during the COVID-19 pandemic (preprint)

There is a trade-off between restrictions on the education sector and other economic sectors in the control of SARS-CoV-2 transmission. Here we integrate a dynamic model of SARS-CoV-2 transmission with a 63-sector economic model reflecting sectoral heterogeneity in transmission and economic interdependence between sectors. We identify control strategies which optimize economic production while keeping schools and universities operational, and constraining infections such that emergency hospital capacity is not exceeded. We estimate an economic gain of between £163bn (24%) and £205bn (31%) for the United Kingdom compared to a blanket lockdown of non-essential activities over six months, depending on hospital capacity. Sectors identified as priorities for closures are contact-intensive, produce few crucial inputs for other sectors and/or are less economically productive. Partial closures over some months are required for retail trade, hospitality, accommodation, creative activities, arts, entertainment, and personal services including hairdressing and beauty treatments under most scenarios.

Using next generation matrices to estimate the proportion of infections that are not detected in an outbreak (preprint)

Contact tracing, where exposed individuals are followed up to break ongoing transmission chains, is a key pillar of outbreak response for infectious disease outbreaks. Unfortunately, these systems are not fully effective, and infections can still go undetected as people may not remember all their contacts or contacts may not be traced successfully. A large proportion of undetected infections suggests poor contact tracing and surveillance systems, which could be a potential area of improvement for a disease response. In this paper, we present a method for estimating the proportion of infections that are not detected during an outbreak. Our method uses next generation matrices that are parameterized by linked contact tracing data and case line-lists. We validate the method using simulated data from an individual-based model and then investigate two case studies: the proportion of undetected infections in the SARS-CoV-2 outbreak in New Zealand during 2020 and the Ebola epidemic in Guinea during 2014. We estimate that only 5.26% of SARS-CoV-2 infections were not detected in New Zealand during 2020 (95% credible interval: 0.243 – 16.0%) but depending on assumptions 39.0% or 37.7% of Ebola infections were not detected in Guinea (95% credible intervals: 1.69 – 87.0% or 1.7 – 80.9%).

The 2020 SARS-CoV-2 epidemic in England: key epidemiological drivers and impact of interventions (preprint)

We fitted a model of SARS-CoV-2 transmission in care homes and the community to regional surveillance data for England. Among control measures implemented, only national lockdown brought the reproduction number below 1 consistently; introduced one week earlier it could have reduced first wave deaths from 36,700 to 15,700 (95%CrI: 8,900–26,800). Improved clinical care reduced the infection fatality ratio from 1.25% (95%CrI: 1.18%–1.33%) to 0.77% (95%CrI: 0.71%–0.84%). The infection fatality ratio was higher in the elderly residing in care homes (35.9%, 95%CrI: 29.1%–43.4%) than those residing in the community (10.4%, 95%CrI: 9.1%–11.5%). England is still far from herd immunity, with regional cumulative infection incidence to 1st December 2020 between 4.8% (95%CrI: 4.4%–5.1%) and 15.4% (95%CrI: 14.9%–15.9%) of the population. One-sentence summary We fit a mathematical model of SARS-CoV-2 transmission to surveillance data from England, to estimate transmissibility, severity, and the impact of interventions

Semi-Mechanistic Bayesian Modeling of COVID-19 with Renewal Processes (preprint)

We propose a general Bayesian approach to modeling epidemics such as COVID-19. The approach grew out of specific analyses conducted during the pandemic, in particular an analysis concerning the effects of non-pharmaceutical interventions (NPIs) in reducing COVID-19 transmission in 11 European countries. The model parameterizes the time varying reproduction number $R_t$ through a regression framework in which covariates can e.g be governmental interventions or changes in mobility patterns. This allows a joint fit across regions and partial pooling to share strength. This innovation was critical to our timely estimates of the impact of lockdown and other NPIs in the European epidemics, whose validity was borne out by the subsequent course of the epidemic. Our framework provides a fully generative model for latent infections and observations deriving from them, including deaths, cases, hospitalizations, ICU admissions and seroprevalence surveys. One issue surrounding our model's use during the COVID-19 pandemic is the confounded nature of NPIs and mobility. We use our framework to explore this issue. We have open sourced an R package epidemia implementing our approach in Stan. Versions of the model are used by New York State, Tennessee and Scotland to estimate the current situation and make policy decisions.

COVID-19 epidemic severity is associated with timing of non-pharmaceutical interventions (preprint)

Background: Unprecedented public health interventions including travel restrictions and national lockdowns have been implemented to stem the COVID-19 epidemic, but the effectiveness of non-pharmaceutical interventions is still debated. International comparisons are hampered by highly variable conditions under which epidemics spread and differences in the timing and scale of interventions. Cumulative COVID-19 morbidity and mortality are functions of both the rate of epidemic growth and the duration of uninhibited growth before interventions were implemented. Incomplete and sporadic testing during the early COVID-19 epidemic makes it difficult to identify how long SARS-CoV-2 was circulating in different places. SARS-CoV-2 genetic sequences can be analyzed to provide an estimate of both the time of epidemic origin and the rate of early epidemic growth in different settings. Methods: We carried out a phylogenetic analysis of more than 29,000 publicly available whole genome SARS-CoV-2 sequences from 57 locations to estimate the time that the epidemic originated in different places. These estimates were cross-referenced with dates of the most stringent interventions in each location as well as the number of cumulative COVID-19 deaths following maximum intervention. Phylodynamic methods were used to estimate the rate of early epidemic growth and proxy estimates of epidemic size. Findings: The time elapsed between epidemic origin and maximum intervention is strongly associated with different measures of epidemic severity and explains 46% of variance in numbers infected at time of maximum intervention. The reproduction number is independently associated with epidemic severity. In multivariable regression models, epidemic severity was not associated with census population size. The time elapsed between detection of initial COVID-19 cases to interventions was not associated with epidemic severity, indicating that many locations experienced long periods of cryptic transmission. Interpretation: Locations where strong non-pharmaceutical interventions were implemented earlier experienced much less severe COVID-19 morbidity and mortality during the period of study.

The Impact of the COVID-19 Epidemic on Patterns of Attendance at Emergency Departments in Two Large London Hospitals: An Observational Study (preprint)

Background Hospitals in England have undergone considerable change to address the surge in demand imposed by the COVID-19 epidemic. The impact of this on emergency department (ED) attendances is unknown, especially for non-COVID-19 related emergencies.Methods We calibrated auto-regressive integrated moving average time-series models of ED attendances to Imperial College Healthcare NHS Trust (ICHNT) using historic (2015–2019) data. Forecasted trends were compared to present year ICHNT data for the period between March 12 (when England implemented the first COVID-19 public health measure) and May 31. We compared ICHTN trends with publicly available regional and national data. Lastly, we compared emergency admissions and in-hospital mortality at ICHNT during the present year to a historic 5-year average.Results ED attendances at ICHNT decreased by 35%, in keeping with the trend for ED attendances across all England regions, which fell by approximately 50%. For ICHNT, the decrease in attendances was mainly amongst those aged < 65 years and those arriving by their own means (e.g. personal or public transport). Increasing distance from postcode of residence to hospital was a significant predictor of reduced attendances. Non-COVID related emergency admissions to hospital after March 12 fell by 48%; there was an indication of a non-significant increase in non-COVID-19 crude mortality risk (RR 1.13, 95%CI 0.94–1.37, p = 0.19).Conclusions Our study finds strong evidence that emergency healthcare seeking has drastically changed across the population in England. At ICHNT, we find that a larger proportion arrived by ambulance and that hospitalisation outcomes of non-COVID patients did not differ from previous years. The extent to which these findings relate to ED avoidance behaviours compared to having sought alternative emergency health services outside of hospital remains unknown. National analyses and strategies to streamline emergency services in England going forward are urgently needed.

Trust in the Time of Corona (preprint)

The global spread of COVID-19 is one of the largest threats to people and governments since the Second World War. The on-going pandemic and its countermeasures have led to varying physical, psychological, and emotional experiences, shaping not just public health and the economy but also societies. We focus on one pillar of society—trust—and explore how trust correlates with the individual experiences of the pandemic. The analysis is based on a new global survey—'Life with Corona'—and uses simple correlational statistics. We show that those who have had contact with sick people and those that are unemployed exhibit lower trust in people, institutions, and in general. By contrast, no such differences exist for those who have personally experienced symptoms of the disease. These associations vary across contexts and are not driven by concerns about personal health or the health of loved ones, but rather by increased levels of worry and stress. Our findings suggest that the effects of the pandemic go well beyond immediate health concerns, leading to important normative changes that are likely to shape how societies will emerge from the pandemic

Evolution and epidemic spread of SARS-CoV-2 in Brazil (preprint)

Brazil currently has one of the fastest growing SARS-CoV-2 epidemics in the world. Due to limited available data, assessments of the impact of non-pharmaceutical interventions (NPIs) on virus transmission and epidemic spread remain challenging. We investigate the impact of NPIs in Brazil using epidemiological, mobility and genomic data. Mobility-driven transmission models for Sao Paulo and Rio de Janeiro cities show that the reproduction number (Rt) reached below 1 following NPIs but slowly increased to values between 1 to 1.3 (1.0 - -1.6). Genome sequencing of 427 new genomes and analysis of a geographically representative genomic dataset from 21 of the 27 Brazilian states identified >100 international introductions of SARS-CoV-2 in Brazil. We estimate that three clades introduced from Europe emerged between 22 and 27 February 2020, and were already well-established before the implementation of NPIs and travel bans. During this first phase of the epidemic establishment of SARS-CoV-2 in Brazil, we find that the virus spread mostly locally and within-state borders. Despite sharp decreases in national air travel during this period, we detected a 25% increase in the average distance travelled by air passengers during this time period. This coincided with the spread of SARS-CoV-2 from large urban centers to the rest of the country. In conclusion, our results shed light on the role of large and highly connected populated centres in the rapid ignition and establishment of SARS-CoV-2, and provide evidence that current interventions remain insufficient to keep virus transmission under control in Brazil.

Estimates of the severity of COVID-19 disease (preprint)

Background: A range of case fatality ratio (CFR) estimates for COVID 19 have been produced that differ substantially in magnitude. Methods: We used individual-case data from mainland China and cases detected outside mainland China to estimate the time between onset of symptoms and outcome (death or discharge from hospital). We next obtained age-stratified estimates of the CFR by relating the aggregate distribution of cases by dates of onset to the observed cumulative deaths in China, assuming a constant attack rate by age and adjusting for the demography of the population, and age and location-based under ascertainment. We additionally estimated the CFR from individual linelist data on 1,334 cases identified outside mainland China. We used data on the PCR prevalence in international residents repatriated from China at the end of January 2020 to obtain age-stratified estimates of the infection fatality ratio (IFR). Using data on age stratified severity in a subset of 3,665 cases from China, we estimated the proportion of infections that will likely require hospitalisation. Findings: We estimate the mean duration from onset-of-symptoms to death to be 17.8 days (95% credible interval, crI 16.9,19.2 days) and from onset-of-symptoms to hospital discharge to be 22.6 days (95% crI 21.1,24.4 days). We estimate a crude CFR of 3.67% (95% crI 3.56%,3.80%) in cases from mainland China. Adjusting for demography and under-ascertainment of milder cases in Wuhan relative to the rest of China, we obtain a best estimate of the CFR in China of 1.38% (95% crI 1.23%,1.53%) with substantially higher values in older ages. Our estimate of the CFR from international cases stratified by age (under 60 or 60 and above) are consistent with these estimates from China. We obtain an overall IFR estimate for China of 0.66% (0.39%,1.33%), again with an increasing profile with age. Interpretation: These early estimates give an indication of the fatality ratio across the spectrum of COVID-19 disease and demonstrate a strong age-gradient in risk.