A comparison of short-term probabilistic forecasts for the incidence of COVID-19 using mechanistic and statistical time series models (preprint)

Short-term forecasts of infectious disease spread are a critical component in risk evaluation and public health decision making. While different models for short-term forecasting have been developed, open questions about their relative performance remain. Here, we compare short-term probabilistic forecasts of popular mechanistic models based on the renewal equation with forecasts of statistical time series models. Our empirical comparison is based on data of the daily incidence of COVID-19 across six large US states over the first pandemic year. We find that, on average, probabilistic forecasts from statistical time series models are overall at least as accurate as forecasts from mechanistic models. Moreover, statistical time series models better capture volatility. Our findings suggest that domain knowledge, which is integrated into mechanistic models by making assumptions about disease dynamics, does not improve short-term forecasts of disease incidence. We note, however, that forecasting is often only one of many objectives and thus mechanistic models remain important, for example, to model the impact of vaccines or the emergence of new variants.

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.

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 Mixed Effects model with auto-regressive terms. 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 and 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.

The uncertainty of infectious disease outbreaks is underestimated (preprint)

Uncertainty can be classified as either aleatoric (intrinsic randomness) or epistemic (imperfect knowledge of parameters). Majority of frameworks assessing infectious disease risk consider only epistemic uncertainty. We only ever observe a single epidemic, and therefore cannot empirically determine aleatoric uncertainty. Here, for the first time, we characterise both epistemic and aleatoric uncertainty using a time-varying general branching processes. Our framework explicitly decomposes aleatoric variance into mechanistic components, quantifying the contribution to uncertainty produced by each factor in the epidemic process, and how these contributions vary over time. The aleatoric variance of an outbreak is itself a renewal equation where past variance affects future variance. Surprisingly, superspreading is not necessary for substantial uncertainty, and profound variation in outbreak size can occur even without overdispersion in the distribution of the number of secondary infections. Aleatoric forecasting uncertainty grows dynamically and rapidly, and so forecasting using only epistemic uncertainty is a significant underestimate. Failure to account for aleatoric uncertainty will ensure that policymakers are misled about the substantially higher true extent of potential risk. We demonstrate our method, and the extent to which potential risk is underestimated, using two historical examples: the 2003 Hong Kong severe acute respiratory syndrome (SARS) outbreak, and the early 2020 UK COVID-19 epidemic. Our framework provides analytical tools to estimate epidemic uncertainty with limited data, to provide reasonable worst-case scenarios and assess both epistemic and aleatoric uncertainty in forecasting, and to retrospectively assess an epidemic and thereby provide a baseline risk estimate for future outbreaks. Our work strongly supports the precautionary principle in pandemic response.

Covid-19 is a leading cause of death in children and young people ages 0-19 years in the United States (preprint)

Covid-19 has caused more than 1 million deaths in the US, including at least 1,433 deaths among children and young people (CYP) aged 0-19 years. Deaths among US CYP are rare in general, and so we argue here that the mortality burden of Covid-19 in CYP is best understood in the context of all other causes of CYP death. Using publicly available data from the National Center for Health Statistics, and comparing to mortality in 2019, the immediate pre-pandemic period, we find that Covid-19 is a leading cause of death in CYP aged 0-19 years in the US, ranking #9 among all causes of deaths, #5 in disease related causes of deaths (excluding accidents, assault and suicide), and #1 in deaths caused by infectious / respiratory diseases. Due to the impact of mitigations such as social distancing and our comparison of a single disease (Covid-19) to groups of causes such as deaths from pneumonia and influenza, these rankings are likely conservative lower bounds. Our findings underscore the importance of continued vaccination campaigns for CYP over 5 years of age in the US and for effective Covid-19 vaccines for under 5 year olds.

Report 46: Factors driving extensive spatial and temporal fluctuations in COVID-19 fatality rates in Brazilian hospitals (preprint)

The SARS-CoV-2 Gamma variant spread rapidly across Brazil, causing substantial infection and death waves. We use individual-level patient records following hospitalisation with suspected or confirmed COVID-19 to document the extensive shocks in hospital fatality rates that followed Gamma’s spread across 14 state capitals, and in which more than half of hospitalised patients died over sustained time periods. We show that extensive fluctuations in COVID-19 in-hospital fatality rates also existed prior to Gamma’s detection, and were largely transient after Gamma’s detection, subsiding with hospital demand. Using a Bayesian fatality rate model, we find that the geographic and temporal fluctuations in Brazil’s COVID-19 in-hospital fatality rates are primarily associated with geographic inequities and shortages in healthcare capacity. We project that approximately half of Brazil’s COVID-19 deaths in hospitals could have been avoided without pre-pandemic geographic inequities and without pandemic healthcare pressure. Our results suggest that investments in healthcare resources, healthcare optimization, and pandemic preparedness are critical to minimize population wide mortality and morbidity caused by highly transmissible and deadly pathogens such as SARS-CoV-2, especially in low- and middle-income countries. Note The following manuscript has appeared as ‘Report 46 - Factors driving extensive spatial and temporal fluctuations in COVID-19 fatality rates in Brazilian hospitals’ at https://spiral.imperial.ac.uk:8443/handle/10044/1/91875 . One sentence summary COVID-19 in-hospital fatality rates fluctuate dramatically in Brazil, and these fluctuations are primarily associated with geographic inequities and shortages in healthcare capacity.

Resurgence of SARS-CoV-2 in India: Potential role of the B.1.617.2 (Delta) variant and delayed interventions (preprint)

India has seen a surge of SARS-CoV-2 infections and deaths in early part of 2021, despite having controlled the epidemic during 2020. Building on a two-strain, semi-mechanistic model that synthesizes mortality and genomic data, we find evidence that altered epidemiological properties of B.1.617.2 (Delta) variant play an important role in this resurgence in India. Under all scenarios of immune evasion, we find an increased transmissibility advantage for B.1617.2 against all previously circulating strains. Using an extended SIR model accounting for reinfections and wanning immunity, we produce evidence in support of how early public interventions in March 2021 would have helped to control transmission in the country. We argue that enhanced genomic surveillance along with constant assessment of risk associated with increased transmission is critical for pandemic responsiveness.

SARS-CoV-2 B.1.617.2 Delta variant emergence and vaccine breakthrough (preprint)

The SARS-CoV-2 B.1.617.2 (Delta) variant was first identified in the state of Maharashtra in late 2020 and has spread throughout India, displacing the B.1.1.7 (Alpha) variant and other pre-existing lineages. Mathematical modelling indicates that the growth advantage is most likely explained by a combination of increased transmissibility and immune evasion. Indeed in vitro, the delta variant is less sensitive to neutralising antibodies in sera from recovered individuals, with higher replication efficiency as compared to the Alpha variant. In an analysis of vaccine breakthrough in over 100 healthcare workers across three centres in India, the Delta variant not only dominates vaccine-breakthrough infections with higher respiratory viral loads compared to non-delta infections (Ct value of 16.5 versus 19), but also generates greater transmission between HCW as compared to B.1.1.7 or B.1.617.1 (p=0.02). In vitro, the Delta variant shows 8 fold approximately reduced sensitivity to vaccine-elicited antibodies compared to wild type Wuhan-1 bearing D614G. Serum neutralising titres against the SARS-CoV-2 Delta variant were significantly lower in participants vaccinated with ChadOx-1 as compared to BNT162b2 (GMT 3372 versus 654, p<0001). These combined epidemiological and in vitro data indicate that the dominance of the Delta variant in India has been most likely driven by a combination of evasion of neutralising antibodies in previously infected individuals and increased virus infectivity. Whilst severe disease in fully vaccinated HCW was rare, breakthrough transmission clusters in hospitals associated with the Delta variant are concerning and indicate that infection control measures need continue in the post-vaccination era.

Understanding the effectiveness of government interventions in Europe’s second wave of COVID-19 (preprint)

As European governments face resurging waves of COVID-19, non-pharmaceutical interventions (NPIs) continue to be the primary tool for infection control. However, updated estimates of their relative effectiveness have been absent for Europe’s second wave, largely due to a lack of collated data that considers the increased subnational variation and diversity of NPIs. We collect the largest dataset of NPI implementation dates in Europe, spanning 114 subnational areas in 7 countries, with a systematic categorisation of interventions tailored to the second wave. Using a hierarchical Bayesian transmission model, we estimate the effectiveness of 17 NPIs from local case and death data. We manually validate the data, address limitations in modelling from previous studies, and extensively test the robustness of our estimates. The combined effect of all NPIs was smaller relative to estimates from the first half of 2020, indicating the strong influence of safety measures and individual protective behaviours--such as distancing--that persisted after the first wave. Closing specific businesses was highly effective. Gathering restrictions were highly effective but only for the strictest limits. We find smaller effects for closing educational institutions compared to the first wave, suggesting that safer operation of schools was possible with a set of stringent safety measures including testing and tracing, preventing mixing, and smaller classes. These results underscore that effectiveness estimates from the early stage of an epidemic are measured relative to pre-pandemic behaviour. Updated estimates are required to inform policy in an ongoing pandemic.

Environmental drivers of SARS-CoV-2 lineage B.1.1.7 transmission in England, October to December 2020 (preprint)

Previous work has shown that environment affects SARS-CoV-2 transmission, but it is unclear whether emerging strains show similar responses. Here we show that lineage B.1.1.7 spread with greater transmission in colder and more densely populated parts of England. We also find evidence of B.1.1.7's transmission advantage at warmer temperatures versus other strains, implying that spring conditions may facilitate B.1.1.7's invasion in Europe and across the Northern hemisphere, undermining the effectiveness of public health interventions.

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.

Impact of the Tier system on SARS-CoV-2 transmission in the UK between the first and second national lockdowns (preprint)

ObjectiveMeasure the effects of the Tier system on the COVID-19 pandemic in the UK between the first and second national lockdowns, before the emergence of the B.1.1.7 variant of concern. DesignModelling study combining estimates of the real-time reproduction number Rt (derived from UK case, death and serological survey data) with publicly available data on regional non-pharmaceutical interventions. We fit a Bayesian hierarchical model with latent factors using these quantities, to account for broader national trends in addition to subnational effects from Tiers. SettingThe UK at Lower Tier Local Authority (LTLA) level. Primary and secondary outcome measuresReduction in real-time reproduction number Rt. ResultsNationally, transmission increased between July and late September, regional differences notwithstanding. Immediately prior to the introduction of the tier system, Rt averaged 1.3 (0.9 - 1.6) across LTLAs, but declined to an average of 1.1 (0.86 - 1.42) two weeks later. Decline in transmission was not solely attributable to Tiers. Tier 1 had negligible effects. Tiers 2 and 3 respectively reduced transmission by 6% (5%-7%) and 23% (21%-25%). 93% of LTLAs would have begun to suppress their epidemics if every LTLA had gone into Tier 3 by the second national lockdown, whereas only 29% did so in reality. ConclusionsThe relatively small effect sizes found in this analysis demonstrate that interventions at least as stringent as Tier 3 are required to suppress transmission, especially considering more transmissible variants, at least until effective vaccination is widespread or much greater population immunity has amassed. Strengths and limitations of this studyO_LIFirst study to measure effects of UK Tier system for SARS-CoV-2 control at national and regional level. C_LIO_LIModel makes minimal assumptions and is primarily data driven. C_LIO_LIInsufficient statistical power to estimate effects of individual interventions that comprise Tiers, or their interaction. C_LIO_LIEstimates show that Tiers 1 and 2 are insufficient to suppress transmission, at least until widespread population immunity has amassed. Emergence of more transmissible variants of concern unfortunately supports this conclusion. C_LI

Gaussian Process Nowcasting: Application to COVID-19 Mortality Reporting (preprint)

Updating observations of a signal due to the delays in the measurement process is a common problem in signal processing, with prominent examples in a wide range of fields. An important example of this problem is the nowcasting of COVID-19 mortality: given a stream of reported counts of daily deaths, can we correct for the delays in reporting to paint an accurate picture of the present, with uncertainty? Without this correction, raw data will often mislead by suggesting an improving situation. We present a flexible approach using a latent Gaussian process that is capable of describing the changing auto-correlation structure present in the reporting time-delay surface. This approach also yields robust estimates of uncertainty for the estimated nowcasted numbers of deaths. We test assumptions in model specification such as the choice of kernel or hyper priors, and evaluate model performance on a challenging real dataset from Brazil. Our experiments show that Gaussian process nowcasting performs favourably against both comparable methods, and against a small sample of expert human predictions. Our approach has substantial practical utility in disease modelling -- by applying our approach to COVID-19 mortality data from Brazil, where reporting delays are large, we can make informative predictions on important epidemiological quantities such as the current effective reproduction number.

A Comparison of Five Epidemiological Models for Transmission of SARS-CoV-2 in India (preprint)

BackgroundMany popular disease transmission models have helped nations respond to the COVID-19 pandemic by informing decisions about pandemic planning, resource allocation, implementation of social distancing measures and other non-pharmaceutical interventions. We study how five epidemiological models forecast and assess the course of the pandemic in India: a baseline model, an extended SIR (eSIR) model, two extended SEIR (SAPHIRE and SEIR-fansy) models, and a semi-mechanistic Bayesian hierarchical model (ICM). MethodsUsing COVID-19 data for India from March 15 to June 18 to train the models, we generate predictions from each of the five models from June 19 to July 18. To compare prediction accuracy with respect to reported cumulative and active case counts and cumulative death counts, we compute the symmetric mean absolute prediction error (SMAPE) for each of the five models. ResultsFor active case counts, SMAPE values are 0.72 (SEIR-fansy) and 33.83 (eSIR). For cumulative case counts, SMAPE values are 1.76 (baseline) 23. (eSIR), 2.07 (SAPHIRE) and 3.20 (SEIR-fansy). For cumulative death counts, the SMAPE values are 7.13 (SEIR-fansy) and 26.30 (eSIR). For cumulative cases and deaths, we compute Pearson’s and Lin’s correlation coefficients to investigate how well the projected and observed reported COVID-counts agree. Three models (SAPHIRE, SEIR-fansy and ICM) return total (sum of reported and unreported) counts as well. We compute underreporting factors as of June 30 and note that the SEIR-fansy model reports the highest underreporting factor for active cases (6.10) and cumulative deaths (3.62), while the SAPHIRE model reports the highest underreporting factor for cumulative cases (27.79).ConclusionsIn this comparative paper we describe five different models used to study full disease transmission of the SARS-Cov-2 disease transmission in India. While simulation studies are the only gold standard way to compare the accuracy of the models, here we were uniquely poised to compare the projected case-counts against observed data on a test period. Prediction of daily active number of cases does show appreciable variation across models. The largest variability across models is observed in predicting the “total” number of infections including reported and unreported cases. The degree of under-reporting has been a major concern in India. 

Transmission of SARS-CoV-2 Lineage B.1.1.7 in England: Insights from linking epidemiological and genetic data (preprint)

The SARS-CoV-2 lineage B.1.1.7, now designated Variant of Concern 202012/01 (VOC) by Public Health England, originated in the UK in late Summer to early Autumn 2020. We examine epidemiological evidence for this VOC having a transmission advantage from several perspectives. First, whole genome sequence data collected from community-based diagnostic testing provides an indication of changing prevalence of different genetic variants through time. Phylodynamic modelling additionally indicates that genetic diversity of this lineage has changed in a manner consistent with exponential growth. Second, we find that changes in VOC frequency inferred from genetic data correspond closely to changes inferred by S-gene target failures (SGTF) in community-based diagnostic PCR testing. Third, we examine growth trends in SGTF and non-SGTF case numbers at local area level across England, and show that the VOC has higher transmissibility than non-VOC lineages, even if the VOC has a different latent period or generation time. Available SGTF data indicate a shift in the age composition of reported cases, with a larger share of under 20 year olds among reported VOC than non-VOC cases. Fourth, we assess the association of VOC frequency with independent estimates of the overall SARS-CoV-2 reproduction number through time. Finally, we fit a semi-mechanistic model directly to local VOC and non-VOC case incidence to estimate the reproduction numbers over time for each. There is a consensus among all analyses that the VOC has a substantial transmission advantage, with the estimated difference in reproduction numbers between VOC and non-VOC ranging between 0.4 and 0.7, and the ratio of reproduction numbers varying between 1.4 and 1.8. We note that these estimates of transmission advantage apply to a period where high levels of social distancing were in place in England; extrapolation to other transmission contexts therefore requires caution.

Quantifying the online news media coverage of the COVID-19 pandemic. (preprint)

BackgroundNon-pharmaceutical interventions such as lockdowns, mask wearing and social distancing have been the primary measures to effectively combat the COVID-19 pandemic. Such measures are highly effective when there is strong population wide adherence which needs to be facilitated by information on the current risks posed by the pandemic alongside a clear exposition of the rules and guidelines in place. Here we address the issue of communication on the pandemic by offering data and analysis of online news media coverage of COVID-19. MethodsWe collected 26 million news articles from the front pages of 172 major online news sources in 11 countries (available at http://sciride.org). Using topic detection we identified COVID-19-related content to quantify the proportion of total coverage pandemic received in 2020. Sentiment analysis tool Vader was employed to stratify the emotional polarity of COVID-19 reporting. Further topic detection and sentiment analysis was performed on COVID-19 articles to reveal the leading themes in pandemic reporting and their respective emotional polarizations. FindingsWe find that COVID-19 coverage accounted for approximately 25% of all front-page online news articles between January and October 2020. Sentiment analysis of English-speaking sources reveals that the overall COVID-19 coverage cannot be simply classified as negative due to the disease subject matter, suggesting a wide heterogeneous reporting of the pandemic. Within this heterogenous coverage, 16% of COVID-19 news articles (or 4% of all English-speaking articles) can be classified as highly negatively polarized, citing issues such as death, fear or crisis. InterpretationThe goal of pandemic public health communication is to increase understanding of distancing rules and maximize the impact of any governmental policy. Our results suggest an information overload in COVID-19 reporting that could risk obscuring effective policy communication. We hope that our data and analysis will inform health communication strategy to minimize the risks of COVID-19 while vaccination regimes are being introduced.

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.

A COVID-19 Model for Local Authorities of the United Kingdom (preprint)

We propose and describe a model for the COVID-19 epidemic of the United Kingdom at the level of local authorities. The model fits within a general framework for semi-mechanistic Bayesian models of the epidemic, with some important innovations: for example, we estimate the proportion of infections resulting in deaths and reported cases and we model the infections explicitly as random variables. The model is designed to be updated daily based on publicly available data. We envisage the model to be useful for short term projections of the epidemic over the next few weeks and to estimate past local values such as the reproduction number of the epidemic in the past. The model fits are available on a public website,https://imperialcollegelondon.github.io/covid19local. The model is currently being used by the Scottish government in their decisions on interventions within Scotland [1,issue 24 to now]

Age groups that sustain resurging COVID-19 epidemics in the United States (preprint)

Following initial declines, in mid 2020, a resurgence in transmission of novel coronavirus disease (COVID-19) has occurred in the United States and parts of Europe. Despite the wide implementation of non-pharmaceutical interventions, it is still not known how they are impacted by changing contact patterns, age and other demographics. As COVID-19 disease control becomes more localised, understanding the age demographics driving transmission and how these impacts the loosening of interventions such as school reopening is crucial. Considering dynamics for the United States, we analyse aggregated, age-specific mobility trends from more than 10 million individuals and link these mechanistically to age-specific COVID-19 mortality data. In contrast to previous approaches, we link mobility to mortality via age-specific contact patterns and use this rich relationship to reconstruct accurate transmission dynamics. Contrary to anecdotal evidence, we find little support for age-shifts in contact and transmission dynamics over time. We estimate that, until August, 63.4% [60.9%-65.5%] of SARS-CoV-2 infections in the United States originated from adults aged 20-49, while 1.2% [0.8%-1.8%] originated from children aged 0- 9. In areas with continued, community-wide transmission, our transmission model predicts that re-opening kindergartens and elementary schools could facilitate spread and lead to additional COVID-19 attributable deaths over a 90-day period. These findings indicate that targeting interventions to adults aged 20-49 are an important consideration in halting resurgent epidemics and preventing COVID-19-attributable deaths when kindergartens and elementary schools reopen.

A comparison of five epidemiological models for transmission of SARS-CoV-2 in India (preprint)

Many popular disease transmission models have helped nations respond to the COVID-19 pandemic by informing decisions about pandemic planning, resource allocation, implementation of social distancing measures and other non-pharmaceutical interventions. We study how five epidemiological models forecast and assess the course of the pandemic in India: a baseline model, an extended SIR (eSIR) model, two extended SEIR (SAPHIRE and SEIR-fansy) models, and a semi-mechanistic Bayesian hierarchical model (ICM). Using COVID-19 data for India from March 15 to June 18 to train the models, we generate predictions from each of the five models from June 19 to July 18. To compare prediction accuracy with respect to reported cumulative and active case counts and cumulative death counts, we compute the symmetric mean absolute prediction error (SMAPE) for each of the five models. For active case counts, SMAPE values are 0.72 (SEIR-fansy) and 33.83 (eSIR). For cumulative case counts, SMAPE values are 1.76 (baseline) 23.10 (eSIR), 2.07 (SAPHIRE) and 3.20 (SEIR-fansy). For cumulative death counts, the SMAPE values are 7.13 (SEIR-fansy) and 26.30 (eSIR). For cumulative cases and deaths, we compute Pearsons and Lins correlation coefficients to investigate how well the projected and observed reported COVID-counts agree. Three models (SAPHIRE, SEIR-fansy and ICM) return total (sum of reported and unreported) counts as well. We compute underreporting factors as of June 30 and note that the SEIR-fansy model reports the highest underreporting factor for active cases (6.10) and cumulative deaths (3.62), while the SAPHIRE model reports the highest underreporting factor for cumulative cases (27.79).