Search

Introduction: OpenAI's ChatGPT, a Large Language Model (LLM), is a powerful tool across domains, designed for text and code generation, fostering collaboration, especially in public health. Investigating the role of this advanced LLM chatbot in assisting public health practitioners in shaping disease transmission models to inform infection control strategies, marks a new era in infectious disease epidemiology research. This study used a case study to illustrate how ChatGPT collaborates with a public health practitioner in co-designing a mathematical transmission model. Methods: Using natural conversation, the practitioner initiated a dialogue involving an iterative process of code generation, refinement, and debugging with ChatGPT to develop a model to fit 10 days of prevalence data to estimate two key epidemiological parameters: i) basic reproductive number (Ro) and ii) final epidemic size. Verification and validation processes are conducted to ensure the accuracy and functionality of the final model. Results: ChatGPT developed a validated transmission model which replicated the epidemic curve and gave estimates of Ro of 4.19 (95 % CI: 4.13- 4.26) and a final epidemic size of 98.3 % of the population within 60 days. It highlighted the advantages of using maximum likelihood estimation with Poisson distribution over least squares method. Conclusion: Integration of LLM in medical research accelerates model development, reducing technical barriers for health practitioners, democratizing access to advanced modeling and potentially enhancing pandemic preparedness globally, particularly in resource-constrained populations.

The COVID-19 pandemic has highlighted the need to upgrade systems for infectious disease surveillance and forecasting and modeling of the spread of infection, both of which inform evidence-based public health guidance and policies. Here, we discuss requirements for an effective surveillance system to support decision making during a pandemic, drawing on the lessons of COVID-19 in the U.S., while looking to jurisdictions in the U.S. and beyond to learn lessons about the value of specific data types. In this report, we define the range of decisions for which surveillance data are required, the data elements needed to inform these decisions and to calibrate inputs and outputs of transmission-dynamic models, and the types of data needed to inform decisions by state, territorial, local, and tribal health authorities. We define actions needed to ensure that such data will be available and consider the contribution of such efforts to improving health equity.

To support the ongoing management of viral respiratory diseases while transitioning out of the acute phase of the COVID-19 pandemic, many countries are moving toward an integrated model of surveillance for SARS-CoV-2, influenza virus, and other respiratory pathogens. Although many surveillance approaches catalyzed by the COVID-19 pandemic provide novel epidemiologic insight, continuing them as implemented during the pandemic is unlikely to be feasible for nonemergency surveillance, and many have already been scaled back. Furthermore, given anticipated cocirculation of SARS-CoV-2 and influenza virus, surveillance activities in place before the pandemic require review and adjustment to ensure their ongoing value for public health. In this report, we highlight key challenges for the development of integrated models of surveillance. We discuss the relative strengths and limitations of different surveillance practices and studies as well as their contribution to epidemiologic assessment, forecasting, and public health decision-making.

Mathematical modelling has played an important role in offering informed advice during the COVID-19 pandemic. In England, a cross government and academia collaboration generated medium-term projections (MTPs) of possible epidemic trajectories over the future 4–6 weeks from a collection of epidemiological models. In this article, we outline this collaborative modelling approach and evaluate the accuracy of the combined and individual model projections against the data over the period November 2021–December 2022 when various Omicron subvariants were spreading across England. Using a number of statistical methods, we quantify the predictive performance of the model projections for both the combined and individual MTPs, by evaluating the point and probabilistic accuracy. Our results illustrate that the combined MTPs, produced from an ensemble of heterogeneous epidemiological models, were a closer fit to the data than the individual models during the periods of epidemic growth or decline, with the 90% confidence intervals widest around the epidemic peaks. We also show that the combined MTPs increase the robustness and reduce the biases associated with a single model projection. Learning from our experience of ensemble modelling during the COVID-19 epidemic, our findings highlight the importance of developing cross-institutional multi-model infectious disease hubs for future outbreak control.

Earlier global detection of novel SARS-CoV-2 variants gives governments more time to respond. However, few countries can implement timely national surveillance, resulting in gaps in monitoring. The United Kingdom implemented large-scale community and hospital surveillance, but experience suggests it might be faster to detect new variants through testing England arrivals for surveillance. We developed simulations of emergence and importation of novel variants with a range of infection hospitalization rates to the United Kingdom. We compared time taken to detect the variant though testing arrivals at England borders, hospital admissions, and the general community. We found that sampling 10%–50% of arrivals at England borders could confer a speed advantage of 3.5–6 weeks over existing community surveillance and 1.5–5 weeks (depending on infection hospitalization rates) over hospital testing. Directing limited global capacity for surveillance to highly connected ports could speed up global detection of novel SARS-CoV-2 variants.

The time-varying reproduction number R(t) measures the number of new infections per infectious individual and is closely correlated with the time series of infection incidence by definition. The timings of actual infections are rarely known, and analysis of epidemics usually relies on time series data for other outcomes such as symptom onset. A common implicit assumption, when estimating R(t) from an epidemic time series, is that R(t) has the same relationship with these downstream outcomes as it does with the time series of incidence. However, this assumption is unlikely to be valid given that most epidemic time series are not perfect proxies of incidence. Rather they represent convolutions of incidence with uncertain delay distributions. Here we define the apparent time-varying reproduction number, RA(t), the reproduction number calculated from a downstream epidemic time series and demonstrate how differences between RA(t) and R(t) depend on the convolution function. The mean of the convolution function sets a time offset between the two signals, whilst the variance of the convolution function introduces a relative distortion between them. We present the convolution functions of epidemic time series that were available during the SARS-CoV-2 pandemic. Infection prevalence, measured by random sampling studies, presents fewer biases than other epidemic time series. Here we show that additionally the mean and variance of its convolution function were similar to that obtained from traditional surveillance based on mass-testing and could be reduced using more frequent testing, or by using stricter thresholds for positivity. Infection prevalence studies continue to be a versatile tool for tracking the temporal trends of R(t), and with additional refinements to their study protocol, will be of even greater utility during any future epidemics or pandemics.

Abstract The value of SARS-CoV-2 lateral flow immunoassay (LFIA) tests for estimating individual disease risk is unclear. The REACT-2 study in England, UK, obtained self-administered SARS-CoV-2 LFIA test results from 361,801 adults in January-May 2021. Here, we link to routine data on subsequent hospitalisation (to September 2021), and death (to December 2021). Among those who had received one or more vaccines, a negative LFIA is associated with increased risk of hospitalisation with COVID-19 (HR: 2.73 [95% confidence interval: 1.15,6.48]), death (all-cause) (HR: 1.59, 95% CI:1.07, 2.37), and death with COVID-19 as underlying cause (20.6 [1.83,232]). For people designated at high risk from COVID-19, who had received one or more vaccines, there is an additional risk of all-cause mortality of 1.9 per 1000 for those testing antibody negative compared to positive. However, the LFIA does not provide substantial predictive information over and above that which is available from detailed sociodemographic and health-related variables. Nonetheless, this simple test provides a marker which could be a valuable addition to understanding population and individual-level risk.

[This corrects the article DOI: 10.1371/journal.pmed.1003793.].

This study presents data from the REACT-1 SARS-CoV-2 community sampling study in England from November 2021 to March 2022. They show that the Omicron variant peaked in January with a prevalence of ~7% and that the BA.2 sublineage had a 1.5x higher reproduction number compared to other Omicron sublineages.

Abstract Background Since the emergence of SARS-CoV-2, evolutionary pressure has driven large increases in the transmissibility of the virus. However, with increasing levels of immunity through vaccination and natural infection the evolutionary pressure will switch towards immune escape. Genomic surveillance in regions of high immunity is crucial in detecting emerging variants that can more successfully navigate the immune landscape. Methods We present phylogenetic relationships and lineage dynamics within England (a country with high levels of immunity), as inferred from a random community sample of individuals who provided a self-administered throat and nose swab for rt-PCR testing as part of the REal-time Assessment of Community Transmission-1 (REACT-1) study. During round 14 (9 September–27 September 2021) and 15 (19 October–5 November 2021) lineages were determined for 1322 positive individuals, with 27.1% of those which reported their symptom status reporting no symptoms in the previous month. Results We identified 44 unique lineages, all of which were Delta or Delta sub-lineages, and found a reduction in their mutation rate over the study period. The proportion of the Delta sub-lineage AY.4.2 was increasing, with a reproduction number 15% (95% CI 8–23%) greater than the most prevalent lineage, AY.4. Further, AY.4.2 was less associated with the most predictive COVID-19 symptoms (p = 0.029) and had a reduced mutation rate (p = 0.050). Both AY.4.2 and AY.4 were found to be geographically clustered in September but this was no longer the case by late October/early November, with only the lineage AY.6 exhibiting clustering towards the South of England. Conclusions As SARS-CoV-2 moves towards endemicity and new variants emerge, genomic data obtained from random community samples can augment routine surveillance data without the potential biases introduced due to higher sampling rates of symptomatic individuals.

The relationship between prevalence of infection and severe outcomes such as hospitalisation and death changed over the course of the COVID-19 pandemic. Reliable estimates of the infection fatality ratio (IFR) and infection hospitalisation ratio (IHR) along with the time-delay between infection and hospitalisation/death can inform forecasts of the numbers/timing of severe outcomes and allow healthcare services to better prepare for periods of increased demand. The REal-time Assessment of Community Transmission-1 (REACT-1) study estimated swab positivity for Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infection in England approximately monthly from May 2020 to March 2022. Here, we analyse the changing relationship between prevalence of swab positivity and the IFR and IHR over this period in England, using publicly available data for the daily number of deaths and hospitalisations, REACT-1 swab positivity data, time-delay models, and Bayesian P-spline models. We analyse data for all age groups together, as well as in 2 subgroups: those aged 65 and over and those aged 64 and under. Additionally, we analysed the relationship between swab positivity and daily case numbers to estimate the case ascertainment rate of England's mass testing programme. During 2020, we estimated the IFR to be 0.67% and the IHR to be 2.6%. By late 2021/early 2022, the IFR and IHR had both decreased to 0.097% and 0.76%, respectively. The average case ascertainment rate over the entire duration of the study was estimated to be 36.1%, but there was some significant variation in continuous estimates of the case ascertainment rate. Continuous estimates of the IFR and IHR of the virus were observed to increase during the periods of Alpha and Delta's emergence. During periods of vaccination rollout, and the emergence of the Omicron variant, the IFR and IHR decreased. During 2020, we estimated a time-lag of 19 days between hospitalisation and swab positivity, and 26 days between deaths and swab positivity. By late 2021/early 2022, these time-lags had decreased to 7 days for hospitalisations and 18 days for deaths. Even though many populations have high levels of immunity to SARS-CoV-2 from vaccination and natural infection, waning of immunity and variant emergence will continue to be an upwards pressure on the IHR and IFR. As investments in community surveillance of SARS-CoV-2 infection are scaled back, alternative methods are required to accurately track the ever-changing relationship between infection, hospitalisation, and death and hence provide vital information for healthcare provision and utilisation.

Brazeau et al. use a statistical modelling approach to estimate COVID-19 infection fatality ratios from seroprevalence data. The authors’ model accounts for seroreversion over the course of the pandemic, as well as other important uncertainties such as serologic test characteristics.

This study characterises Long COVID using data from the REACT-2 community-based study in England. It estimates that 38% (in autumn/winter 2020/21) and 22% (in spring 2021) of people reported at least one symptom 12 weeks after symptom onset; identifies risk factors for persistent symptoms; and finds evidence of symptom clustering.

Since 8th March 2020 up to the time of writing, we have been producing near real-time weekly estimates of SARS-CoV-2 transmissibility and forecasts of deaths due to COVID-19 for all countries with evidence of sustained transmission, shared online. We also developed a novel heuristic to combine weekly estimates of transmissibility to produce forecasts over a 4-week horizon. Here we present a retrospective evaluation of the forecasts produced between 8th March to 29th November 2020 for 81 countries. We evaluated the robustness of the forecasts produced in real-time 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. The retrospective evaluation of our models shows that simple transmission models calibrated using routine disease surveillance data can reliably capture the epidemic trajectory in multiple countries. 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.

The authors present results from the REACT-2 study, a series of cross-sectional community surveys during the SARS-CoV-2 pandemic in England. They measure antibodies by self-administered lateral flow tests and describe antibody positivity by time since vaccination, age, sex, co-morbidities, infection history, and vaccine type.

Introduction: Two years into the coronavirus 2019 (COVID-19) pandemic, populations with less built-up immunity continued to devise ways to optimize social distancing measures (SDMs) relaxation levels for outbreaks triggered by SARS-CoV-2 and its variants to resume minimal economics activities while avoiding hospital system collapse. Method: An age-stratified compartmental model featuring social mixing patterns was first fitted the incidence data in second wave in Hong Kong. Hypothetical scenario analysis was conducted by varying population mobility and vaccination coverages (VCs) to predict the number of hospital and intensive-care unit admissions in outbreaks initiated by ancestral strain and its variants (Alpha, Beta, Gamma, Delta and Omicron). Scenarios were “unsustainable” if either of admissions was larger than the maximum of its occupancy. Results: At VC of 65%, scenarios of full SDMs relaxation (mean daily social encounters prior to COVID-19 pandemic = 14.1 contacts) for outbreaks triggered by ancestral strain, Alpha and Beta were sustainable. Restricting levels of SDMs was required such that the optimal population mobility had to be reduced to 0.9, 0.65 and 0.37 for Gamma, Delta and Omicron associated outbreaks respectively. VC improvement from 65% to 75% and 95% allowed complete SDMs relaxation in Gamma-, and Delta-driven epidemic respectively. However, this was not supported for Omicron-triggered epidemic. Discussion: To seek a path to normality, speedy vaccine and booster distribution to the majority across all age groups is the first step. Gradual or complete SDMs lift could be considered if the hybrid immunity could be achieved due to high vaccination coverage and natural infection rate among vaccinated or the COVID-19 case fatality rate could be reduced similar to that for seasonal influenza to secure hospital system sustainability.

Background: Over a life course, human adaptive immunity to antigenically mutable pathogens exhibits competitive and facilitative interactions. We hypothesize that such interactions may lead to cyclic dynamics in immune responses over a lifetime. Methods: To investigate the cyclic behavior, we analyzed hemagglutination inhibition titers against 21 historical influenza A(H3N2) strains spanning 47 years from a cohort in Guangzhou, China, and applied Fourier spectrum analysis. To investigate possible biological mechanisms, we simulated individual antibody profiles encompassing known feedbacks and interactions due to generally recognized immunological mechanisms. Results: We demonstrated a long-term periodicity (about 24 years) in individual antibody responses. The reported cycles were robust to analytic and sampling approaches. Simulations suggested that individual-level cross-reaction between antigenically similar strains likely explains the reported cycle. We showed that the reported cycles are predictable at both individual and birth cohort level and that cohorts show a diversity of phases of these cycles. Phase of cycle was associated with the risk of seroconversion to circulating strains, after accounting for age and pre-existing titers of the circulating strains. Conclusions: Our findings reveal the existence of long-term periodicities in individual antibody responses to A(H3N2). We hypothesize that these cycles are driven by preexisting antibody responses blunting responses to antigenically similar pathogens (by preventing infection and/or robust antibody responses upon infection), leading to reductions in antigen-specific responses over time until individual’s increasing risk leads to an infection with an antigenically distant enough virus to generate a robust immune response. These findings could help disentangle cohort effects from individual-level exposure histories, improve our understanding of observed heterogeneous antibody responses to immunizations, and inform targeted vaccine strategy. Funding: This study was supported by grants from the NIH R56AG048075 (DATC, JL), NIH R01AI114703 (DATC, BY), the Wellcome Trust 200861/Z/16/Z (SR), and 200187/Z/15/Z (SR). This work was also supported by research grants from Guangdong Government HZQB-KCZYZ-2021014 and 2019B121205009 (YG and HZ). DATC, JMR and SR acknowledge support from the National Institutes of Health Fogarty Institute (R01TW0008246). JMR acknowledges support from the Medical Research Council (MR/S004793/1) and the Engineering and Physical Sciences Research Council (EP/N014499/1). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

BackgroundFollowing rapidly rising COVID-19 case numbers, England entered a national lockdown on 6 January 2021, with staged relaxations of restrictions from 8 March 2021 onwards.AimWe characterise how the lockdown and subsequent easing of restrictions affected trends in SARS-CoV-2 infection prevalence.MethodsOn average, risk of infection is proportional to infection prevalence. The REal-time Assessment of Community Transmission-1 (REACT-1) study is a repeat cross-sectional study of over 98,000 people every round (rounds approximately monthly) that estimates infection prevalence in England. We used Bayesian P-splines to estimate prevalence and the time-varying reproduction number (Rt) nationally, regionally and by age group from round 8 (beginning 6 January 2021) to round 13 (ending 12 July 2021) of REACT-1. As a comparator, a separate segmented-exponential model was used to quantify the impact on Rt of each relaxation of restrictions.ResultsFollowing an initial plateau of 1.54% until mid-January, infection prevalence decreased until 13 May when it reached a minimum of 0.09%, before increasing until the end of the study to 0.76%. Following the first easing of restrictions, which included schools reopening, the reproduction number Rt increased by 82% (55%, 108%), but then decreased by 61% (82%, 53%) at the second easing of restrictions, which was timed to match the Easter school holidays. Following further relaxations of restrictions, the observed Rt increased steadily, though the increase due to these restrictions being relaxed was offset by the effects of vaccination and also affected by the rapid rise of Delta. There was a high degree of synchrony in the temporal patterns of prevalence between regions and age groups.ConclusionHigh-resolution prevalence data fitted to P-splines allowed us to show that the lockdown was effective at reducing risk of infection with school holidays/closures playing a significant part.

Abstract SARS-CoV-2 infections have been reported in all age groups including infants, children, and adolescents. However, the role of children in the COVID-19 pandemic is still uncertain. This systematic review of early studies synthesises evidence on the susceptibility of children to SARS-CoV-2 infection, the severity and clinical outcomes in children with SARS-CoV-2 infection, and the transmissibility of SARS-CoV-2 by children in the initial phases of the COVID-19 pandemic. A systematic literature review was conducted in PubMed. Reviewers extracted data from relevant, peer-reviewed studies published up to July 4th 2020 during the first wave of the SARS-CoV-2 outbreak using a standardised form and assessed quality using the NIH Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies. For studies included in the meta-analysis, we used a random effects model to calculate pooled estimates of the proportion of children considered asymptomatic or in a severe or critical state. We identified 2775 potential studies of which 128 studies met our inclusion criteria; data were extracted from 99, which were then quality assessed. Finally, 29 studies were considered for the meta-analysis that included information of symptoms and/or severity, these were further assessed based on patient recruitment. Our pooled estimate of the proportion of test positive children who were asymptomatic was 21.1% (95% CI: 14.0–28.1%), based on 13 included studies, and the proportion of children with severe or critical symptoms was 3.8% (95% CI: 1.5–6.0%), based on 14 included studies. We did not identify any studies designed to assess transmissibility in children and found that susceptibility to infection in children was highly variable across studies. Children’s susceptibility to infection and onward transmissibility relative to adults is still unclear and varied widely between studies. However, it is evident that most children experience clinically mild disease or remain asymptomatically infected. More comprehensive contact-tracing studies combined with serosurveys are needed to quantify children’s transmissibility relative to adults. With children back in schools, testing regimes and study protocols that will allow us to better understand the role of children in this pandemic are critical.

The time-varying reproduction number (Rt) can change rapidly over the course of a pandemic due to changing restrictions, behaviours, and levels of population immunity. Many methods exist that allow the estimation of Rtfrom case data. However, these are not easily adapted to point prevalence data nor can they infer Rtacross periods of missing data. We developed a Bayesian P-spline model suitable for fitting to a wide range of epidemic time-series, including point-prevalence data. We demonstrate the utility of the model by fitting to periodic daily SARS-CoV-2 swab-positivity data in England from the first 7 rounds (May 2020–December 2020) of the REal-time Assessment of Community Transmission-1 (REACT-1) study. Estimates of Rtover the period of two subsequent rounds (6–8 weeks) and single rounds (2–3 weeks) inferred using the Bayesian P-spline model were broadly consistent with estimates from a simple exponential model, with overlapping credible intervals. However, there were sometimes substantial differences in point estimates. The Bayesian P-spline model was further able to infer changes in Rtover shorter periods tracking a temporary increase above one during late-May 2020, a gradual increase in Rtover the summer of 2020 as restrictions were eased, and a reduction in Rtduring England’s second national lockdown followed by an increase as the Alpha variant surged. The model is robust against both under-fitting and over-fitting and is able to interpolate between periods of available data; it is a particularly versatile model when growth rate can change over small timescales, as in the current SARS-CoV-2 pandemic. This work highlights the importance of pairing robust methods with representative samples to track pandemics.

Background. Even with good progress on vaccination, SARS-CoV-2 infections in the UK may continue to impose a high burden of disease and therefore pose substantial challenges for health policy decision makers. Stringent government-mandated physical distancing measures (lockdown) have been demonstrated to be epidemiologically effective, but can have both positive and negative economic consequences. The duration and frequency of any intervention policy could, in theory, be optimized to maximize economic benefits while achieving substantial reductions in disease. Methods. Here, we use a pre-existing SARS-CoV-2 transmission model to assess the health and economic implications of different strengths of control through time in order to identify optimal approaches to non-pharmaceutical intervention stringency in the UK, considering the role of vaccination in reducing the need for future physical distancing measures. The model is calibrated to the COVID-19 epidemic in England and we carry out retrospective analysis of the optimal timing of precautionary breaks in 2020 and the optimal relaxation policy from the January 2021 lockdown, considering the willingness to pay (WTP) for health improvement. Results. We find that the precise timing and intensity of interventions is highly dependent upon the objective of control. As intervention measures are relaxed, we predict a resurgence in cases, but the optimal intervention policy can be established dependent upon the WTP per quality adjusted life year loss avoided. Our results show that establishing an optimal level of control can result in a reduction in net monetary loss of billions of pounds, dependent upon the precise WTP value. Conclusion. It is vital, as the UK emerges from lockdown, but continues to face an on-going pandemic, to accurately establish the overall health and economic costs when making policy decisions. We demonstrate how some of these can be quantified, employing mechanistic infectious disease transmission models to establish optimal levels of control for the ongoing COVID-19 pandemic.

To define appropriate planning scenarios for future pandemics of respiratory pathogens, it is important to understand the initial transmission dynamics of COVID-19 during 2020. Here, we fit an age-stratified compartmental model with a flexible underlying transmission term to daily COVID-19 death data from states in the contiguous U.S. and to national and sub-national data from around the world. The daily death data of the first months of the COVID-19 pandemic was qualitatively categorized into one of four main profile types: "spring single-peak", "summer single-peak", "spring/summer two-peak" and "broad with shoulder". We estimated a reproduction number R as a function of calendar time tc and as a function of time since the first death reported in that population (local pandemic time, tp). Contrary to the diversity of categories and range of magnitudes in death incidence profiles, the R(tp) profiles were much more homogeneous. We found that in both the contiguous U.S. and globally, the initial value of both R(tc) and R(tp) was substantial: at or above two. However, during the early months, pandemic time R(tp) decreased exponentially to a value that hovered around one. This decrease was accompanied by a reduction in the variance of R(tp). For calendar time R(tc), the decrease in magnitude was slower and non-exponential, with a smaller reduction in variance. Intriguingly, similar trends of exponential decrease and reduced variance were not observed in raw death data. Our findings suggest that the combination of specific government responses and spontaneous changes in behaviour ensured that transmissibility dropped, rather than remaining constant, during the initial phases of a pandemic. Future pandemic planning scenarios should include models that assume similar decreases in transmissibility, which lead to longer epidemics with lower peaks when compared with models based on constant transmissibility.

Background: Lateral flow immunoassays (LFIAs) are able to achieve affordable, large scale antibody testing and provide rapid results without the support of central laboratories. As part of the development of the REACT programme extensive evaluation of LFIA performance was undertaken with individuals following natural infection. Here we assess the performance of the selected LFIA to detect antibody responses in individuals who have received at least one dose of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccine. Methods: This was a prospective diagnostic accuracy study. Sampling was carried out at renal outpatient clinic and healthcare worker testing sites at Imperial College London NHS Trust. Two cohorts of patients were recruited; the first was a cohort of 108 renal transplant patients attending clinic following two doses of SARS-CoV-2 vaccine, the second cohort comprised 40 healthcare workers attending for first SARS-CoV-2 vaccination and subsequent follow up. During the participants visit, finger-prick blood samples were analysed on LFIA device, while paired venous sampling was sent for serological assessment of antibodies to the spike protein (anti-S) antibodies. Anti-S IgG was detected using the Abbott Architect SARS-CoV-2 IgG Quant II CMIA. A total of 186 paired samples were collected. The accuracy of Fortress LFIA in detecting IgG antibodies to SARS-CoV-2 compared to anti-spike protein detection on Abbott Assay Results: The LFIA had an estimated sensitivity of 92.0% (114/124; 95% confidence interval [CI] 85.7% to 96.1%) and specificity of 93.6% (58/62; 95% CI 84.3% to 98.2%) using the Abbott assay as reference standard (using the threshold for positivity of 7.10 BAU/ml) Conclusions: Fortress LFIA performs well in the detection of antibody responses for intended purpose of population level surveillance but does not meet criteria for individual testing.

REACT-2 is a large-scale community study of SARS-CoV-2 seroprevalence in England. Here, the authors estimate that 6% of adults in England had been infected by mid-July 2020, with health and long-term care workers and those of Black or South Asian ethnicity disproportionately affected.

Social distancing policies aiming to reduce COVID-19 transmission have been reflected in reductions in human mobility. Here, the authors show that reduced mobility is correlated with decreased transmission, but that this relationship weakened over time as social distancing measures were relaxed.

Objectives: In this data collation study, we aimed to provide a comprehensive database describing the epidemic trends and responses during the first wave of coronavirus disease 2019 (COVID-19) throughout the main provinces in China. Methods: From mid-January to March 2020, we extracted publicly available data regarding the spread and control of COVID-19 from 31 provincial health authorities and major media outlets in mainland China. Based on these data, we conducted descriptive analyses of the epidemic in the six most-affected provinces. Results: School closures, travel restrictions, community-level lockdown, and contact tracing were introduced concurrently around late January but subsequent epidemic trends differed among provinces. Compared with Hubei, the other five most-affected provinces reported a lower crude case fatality ratio and proportion of critical and severe hospitalised cases. From March 2020, as the local transmission of COVID-19 declined, switching the focus of measures to the testing and quarantine of inbound travellers may have helped to sustain the control of the epidemic. Conclusions: Aggregated indicators of case notifications and severity distributions are essential for monitoring an epidemic. A publicly available database containing these indicators and information regarding control measures is a useful resource for further research and policy planning in response to the COVID-19 epidemic.

High numbers of COVID-19-related deaths have been reported in the United States, but estimation of the true numbers of infections is challenging. Here, the authors estimate that on 1 June 2020, 3.7% of the US population was infected with SARS-CoV-2, and 0.01% was infectious, with wide variation by state.

Abstract Background After experiencing a sharp growth in COVID-19 cases early in the pandemic, South Korea rapidly controlled transmission while implementing less stringent national social distancing measures than countries in Europe and the USA. This has led to substantial interest in their “test, trace, isolate” strategy. However, it is important to understand the epidemiological peculiarities of South Korea’s outbreak and characterise their response before attempting to emulate these measures elsewhere. Methods We systematically extracted numbers of suspected cases tested, PCR-confirmed cases, deaths, isolated confirmed cases, and numbers of confirmed cases with an identified epidemiological link from publicly available data. We estimated the time-varying reproduction number, R t , using an established Bayesian framework, and reviewed the package of interventions implemented by South Korea using our extracted data, plus published literature and government sources. Results We estimated that after the initial rapid growth in cases, R t dropped below one in early April before increasing to a maximum of 1.94 (95%CrI, 1.64–2.27) in May following outbreaks in Seoul Metropolitan Region. By mid-June, R t was back below one where it remained until the end of our study (July 13th). Despite less stringent “lockdown” measures, strong social distancing measures were implemented in high-incidence areas and studies measured a considerable national decrease in movement in late February. Testing the capacity was swiftly increased, and protocols were in place to isolate suspected and confirmed cases quickly; however, we could not estimate the delay to isolation using our data. Accounting for just 10% of cases, individual case-based contact tracing picked up a relatively minor proportion of total cases, with cluster investigations accounting for 66%. Conclusions Whilst early adoption of testing and contact tracing is likely to be important for South Korea’s successful outbreak control, other factors including regional implementation of strong social distancing measures likely also contributed. The high volume of testing and the low number of deaths suggest that South Korea experienced a small epidemic relative to other countries. Caution is needed in attempting to replicate the South Korean response in populations with larger more geographically widespread epidemics where finding, testing, and isolating cases that are linked to clusters may be more difficult.

BackgroundThe importance of infectious disease epidemic forecasting and prediction research is underscored by decades of communicable disease outbreaks, including COVID-19. Unlike other fields of medical research, such as clinical trials and systematic reviews, no reporting guidelines exist for reporting epidemic forecasting and prediction research despite their utility. We therefore developed the EPIFORGE checklist, a guideline for standardized reporting of epidemic forecasting research.Methods and findingsWe developed this checklist using a best-practice process for development of reporting guidelines, involving a Delphi process and broad consultation with an international panel of infectious disease modelers and model end users. The objectives of these guidelines are to improve the consistency, reproducibility, comparability, and quality of epidemic forecasting reporting. The guidelines are not designed to advise scientists on how to perform epidemic forecasting and prediction research, but rather to serve as a standard for reporting critical methodological details of such studies.ConclusionsThese guidelines have been submitted to the EQUATOR network, in addition to hosting by other dedicated webpages to facilitate feedback and journal endorsement.

We measured contact patterns using online diaries for 304 employees of 3 U.S. companies working remotely. The median number of daily contacts was 2 (IQR 1-4); majority were conversation (55 %), occurred at home (64 %) and lasted >4 h (38 %). These data are crucial for modeling outbreak control among the workforces.

Background: As of August 2021, every region of the world has been affected by the COVID-19 pandemic, with more than 196,000,000 cases worldwide. Methods: We analysed COVID-19 cases among travellers from mainland China to different regions and countries, comparing the region- and country-specific rates of detected and confirmed cases per flight volume to estimate the relative sensitivity of surveillance in different regions and countries. Results: Although travel restrictions from Wuhan City and other cities across China may have reduced the absolute number of travellers to and from China, we estimated that up to 70% (95% CI: 54% - 80%) of imported cases could remain undetected relative to the sensitivity of surveillance in Singapore. The percentage of undetected imported cases rises to 75% (95% CI 66% - 82%) when comparing to the surveillance sensitivity in multiple countries. Conclusions: Our analysis shows that a large number of COVID-19 cases remain undetected across the world. These undetected cases potentially resulted in multiple chains of human-to-human transmission outside mainland China.

BackgroundRapid detection, isolation, and contact tracing of community COVID-19 cases are essential measures to limit the community spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). We aimed to identify a parsimonious set of symptoms that jointly predict COVID-19 and investigated whether predictive symptoms differ between the B.1.1.7 (Alpha) lineage (predominating as of April 2021 in the US, UK, and elsewhere) and wild type.Methods and findingsWe obtained throat and nose swabs with valid SARS-CoV-2 PCR test results from 1,147,370 volunteers aged 5 years and above (6,450 positive cases) in the REal-time Assessment of Community Transmission-1 (REACT-1) study. This study involved repeated community-based random surveys of prevalence in England (study rounds 2 to 8, June 2020 to January 2021, response rates 22%-27%). Participants were asked about symptoms occurring in the week prior to testing. Viral genome sequencing was carried out for PCR-positive samples with N-gene cycle threshold value < 34 (N = 1,079) in round 8 (January 2021). In univariate analysis, all 26 surveyed symptoms were associated with PCR positivity compared with non-symptomatic people. Stability selection (1,000 penalized logistic regression models with 50% subsampling) among people reporting at least 1 symptom identified 7 symptoms as jointly and positively predictive of PCR positivity in rounds 2-7 (June to December 2020): loss or change of sense of smell, loss or change of sense of taste, fever, new persistent cough, chills, appetite loss, and muscle aches. The resulting model (rounds 2-7) predicted PCR positivity in round 8 with area under the curve (AUC) of 0.77. The same 7 symptoms were selected as jointly predictive of B.1.1.7 infection in round 8, although when comparing B.1.1.7 with wild type, new persistent cough and sore throat were more predictive of B.1.1.7 infection while loss or change of sense of smell was more predictive of the wild type. The main limitations of our study are (i) potential participation bias despite random sampling of named individuals from the National Health Service register and weighting designed to achieve a representative sample of the population of England and (ii) the necessary reliance on self-reported symptoms, which may be prone to recall bias and may therefore lead to biased estimates of symptom prevalence in England.ConclusionsWhere testing capacity is limited, it is important to use tests in the most efficient way possible. We identified a set of 7 symptoms that, when considered together, maximize detection of COVID-19 in the community, including infection with the B.1.1.7 lineage.

Influenza incidence forecasting is used to facilitate better health system planning and could potentially be used to allow at-risk individuals to modify their behavior during a severe seasonal influenza epidemic or a novel respiratory pandemic. For example, the US Centers for Disease Control and Prevention (CDC) runs an annual competition to forecast influenza-like illness (ILI) at the regional and national levels in the US, based on a standard discretized incidence scale. Here, we use a suite of forecasting models to analyze type-specific incidence at the smaller spatial scale of clusters of nearby counties. We used data from point-of-care (POC) diagnostic machines over three seasons, in 10 clusters, capturing: 57 counties; 1,061,891 total specimens; and 173,909 specimens positive for Influenza A. Total specimens were closely correlated with comparable CDC ILI data. Mechanistic models were substantially more accurate when forecasting influenza A positive POC data than total specimen POC data, especially at longer lead times. Also, models that fit subpopulations of the cluster (individual counties) separately were better able to forecast clusters than were models that directly fit to aggregated cluster data. Public health authorities may wish to consider developing forecasting pipelines for type-specific POC data in addition to ILI data. Simple mechanistic models will likely improve forecast accuracy when applied at small spatial scales to pathogen-specific data before being scaled to larger geographical units and broader syndromic data. Highly local forecasts may enable new public health messaging to encourage at-risk individuals to temporarily reduce their social mixing during seasonal peaks and guide public health intervention policy during potentially severe novel influenza pandemics.

Background: The time-concentrated nature of the first wave of the COVID-19 epidemic in England in March and April 2020 provides a natural experiment to measure changes in antibody positivity at the population level before onset of the second wave and initiation of the vaccination programme. Methods: Three cross-sectional national surveys with non-overlapping random samples of the population in England undertaken between late June and September 2020 (REACT-2 study). 365,104 adults completed questionnaires and self-administered lateral flow immunoassay (LFIA) tests for IgG against SARS-CoV-2. Findings: Overall, 17,576 people had detectable antibodies, a prevalence of 4.9% (95% confidence intervals 4.9, 5.0) when adjusted for test characteristics and weighted to the adult population of England. The prevalence declined from 6.0% (5.8, 6.1), to 4.8% (4.7, 5.0) and 4.4% (4.3, 4.5), over the three rounds of the study a difference of -26.5% (-29.0, -23.8). The highest prevalence and smallest overall decline in positivity was in the youngest age group (18-24 years) at -14.9% (-21.6, -8.1), and lowest prevalence and largest decline in the oldest group (>74 years) at -39.0% (-50.8, -27.2). The decline from June to September 2020 was largest in those who did not report a history of COVID-19 at -64.0% (-75.6, -52.3), compared to -22.3% (-27.0, -17.7) in those with SARS-CoV-2 infection confirmed on PCR. Interpretation: A large proportion of the population remained susceptible to SARS-CoV-2 infection in England based on naturally acquired immunity from the first wave. Widespread vaccination is needed to confer immunity and control the epidemic at population level. Funding: This work was funded by the Department of Health and Social Care in England.

Background: England, UK has one of the highest rates of confirmed COVID-19 mortality globally. Until recently, testing for the SARS-CoV-2 virus focused mainly on healthcare and care home settings. As such, there is far less understanding of community transmission. Protocol: The REal-time Assessment of Community Transmission (REACT) programme is a major programme of home testing for COVID-19 to track progress of the infection in the community. REACT-1 involves cross-sectional surveys of viral detection (virological swab for RT-PCR) tests in repeated samples of 100,000 to 150,000 randomly selected individuals across England. This examines how widely the virus has spread and how many people are currently infected. The age range is 5 years and above. Individuals are sampled from the England NHS patient list. REACT-2 is a series of five sub-studies towards establishing the seroprevalence of antibodies to SARS-CoV-2 in England as an indicator of historical infection. The main study (study 5) uses the same design and sampling approach as REACT-1 using a self-administered lateral flow immunoassay (LFIA) test for IgG antibodies in repeated samples of 100,000 to 200,000 adults aged 18 years and above. To inform study 5, studies 1-4 evaluate performance characteristics of SARS-CoV-2 LFIAs (study 1) and different aspects of feasibility, usability and application of LFIAs for home-based testing in different populations (studies 2-4). Ethics and dissemination: The study has ethical approval. Results are reported using STROBE guidelines and disseminated through reports to public health bodies, presentations at scientific meetings and open access publications. Conclusions: This study provides robust estimates of the prevalence of both virus (RT-PCR, REACT-1) and seroprevalence (antibody, REACT-2) in the general population in England. We also explore acceptability and usability of LFIAs for self-administered testing for SARS-CoV-2 antibody in a home-based setting, not done before at such scale in the general population.

The emergence and reemergence of coronavirus epidemics sparked renewed concerns from global epidemiology researchers and public health administrators. Mathematical models that represented how contact tracing and follow-up may control Severe Acute Respiratory Syndrome (SARS) and Middle East Respiratory Syndrome (MERS) transmissions were developed for evaluating different infection control interventions, estimating likely number of infections as well as facilitating understanding of their likely epidemiology. We reviewed mathematical models for contact tracing and follow-up control measures of SARS and MERS transmission. Model characteristics, epidemiological parameters and intervention parameters used in the mathematical models from seven studies were summarized. A major concern identified in future epidemics is whether public health administrators can collect all the required data for building epidemiological models in a short period of time during the early phase of an outbreak. Also, currently available models do not explicitly model constrained resources. We urge for closed-loop communication between public health administrators and modelling researchers to come up with guidelines to delineate the collection of the required data in the midst of an outbreak and the inclusion of additional logistical details in future similar models. Keywords: Contact Tracing, Coronavirus Epidemics, Transmission Modelling, SARS, MERS

When analysing in vitro data, growth kinetics of influenza virus strains are often compared by computing their growth rates, which are sometimes used as proxies for fitness. However, analogous to mathematical models for epidemics, the growth rate can be defined as a function of mechanistic traits: the basic reproduction number (the average number of cells each infected cell infects) and the mean generation time (the average length of a replication cycle). Fitting a model to previously published and newly generated data from experiments in human lung cells, we compared estimates of growth rate, reproduction number and generation time for six influenza A strains. Of four strains in previously published data, A/Canada/RV733/2003 (seasonal H1N1) had the lowest basic reproduction number, followed by A/Mexico/INDRE4487/2009 (pandemic H1N1), then A/Indonesia/05/2005 (spill-over H5N1) and A/Anhui/1/2013 (spill-over H7N9). This ordering of strains was preserved for both generation time and growth rate, suggesting a positive biological correlation between these quantities which have not been previously observed. We further investigated these potential correlations using data from reassortant viruses with different internal proteins (from A/England/195/2009 (pandemic H1N1) and A/Turkey/05/2005 (H5N1)), and the same surface proteins (from A/Puerto Rico/8/34 (lab-adapted H1N1)). Similar correlations between traits were observed for these viruses, confirming our initial findings and suggesting that these patterns were related to the degree of human adaptation of internal genes. Also, the model predicted that strains with a smaller basic reproduction number, shorter generation time and slower growth rate underwent more replication cycles by the time of peak viral load, potentially accumulating mutations more quickly. These results illustrate the utility of mathematical models in inferring traits driving observed differences in in vitro growth of influenza strains.

Background: The COVID-19 epidemic was declared a Global Pandemic by WHO on 11 March 2020. By 24 March 2020, over 440,000 cases and almost 20,000 deaths had been reported worldwide. In response to the fast-growing epidemic, which began in the Chinese city of Wuhan, Hubei, China imposed strict social distancing in Wuhan on 23 January 2020 followed closely by similar measures in other provinces. These interventions have impacted economic productivity in China, and the ability of the Chinese economy to resume without restarting the epidemic was not clear. Methods: Using daily reported cases from mainland China and Hong Kong SAR, we estimated transmissibility over time and compared it to daily within-city movement, as a proxy for economic activity. Results: Initially, within-city movement and transmission were very strongly correlated in the five mainland provinces most affected by the epidemic and Beijing. However, that correlation decreased rapidly after the initial sharp fall in transmissibility. In general, towards the end of the study period, the correlation was no longer apparent, despite substantial increases in within-city movement. A similar analysis for Hong Kong shows that intermediate levels of local activity were maintained while avoiding a large outbreak. At the very end of the study period, when China began to experience the re-introduction of a small number of cases from Europe and the United States, there is an apparent up-tick in transmission. Conclusions: Although these results do not preclude future substantial increases in incidence, they suggest that after very intense social distancing (which resulted in containment), China successfully exited its lockdown to some degree. Elsewhere, movement data are being used as proxies for economic activity to assess the impact of interventions. The results presented here illustrate how the eventual decorrelation between transmission and movement is likely a key feature of successful COVID-19 exit strategies.

Background: Since early March 2020, the COVID-19 epidemic across the United Kingdom has led to a range of social distancing policies, which have resulted in reduced mobility across different regions. Crowd level data on mobile phone usage can be used as a proxy for actual population mobility patterns and provide a way of quantifying the impact of social distancing measures on changes in mobility. Methods: Here, we use two mobile phone-based datasets (anonymised and aggregated crowd level data from O2 and from the Facebook app on mobile phones) to assess changes in average mobility, both overall and broken down into high and low population density areas, and changes in the distribution of journey lengths. Results: We show that there was a substantial overall reduction in mobility, with the most rapid decline on the 24th March 2020, the day after the Prime Minister’s announcement of an enforced lockdown. The reduction in mobility was highly synchronized across the UK. Although mobility has remained low since 26th March 2020, we detect a gradual increase since that time. We also show that the two different datasets produce similar trends, albeit with some location-specific differences. We see slightly larger reductions in average mobility in high-density areas than in low-density areas, with greater variation in mobility in the high-density areas: some high-density areas eliminated almost all mobility. Conclusions: These analyses form a baseline from which to observe changes in behaviour in the UK as social distancing is eased and inform policy towards the future control of SARS-CoV-2 in the UK.

Complex exposure histories and immune mediated interactions between influenza strains contribute to the life course of human immunity to influenza. Antibody profiles can be generated by characterizing immune responses to multiple antigenically variant strains, but how these profiles vary across individuals and determine future responses is unclear. We used hemagglutination inhibition titers from 21 H3N2 strains to construct 777 paired antibody profiles from people aged 2 to 86, and developed novel metrics to capture features of these profiles. Total antibody titer per potential influenza exposure increases in early life, then decreases in middle age. Increased titers to one or more strains were seen in 97.8% of participants during a roughly four-year interval, suggesting widespread influenza exposure. While titer changes were seen to all strains, recently circulating strains exhibited the greatest titer rise. Higher pre-existing, homologous titers at baseline reduced the risk of seroconversion to recent strains. After adjusting for homologous titer, we also found an increased frequency of seroconversion against recent strains among those with higher immunity to older previously exposed strains. Including immunity to previously exposures also improved the deviance explained by the models. Our results suggest that a comprehensive quantitative description of immunity encompassing past exposures could lead to improved correlates of risk of influenza infection.

Background: The COVID-19 epidemic was declared a Global Pandemic by WHO on 11 March 2020. By 24 March 2020, over 440,000 cases and almost 20,000 deaths had been reported worldwide. In response to the fast-growing epidemic, which began in the Chinese city of Wuhan, Hubei, China imposed strict social distancing in Wuhan on 23 January 2020 followed closely by similar measures in other provinces. These interventions have impacted economic productivity in China, and the ability of the Chinese economy to resume without restarting the epidemic was not clear. Methods: Using daily reported cases from mainland China and Hong Kong SAR, we estimated transmissibility over time and compared it to daily within-city movement, as a proxy for economic activity. Results: Initially, within-city movement and transmission were very strongly correlated in the five mainland provinces most affected by the epidemic and Beijing. However, that correlation decreased rapidly after the initial sharp fall in transmissibility. In general, towards the end of the study period, the correlation was no longer apparent, despite substantial increases in within-city movement. A similar analysis for Hong Kong shows that intermediate levels of local activity were maintained while avoiding a large outbreak. At the very end of the study period, when China began to experience the re-introduction of a small number of cases from Europe and the United States, there is an apparent up-tick in transmission. Conclusions: Although these results do not preclude future substantial increases in incidence, they suggest that after very intense social distancing (which resulted in containment), China successfully exited its lockdown to some degree. Elsewhere, movement data are being used as proxies for economic activity to assess the impact of interventions. The results presented here illustrate how the eventual decorrelation between transmission and movement is likely a key feature of successful COVID-19 exit strategies.

The study of infectious disease outbreaks is required to train today’s epidemiologists. A typical way to introduce and explain key epidemiological concepts is through the analysis of a historical outbreak. There are, however, few training options that explicitly utilise real-time simulated stochastic outbreaks where the participants themselves comprise the dataset they subsequently analyse. In this paper, we present a teaching exercise in which an infectious disease outbreak is simulated over a five-day period and subsequently analysed. We iteratively developed the teaching exercise to offer additional insight into analysing an outbreak. An R package for visualisation, analysis and simulation of the outbreak data was developed to accompany the practical to reinforce learning outcomes. Computer simulations of the outbreak revealed deviations from observed dynamics, highlighting how simplifying assumptions conventionally made in mathematical models often differ from reality. Here we provide a pedagogical tool for others to use and adapt in their own settings. Keywords: Teaching, Outbreak analysis, Pedagogical tool, Simulation analysis, Network reconstruction

Abstract Background Non-hospital residential facilities are important reservoirs for MRSA transmission. However, conclusions and public health implications drawn from the many mathematical models depicting nosocomial MRSA transmission may not be applicable to these settings. Therefore, we reviewed the MRSA transmission dynamics studies in defined non-hospital residential facilities to: (1) provide an overview of basic epidemiology which has been addressed; (2) identify future research direction; and (3) improve future model implementation. Methods A review was conducted by searching related keywords in PUBMED without time restriction as well as internet searches via Google search engine. We included only articles describing the epidemiological transmission pathways of MRSA/community-associated MRSA within and between defined non-hospital residential settings. Results Among the 10 included articles, nursing homes (NHs) and correctional facilities (CFs) were two settings considered most frequently. Importation of colonized residents was a plausible reason for MRSA outbreaks in NHs, where MRSA was endemic without strict infection control interventions. The importance of NHs over hospitals in increasing nosocomial MRSA prevalence was highlighted. Suggested interventions in NHs included: appropriate staffing level, screening and decolonizing, and hand hygiene. On the other hand, the small population amongst inmates in CFs has no effect on MRSA community transmission. Included models ranged from system-level compartmental models to agent-based models. There was no consensus over the course of disease progression in these models, which were mainly featured with NH residents /CF inmates/ hospital patients as transmission pathways. Some parameters used by these models were outdated or unfit. Conclusions Importance of NHs has been highlighted from these current studies addressing scattered aspects of MRSA epidemiology. However, the wide variety of non-hospital residential settings suggest that more work is needed before robust conclusions can be drawn. Learning from existing work for hospitals, we identified critical future research direction in this area from infection control, ecological and economic perspectives. From current model deficiencies, we suggest more transmission pathways be specified to depict MRSA transmission, and further empirical studies be stressed to support evidence-based mathematical models of MRSA in non-hospital facilities. Future models should be ready to cope with the aging population structure.

Outbreaks of novel pathogens such as SARS, pandemic influenza and Ebola require substantial investments in reactive interventions, with consequent implementation plans sometimes revised on a weekly basis. Therefore, short-term forecasts of incidence are often of high priority. In light of the recent Ebola epidemic in West Africa, a forecasting exercise was convened by a network of infectious disease modellers. The challenge was to forecast unseen “future” simulated data for four different scenarios at five different time points. In a similar method to that used during the recent Ebola epidemic, we estimated current levels of transmissibility, over variable time-windows chosen in an ad hoc way. Current estimated transmissibility was then used to forecast near-future incidence. We performed well within the challenge and often produced accurate forecasts. A retrospective analysis showed that our subjective method for deciding on the window of time with which to estimate transmissibility often resulted in the optimal choice. However, when near-future trends deviated substantially from exponential patterns, the accuracy of our forecasts was reduced. This exercise highlights the urgent need for infectious disease modellers to develop more robust descriptions of processes – other than the widespread depletion of susceptible individuals – that produce non-exponential patterns of incidence. Keywords: Forecasting, Rapid response, Branching process, Renewal equation, MCMC