Epidemiology and Infection

Since the first analysis was published on 7 April 2021 the PCR test positivity rate has dropped significantly below the then estimated false positive rate (FPR) of 1.16% using the exponential decay to FPR model. Therefore, the estimate has been rejected and a new model was developed. Using the ONS infection surveys assumption (PCR FPR rate below 0.1%) the new model splits the test time series data into two periods based on a change in transmissibility that coincides with the reopening of England schools on 8 March. The new model provides for two base levels of exponential decay (for each periods transmissibility) combined with a single decay rate increase dependent on vaccination. Because the FPR is relatively insignificant compared to current PCR test positives, it cannot be statistically separated using currently available England epidemic time series data by the non-linear least squares estimation technique. Therefore, the FPR factor is temporarily dropped in the least squares regression. The new model is stable in that it reasonably predicts through the most current available data (25 April) the future test prevalence using parameters estimated with 29 March data. Thus far, the estimate parameters remain within their original confidence intervals as successive days are added to the time series. Of potential usefulness is the current estimate for change in decay rate per mean vaccination rate, currently estimated at approximately 10.7% (CI: 8.8% - 12.6%). The estimate should be used with caution as other unforeseen factors could cause the model to misestimate.

BackgroundNational epidemic dynamics of SARS-CoV-2 infections are being driven by: the degree of recent indoor mixing (both social and workplace), vaccine coverage, intrinsic properties of the circulating lineages, and prior history of infection (via natural immunity). In England, infections, hospitalisations and deaths fell during the first two steps of the "roadmap" for exiting the third national lockdown. The third step of the roadmap in England takes place on 17 May 2021. MethodsWe report the most recent findings on community infections from the REal-time Assessment of Community Transmission-1 (REACT-1) study in which a swab is obtained from a representative cross-sectional sample of the population in England and tested using PCR. Round 11 of REACT-1 commenced self-administered swab-collection on 15 April 2021 and completed collections on 3 May 2021. We compare the results of REACT-1 round 11 to round 10, in which swabs were collected from 11 to 30 March 2021. ResultsBetween rounds 10 and 11, prevalence of swab-positivity dropped by 50% in England from 0.20% (0.17%, 0.23%) to 0.10% (0.08%, 0.13%), with a corresponding R estimate of 0.90 (0.87, 0.94). Rates of swab-positivity fell in the 55 to 64 year old group from 0.17% (0.12%, 0.25%) in round 10 to 0.06% (0.04%, 0.11%) in round 11. Prevalence in round 11 was higher in the 25 to 34 year old group at 0.21% (0.12%, 0.38%) than in the 55 to 64 year olds and also higher in participants of Asian ethnicity at 0.31% (0.16%, 0.60%) compared with white participants at 0.09% (0.07%, 0.11%). Based on sequence data for positive samples for which a lineage could be identified, we estimate that 92.3% (75.9%, 97.9%, n=24) of infections were from the B.1.1.7 lineage compared to 7.7% (2.1%, 24.1%, n=2) from the B.1.617.2 lineage. Both samples from the B.1.617.2 lineage were detected in London from participants not reporting travel in the previous two weeks. Also, allowing for suitable lag periods, the prior close alignment between prevalence of infections and hospitalisations and deaths nationally has diverged. DiscussionWe observed marked reductions in prevalence from March to April and early May 2021 in England reflecting the success of the vaccination programme and despite easing of restrictions during lockdown. However, there is potential upwards pressure on prevalence from the further easing of lockdown regulations and presence of the B.1.617.2 lineage. If prevalence rises in the coming weeks, policy-makers will need to assess the possible impact on hospitalisations and deaths. In addition, consideration should be given to other health and economic impacts if increased levels of community transmission occur.

To describe characteristics of COVID-19 outbreaks in Australia to guide policy development for mitigation of future outbreaks, we conducted a retrospective analysis of COVID-19 outbreaks affecting two or more people reported to COVID-Net--an Australian national surveillance network--from 28 January until 27 December 2020. The COVID-Net surveillance network covered all Australian states and territories, with an estimated population of 25,649,985 persons as at 31 June 2020. We reported the epidemiology of COVID-19 outbreaks in Australia, including the setting in which they occurred, size, and duration. 853 outbreaks of COVID-19 were reported; associated with 13,957 confirmed cases, of whom 2,047 were hospitalised, and 800 died. The pattern of outbreaks followed a similar trend to the epidemic in Australia, defined by two distinct peaks in mid-March and July. Victoria reported the greatest number of outbreaks across all settings aligned with the second wave of infections. Outbreaks most commonly occurred in the workplace/industry setting (22%, 190/853), followed by education (14%, 122/853), residential aged care (13%, 114/853) and hospitals (10%, 83/853). The majority (40%, 340/853) of outbreaks had 6 to 24 cases, and the median outbreak duration increased in proportion with the number of associated cases. This report summarising COVID-19 outbreaks in Australia identifies settings of highest risk. Surveillance of outbreaks informs our understanding of transmission dynamics in Australia relative to national and jurisdictional interventions. For settings that are high risk for COVID-19, it is important to prioritise planning, surveillance, and implementation of control measures.

Objectivesto estimate the current number of total infections in a region in order to measure the progress of the epidemic with the purpose of reopening activities and planning the deployment of vaccines. Study designWe recovered estimates of the basic reproductive number (R0) and the Infection Fatality Risk (IFR) as well as the number of confirmed cases and deaths in several countries. Methodsthis works presents an expression to estimate the number of remaining susceptible in a population using the observed number of SARS-CoV-2 related deaths and current estimates of R0 and IFR. Resultsthe epidemic will infect most of the population causing 2.5 deaths per thousand inhabitants on average, and herd immunity will be achieved when the number of deaths per thousand inhabitants is close to two. This work introduces an expression to provide estimates of the number of remaining susceptible in a region using the reported number of deaths. Conclusionsany region with fewer than 2.5 deaths per thousand individuals will continue accumulating deaths until this average is achieved, and the infection rate will exceed the removal rate until the number of deaths is about two deaths per thousand, when herd immunity is reached. Waves may occur in any region where the number of deaths is below the herd immunity level.

1ObjectiveTo track the statistical case fatality rate (CFR) in the second wave of the UK coronavirus outbreak, and to understand its variations over time. DesignPublicly available UK government data and clinical evidence on the time between first positive PCR test and death are used to determine the relationships between reported cases and deaths, according to age groups and across regions in England. Main Outcome MeasuresEstimates of case fatality rates and their variations over time. ResultsThroughout October and November 2020, deaths in England can be broadly understood in terms of CFRs which are approximately constant over time. The same CFRs prove a poor predictor of deaths when applied back to September, when prevalence of the virus was comparatively low, suggesting that the potential effect of false positive tests needs to be taken into account. Similarly, increasing CFRs are needed to match cases to deaths when projecting the model forwards into December. The growth of the S gene dropout VOC in December occurs too late to explain this increase in CFR alone, but at 33% increased mortality, it can explain the peak in deaths in January. Seasonal effects could be in part responsible for the early December increase in CFR, and if so, the estimate of increased mortality would be reduced. There is also evidence that the prevalence of B.1.1.7 may have been slower amongst older age groups, and if this is a factor, then 33% could be an underestimate of mortality. From the second half of January, the CFRs for older age groups show a marked decline. Since the fraction of the VOC has not decreased, this decline is likely to be the result of the rollout of vaccination. However, due to the rapidly decreasing nature of the raw cases, any imprecisions in the time-to-death distribution are magnified in this time period, rendering estimates of vaccinations effect less precise. ConclusionsThe relationship between cases and deaths, even when controlling for age, is not static through the second wave of coronavirus in England. An apparently anomalous low case-fatality ratio in September can be accounted for by a 0.4% false-positive fraction. The rapid growth in CFR in December can be understood in part in terms of a more deadly new variant B.1.1.7, while a decline in January correlates with vaccine roll-out, suggesting that vaccine reduce the severity of infection, as well as the risk. Summary BoxO_ST_ABSWhat is already known on this topicC_ST_ABSThe case fatality rate (CFR) is a useful measure which enables one to estimate future deaths based on current infections. In England, there was a surge in Covid-19 CFR around the beginning of December. What the study addsUsing it, we monitor the case-fatality rate across time, region and age group from publicly available case data. This quantity is related to the lethality of the virus. It shows a sharp increase in December 2020, which parallels the spread of the B.1.1.7 variant. The January peak in actual deaths matches that predicted by cases if B.1.1.7 is about 33% more deadly; this estimate would be lower if there is a seasonal effect on deaths, and higher if at the peak the variant was less pervasive amongst older age groups. A steady drop in CFR from January suggests that vaccination not only reduces transmission but also the risk of serious illness among those infected. It is notable that these conclusions are reached with publicly available data independent of clinical studies.

BackgroundOn a Friday evening in July 2021, Public Health Wales received notification of two cases of salmonella gastroenteritis. Both cases reported attending an Eid al-Adha celebration barbecue in a public park in Cardiff, UK two days earlier. Case finding over the weekend indicated further cases in those who had attended this event and an outbreak investigation was initiated. MethodsCases were defined as an individual with diarrhoea and/or vomiting with date of onset on the day of the barbecue, with an epidemiological link to the Eid-al Adha celebration meal. We undertook a cohort study in 36 people attending the barbecue and an associated lunchtime event, and a nested case-control study using Firth logistic regression. A communication campaign that was sensitive towards cultural practices was developed in collaboration with the community, community liaison staff and the Public Health Wales communications team. ResultsConsumption of a traditional raw liver dish, marrara, at the barbecue was the most likely vehicle for infection (Firth logistic regression, aOR: 49.99, 95%CI 1.71-1461.54, p=0.02). Meat came from two local butchers with the same supplier and food samples yielded identical whole genome sequences to those from cases. This outbreak highlights the need for appropriate food hygiene advice in communities preparing traditional dishes. ConclusionsThis outbreak identified a new vehicle of interest, a traditional raw offal dish. Future outbreak investigations, particularly where cultural events are associated with particular foods, should consider dishes beyond those on routine questionnaires to those which may be relevant to the community in question.

Risk estimators for COVID-19 propagation based on the incidence rate of new cases can be misleading as they usually fail to account for the fraction of population immunized by infection or vaccination.This misconception yields different incidence rates, as we illustrate using the daily number of COVID-19 reported cases in Spain during the pre-vaccine period, between 15/01/2020 and 11/07/2021. An increase in the incidence rate of about 7% is found when properly accounting for the population at risk. Our results demonstrate that accounting for dynamic changes to the immunized fraction of the population is necessary for accurate risk estimation. We hope that our findings can lead to more effective strategies for pandemic response.

BackgroundThere is limited information on the role of individuals in different age groups in the spread of infection during the Omicron epidemics, especially ones beyond the winter epidemic wave in 2021-2022. COVID-19 booster vaccination in England during the Autumn 2022 was restricted to persons aged over 50y, and persons in clinical risk groups. MethodsWe used previously developed methodology to evaluate the role of individuals in different age groups in propagating the Spring, Summer, and Autumn waves of the Omicron epidemic in England. This methodology utilizes the relative risk (RR) statistic that measures the change in the proportion of cases in each age group among all COVID-19 cases in the population before the peak of an epidemic wave vs. after the peak of an epidemic wave. Higher values for the RR statistic represent age groups that experienced a disproportionate depletion of susceptible individuals during the ascent of the epidemic (due to increased contact rates and/or susceptibility to infection). ResultsFor the 2022 Spring wave, the highest RR estimate belonged to children aged 5-9y (RR=2.05 (95%CI (2.02,2.08)), followed by children aged 10-14y (RR=1.68 (1.66,1.7)) and children aged 0-4y (RR=1.38 (1.36,1.41)). For the Summer wave, the highest RR estimates belonged to persons aged 20-34y: (RR=1.09 (1.07,1.12) in aged 20-24y, RR=1.09 (1.07,1.11) in aged 25-29y, RR=1.09(1.07,1.11) in aged 30-34y). For the Autumn wave, the highest RR estimates belonged to those aged 70-74y (RR=1.10 (1.07,1.14)), followed by adults aged 35-39y (RR=1.09 (1.06,1.12)), adults aged 40-44y (RR=1.09 (1.06,1.12)), and adults aged 65-69y (RR=1.08 (1.05,1.11)). ConclusionsAs time progressed, the greatest relative roles in propagating different waves of the Omicron epidemic in England shifted from school-age children to younger adults to adults aged 35-44y and 65-74y. Extending booster vaccination to all adults, and possibly to children should help limit the spread of Omicron infections in the community.

BackgroundThe COVID-19 pandemic has led to many studies of seroprevalence. A number of methods exist in the statistical literature to correctly estimate disease prevalence or seroprevalence in the presence of diagnostic test misclassification, but these methods seem to be less known and not routinely used in the public health literature. We aimed to examine how widespread the problem is in recent publications, and to quantify the magnitude of bias introduced when correct methods are not used. MethodsA systematic review was performed to estimate how often public health researchers accounted for diagnostic test performance in estimates of seroprevalence. Using straightforward calculations, we estimated the amount of bias introduced when reporting the proportion of positive test results instead of using sensitivity and specificity to estimate disease prevalence. ResultsOf the seroprevalence studies sampled, 78% (95% CI 72% to 82%) failed to account for sensitivity and specificity. Expected bias is often more than is desired in practice, ranging from 1% to 12%. ConclusionsResearchers conducting studies of prevalence should correctly account for test sensitivity and specificity in their statistical analysis.

BackgroundThe management of Covid-19 outbreaks presented particular challenges in the prison setting. In this study we describe the results from the implementation of a serial mass testing approach in two adult prisons in northern England. The overall aim was to examine the epidemiology of Covid-19 outbreaks in prisons and help inform public health policy and practice during the pandemic. MethodsRepeat mass testing was offered to all eligible staff and residents in a womens (nresidents=239; nstaff=246) and a mens (nresidents=703; nstaff=340) prison in February and March 2021 at days 0, 7 and 28 after Covid-19 outbreaks were declared. Positive swab samples were sent for viral whole genome sequencing by COG-UK. FindingsParticipation in at least one testing round ranged from a low of 67% of staff in the mens prison to a high of 98% of residents in the womens prison. The largest outbreak, in the mens prison (261 cases in residents and 37 cases in staff), continued to see new cases identified at the last testing round on day 28. Test positivity in residents of both prisons was significantly lower (p<0.05) at day 28 than on preceding test days, but no significant difference was observed for staff (p>0.05). Epidemiological data in conjunction with sequencing information provided evidence for multiple introductions of the SARS-CoV-2 virus from the local community into the prisons, with transmission identified both within wings and between wings among residents and staff. Two distinct SARS-CoV-2 lineages were identified in the womens and mens prisons, B.1.177 and B.1.17, respectively. ConclusionsDuring a Covid-19 outbreak, timely implementation of a whole prison testing regime can serve to inform a targeted approach to infection prevention and control by identifying the true extent of disease transmission in all (including asymptomatic) individuals. Staff, in particular, should be tested regularly and testing uptake should be as high as possible to minimise the risk of infection incursion. Ensuring high testing uptake across all testing rounds remains a challenge.

BackgroundLocal estimates of the time-varying effective reproduction number (Rt) of COVID-19 in England became increasingly heterogeneous during April and May 2021. This may have been attributable to the spread of the Delta SARS-CoV-2 variant. This paper documents real-time analysis that aimed to investigate the association between changes in the proportion of positive cases that were S-gene positive, an indicator of the Delta variant against a background of the previously predominant Alpha variant, and the estimated time-varying Rt at the level of upper-tier local authorities (UTLA). MethodWe explored the relationship between the proportion of samples that were S-gene positive and the Rt of test-positive cases over time from the 23 February 2021 to the 25 May 2021. Effective reproduction numbers were estimated using the EpiNow2 R package independently for each local authority using two different estimates of the generation time. We then fit a range of regression models to estimate a multiplicative relationship between S-gene positivity and weekly mean Rt estimate. ResultsWe found evidence of an association between increased mean Rt estimates and the proportion of S-gene positives across all models evaluated with the magnitude of the effect increasing as model flexibility was decreased. Models that adjusted for either national level or NHS region level time-varying residuals were found to fit the data better, suggesting potential unexplained confounding. ConclusionsOur results indicated that even after adjusting for time-varying residuals between NHS regions, S-gene positivity was associated with an increase in the effective reproduction number of COVID-19. These findings were robust across a range of models and generation time assumptions, though the specific effect size was variable depending on the assumptions used. The lower bound of the estimated effect indicated that the reproduction number of Delta was above 1 in almost all local authorities throughout the period of investigation.

1Objectiveto determine the statistical relationship between reported deaths and infections in the UK coronavirus outbreak DesignPublicly available UK government data is used to determine a relationship between reported cases and deaths, taking into account various UK regions, age profiles and prevalence of the variant of concern (VOC) B.1.1.7. Main Outcome MeasuresEstablishing a simple statistical relationship between detected cases and subsequent mortality. ResultsThroughout October and November 2020, deaths in England are well described as 1/55th of detected cases from 12 days previously. After that, the relationship no longer holds and deaths are significantly higher. This is especially true in regions affected by the VOC B.1.1.7 ConclusionsIn early December, some new factor emerged to increase the case-fatality rate in the UK. Summary BoxO_ST_ABSWhat is already known on this topicC_ST_ABSThe infection-mortality ratio enables one to predict future deaths based on current infections. Incomplete monitoring of infection may be sufficient to predict future trends. What the study addsFor the specific case of the second wave of coronavirus infection in the UK, we show a clear mathematical relationship between detected infections (positive tests) and subsequent deaths. This relationship begins to fail in December, with unexpectedly high death rates. This may be correlated in time and region with the emergence of the Variant of Concern B 1.1.7.

Background and MethodsEngland entered its third national lockdown of the COVID-19 pandemic on 6th January 2021 with the aim of reducing the daily number of deaths and pressure on healthcare services. The real-time assessment of community transmission study (REACT-1) obtains throat and nose swabs from randomly selected people in England in order to describe patterns of SARS-CoV-2 prevalence. Here, we report data from round 9a of REACT-1 for swabs collected between 4th and 13th February 2021. ResultsOut of 85,473 tested-swabs, 378 were positive. Overall weighted prevalence of infection in the community in England was 0.51%, a fall of more than two thirds since our last report (round 8) in January 2021 when 1.57% of people tested positive. We estimate a halving time of 14.6 days and a reproduction number R of 0.72, based on the difference in prevalence between the end of round 8 and the beginning of round 9. Although prevalence fell in all nine regions of England over the same period, there was greater uncertainty in the trend for North West, North East, and Yorkshire and The Humber. Prevalence fell substantially across all age groups with highest prevalence among 18- to 24-year olds at 0.89% (0.47%, 1.67%) and those aged 5 to12 years at 0.86% (0.60%, 1.24%). Large household size, living in a deprived neighbourhood, and Asian ethnicity were all associated with increased prevalence. Healthcare and care home workers were more likely to test positive compared to other workers. ConclusionsThere is a strong decline in prevalence of SARS-CoV-2 in England among the general population five to six weeks into lockdown, but prevalence remains high: at levels similar to those observed in late September 2020. Also, the number of COVID-19 cases in hospitals is higher than at the peak of the first wave in April 2020. The effects of easing of social distancing when we transition out of lockdown need to be closely monitored to avoid a resurgence in infections and renewed pressure on health services.

In recent weeks, several seroprevalence studies have appeared which attempt to determine the prevalence of antibodies against SARS-CoV-2 in the population of certain European and American locations. Many of these studies find an antibody prevalence comparable to the false positive rate of their respective serology tests and the relatively low statistical power associated with each study has invited criticism. To determine the strength of the signal, we perform a meta-analysis on the publicly available seroprevalence data based on Bayesian hierarchical modelling with Markov Chain Monte Carlo and Generalized Linear Mixed Modelling with prediction sampling. We examine studies with results from Santa Clara County (CA), Los Angeles County (CA), San Miguel County (CO), Chelsea (MA), the Comte de Geneve (Switzerland), and Gangelt (Germany). Our results are in broad agreement with the conclusions of the studies; we find that there is evidence for non-trivial levels of antibody prevalence across all study locations. However, we also find that a significant probability mass exists for antibody prevalence at levels lower than the reported figures. The results of our meta-analysis on the recent seroprevalence studies point to an important and strongly suggestive signal.

If the knowledge of the incubation time is helpful in designing non-pharmaceutical interventions such as quarantine measures, can one use the number of cases arising after a lockdown to check (a) if our assumptions of the incubation time were correct and (b) if the quarantine measures were as successful as they could theoretically be. These are the two questions we raise by studying the number of new cases arising after lockdowns in a few European countries. The analysis which purely relies on the publicly available data of the numbers of new infections, rather than extensive contact tracing of individual patients, suggests a "memory" of the infections in the community with a median of 13.3 days. This distribution of the memory of infections which may even be considered as a surrogate of the incubation time in a perfect lockdown, suggests that even a perfect quarantine of 30 days is only 90% complete.

BackgroundThe extent to which the oro-faecal route contributes to the transmission of SARS-CoV-2 is not established. MethodsWe systematically reviewed the evidence on the presence of infectious SARS-CoV-2 in faeces and other gastrointestinal sources by examining studies that used viral culture to investigate the presence of replication-competent virus in these samples. We conducted searches in the WHO Covid-19 Database, LitCovid, medRxiv, and Google Scholar for SARS-CoV-2 using keywords and associated synonyms, with a search date up to 28th of November 2023. ResultsWe included 13 studies involving 328 COVID-19 subjects - providing 314 faecal or rectal swab SARS-CoV2 positive samples tested also with viral culture. The methods used for viral culture across the studies were heterogeneous. Three studies (2 cohorts and 1 case-series) reported observing replication-competent SARS-CoV-2 confirmed by quantitative RT-PCR (qPCR) and whole genome sequencing, and qPCR including appropriate cycle threshold changes. Overall, six (1.9%) of 314 faecal samples subjected to cell culture showed replication-competent virus. One study found replication competent samples from one immunocompromised patient. No studies were identified demonstrating direct evidence of oro-faecal transmission to humans. ConclusionsOur review found a relatively low frequency of replication-competent SARS-CoV-2 in faecal and other gastrointestinal sources. Although it is biologically plausible, more research is needed, using standardized cell culture methods, control groups, adequate follow-up and robust epidemiologic methods, including whether secondary infections occurred, to determine the role of the oro-faecal route in the transmission of SARS-CoV-2.

PurposeThe Gamma variant of SARS-CoV-2, first detected in travellers from Brazil, was found to have high transmissibility and virulence; following this finding, this paper aims to describe the epidemiology of Gamma cases in England from its first detection on 12 February 2021 to 31 August 2021. MethodsThe demographic analysis of Gamma cases was stratified by travel exposure. Travel-associated cases were further analysed by countries travelled from, stratified by categories set in place by the Red (highest risk countries), Amber, Green (lowest risk countries) travel policy, which was implemented from May to October 2021. ResultsThere were 251 confirmed Gamma cases detected in England in the study period. 35.1% were imported, 5.6% were secondary, and 29.5% were not travel associated. Early cases were predominantly travel-associated, with later cases likely obtained through community transmission. 51.0% of travel-related cases were travellers from Amber countries, and 40.2% had at least one Red country in their journey. ConclusionThe Gamma variant has not seen the same expansion as other variants such as Delta, most likely due to Delta out-competing community transmission of Gamma. Findings indicate the travel policy requiring quarantine for Red and Amber list travellers may have also contributed to preventing onward transmission of Gamma.

BackgroundBiases in COVID-19 burden and uncertainty in estimation of the corresponding epidemiologic indexes is a known and common phenomenon in infectious diseases. We investigated to what extent healthcare access (HCA) related supply/demand interfered with registered data on COVID-19 in Poland. Material and methodsWe run a multiple linear regression model with interactions to explain geographic variation in seroprevalence, hospitalizations (on voivodeship - NUTS-2 level) and current (beginning of the 4th wave - 15.09-21.11.2021) case notifications/crude mortality (on poviat - old NUTS-4 level). We took vaccination coverage and cumulative case notifications up to the so called 3rd wave as predictor variables and supply/demand (HCA) as moderating variables. ResultsHCA with interacting terms (mainly demand) explained to the great extent the variance of current incidence and most variance of current mortality. HCA (mainly supply) is significantly moderating cumulative case notifications till the 3rd wave explaining the variance in seroprevalence and hospitalization. ConclusionsSeeking causal relations between vaccination-or infection-gained immunity level and current infection dynamics could be misleading without understanding socio-epidemiologic context such as the moderating role of HCA (sensu lato). After quantification, HCA could be incorporated into epidemiologic models for improved prediction of real disease burden.

BackgroundDengue fever represents a persistent public health challenge in Brazil. Traditional outbreak prediction models prioritize high-incidence areas, potentially overlooking municipalities that serve as critical transmission bridges. MethodsWe analyzed dengue surveillance data from Brazils SINAN across three epidemic regimes: 2023 (1.51M cases), 2024 (6.43M cases, hyperendemic), and partial 2025 (1.50M cases). We constructed transmission networks using documented importation flows and temporal cross-correlations with regime-adaptive thresholds. Discrete Morse theory classified municipalities as transmission sources (maxima), bridges (saddles), or sinks (minima) based on composite risk scores incorporating case counts, connectivity, and importation patterns. ResultsDespite 4.3-fold case variation across years, network density remained stable (0.0024-0.0027), with edge counts scaling proportionally to municipality coverage. Critical point distributions varied systematically: 2023 had 449 critical nodes; hyperendemic 2024 showed only 274 despite highest case burden; partial 2025 revealed 414 critical nodes. Critical municipalities exhibited significantly higher hub scores (M=2.08-2.16) versus non-critical nodes (M=0.47-0.91, Cohens d=4.2-6.8, p<0.001). Hub scores correlated modestly with case counts ({rho}=0.35-0.42), confirming structural criticality diverges from epidemic volume. ConclusionsDiscrete Morse theory successfully identifies transmission-critical municipalities across varying epidemic intensities. The paradoxical reduction in critical points during hyperendemic transmission (274 vs. 449 in moderate years) suggests topological simplification rather than elaboration during peak transmission. Stable network density across 4.3-fold case variation indicates resilient transmission architecture where epidemic intensity affects volume rather than structure. This provides actionable surveillance tools for public health systems managing fluctuating dengue transmission, suggesting authorities to prioritize action areas structure-based rather than volume-based.

IntroductionIdentifying likely COVID-19 cases with good accuracy is essential for epidemiological research on the pandemics health effects. Therefore, evaluating case detection methods for cost-effectiveness and reliability is important. We investigated the agreement between a validated self-report questionnaire for COVID-19 and dried blood spot serology for SARS-CoV-2 antibodies. MethodsBetween June and October 2023, 311 adults completed the self-report COVID-19 questionnaire validated by the Avon Longitudinal Study of Parents and Children, and provided fingertip blood samples which underwent Enzyme Linked Immunosorbent Assay to quantify IgG antibodies to SARS-CoV-2 nucleocapsid (N)-protein. We applied several statistical approaches to assess agreement: Cohens Kappa measure of inter-rater reliability; positive (PPV) and negative (NPV) predictive values; logistic regressions of the year of most recent self-reported COVID-19 on serostatus; and linear regressions of N-protein antibody concentrations. ResultsTwo-thirds (203, 65%) of participants self-reported a history of COVID-19 whereas one-third (98, 32%) were seropositive for SARS-CoV-2 N-protein antibodies. Across all years, there was only "fair" agreement ({kappa} = 0.23 [95%CI 0.15 - 0.31]). Self-reported COVID-19 had a PPV of 41% and an NPV of 87% for SARS-CoV-2 seropositivity. PPV was low for 2020-21 (36%) and 2022 (33%), but higher (75%) for participants whose most recent case was in 2023. Compared to participants with no self-reported history of COVID-19, those reporting SARS-CoV-2 infection in 2023 had 23 times greater odds of being seropositive (95%CI: 9.18-62.5), and had 1,079% (617-1,838%) higher N-protein concentrations, after adjustment for confounders. ConclusionA validated COVID-19 self-report questionnaire is useful for identifying people who have previously had COVID-19 and may be reasonably accurate in identifying when an infection occurred, at least within a calendar year. Furthermore, consistent with existing evidence, serological testing becomes much less sensitive over time.