Comparative Application of the Fluctuation Test to the data of Morbidity by COVID-19 in United States of America, United Kingdom, Taiwan and China 2020-2023 (preprint)

In this work the Luria and Delbruck Fluctuation Test was comparatively applied to the data of Morbidity by COVID-19 in the United States of America (USA), United Kingdom (UK), Taiwan and China from 2020 to 2023. Three types of data were used: es.statista.com, datosmacro.expansion.com and larepublica.co without modification, but trying to avoid and justify the anomalies and inconsistencies observed. The methods originally used to establish the interactions of two populations were evaluated: the viral population with that of its host and the drift of both organisms. Only the interactive fluctuations of the weekly Variance of daily increase of Cases (Morbidity) were studied. The results showed that the Fluctuation Test is applicable to the selected data from USA, UK, Taiwan and China and other data from several countries used as controls. The study was separated into two approaches: First, comparison of the total or partial logarithmic profile of fluctuations of Variance of Cases (Morbidity) of USA, UK, Taiwan and China. Second, comparison of the values of the first fluctuation of Variance of Cases (Morbidity) in the boreal winter of 2020 for USA, UK, Taiwan, China and several countries used as controls. The results obtained for Morbidity demonstrate that USA and UK present a similar bimodal profile. China shows an inverted profile and Taiwan shows an intermediate profile between both tendencies. However, it was possible to detect some anomalies and uncertainties that were possibly derived from inconsistencies in the original data. Only USA shows a value of the first fluctuation comparable to the order of magnitude of the value of the first fluctuation of the Variance of Cases of China, in the northern winter of 2020. In the First Approach USA, UK and China had two important fluctuations: the first in the northern winter of 2020 before week 16 and the second at the beginning of northern winter of 2022, more than 100 weeks later. Taiwan showed only the latter. This latest fluctuation coincides with two events: the possible achievement of herd immunity and the emergence of Omicron variant. In this work we have evaluated whether this coincidence is casual or causal. The results obtained in the Second Approach aim to confirm the hypothesis of the animal origin of the first variant of SARS CoV-2.

Covid19 and Pregnancy: Infected Mothers and Premature Babies? (preprint)

Introduction: The emergency of the SARS-CoV-2 virus spread and its subsequent global pan-demic have raised significant concerns regarding its impact on pregnancy outcomes. This review aims to summarize the emerging data on the risk of preterm delivery in pregnant women infected with SARS-CoV-2. Materials and Methods: A systematic search was conducted from March 2020 to December 2023 using PubMed, following PRISMA guidelines. Studies correlating maternal COVID-19 infection with preterm birth were included. Results: Thirteen studies were analyzed, indicating a higher incidence of preterm birth in SARS-CoV-2 positive pregnant women compared to controls. The average incidence rate of pre-term birth in infected patients was 18.5%, with a median of 12.75%, while non infected women showed an average incidence of preterm birth of 10% with a median of 8.2%. Discussion: Studies suggest an association between SARS-CoV-2 infection during pregnancy and increased risk of preterm birth and cesarean section. Severity of symptoms and underlying comorbidities further elevate this risk. Notably, infections during the third trimester pose the highest risk of preterm birth. Conclusion: Preventing SARS-CoV-2 infection during pregnancy is crucial to mitigate adverse obstetric outcomes. Close monitoring and tailored interventions for infected pregnant women, particularly those in later trimesters and with comorbidities, are imperative to reduce the risk of preterm birth and improve maternal-fetal outcomes.

Fine structure of a daytime mid-latitude sporadic-E revealed by GNSS TEC, InSAR, and MAI (preprint)

Combining the total electron content (TEC) data from two nationwide Global Navigation Satellite System (GNSS) networks in Japan with the L-band synthetic aperture radar (SAR) data, we reveal the fine spatial and temporal structure of a daytime sporadic-E (Es) episode in Shikoku, Japan. The snapshot of the Es is derived not only from interferometric SAR (InSAR) but also from multiple aperture interferometry (MAI). The Es episode is accompanied by east-west elongated (up to ~ 180 km) multiple southward migrating TEC striations with a speed of ~ 90 m/s and ~ 10–20 km widths in the north-south direction. As previously suggested by the GNSS TEC time series, the present InSAR and MAI data independently confirm that electron density gradually increases from the frontal leading edge but abruptly drops in the trailing edge. The asymmetric electron density distribution is consistent with a previous study but requires further clarifications that can account for the occurrence in the daytime. The multiple TEC striations are reminiscent of the quasi-periodic (QP) echoes in nighttime Es detected by the Middle and Upper Atmosphere (MU) radar. Still, no vertically extended anomalies are suggested in the present daytime Es. The Kelvin-Helmholtz instabilities around the wind shear of neutral winds could be responsible for the QP TEC striations.

Indigenous Peoples’ evaluation of health risks when facing mandatory evacuation for birth during the COVID-19 pandemic: An Indigenous feminist analysis (preprint)

Background Indigenous Peoples living on Turtle Island are comprised of First Nations, Inuit, and Métis people and because of the Government of Canada’s mandatory evacuation policy, those living in rural and remote regions of Ontario are required to travel to urban, tertiary care centres to give birth. When evaluating the risk of travelling for birth, Indigenous Peoples understand, evaluate, and conceptualise health risks differently than Eurocentric biomedical models of health. Also, the global COVID-19 pandemic changed how people perceived risks to their health. Our research goal was to better understand how Indigenous parturients living in rural and remote communities conceptualised the risks associated with evacuation for birth before and during the COVID-19 pandemic.Methods To achieve this goal, we conducted semi-structured interviews with 11 parturients who travelled for birth during the pandemic and with 5 family members of those who were evacuated for birth.Results Participants conceptualised evacuation for birth as riskier during the COVID-19 pandemic and identified how the pandemic exacerbated existing risks of travelling for birth. In fact, Indigenous parturients noted the increased risk of contracting COVID-19 when travelling to urban centres for perinatal care, the impact of public health restrictions on increased isolation from family and community, the emotional impact of fear during the pandemic, and the decreased availability of quality healthcare.Conclusions Using Indigenous Feminist Methodology and Indigenous Feminist Theory, we critically analysed how mandatory evacuation for birth functions as a colonial tool and how conceptualizations of risk empowered Indigenous Peoples to make decisions that reduced risks to their health during the pandemic. With the results of this study, policy makers and governments can better understand how Indigenous Peoples conceptualise risk related to evacuation for birth before and during the pandemic, and prioritise further consultation with Indigenous Peoples to collaborate in the delivery of the health and care they need and desire.

Rates of Positive M-CHAT-R Screenings by Pandemic Birth and Prenatal SARS-CoV-2 Exposure (preprint)

Maternal stress and viral illness during pregnancy are associated with neurodevelopmental conditions in offspring. Children born during the COVID-19 pandemic, including those exposed prenatally to maternal SARS-CoV-2 infections, are reaching the developmental age for the assessment of risk for neurodevelopmental conditions. We examined associations between birth during the COVID-19 pandemic, prenatal exposure to maternal SARS-CoV-2 infection, and rates of positive screenings on the Modified Checklist for Autism in Toddlers-Revised (M-CHAT-R). Data were drawn from the COVID-19 Mother Baby Outcomes (COMBO) Initiative. Participants completed the M-CHAT-R as part of routine clinical care (COMBO-EHR cohort) or for research purposes (COMBO-RSCH cohort). Maternal SARS-CoV-2 status during pregnancy was determined through electronic health records. The COMBO-EHR cohort includes n=1664 children (n=442 historical cohort, n=1222 pandemic cohort; n=997 SARS-CoV-2 unexposed prenatally, n=130 SARS-CoV-2 exposed prenatally) who were born at affiliated hospitals between 2018-2023 and who had a valid M-CHAT-R score in their health record. The COMBO-RSCH cohort consists of n=359 children (n=268 SARS-CoV-2 unexposed prenatally, n=91 SARS-CoV-2 exposed prenatally) born at the same hospitals who enrolled into a prospective cohort study that included administration of the M-CHAT-R at 18-months. Birth during the pandemic was not associated with greater likelihood of a positive M-CHAT-R screen in the COMBO-EHR cohort. Maternal SARS-CoV-2 was associated with lower likelihood of a positive M-CHAT-R screening in adjusted models in the COMBO-EHR cohort (OR=0.40, 95% CI=0.22 - 0.68, p=0.001), while analyses in the COMBO-RSCH cohort yielded similar but non-significant results (OR=0.67, 95% CI=0.31-1.37, p=0.29).These results suggest that children born during the first 18 months of the COVID-19 pandemic and those exposed prenatally to a maternal SARS-CoV-2 infection are not at greater risk for screening positive on the M-CHAT-R.

Becoming a New Parent During the Pandemic: Experiences of Pregnancy, Birth, and the Post-Natal Period (preprint)

New parenthood in ordinary times can be a sensitive and unpredictable time. The Covid-19 pandemic brought additional, unprecedented changes to policy and practice that drastically impacted on the experiences of parents. This study aimed to generate data to enhance understanding of the experiences of new parents during the pandemic by qualitatively analysing their experiences. New parents from the UK (N = 303; female = 297; male = 7) responded to a survey asking about experiences of pregnancy, birth and the postnatal period and their experiences of digital technology during this time. Responses were analysed thematically by drawing across the three time periods. Parents reported conflicting feelings, negative feelings and silver linings cutting across eight themes that we developed, including: impacts on well-being, feeling without a village, changes to healthcare, atypical social experiences as a new parent, differential impacts on financial and working lives, conflicting feelings around digital technology, anger and worry regarding contradictory government guidance and recommendations for other parents. The findings offer much needed insights into the experiences of new parents during this time and provide some context to elevated levels of perinatal mental health difficulties in new parents during the pandemic. We suggest key recommendations going forwards in the care of new parents both now, and in times of future national crisis.

Estimation of Dynamic Origin-Destination Matrices in a Railway Transportation Network integrating Ticket Sales and Passenger Count Data (preprint)

Accurately estimating Origin-Destination (OD) matrices is a topic of increasing interest for efficient transportation network management and sustainable urban planning. Traditionally, travel surveys have supported this process; however, their availability and comprehensiveness can be limited. Moreover, the recent COVID-19 pandemic has triggered unprecedented shifts in mobility patterns, underscoring the urgency of accurate and dynamic mobility data supporting policies and decisions with data-driven evidence. In this study, we tackle these challenges by introducing an innovative pipeline for estimating dynamic OD matrices. The real motivating problem behind this is based on the Trenord railway transportation network in Lombardy, Italy. We apply a novel approach that integrates ticket and subscription sales data with passenger counts obtained from Automated Passenger Counting (APC) systems, making use of the Iterative Proportional Fitting (IPF) algorithm. Our work effectively addresses the complexities posed by incomplete and diverse data sources, showcasing the adaptability of our pipeline across various transportation contexts. Ultimately, this research bridges the gap between available data sources and the escalating need for precise OD matrices. The proposed pipeline fosters a comprehensive grasp of transportation network dynamics, providing a valuable tool for transportation operators, policymakers, and researchers. Indeed, to highlight the potentiality of dynamic OD matrices, we showcase some methods to perform anomaly detection of mobility trends in the network through such matrices and interpret them in light of events that happened in the last months of 2022.

Early warning system using primary healthcare data in the post-COVID-19-pandemic era: Brazil nationwide case-study (preprint)

BackgroundSyndromic surveillance utilising primary health care (PHC) data is a valuable tool for early outbreak detection, as demonstrated in the potential to identify COVID-19 outbreaks. However, the potential of such an early warning system in the post-COVID-19 era remains largely unexplored. MethodsWe analysed PHC encounter counts due to respiratory complaints registered in the Brazilian database of the Universal Health System between January and July 2023. We applied EARS (variation C1-C2-C3) and EVI to estimate the weekly thresholds. An alarm was determined when the number of encounters exceeded the week-specific threshold. We used data on hospitalisation due to respiratory disease to classify weeks in which the number of cases surpassed predetermined thresholds as anomalies. We compared EARS and EVIs efficacy in anticipating anomalies. FindingsA total of 119 anomalies were identified across 116 immediate regions during the study period. The EARS-C2 presented the highest early alarm rate, with 81/119 (68%) early alarms, and C1 the lowest, with 71 (60%) early alarms. The lowest true positivity was the EARS-C1 118/1354 (8.7%) and the highest EARS-C3 99/856 (11.6%). ConclusionRoutinely collected PHC data can be successfully used to detect respiratory disease outbreaks in Brazil. Syndromic surveillance enhances timeliness in surveillance strategies, albeit with lower specificity. A combined approach with other strategies is essential to strengthen accuracy, offering a proactive and effective public health response against future outbreaks.

Forecasting dominance of SARS-CoV-2 lineages by anomaly detection using deep AutoEncoders (preprint)

The COVID-19 pandemic exemplified the need for a rapid, effective genomic-based surveillance system to predict emerging SARS-CoV-2 variants and lineages. Traditional molecular epidemiology methods, which leverage public health surveillance or integrated sequence data repositories, are able to characterize the evolutionary history of infection waves and genetic evolution but fall short in predicting future outlooks in promptly anticipating viral genetic alterations. To bridge this gap, we introduce a novel Deep learning, autoencoder-based method for anomaly detection in SARS-CoV-2 (DeepAutoCov). Trained and updated on the public global SARS-CoV-2 GISAID database. DeepAutoCov identifies Future Dominant Lineages (FDLs), defined as lineages comprising at least 25% of SARS-CoV-2 genomes added on a given week, on a weekly basis, using the Spike (S) protein. Our algorithm is grounded on anomaly detection via an unsupervised approach, which is necessary given that FDLs can be known only a posteriori (i.e., after they have become dominant). We developed two concurrent approaches (a linear unsupervised and a posteriori supervised) to evaluate DeepAutoCoV performance. DeepAutoCoV identifies FDL, using the spike (S) protein, with a median lead time of 31 weeks on global data and achieves a positive predictive value ~7x better and 23% higher than the other approaches. Furthermore, it predicts vaccine related FDLs up to 17 months in advance. Finally, DeepAutoCoV is not only predictive but also interpretable, since it can pinpoint specific mutations within FDLs, generating hypotheses on the potential increases in virulence or transmissibility of a lineage. By integrating genomic surveillance with artificial intelligence, our work marks a transformative step that may provide valuable insights for the optimization of public health prevention and intervention strategies.

Fast and Accurate Dual-Way Streaming PARAFAC2 for Irregular Tensors -- Algorithm and Application (preprint)

How can we efficiently and accurately analyze an irregular tensor in a dual-way streaming setting where the sizes of two dimensions of the tensor increase over time? What types of anomalies are there in the dual-way streaming setting? An irregular tensor is a collection of matrices whose column lengths are the same while their row lengths are different. In a dual-way streaming setting, both new rows of existing matrices and new matrices arrive over time. PARAFAC2 decomposition is a crucial tool for analyzing irregular tensors. Although real-time analysis is necessary in the dual-way streaming, static PARAFAC2 decomposition methods fail to efficiently work in this setting since they perform PARAFAC2 decomposition for accumulated tensors whenever new data arrive. Existing streaming PARAFAC2 decomposition methods work in a limited setting and fail to handle new rows of matrices efficiently. In this paper, we propose Dash, an efficient and accurate PARAFAC2 decomposition method working in the dual-way streaming setting. When new data are given, Dash efficiently performs PARAFAC2 decomposition by carefully dividing the terms related to old and new data and avoiding naive computations involved with old data. Furthermore, applying a forgetting factor makes Dash follow recent movements. Extensive experiments show that Dash achieves up to 14.0x faster speed than existing PARAFAC2 decomposition methods for newly arrived data. We also provide discoveries for detecting anomalies in real-world datasets, including Subprime Mortgage Crisis and COVID-19.

Reduction in Spontaneous and Iatrogenic Preterm Births in Twin Pregnancies During COVID-19 Lockdown in Melbourne, Australia: A Multicenter Cohort Study (preprint)

Background: Melbourne, Australia, recorded one of the longest and most stringent pandemic lockdowns in 2020, which was associated with an increase in preterm stillbirths among singleton pregnancies. Twin pregnancies may be particularly susceptible to the impacts of pandemic disruptions to maternity care due to their higher background risk of adverse perinatal outcomes. Objective: To compare the rates of adverse perinatal outcomes in twin pregnancies exposed and unexposed to lockdown restrictions in Melbourne. Study Design: Multicenter retrospective cohort study of all twin pregnancies birthing in public maternity hospitals in Melbourne. We compared perinatal outcomes between a pre-pandemic group ('unexposed') and two lockdown-exposed groups: exposure 1 from 22 March 2020 to 21 March 2021 and exposure 2 from 22 March 2021 to 27 March 2022. We analyzed routinely-collected maternity data on all twin births >20 weeks where outcomes were available for both infants. The primary outcomes were rates of preterm birth<37 weeks and all-cause stillbirth. Multivariable log-binomial regression models were used to compare perinatal outcomes between the pre-pandemic group and women in whom weeks 20+0 to 40+0 of their pregnancy occurred entirely during each lockdown-exposure period. Perinatal outcomes were calculated per infant; maternal outcomes were calculated per pregnancy. Results: We included 2267 women birthing twins. Total preterm births<37 weeks were significantly lower in the exposure 1 group compared with the pre-pandemic group (63.1% vs 68.3% respectively; adjusted risk ratio, aRR 0.92 95% CI 0.87-0.98, p=0.01). This was driven by both fewer iatrogenic preterm births (44.1% vs 48.1%; aRR 0.97 95% CI 0.92-1.03, p=0.39) and fewer spontaneous preterm births (18.9% vs 20.3%; aRR 0.95 95% CI 0.90-0.99, p=0.04). There were also significantly lower rates of preterm birth<34 weeks in the exposure 1 group compared with the pre-pandemic group (19.9% vs 23.0%, aRR 0.93 95% CI 0.89-0.98 p=0.01). Total iatrogenic births for fetal compromise were significantly lower (13.4% vs 20.4%; aRR 0.94 95% CI 0.89-0.98, p=0.01). There were fewer special care nursery admissions (38.5% vs 43.4%; aRR 0.91 95% CI 0.87-0.95, p<0.001). There was no associated difference in all-cause stillbirths (1.5% vs 1.6%; aRR 1.00 95% CI 0.99-1.01, p=0.82), adjusted stillbirths, birthweight<3rd centile (5.7% vs 6.0%; aRR 1.00, 95% CI 0.98-1.02 p=0.74) or neonatal intensive care unit admissions in the exposure 1 group compared to the pre-pandemic group. In contrast, when comparing the pre-pandemic group with exposure 2 group, there was no significant difference in the rates of preterm birth<37 or <34 weeks (p>0.05). However, during exposure 2 the rate of preterm birth<28 weeks was significantly higher (7.2% vs 4.8%; aRR 1.03 95% CI 1.01-1.05, p=0.04) and infants were more likely to be admitted to a neonatal intensive care unit (25.0% vs 19.6%; aRR 1.06 95% CI 1.03-1.10, p<0.0001) compared with the pre-pandemic period. Conclusions: Melbourne's first lockdown-exposure period was associated with fewer twin preterm births<34 and <37 weeks without significant differences in stillbirths or adverse newborn outcomes. These lower rates were not sustained into the second exposure period. Pandemic conditions may provide important lessons for future antenatal care of twin pregnancies, including prevention of preterm birth and optimal timing of birth.

Multisensory Stimulation Bundles on Sleep and Neurobehavioral Development in the First Year After Birth in Very Preterm Infants: A Randomized Crossover Controlled Study Protocol (preprint)

Background This paper is a protocol for an evaluator-blinded, randomized crossover trial. It aims to assess the sleep efficiency of hospitalized very preterm infants (VPIs) by providing multisensory stimulation bundles. In addition, it will observe the intervention impacts on sleep during hospitalization and the sleep and neurodevelopmental outcomes during the first year of post-discharge follow-up.Methods The study will be conducted in the neonatology department of a tertiary pediatric teaching hospital. All eligible VPIs will undergo two types of care in random order: “standard care” (2 weeks) and “standard care plus multisensory stimulation bundles” (2 weeks). A generated list of random numbers will be used for case sequence allocation. Sleep outcomes will be evaluated using the Actiwatch-2 Actigraph. Moreover, the amplitude-integrated electroencephalography and the Griffiths Mental Development Scales will be used to measure the neurodevelopmental outcomes during hospitalization and in the first year of follow-up of VPIs.Discussion The intervention protocol of this study differs from other traditional interventions by producing precise and consistent supportive stimulations, similar to the maternal tactile, auditory, posture, and visual effects for hospitalized preterm infants. This protocol could be an effective measure to facilitate sleep and early neurodevelopment of VPIs. The expected outcomes will help confirm implementing and generalizing of the multisensory stimulation bundles care protocol in neonatology departments. We expect the study to positively impact hospitalized VPIs, specifically for their sleep and early neurodevelopmental outcomes. The study will also provide a new perspective regarding parent and infant interaction strategies, particularly for newborn intensive care units that limit visits due to the global spread of COVID-19.Trial registration: Chinese Clinical Trial Registry (Registry Number: ChiCTR 2200059099), Registered 25 April 2022, https://www.chictr.org.cn/showproj.html?proj=166980

Long COVID: Cognitive and FDG PET evolutions in six patients (preprint)

Long COVID is often characterized by cognitive complaints and deficits occurring immediately or several weeks after the infectious disease. Neuropsychological tests can revealed attention and executive function anomalies and FDG PET can display hypometabolic areas affecting various regions including frontal and cingulate cortices as well as precuneus and brainstem. We report here the cognitive and FDG PET evolutions over one year in 6 patients suffering from long COVID. Our study shows cognitive and FDG PET improvements in most of the cases and highlight the importance of a careful neurological follow-up in these patients.

Respiratory Function at 3 Months After Hospital Discharge in Critically Ill Patients With Covid-19 (preprint)

Background: Up to 20% of patients with COVID-19 get critically ill and require intensive care unit (ICU)admission. At hospital discharge, most patients still have physical and mental limitations, which affect their quality of life. Pulmonaryfunctional alterations in patients with COVID-19 vary from the absence of functional abnormalities to restrictive and diffusion impairments. We aimed to describe pulmonary function abnormalities as well as their impact on the 6-minute walk test (6 MWT) and SF-36 physical component summary (PCS) score in patients with COVID-19 at ≥ 3 months after hospital discharge. Methods: Weincluded 65 patients aged ≥ 18 years with severe COVID-19 confirmed throughreal-time reverse transcriptase-polymerase chain reaction andadmitted to the ICU between April 2020 and October 2021. Patients were evaluated at ≥ 3 months after hospital discharge using the 6 MWT, pulmonary function tests (PFTs), and the PCS score. Results: Among the included patients, 27patients had abnormal PFT findings, 21 (32.3%) had forced vital capacity < 80%, 17 (26.1%) had forced expiratory volume in 1 s< 80%, and 4 (6.1%) had a maximal mid-expiratory flow< 65%. Compared with patients without abnormal PFT findings, patientswith abnormal PFT findings were older and had significantlyhigher ferritin levels. There were no significant between-group differences ininvasive and noninvasive respiratory support, mechanical ventilation duration, vasopressor use,and renal replacement therapy. However, compared with patients with normalPFT findings, patients with abnormal PFT findings showed asignificantly lower 6-MWT score [78% (0.0–92) vs.95% (75–100), p = 0.01] and worse PCS scores [39.4 (32.1–51.3) vs. 52.0 (47.4–57.3), p = 0.007]. There was an independent association between the PCS scores and PFT findings. Conclusions: We found that a significant proportion of patients present pulmonary functional alterations ≥ 3 months after discharge from the hospital after treatment forsevere COVID-19; further, these alterations affectphysical functional capacity and quality of life. Trial registration: The trial protocol was approved by the Research Ethics Committee of the Hospital Sao Domingos (Number 5.403.663) in May 12, 2022 and registered in clinical trials. Gov(NCT05249842), February 22, 2022.

Adult mortality patterns in Yemen before and during armed conflict and the COVID-19 pandemic: Evidence from a web survey of the global diaspora (preprint)

Background The ongoing war in Yemen has created a severe and protracted crisis that has left nearly three-quarters of the population in need of urgent humanitarian assistance. Despite eight years of conflict there exist few robust estimates of how the crisis and COVID-19 pandemic have affected mortality in Yemen. The security situation has limited access to affected populations and thus required novel alternatives to local mortality surveys.Methods We used a web-based, respondent-driven sampling method to disseminate a mortality survey amongst the global Yemeni diaspora. We used mortality estimation methods and survival analysis to calculate mortality and/or survivorship amongst respondents’ close family members in Yemen including adults aged 50+, siblings, and children under five years.Results Eighty-nine eligible respondents completed the survey. Respondents provided data on the status of 1704 individuals of whom 85 (5%) had died; of these 65 (3.8%) were reported to have died in Yemen. An analysis of survivorship of respondents’ parents after their 50th birthday (adjusted for birth cohort and gender) provided weak evidence that the war and pandemic periods were associated with higher mortality when compared to the pre-war period. Analysis of the subset of individuals who died in Yemen also suggested an increased hazard of dying during the war/pandemic period; however, these results were non-significant. Sibling mortality amongst those aged 15–49 was 0.7 per 1000 person-years during the pre-war period compared to 1.1 during the war/pandemic period amongst males, and 0.8 versus 0.0 amongst females; however, these estimates reflected small numbers of deaths. The number of deaths amongst children under five in Yemen was too low to allow meaningful analysis; only three of the seven deaths in this group occurred during the analysis period.Conclusions Our data suggest increased mortality during the war/pandemic period, compared to the pre-war period, among elderly Yemenis. Our findings require careful interpretation as our small and non-representative sample appeared skewed towards higher-income, urban communities. Surveys of diaspora populations offer a promising means of describing mortality patterns in crisis-affected populations; however, large numbers of respondents are likely required to achieve accurate mortality estimates and attempt adjustment for selection bias.

An economic evaluation of presymptomatic Sars-Cov-2 (COVID-19) detection using anomaly detection models in the COVID-RED Trial (preprint)

Background: There is an average 5 day delay between a COVID-19 infection and symptom onset. Evidence suggests between 35% and 50% of all transmissions occur within this time-frame. Presymptomatic COVID-19 detection, therefore, serves a crucial role in containing the spread of COVID-19. An early COVID-19 detection algorithm were developed in the COVID-Red trial, which based on anomalies in biophysical markers, measured via a bracelet, predicts COVID-19 infection.  Method: This paper is an economic evaluation of the anomaly detection model developed in the COVID-Red trial with a health sector perspective. The primary outcome is the incremental cost effectiveness ratio (ICER), where the effectiveness is measured as early COVID-19 detection. Data on healthcare utilisation and COVID-19 testing are collected through biweekly surveys, structured phone interviews and a daily symptom diary. All data is self reported.  Results: COVID-19 can be detected 3.50 days earlier using an anomaly detection model, as opposed to a deterministic approach. The associated costs are ­≈54€ and ≈50€, on average, in the intervention and control group respectively. The ICER is 1.15€, and implies a marginal cost of 1.15€ per days of early detection.  Conclusion: The anomaly detection model detects COVID-19 3.5 earlier than otherwise possible, with a marginal cost of 4€. The ICER is 1.15€. Compared to the existing test strategies, the bracelet and algorithm combination is expensive. It can be considered inexpensive amongst, for example, healthcare workers where structured COVID-19 screenings are common.  Trial registration: Dutch Trial Register, NL9320. Registered 18/02/2021, https://clinicaltrialregister.nl/en/trial/23180

Laplacian Change Point Detection for Single and Multi-view Dynamic Graphs (preprint)

Dynamic graphs are rich data structures that are used to model complex relationships between entities over time. In particular, anomaly detection in temporal graphs is crucial for many real world applications such as intrusion identification in network systems, detection of ecosystem disturbances and detection of epidemic outbreaks. In this paper, we focus on change point detection in dynamic graphs and address three main challenges associated with this problem: i). how to compare graph snapshots across time, ii). how to capture temporal dependencies, and iii). how to combine different views of a temporal graph. To solve the above challenges, we first propose Laplacian Anomaly Detection (LAD) which uses the spectrum of graph Laplacian as the low dimensional embedding of the graph structure at each snapshot. LAD explicitly models short term and long term dependencies by applying two sliding windows. Next, we propose MultiLAD, a simple and effective generalization of LAD to multi-view graphs. MultiLAD provides the first change point detection method for multi-view dynamic graphs. It aggregates the singular values of the normalized graph Laplacian from different views through the scalar power mean operation. Through extensive synthetic experiments, we show that i). LAD and MultiLAD are accurate and outperforms state-of-the-art baselines and their multi-view extensions by a large margin, ii). MultiLAD's advantage over contenders significantly increases when additional views are available, and iii). MultiLAD is highly robust to noise from individual views. In five real world dynamic graphs, we demonstrate that LAD and MultiLAD identify significant events as top anomalies such as the implementation of government COVID-19 interventions which impacted the population mobility in multi-view traffic networks.

Regime type and Data Manipulation: Evidence from the COVID-19 Pandemic (preprint)

Autocratic and democratic leaders have an incentive to misreport data that may reveal policy failure. However, it is easier for autocratic leaders to fabricate data because they are not subject to scrutiny from media, opposition parties, and civil society. This suggests that autocratic governments are more likely to manipulate policy-relevant statistics than democratic governments. It is inherently difficult to test that claim because researchers typically do not have access to data from sources other than the government. The COVID-19 pandemic represents a unique opportunity to examine the relationship between regime type and data manipulation because of its widespread impact, as well as the ability to compare reported with excess deaths and test for statistical anomalies in reported data. Based on regressions for undercounting and statistical irregularities that take into account unintentional mismeasurement, I find that autocratic governments are more likely to deliberately under-report the impact of COVID-19 than their democratic counterparts.

Trajectories of child mental health, physical activity and screen-time during the COVID-19 pandemic considering different family situations: results from a longitudinal birth cohort. (preprint)

Background: Many authors described a significant mental health burden of children and adolescents during the COVID-19 pandemic, possibly moderated by social disparities. This analysis explores if pre-pandemic family circumstances might be related to different aspects of child health during the pandemic. Methods: We analyzed trajectories of health related outcomes in children aged 5 to 9 years (T7 to T11) using the Ulm SPATZ Health study, a population based birth cohort study (baseline 04/2012-05/2013) conducted in the South of Germany. Outcomes were children`s mental health, quality of life, and lifestyle, such as screen-time, and physical activity. We conducted descriptive statistics of maternal and child characteristics before and throughout the pandemic. We defined three different groups of pre-pandemic family situations and used adjusted mixed models to estimate differences in means associated to the time during the pandemic vs. before the pandemic in (a) all children and in (b) children belonging to specific pre-pandemic family situations. Results: We analyzed data of n=588 children from whom at least one questionnaire was completed between T7 and T11. When not considering pre-pandemic family situation, adjusted mixed models showed statistically significant lower mean scores of health-related quality of life among girls during vs. before the COVID-19 pandemic (difference in mean (b): -3.9 (95% confidence interval (CI): -6.4, -1.4). There were no substantial differences in mental health, in screen-time, and in physical activity neither in boys nor in girls. When considering pre-pandemic family situation boys with mothers having symptoms of depression or anxiety showed a substantial loss of health-related quality of life in the subscale friends (b: -10.5 (95% CI: -19.7, -1.4). Among girls of this group 60% of the 15 assessed outcomes were negatively associated with a remarkable loss in health related quality of life (e.g. KINDL-physical well-being difference in means: -12.2 (95% CI: -18.9, -5.4)). Furthermore, a substantial increase in screen-time was found (+2.9 h (95% CI: 0.3, 5.6)). Conclusion: Our results suggest that the health (and behavior) of primary school-aged children is possibly impacted by the COVID-19 pandemic with adverse consequences differing by gender and very likely by pre-pandemic family situation. Especially in girls having a mother with depression or anxiety symptoms the adverse consequences of the pandemic on mental health seem to be aggregated. Boys showed fewer adverse trajectories and it needs to be further assessed which factors exactly are behind the (socioeconomic) factors such as maternal working habits and limited living space when analyzing the effect of the pandemic on children’s health.