Spatial Optimization Methods for Malaria Risk Mapping in Sub-Saharan African Cities Using Demographic and Health Surveys.

Vector-borne diseases, such as malaria, are affected by the rapid urban growth and climate change in sub-Saharan Africa (SSA). In this context, intra-urban malaria risk maps act as a key decision-making tool for targeting malaria control interventions, especially in resource-limited settings. The Demographic and Health Surveys (DHS) provide a consistent malaria data source for mapping malaria risk at the national scale, but their use is limited at the intra-urban scale because survey cluster coordinates are randomly displaced for ethical reasons. In this research, we focus on predicting intra-urban malaria risk in SSA cities-Dakar, Dar es Salaam, Kampala and Ouagadougou-and investigate the use of spatial optimization methods to overcome the effect of DHS spatial displacement. We modeled malaria risk using a random forest regressor and remotely sensed covariates depicting the urban climate, the land cover and the land use, and we tested several spatial optimization approaches. The use of spatial optimization mitigated the effects of DHS spatial displacement on predictive performance. However, this comes at a higher computational cost, and the percentage of variance explained in our models remained low (around 30%-40%), which suggests that these methods cannot entirely overcome the limited quality of epidemiological data. Building on our results, we highlight potential adaptations to the DHS sampling strategy that would make them more reliable for predicting malaria risk at the intra-urban scale.

Drivers of the dynamics of the spread of cholera in the Democratic Republic of the Congo, 2000-2018: An eco-epidemiological study.

BACKGROUND: The dynamics of the spread of cholera epidemics in the Democratic Republic of the Congo (DRC), from east to west and within western DRC, have been extensively studied. However, the drivers of these spread processes remain unclear. We therefore sought to better understand the factors associated with these spread dynamics and their potential underlying mechanisms. METHODS: In this eco-epidemiological study, we focused on the spread processes of cholera epidemics originating from the shores of Lake Kivu, involving the areas bordering Lake Kivu, the areas surrounding the lake areas, and the areas out of endemic eastern DRC (eastern and western non-endemic provinces). Over the period 2000-2018, we collected data on suspected cholera cases, and a set of several variables including types of conflicts, the number of internally displaced persons (IDPs), population density, transportation network density, and accessibility indicators. Using multivariate ordinal logistic regression models, we identified factors associated with the spread of cholera outside the endemic eastern DRC. We performed multivariate Vector Auto Regressive models to analyze potential underlying mechanisms involving the factors associated with these spread dynamics. Finally, we classified the affected health zones using hierarchical ascendant classification based on principal component analysis (PCA). FINDINGS: The increase in the number of suspected cholera cases, the exacerbation of conflict events, and the number of IDPs in eastern endemic areas were associated with an increased risk of cholera spreading outside the endemic eastern provinces. We found that the increase in suspected cholera cases was influenced by the increase in battles at lag of 4 weeks, which were influenced by the violence against civilians with a 1-week lag. The violent conflict events influenced the increase in the number of IDPs 4 to 6 weeks later. Other influences and uni- or bidirectional causal links were observed between violent and non-violent conflicts, and between conflicts and IDPs. Hierarchical clustering on PCA identified three categories of affected health zones: densely populated urban areas with few but large and longer epidemics; moderately and accessible areas with more but small epidemics; less populated and less accessible areas with more and larger epidemics. CONCLUSION: Our findings argue for monitoring conflict dynamics to predict the risk of geographic expansion of cholera in the DRC. They also suggest areas where interventions should be appropriately focused to build their resilience to the disease.

Time series modelling for wastewater-based epidemiology of COVID-19: A nationwide study in 40 wastewater treatment plants of Belgium, February 2021 to June 2022.

BACKGROUND: Wastewater-based epidemiology (WBE) has been implemented to monitor surges of COVID-19. Yet, multiple factors impede the usefulness of WBE and quantitative adjustment may be required. AIM: We aimed to model the relationship between WBE data and incident COVID-19 cases, while adjusting for confounders and autocorrelation. METHODS: This nationwide WBE study includes data from 40 wastewater treatment plants (WWTPs) in Belgium (02/2021-06/2022). We applied ARIMA-based modelling to assess the effect of daily flow rate, pepper mild mottle virus (PMMoV) concentration, a measure of human faeces in wastewater, and variants (alpha, delta, and omicron strains) on SARS-CoV-2 RNA levels in wastewater. Secondly, adjusted WBE metrics at different lag times were used to predict incident COVID-19 cases. Model selection was based on AICc minimization. RESULTS: In 33/40 WWTPs, RNA levels were best explained by incident cases, flow rate, and PMMoV. Flow rate and PMMoV were associated with -13.0 % (95 % prediction interval: -26.1 to +0.2 %) and +13.0 % (95 % prediction interval: +5.1 to +21.0 %) change in RNA levels per SD increase, respectively. In 38/40 WWTPs, variants did not explain variability in RNA levels independent of cases. Furthermore, our study shows that RNA levels can lead incident cases by at least one week in 15/40 WWTPs. The median population size of leading WWTPs was 85.1 % larger than that of non­leading WWTPs. In 17/40 WWTPs, however, RNA levels did not lead or explain incident cases in addition to autocorrelation. CONCLUSION: This study provides quantitative insights into key determinants of WBE, including the effects of wastewater flow rate, PMMoV, and variants. Substantial inter-WWTP variability was observed in terms of explaining incident cases. These findings are of practical importance to WBE practitioners and show that the early-warning potential of WBE is WWTP-specific and needs validation.

Spatial accessibility to health facilities in Sub-Saharan Africa: comparing existing models with survey-based perceived accessibility.

BACKGROUND: Mapping geographical accessibility to health services is essential to improve access to public health in sub-Saharan Africa. Different methods exist to estimate geographical accessibility, but little is known about the ability of these methods to represent the experienced accessibility of the population, and about the added-value of sophisticated and data-demanding methods over simpler ones. Here we compare the most commonly used methods to survey-based perceived accessibility in different geographical settings. METHODS: Modelled accessibility maps are computed for 12 selected sub-Saharan African countries using four methods: Euclidean distance, cost-distance considering walking and motorized speed, and Kernel density. All methods are based on open and large-scale datasets to allow replication. Correlation coefficients are computed between the four modelled accessibility indexes and the perceived accessibility index extracted from Demographic and Health Surveys (DHS), and compared across different socio-geographical contexts (rural and urban, population with or without access to motorized transports, per country). RESULTS: Our analysis suggests that, at medium spatial resolution and using globally-consistent input datasets, the use of sophisticated and data-demanding methods is difficult to justify as their added value over a simple Euclidian distance method is not clear. We also highlight that all modelled accessibilities are better correlated with perceived accessibility in rural than urban contexts and for population who do not have access to motorized transportation. CONCLUSIONS: This paper should guide researchers in the public health domain for knowing strengths and limits of different methods to evaluate disparities in health services accessibility. We suggest that using cost-distance accessibility maps over Euclidean distance is not always relevant, especially when based on low resolution and/or non-exhaustive geographical datasets, which is often the case in low- and middle-income countries.

SARS-CoV-2 Surveillance in Belgian Wastewaters.

Wastewater-based surveillance was conducted by the national public health authority to monitor SARS-CoV-2 circulation in the Belgian population. Over 5 million inhabitants representing 45% of the Belgian population were monitored throughout 42 wastewater treatment plants for 15 months comprising three major virus waves. During the entire period, a high correlation was observed between the daily new COVID-19 cases and the SARS-CoV-2 concentration in wastewater corrected for rain impact and covered population size. Three alerting indicators were included in the weekly epidemiological assessment: High Circulation, Fast Increase, and Increasing Trend. These indicators were computed on normalized concentrations per individual treatment plant to allow for a comparison with a reference period as well as between analyses performed by distinct laboratories. When the indicators were not corrected for rain impact, rainy events caused an underestimation of the indicators. Despite this negative impact, the indicators permitted us to effectively monitor the evolution of the fourth virus wave and were considered complementary and valuable information to conventional epidemiological indicators in the weekly wastewater reports communicated to the National Risk Assessment Group.

Residential green space in association with the methylation status in a CpG site within the promoter region of the placental serotonin receptor HTR2A.

Green space could influence adult cognition and childhood neurodevelopment , and is hypothesized to be partly driven by epigenetic modifications. However, it remains unknown whether some of these associations are already evident during foetal development. Similar biological signals shape the developmental processes in the foetal brain and placenta.Therefore, we hypothesize that green space can modify epigenetic processes of cognition-related pathways in placental tissue, such as DNA-methylation of the serotonin receptor HTR2A. HTR2A-methylation was determined within 327 placentas from the ENVIRONAGE (ENVIRonmental influence ON early AGEing) birth cohort using bisulphite-PCR-pyrosequencing. Total green space exposure was calculated using high-resolution land cover data derived from the Green Map of Flanders in seven buffers (50 m-3 km) and stratified into low (<3 m) and high (≥3 m) vegetation. Residential nature was calculated using the Land use Map of Flanders. We performed multivariate regression models adjusted for several a priori chosen covariables. For an IQR increment in total green space within a 1,000 m, 2,000 m and 3,000 m buffer the methylation of HTR2A increased with 1.47% (95%CI:0.17;2.78), 1.52% (95%CI:0.21;2.83) and 1.42% (95%CI:0.15;2.69), respectively. Additionally,, we found 3.00% (95%CI:1.09;4.91) and 1.98% (95%CI:0.28;3.68) higher HTR2A-methylation when comparing residences with and without the presence of nature in a 50 m and 100 m buffer, respectively. The methylation status of HTR2A in placental tissue is positively associated with maternal green space exposure. Future research is needed to understand better how these epigenetic changes are related to functional modifications in the placenta and the consequent implications for foetal development.

Modalities and preferred routes of geographic spread of cholera from endemic areas in eastern Democratic Republic of the Congo.

Cholera is endemic along the Great Lakes Region, in eastern Democratic Republic of the Congo (DRC). From these endemic areas, also under perpetual conflicts, outbreaks spread to other areas. However, the main routes of propagation remain unclear. This research aimed to explore the modalities and likely main routes of geographic spread of cholera from endemic areas in eastern DRC. We used historical reconstruction of major outbreak expansions of cholera since its introduction in eastern DRC, maps of distribution and spatiotemporal cluster detection analyses of cholera data from passive surveillance (2000-2017) to describe the spread dynamics of cholera from eastern DRC. Four modalities of geographic spread and their likely main routes from the source areas of epidemics to other areas were identified: in endemic eastern provinces, and in non-endemic provinces of eastern, central and western DRC. Using non-parametric statistics, we found that the higher the number of conflict events reported in eastern DRC, the greater the geographic spread of cholera across the country. The present study revealed that the dynamics of the spread of cholera follow a fairly well-defined spatial logic and can therefore be predicted.

The spread of cholera in western Democratic Republic of the Congo is not unidirectional from East-West: a spatiotemporal analysis, 1973-2018.

BACKGROUND: Cholera outbreaks in western Democratic Republic of the Congo (DRC) are thought to be primarily the result of westward spread of cases from the Great Lakes Region. However, other patterns of spatial spread in this part of the country should not be excluded. The aim of this study was to explore alternative routes of spatial spread in western DRC. METHODS: A literature review was conducted to reconstruct major outbreak expansions of cholera in western DRC since its introduction in 1973. We also collected data on cholera cases reported at the health zone (HZ) scale by the national surveillance system during 2000-2018. Based on data from routine disease surveillance, we identified two subperiods (week 45, 2012-week 42, 2013 and week 40, 2017-week 52, 2018) for which the retrospective space-time permutation scan statistic was implemented to detect spatiotemporal clusters of cholera cases and then to infer the spread patterns in western DRC other than that described in the literature. RESULTS: Beyond westward and cross-border spread in the West Congo Basin from the Great Lakes Region, other dynamics of cholera epidemic propagation were observed from neighboring countries, such as Angola, to non-endemic provinces of southwestern DRC. Space-time clustering analyses sequentially detected clusters of cholera cases from southwestern DRC to the northern provinces, demonstrating a downstream-to-upstream spread along the Congo River. CONCLUSIONS: The spread of cholera in western DRC is not one-sided. There are other patterns of spatial spread, including a propagation from downstream to upstream areas along the Congo River, to be considered as preferential trajectories of cholera in western DRC.

Investigating the drivers of the spatio-temporal heterogeneity in COVID-19 hospital incidence-Belgium as a study case.

BACKGROUND: The COVID-19 pandemic is affecting nations globally, but with an impact exhibiting significant spatial and temporal variation at the sub-national level. Identifying and disentangling the drivers of resulting hospitalisation incidence at the local scale is key to predict, mitigate and manage epidemic surges, but also to develop targeted measures. However, this type of analysis is often not possible because of the lack of spatially-explicit health data and spatial uncertainties associated with infection. METHODS: To overcome these limitations, we propose an analytical framework to investigate potential drivers of the spatio-temporal heterogeneity in COVID-19 hospitalisation incidence when data are only available at the hospital level. Specifically, the approach is based on the delimitation of hospital catchment areas, which allows analysing associations between hospitalisation incidence and spatial or temporal covariates. We illustrate and apply our analytical framework to Belgium, a country heavily impacted by two COVID-19 epidemic waves in 2020, both in terms of mortality and hospitalisation incidence. RESULTS: Our spatial analyses reveal an association between the hospitalisation incidence and the local density of nursing home residents, which confirms the important impact of COVID-19 in elderly communities of Belgium. Our temporal analyses further indicate a pronounced seasonality in hospitalisation incidence associated with the seasonality of weather variables. Taking advantage of these associations, we discuss the feasibility of predictive models based on machine learning to predict future hospitalisation incidence. CONCLUSION: Our reproducible analytical workflow allows performing spatially-explicit analyses of data aggregated at the hospital level and can be used to explore potential drivers and dynamic of COVID-19 hospitalisation incidence at regional or national scales.

Neighbourhood-level housing quality indices for health assessment in Dakar, Senegal.

In sub-Saharan African cities, the dearth of accurate and detailed data is a major problem in the study of health and socioeconomic changes driven by rapid urbanization. Data on both health determinants and health outcomes are often lacking or are of poor quality. Proxies associated with socioeconomic differences are needed to compensate the lack of data. One of the most straightforward proxies is housing quality, which is a multidimensional concept including characteristics of both the built and natural environments. In this work, we combined the 2013 census data with remotely sensed land cover and land use data at a very high resolution in order to develop an integrated housing quality-based typology of the neighbourhoods in Dakar, Senegal. Principal component analysis and hierarchical classification were used to derive neighbourhood housing quality indices and four neighbourhood profiles. Paired tests revealed significant variations in the censusderived mortality rates between profile 1, associated with the lowest housing quality, and the three other profiles. These findings demonstrate the importance of housing quality as an important health risk factor. From a public health perspective, it should be a useful contribution for geographically targeted planning health policies, at the neighbourhood spatial level, which is the most appropriate administrative level for interventions.

Exposure to green space and pollen allergy symptom severity: A case-crossover study in Belgium.

BACKGROUND: The prevalence of pollen allergy has increased due to urbanization, climate change and air pollution. The effects of green space and air pollution on respiratory health of pollen allergy patients are complex and best studied in spatio-temporal detail. METHODS: We tracked 144 adults sensitized to Betulaceae pollen during the tree pollen season (January-May) of 2017 and 2018 and assessed their spatio-temporal exposure to green space, allergenic trees, air pollutants and birch pollen. Participants reported daily symptom severity scores. We extracted 404 case days with high symptom severity scores and matched these to 404 control days. The data were analyzed using conditional logistic regression with a 1:1 case-crossover design. RESULTS: Case days were associated with exposure to birch pollen concentration (100 grains/m3) [adjusted odds ratio 1.045 and 95% confidence interval (1.014-1.078)], O3 concentration (10 µg/m3) [1.504 (1.281-1.766)] and PM10 concentration (10 µg/m3) [1.255 (1.007-1.565)] on the day of the severe allergy event and with the cumulative exposure of one and two days before. Exposure to grass cover (10% area fraction) [0.655 (0.446-0.960)], forest cover (10% area fraction) [0.543 (0.303-0.973)] and density of Alnus (10%) [0.622 (0.411-0.942)] were protective for severe allergy, but only on the day of the severe allergy event. Increased densities of Betula trees (10%) were a risk factor [unadjusted OR: 2.014 (1.162-3.490)]. CONCLUSION: Exposure to green space may mitigate tree pollen allergy symptom severity but only when the density of allergenic trees is low. Air pollutants contribute to more severe allergy symptoms. Spatio-temporal tracking allows for a more realistic exposure assessment.

SARS-CoV-2 emergence and diffusion: a new disease manifesting human-environment interactions and a global geography of health.

SARS-CoV-2, and the disease it causes, COVID-19, is sweeping through the world, disrupting human activities everywhere. The consequences of this on-going event on societies are yet to be fully understood. The emergence of SARS-CoV-2 illustrates how human-environment interaction should be framing research on pathogen spillover. Furthermore, the geography of human contacts at various scales in our globalized and urbanized world affects its diffusion. Both elements plead for a robust backbone of geography of health, including land use, to understanding disease emergence and diffusion.

Modelling and mapping the intra-urban spatial distribution of Plasmodium falciparum parasite rate using very-high-resolution satellite derived indicators.

BACKGROUND: The rapid and often uncontrolled rural-urban migration in Sub-Saharan Africa is transforming urban landscapes expected to provide shelter for more than 50% of Africa's population by 2030. Consequently, the burden of malaria is increasingly affecting the urban population, while socio-economic inequalities within the urban settings are intensified. Few studies, relying mostly on moderate to high resolution datasets and standard predictive variables such as building and vegetation density, have tackled the topic of modeling intra-urban malaria at the city extent. In this research, we investigate the contribution of very-high-resolution satellite-derived land-use, land-cover and population information for modeling the spatial distribution of urban malaria prevalence across large spatial extents. As case studies, we apply our methods to two Sub-Saharan African cities, Kampala and Dar es Salaam. METHODS: Openly accessible land-cover, land-use, population and OpenStreetMap data were employed to spatially model Plasmodium falciparum parasite rate standardized to the age group 2-10 years (PfPR2-10) in the two cities through the use of a Random Forest (RF) regressor. The RF models integrated physical and socio-economic information to predict PfPR2-10 across the urban landscape. Intra-urban population distribution maps were used to adjust the estimates according to the underlying population. RESULTS: The results suggest that the spatial distribution of PfPR2-10 in both cities is diverse and highly variable across the urban fabric. Dense informal settlements exhibit a positive relationship with PfPR2-10 and hotspots of malaria prevalence were found near suitable vector breeding sites such as wetlands, marshes and riparian vegetation. In both cities, there is a clear separation of higher risk in informal settlements and lower risk in the more affluent neighborhoods. Additionally, areas associated with urban agriculture exhibit higher malaria prevalence values. CONCLUSIONS: The outcome of this research highlights that populations living in informal settlements show higher malaria prevalence compared to those in planned residential neighborhoods. This is due to (i) increased human exposure to vectors, (ii) increased vector density and (iii) a reduced capacity to cope with malaria burden. Since informal settlements are rapidly expanding every year and often house large parts of the urban population, this emphasizes the need for systematic and consistent malaria surveys in such areas. Finally, this study demonstrates the importance of remote sensing as an epidemiological tool for mapping urban malaria variations at large spatial extents, and for promoting evidence-based policy making and control efforts.

Annually modelling built-settlements between remotely-sensed observations using relative changes in subnational populations and lights at night.

Mapping urban features/human built-settlement extents at the annual time step has a wide variety of applications in demography, public health, sustainable development, and many other fields. Recently, while more multitemporal urban features/human built-settlement datasets have become available, issues still exist in remotely-sensed imagery due to spatial and temporal coverage, adverse atmospheric conditions, and expenses involved in producing such datasets. Remotely-sensed annual time-series of urban/built-settlement extents therefore do not yet exist and cover more than specific local areas or city-based regions. Moreover, while a few high-resolution global datasets of urban/built-settlement extents exist for key years, the observed date often deviates many years from the assigned one. These challenges make it difficult to increase temporal coverage while maintaining high fidelity in the spatial resolution. Here we describe an interpolative and flexible modelling framework for producing annual built-settlement extents. We use a combined technique of random forest and spatio-temporal dasymetric modelling with open source subnational data to produce annual 100 m × 100 m resolution binary built-settlement datasets in four test countries located in varying environmental and developmental contexts for test periods of five-year gaps. We find that in the majority of years, across all study areas, the model correctly identified between 85 and 99% of pixels that transition to built-settlement. Additionally, with few exceptions, the model substantially out performed a model that gave every pixel equal chance of transitioning to built-settlement in each year. This modelling framework shows strong promise for filling gaps in cross-sectional urban features/built-settlement datasets derived from remotely-sensed imagery, provides a base upon which to create urban future/built-settlement extent projections, and enables further exploration of the relationships between urban/built-settlement area and population dynamics.

Malaria risk assessment and mapping using satellite imagery and boosted regression trees in the Peruvian Amazon.

This is the first study to assess the risk of co-endemic Plasmodium vivax and Plasmodium falciparum transmission in the Peruvian Amazon using boosted regression tree (BRT) models based on social and environmental predictors derived from satellite imagery and data. Yearly cross-validated BRT models were created to discriminate high-risk (annual parasite index API > 10 cases/1000 people) and very-high-risk for malaria (API > 50 cases/1000 people) in 2766 georeferenced villages of Loreto department, between 2010-2017 as other parts in the article (graphs, tables, and texts). Predictors were cumulative annual rainfall, forest coverage, annual forest loss, annual mean land surface temperature, normalized difference vegetation index (NDVI), normalized difference water index (NDWI), shortest distance to rivers, time to populated villages, and population density. BRT models built with predictor data of a given year efficiently discriminated the malaria risk for that year in villages (area under the ROC curve (AUC) > 0.80), and most models also effectively predicted malaria risk in the following year. Cumulative rainfall, population density and time to populated villages were consistently the top three predictors for both P. vivax and P. falciparum incidence. Maps created using the BRT models characterize the spatial distribution of the malaria incidence in Loreto and should contribute to malaria-related decision making in the area.

Correction to: Extending Data for Urban Health Decision-Making: a Menu of New and Potential Neighborhood-Level Health Determinants Datasets in LMICs.

Readers should note an additional Acknowledgment for this article: Dana Thomson is funded by the Economic and Social Research Council grant number ES/5500161/1.

Extending Data for Urban Health Decision-Making: a Menu of New and Potential Neighborhood-Level Health Determinants Datasets in LMICs.

Area-level indicators of the determinants of health are vital to plan and monitor progress toward targets such as the Sustainable Development Goals (SDGs). Tools such as the Urban Health Equity Assessment and Response Tool (Urban HEART) and UN-Habitat Urban Inequities Surveys identify dozens of area-level health determinant indicators that decision-makers can use to track and attempt to address population health burdens and inequalities. However, questions remain as to how such indicators can be measured in a cost-effective way. Area-level health determinants reflect the physical, ecological, and social environments that influence health outcomes at community and societal levels, and include, among others, access to quality health facilities, safe parks, and other urban services, traffic density, level of informality, level of air pollution, degree of social exclusion, and extent of social networks. The identification and disaggregation of indicators is necessarily constrained by which datasets are available. Typically, these include household- and individual-level survey, census, administrative, and health system data. However, continued advancements in earth observation (EO), geographical information system (GIS), and mobile technologies mean that new sources of area-level health determinant indicators derived from satellite imagery, aggregated anonymized mobile phone data, and other sources are also becoming available at granular geographic scale. Not only can these data be used to directly calculate neighborhood- and city-level indicators, they can be combined with survey, census, administrative and health system data to model household- and individual-level outcomes (e.g., population density, household wealth) with tremendous detail and accuracy. WorldPop and the Demographic and Health Surveys (DHS) have already modeled dozens of household survey indicators at country or continental scales at resolutions of 1 × 1 km or even smaller. This paper aims to broaden perceptions about which types of datasets are available for health and development decision-making. For data scientists, we flag area-level indicators at city and sub-city scales identified by health decision-makers in the SDGs, Urban HEART, and other initiatives. For local health decision-makers, we summarize a menu of new datasets that can be feasibly generated from EO, mobile phone, and other spatial data-ideally to be made free and publicly available-and offer lay descriptions of some of the difficulties in generating such data products.

Publisher Correction: Past and future spread of the arbovirus vectors Aedes aegypti and Aedes albopictus.

This Article was mistakenly not made Open Access when originally published; this has now been amended, and information about the Creative Commons Attribution 4.0 International License has been added into the 'Additional information' section.

Publisher Correction: Past and future spread of the arbovirus vectors Aedes aegypti and Aedes albopictus.

In the version of this Article originally published, the affiliation for author Catherine Linard was incorrectly stated as '6Department of Infectious Disease Epidemiology, London School of Hygiene and Tropical Medicine, London, UK'. The correct affiliation is '9Spatial Epidemiology Lab (SpELL), Universite Libre de Bruxelles, Brussels, Belgium'. The affiliation for author Hongjie Yu was also incorrectly stated as '11Department of Statistics, Harvard University, Cambridge, MA, USA'. The correct affiliation is '15School of Health, Fudan University, Key Laboratory of Public Health Safety, Ministry of Education, Shanghai, China'. This has now been amended in all versions of the Article.