Oscar Patterson-Lomba

Temporal Variations in the Effective Reproduction Number of the 2014 West Africa Ebola Outbreak

Background The rapidly evolving 2014 Ebola virus disease (EVD) outbreak in West Africa is the largest documented in history, both in terms of the number of people infected and in the geographic spread. The high morbidity and mortality have inspired response strategies to the outbreak at the individual, regional, and national levels. Methods to provide real-time assessment of changing transmission dynamics are critical to the understanding of how these adaptive intervention measures have affected the spread of the outbreak. Methods In this analysis, we use the time series of EVD cases in Guinea, Sierra Leone, and Liberia up to September 8, 2014, and employ novel methodology to estimate how the rate of exponential rise of new cases has changed over the outbreak using piecewise fits of exponential curves to the outbreak data. Results We find that for Liberia and Guinea, the effective reproduction number rose, rather than fell, around the time that the outbreak spread to densely populated cities, and enforced quarantine was imposed on several regions in the countries; this may indicate that enforced quarantine may not be an effective control measure. Conclusions If effective control measures are not put in place, and the current rate of exponential rise of new cases continues, we predict 4400 new Ebola cases in West Africa during the last half of the month of September, with an upper 95% confidence level of 6800 new cases.

Estimating influenza attack rates in the United States using a participatory cohort

We considered how participatory syndromic surveillance data can be used to estimate influenza attack rates during the 2012–2013 and 2013–2014 seasons in the United States. Our inference is based on assessing the difference in the rates of self-reported influenza-like illness (ILI, defined as presence of fever and cough/sore throat) among the survey participants during periods of active vs. low influenza circulation as well as estimating the probability of self-reported ILI for influenza cases. Here, we combined Flu Near You data with additional sources (Hong Kong household studies of symptoms of influenza cases and the U.S. Centers for Disease Control and Prevention estimates of vaccine coverage and effectiveness) to estimate influenza attack rates. The estimated influenza attack rate for the early vaccinated Flu Near You members (vaccination reported by week 45) aged 20–64 between calendar weeks 47–12 was 14.7%(95% CI(5.9%,24.1%)) for the 2012–2013 season and 3.6%(−3.3%,10.3%) for the 2013–2014 season. The corresponding rates for the US population aged 20–64 were 30.5% (4.4%, 49.3%) in 2012–2013 and 7.1%(−5.1%, 32.5%) in 2013–2014. The attack rates in women and men were similar each season. Our findings demonstrate that participatory syndromic surveillance data can be used to gauge influenza attack rates during future influenza seasons.

Utilizing Syndromic Surveillance Data for Estimating Levels of Influenza Circulation

The availability of weekly Web-based participatory surveillance data on self-reported influenza-like illness (ILI), defined here as self-reported fever and cough/sore throat, over several influenza seasons allows for estimation of the incidence of influenza infection in population cohorts. We demonstrate this using syndromic data reported through the Influenzanet surveillance platform in the Netherlands. We used the 2011-2012 influenza season, a low-incidence season that began late, to assess the baseline rates of self-reported ILI during periods of low influenza circulation, and we used ILI rates above that baseline level from the 2012-1013 season, a major influenza season, to estimate influenza attack rates for that period. The latter conversion required estimates of age-specific probabilities of self-reported ILI given influenza (Flu) infection (P(ILI | Flu)), which were obtained from separate data (extracted from Hong Kong, China, household studies). For the 2012-2013 influenza season in the Netherlands, we estimated combined influenza A/B attack rates of 29.2% (95% credible interval (CI): 21.6, 37.9) among survey participants aged 20-49 years, 28.3% (95% CI: 20.7, 36.8) among participants aged 50-60 years, and 5.9% (95% CI: 0.4, 11.8) among participants aged ≥61 years. Estimates of influenza attack rates can be obtained in other settings using analogous, multiseason surveillance data on self-reported ILI together with separate, context-specific estimates of P(ILI | Flu).

Explaining the Prevalence, Scaling and Variance of Urban Phenomena

The prevalence of many urban phenomena changes systematically with population size1. We propose a theory that unifies models of economic complexity2,3 and cultural evolution4 to derive urban scaling. The theory accounts for the difference in scaling exponents and average prevalence across phenomena, as well as the difference in the variance within phenomena across cities of similar size. The central ideas are that a number of necessary complementary factors must be simultaneously present for a phenomenon to occur, and that the diversity of factors is logarithmically related to population size. The model reveals that phenomena that require more factors will be less prevalent, scale more superlinearly and show larger variance across cities of similar size. The theory applies to data on education, employment, innovation, disease and crime, and it entails the ability to predict the prevalence of a phenomenon across cities, given information about the prevalence in a single city.

Per capita incidence of sexually transmitted infections increases systematically with urban population size: a cross-sectional study

Objectives Rampant urbanisation rates across the globe demand that we improve our understanding of how infectious diseases spread in modern urban landscapes, where larger and more connected host populations enhance the thriving capacity of certain pathogens. Methods A data-driven approach is employed to study the ability of sexually transmitted diseases (STDs) to thrive in urban areas. The conduciveness of population size of urban areas and their socioeconomic characteristics are used as predictors of disease incidence, using confirmed-case data on STDs in the USA as a case study. Results A superlinear relation between STD incidence and urban population size is found, even after controlling for various socioeconomic aspects, suggesting that doubling the population size of a city results in an expected increase in STD incidence larger than twofold, provided that all other socioeconomic aspects remain fixed. Additionally, the percentage of African–Americans, income inequalities, education and per capita income are found to have a significant impact on the incidence of each of the three STDs studied. Conclusions STDs disproportionately concentrate in larger cities. Hence, larger urban areas merit extra prevention and treatment efforts, especially in low-income and middle-income countries where urbanisation rates are higher.

A Generalized Entropy Measure of Within-Host Viral Diversity for Identifying Recent HIV-1 Infections

Abstract There is a need for incidence assays that accurately estimate HIV incidence based on cross-sectional specimens. Viral diversity-based assays have shown promises but are not particularly accurate. We hypothesize that certain viral genetic regions are more predictive of recent infection than others and aim to improve assay accuracy by using classification algorithms that focus on highly informative regions (HIRs). We analyzed HIV gag sequences from a cohort in Botswana. Forty-two subjects newly infected by HIV-1 Subtype C were followed through 500 days post-seroconversion. Using sliding window analysis, we screened for genetic regions within gag that best differentiate recent versus chronic infections. We used both nonparametric and parametric approaches to evaluate the discriminatory abilities of sequence regions. Segmented Shannon Entropy measures of HIRs were aggregated to develop generalized entropy measures to improve prediction of recency. Using logistic regression as the basis for our classification algorithm, we evaluated the predictive power of these novel biomarkers and compared them with recently reported viral diversity measures using area under the curve (AUC) analysis. Change of diversity over time varied across different sequence regions within gag. We identified the top 50% of the most informative regions by both nonparametric and parametric approaches. In both cases, HIRs were in more variable regions of gag and less likely in the p24 coding region. Entropy measures based on HIRs outperformed previously reported viral-diversity-based biomarkers. These methods are better suited for population-level estimation of HIV recency. The patterns of diversification of certain regions within the gag gene are more predictive of recency of infection than others. We expect this result to apply in other HIV genetic regions as well. Focusing on these informative regions, our generalized entropy measure of viral diversity demonstrates the potential for improving accuracy when identifying recent HIV-1 infections.

Predicting Subnational Ebola Virus Disease Epidemic Dynamics from Sociodemographic Indicators.

Background The recent Ebola virus disease (EVD) outbreak in West Africa has spread wider than any previous human EVD epidemic. While individual-level risk factors that contribute to the spread of EVD have been studied, the population-level attributes of subnational regions associated with outbreak severity have not yet been considered. Methods To investigate the area-level predictors of EVD dynamics, we integrated time series data on cumulative reported cases of EVD from the World Health Organization and covariate data from the Demographic and Health Surveys. We first estimated the early growth rates of epidemics in each second-level administrative district (ADM2) in Guinea, Sierra Leone and Liberia using exponential, logistic and polynomial growth models. We then evaluated how these growth rates, as well as epidemic size within ADM2s, were ecologically associated with several demographic and socio-economic characteristics of the ADM2, using bivariate correlations and multivariable regression models. Results The polynomial growth model appeared to best fit the ADM2 epidemic curves, displaying the lowest residual standard error. Each outcome was associated with various regional characteristics in bivariate models, however in stepwise multivariable models only mean education levels were consistently associated with a worse local epidemic. Discussion By combining two common methods—estimation of epidemic parameters using mathematical models, and estimation of associations using ecological regression models—we identified some factors predicting rapid and severe EVD epidemics in West African subnational regions. While care should be taken interpreting such results as anything more than correlational, we suggest that our approach of using data sources that were publicly available in advance of the epidemic or in real-time provides an analytic framework that may assist countries in understanding the dynamics of future outbreaks as they occur.

Modeling the role of healthcare access inequalities in epidemic outcomes

Urban areas, with large and dense populations, offer conditions that favor the emergence and spread of certain infectious diseases. One common feature of urban populations is the existence of large socioeconomic inequalities which are often mirrored by disparities in access to healthcare. Recent empirical evidence suggests that higher levels of socioeconomic inequalities are associated with worsened public health outcomes, including higher rates of sexually transmitted diseases (STDs) and lower life expectancy. However, the reasons for these associations are still speculative. Here we formulate a mathematical model to study the effect of healthcare disparities on the spread of an infectious disease that does not confer lasting immunity, such as is true of certain STDs. Using a simple epidemic model of a population divided into two groups that differ in their recovery rates due to different levels of access to healthcare, we find that both the basic reproductive number (R0) of the disease and its endemic prevalence are increasing functions of the disparity between the two groups, in agreement with empirical evidence. Unexpectedly, this can be true even when the fraction of the population with better access to healthcare is increased if this is offset by reduced access within the disadvantaged group. Extending our model to more than two groups with different levels of access to healthcare, we find that increasing the variance of recovery rates among groups, while keeping the mean recovery rate constant, also increases R0 and disease prevalence. In addition, we show that these conclusions are sensitive to how we quantify the inequalities in our model, underscoring the importance of basing analyses on appropriate measures of inequalities. These insights shed light on the possible impact that increasing levels of inequalities in healthcare access can have on epidemic outcomes, while offering plausible explanations for the observed empirical patterns.

Assessing Biases in the Evaluation of Classification Assays for HIV Infection Recency

Identifying recent HIV infection cases has important public health and clinical implications. It is essential for estimating incidence rates to monitor epidemic trends and evaluate the effectiveness of interventions. Detecting recent cases is also important for HIV prevention given the crucial role that recently infected individuals play in disease transmission, and because early treatment onset can improve the clinical outlook of patients while reducing transmission risk. Critical to this enterprise is the development and proper assessment of accurate classification assays that, based on cross-sectional samples of viral sequences, help determine infection recency status. In this work we assess some of the biases present in the evaluation of HIV recency classification algorithms that rely on measures of within-host viral diversity. Particularly, we examine how the time since infection (TSI) distribution of the infected subjects from which viral samples are drawn affect performance metrics (e.g., area under the ROC curve, sensitivity, specificity, accuracy and precision), potentially leading to misguided conclusions about the efficacy of classification assays. By comparing the performance of a given HIV recency assay using six different TSI distributions (four simulated TSI distributions representing different epidemic scenarios, and two empirical TSI distributions), we show that conclusions about the overall efficacy of the assay depend critically on properties of the TSI distribution. Moreover, we demonstrate that an assay with high overall classification accuracy, mainly due to properly sorting members of the well-represented groups in the validation dataset, can still perform notoriously poorly when sorting members of the less represented groups. This is an inherent issue of classification and diagnostics procedures that is often underappreciated. Thus, this work underscores the importance of acknowledging and properly addressing evaluation biases when proposing new HIV recency assays.