Probabilistic seasonal dengue forecasting in Vietnam: A modelling study using superensembles

BackgroundWith enough advanced notice, dengue outbreaks can be mitigated. As a climate-sensitive disease, environmental conditions and past patterns of dengue can be used to make predictions about future outbreak risk. These predictions improve public health planning and decision-making to ultimately reduce the burden of disease. Past approaches to dengue forecasting have used seasonal climate forecasts, but the predictive ability of a system using different lead times in a year-round prediction system has been seldom explored. Moreover, the transition from theoretical to operational systems integrated with disease control activities is rare.Methods and findingsWe introduce an operational seasonal dengue forecasting system for Vietnam where Earth observations, seasonal climate forecasts, and lagged dengue cases are used to drive a superensemble of probabilistic dengue models to predict dengue risk up to 6 months ahead. Bayesian spatiotemporal models were fit to 19 years (2002–2020) of dengue data at the province level across Vietnam. A superensemble of these models then makes probabilistic predictions of dengue incidence at various future time points aligned with key Vietnamese decision and planning deadlines. We demonstrate that the superensemble generates more accurate predictions of dengue incidence than the individual models it incorporates across a suite of time horizons and transmission settings. Using historical data, the superensemble made slightly more accurate predictions (continuous rank probability score [CRPS] = 66.8, 95% CI 60.6–148.0) than a baseline model which forecasts the same incidence rate every month (CRPS = 79.4, 95% CI 78.5–80.5) at lead times of 1 to 3 months, albeit with larger uncertainty. The outbreak detection capability of the superensemble was considerably larger (69%) than that of the baseline model (54.5%). Predictions were most accurate in southern Vietnam, an area that experiences semi-regular seasonal dengue transmission. The system also demonstrated added value across multiple areas compared to previous practice of not using a forecast. We use the system to make a prospective prediction for dengue incidence in Vietnam for the period May to October 2020. Prospective predictions made with the superensemble were slightly more accurate (CRPS = 110, 95% CI 102–575) than those made with the baseline model (CRPS = 125, 95% CI 120–168) but had larger uncertainty. Finally, we propose a framework for the evaluation of probabilistic predictions. Despite the demonstrated value of our forecasting system, the approach is limited by the consistency of the dengue case data, as well as the lack of publicly available, continuous, and long-term data sets on mosquito control efforts and serotype-specific case data.ConclusionsThis study shows that by combining detailed Earth observation data, seasonal climate forecasts, and state-of-the-art models, dengue outbreaks can be predicted across a broad range of settings, with enough lead time to meaningfully inform dengue control. While our system omits some important variables not currently available at a subnational scale, the majority of past outbreaks could be predicted up to 3 months ahead. Over the next 2 years, the system will be prospectively evaluated and, if successful, potentially extended to other areas and other climate-sensitive disease systems.     

Progress and opportunities in advancing near-term forecasting of freshwater quality

Near-term freshwater forecasts, defined as sub-daily to decadal future predictions of a freshwater variable with quantified uncertainty, are urgently needed to improve water quality management as freshwater ecosystems exhibit greater variability due to global change. Shifting baselines in freshwater ecosystems due to land use and climate change prevent managers from relying on historical averages for predicting future conditions, necessitating near-term forecasts to mitigate freshwater risks to human health and safety (e.g., flash floods, harmful algal blooms) and ecosystem services (e.g., water-related recreation and tourism). To assess the current state of freshwater forecasting and identify opportunities for future progress, we synthesized freshwater forecasting papers published in the past 5 years. We found that freshwater forecasting is currently dominated by near-term forecasts of water quantity and that near-term water quality forecasts are fewer in number and in the early stages of development (i.e., non-operational) despite their potential as important preemptive decision support tools. We contend that more freshwater quality forecasts are critically needed and that near-term water quality forecasting is poised to make substantial advances based on examples of recent progress in forecasting methodology, workflows, and end-user engagement. For example, current water quality forecasting systems can predict water temperature, dissolved oxygen, and algal bloom/toxin events 5 days ahead with reasonable accuracy. Continued progress in freshwater quality forecasting will be greatly accelerated by adapting tools and approaches from freshwater quantity forecasting (e.g., machine learning modeling methods). In addition, future development of effective operational freshwater quality forecasts will require substantive engagement of end users throughout the forecast process, funding, and training opportunities. Looking ahead, near-term forecasting provides a hopeful future for freshwater management in the face of increased variability and risk due to global change, and we encourage the freshwater scientific community to incorporate forecasting approaches in water quality research and management.     

Forecasting and chronic illness

Biomedical data in the form of series of observations made on a single process at regular intervals constitute a discrete time series and are eligible for time series methods of analysis. The models yielded by this analysis provide the framework within which exponential smoothing methods may operate on the data to provide recurrent forecasts of future states of the process. Because the forecasts may be made on an individual basis and are sensitive to the past behavior of the individual process, the methods are presented as being potentially of great utility in the management of chronic and progressive illnesses. When incorporated into automated testing and diagnostic systems, the forecasting method will provide the capability of making prognoses for large numbers of individuals, quickly, routinely and reproducibly.     

Correcting delayed reporting of COVID-19 using the generalized-Dirichlet-multinomial method

The COVID-19 pandemic has highlighted delayed reporting as a significant impediment to effective disease surveillance and decision-making. In the absence of timely data, statistical models which account for delays can be adopted to nowcast and forecast cases or deaths. We discuss the four key sources of systematic and random variability in available data for COVID-19 and other diseases, and critically evaluate current state-of-the-art methods with respect to appropriately separating and capturing this variability. We propose a general hierarchical approach to correcting delayed reporting of COVID-19 and apply this to daily English hospital deaths, resulting in a flexible prediction tool which could be used to better inform pandemic decision-making. We compare this approach to competing models with respect to theoretical flexibility and quantitative metrics from a 15-month rolling prediction experiment imitating a realistic operational scenario. Based on consistent leads in predictive accuracy, bias, and precision, we argue that this approach is an attractive option for correcting delayed reporting of COVID-19 and future epidemics.     

Complexity is costly: a meta‐analysis of parametric and non‐parametric methods for short‐term population forecasting

Short‐term forecasts based on time series of counts or survey data are widely used in population biology to provide advice concerning the management, harvest and conservation of natural populations. A common approach to produce these forecasts uses time‐series models, of different types, fit to time series of counts. Similar time‐series models are used in many other disciplines, however relative to the data available in these other disciplines, population data are often unusually short and noisy and models that perform well for data from other disciplines may not be appropriate for population data. In order to study the performance of time‐series forecasting models for natural animal population data, we assembled 2379 time series of vertebrate population indices from actual surveys. Our data were comprised of three vastly different types: highly variable (marine fish productivity), strongly cyclic (adult salmon counts), and small variance but long‐memory (bird and mammal counts). We tested the predictive performance of 49 different forecasting models grouped into three broad classes: autoregressive time‐series models, non‐linear regression‐type models and non‐parametric time‐series models. Low‐dimensional parametric autoregressive models gave the most accurate forecasts across a wide range of taxa; the most accurate model was one that simply treated the most recent observation as the forecast. More complex parametric and non‐parametric models performed worse, except when applied to highly cyclic species. Across taxa, certain life history characteristics were correlated with lower forecast error; specifically, we found that better forecasts were correlated with attributes of slow growing species: large maximum age and size for fishes and high trophic level for birds. Synthesis Evaluating the data support for multiple plausible models has been an integral focus of many ecological analyses. However, the most commonly used tools to quantify support have weighted models’ hindcasting and forecasting abilities. For many applications, predicting the past may be of little interest. Concentrating only on the future predictive performance of time series models, we performed a forecasting competition among many different kinds of statistical models, applying each to many different kinds of vertebrate time series of population abundance. Low‐dimensional (simple) models performed well overall, but more complex models did slightly better when applied to time series of cyclic species (e.g. salmon).     

The heritability of fertility makes world population stabilization unlikely in the foreseeable future

The forecasting of the future growth of world population is of critical importance to anticipate and address a wide range of global challenges. The United Nations produces forecasts of fertility and world population every two years. As part of these forecasts, they model fertility levels in post-demographic transition countries as tending toward a long-term mean, leading to forecasts of flat or declining population in these countries. We substitute this assumption of constant long-term fertility with a dynamic model, theoretically founded in evolutionary biology, with heritable fertility. Rather than stabilizing around a long-term level for post-demographic transition countries, fertility tends to increase as children from larger families represent a larger share of the population and partly share their parents' trait of having more offspring. Our results suggest that world population will grow larger in the future than currently anticipated.     

An ensemble data assimilation modeling system for operational outdoor microalgae growth forecasting

Microalgae have received increasing attention as a potential feedstock for biofuel or biobased products. Forecasting the microalgae growth is beneficial for managers in planning pond operations and harvesting decisions. This study proposed a biomass forecasting system comprised of the Huesemann Algae Biomass Growth Model (BGM), the Modular Aquatic Simulation System in Two Dimensions (MASS2), ensemble data assimilation (DA), and numerical weather prediction Global Ensemble Forecast System (GEFS) ensemble meteorological forecasts. The novelty of this study is to seek the use of ensemble DA to improve both BGM and MASS2 model initial conditions with the assimilation of biomass and water temperature measurements and consequently improve short-term biomass forecasting skills. This study introduces the theory behind the proposed integrated biomass forecasting system, with an application undertaken in pseudo-real-time in three outdoor ponds cultured with Chlorella sorokiniana in Delhi, California, United States. Results from all three case studies demonstrate that the biomass forecasting system improved the short-term (i.e., 7-day) biomass forecasting skills by about 60% on average, comparing to forecasts without using the ensemble DA method. Given the satisfactory performances achieved in this study, it is probable that the integrated BGM-MASS2-DA forecasting system can be used operationally to inform managers in making pond operation and harvesting planning decisions.     

Nowcasting and Forecasting the Monthly Food Stamps Data in the US Using Online Search Data

We propose the use of Google online search data for nowcasting and forecasting the number of food stamps recipients. We perform a large out-of-sample forecasting exercise with almost 3000 competing models with forecast horizons up to 2 years ahead, and we show that models including Google search data statistically outperform the competing models at all considered horizons. These results hold also with several robustness checks, considering alternative keywords, a falsification test, different out-of-samples, directional accuracy and forecasts at the state-level.     

A scalable deep learning system for monitoring and forecasting pollutant concentration levels on UK highways

The construction of intercity highways by the government has resulted in a progressive increase in vehicle emissions and pollution from noise, dust, and vibrations despite its recognition of the air pollution menace. Efforts that have targeted roadside pollution still do not accurately monitor deadly pollutants such as nitrogen oxides and particulate matter. Reports on regional highways across the country are based on a limited number of fixed monitoring stations that are sometimes located far from the highway. These periodic and coarse-grained measurements cause inefficient highway air quality reporting, leading to inaccurate air quality forecasts. This paper, therefore, proposes and validates a scalable deep learning framework for efficiently capturing fine-grained highway data and forecasting future concentration levels. Highways in four different UK regions - Newport, Lewisham, Southwark, and Chepstow were used as case studies to develop a REVIS system and validate the proposed framework. REVIS examined the framework's ability to capture granular pollution data, scale up its storage facility to rapid data growth and translate high-level user queries to structured query language (SQL) required for exploratory data analysis. Finally, the framework's suitability for predictive analytics was tested using fastai's library for tabular data, and automated hyperparameter tuning was implemented using bayesian optimisation. The results of our experiments demonstrate the suitability of the proposed framework in building end-to-end systems for extensive monitoring and forecasting of pollutant concentration levels on highways. The study serves as a background for future related research looking to improve the overall performance of roadside and highway air quality forecasting models.     

Bayesian back-calculation and nowcasting for line list data during the COVID-19 pandemic

Surveillance is critical to mounting an appropriate and effective response to pandemics. However, aggregated case report data suffers from reporting delays and can lead to misleading inferences. Different from aggregated case report data, line list data is a table contains individual features such as dates of symptom onset and reporting for each reported case and a good source for modeling delays. Current methods for modeling reporting delays are not particularly appropriate for line list data, which typically has missing symptom onset dates that are non-ignorable for modeling reporting delays. In this paper, we develop a Bayesian approach that dynamically integrates imputation and estimation for line list data. Specifically, this Bayesian approach can accurately estimate the epidemic curve and instantaneous reproduction numbers, even with most symptom onset dates missing. The Bayesian approach is also robust to deviations from model assumptions, such as changes in the reporting delay distribution or incorrect specification of the maximum reporting delay. We apply the Bayesian approach to COVID-19 line list data in Massachusetts and find the reproduction number estimates correspond more closely to the control measures than the estimates based on the reported curve.     

An approximate Bayesian approach for estimation of the instantaneous reproduction number under misreported epidemic data

In epidemic models, the effective reproduction number is of central importance to assess the transmission dynamics of an infectious disease and to orient health intervention strategies. Publicly shared data during an outbreak often suffers from two sources of misreporting (underreporting and delay in reporting) that should not be overlooked when estimating epidemiological parameters. The main statistical challenge in models that intrinsically account for a misreporting process lies in the joint estimation of the time-varying reproduction number and the delay/underreporting parameters. Existing Bayesian approaches typically rely on Markov chain Monte Carlo algorithms that are extremely costly from a computational perspective. We propose a much faster alternative based on Laplacian-P-splines (LPS) that combines Bayesian penalized B-splines for flexible and smooth estimation of the instantaneous reproduction number and Laplace approximations to selected posterior distributions for fast computation. Assuming a known generation interval distribution, the incidence at a given calendar time is governed by the epidemic renewal equation and the delay structure is specified through a composite link framework. Laplace approximations to the conditional posterior of the spline vector are obtained from analytical versions of the gradient and Hessian of the log-likelihood, implying a drastic speed-up in the computation of posterior estimates. Furthermore, the proposed LPS approach can be used to obtain point estimates and approximate credible intervals for the delay and reporting probabilities. Simulation of epidemics with different combinations for the underreporting rate and delay structure (one-day, two-day, and weekend delays) show that the proposed LPS methodology delivers fast and accurate estimates outperforming existing methods that do not take into account underreporting and delay patterns. Finally, LPS is illustrated in two real case studies of epidemic outbreaks.     

Zero-inflated Poisson models with measurement error in the response

Zero-inflated count data arise frequently from genomics studies. Analysis of such data is often based on a mixture model which facilitates excess zeros in combination with a Poisson distribution, and various inference methods have been proposed under such a model. Those analysis procedures, however, are challenged by the presence of measurement error in count responses. In this article, we propose a new measurement error model to describe error-contaminated count data. We show that ignoring the measurement error effects in the analysis may generally lead to invalid inference results, and meanwhile, we identify situations where ignoring measurement error can still yield consistent estimators. Furthermore, we propose a Bayesian method to address the effects of measurement error under the zero-inflated Poisson model and discuss the identifiability issues. We develop a data-augmentation algorithm that is easy to implement. Simulation studies are conducted to evaluate the performance of the proposed method. We apply our method to analyze the data arising from a prostate adenocarcinoma genomic study.     

Ecological forecasts to inform near‐term management of threats to biodiversity

Ecosystems are being altered by rapid and interacting changes in natural processes and anthropogenic threats to biodiversity. Uncertainty in historical, current and future effectiveness of actions hampers decisions about how to mitigate changes to prevent biodiversity loss and species extinctions. Research in resource management, agriculture and health indicates that forecasts predicting the effects of near‐term or seasonal environmental conditions on management greatly improve outcomes. Such forecasts help resolve uncertainties about when and how to operationalize management. We reviewed the scientific literature on environmental management to investigate whether near‐term forecasts are developed to inform biodiversity decisions in Australia, a nation with one of the highest recent extinction rates across the globe. We found that forecasts focused on economic objectives (e.g. fisheries management) predict on significantly shorter timelines and answer a broader range of management questions than forecasts focused on biodiversity conservation. We then evaluated scientific literature on the effectiveness of 484 actions to manage seven major terrestrial threats in Australia, to identify opportunities for near‐term forecasts to inform operational conservation decisions. Depending on the action, between 30% and 80% threat management operations experienced near‐term weather impacts on outcomes before, during or after management. Disease control, species translocation/reintroduction and habitat restoration actions were most frequently impacted, and negative impacts such as increased species mortality and reduced recruitment were more likely than positive impacts. Drought or dry conditions, and rainfall, were the most frequently reported weather impacts, indicating that near‐term forecasts predicting the effects of low or excessive rainfall on management outcomes are likely to have the greatest benefits. Across the world, many regions are, like Australia, becoming warmer and drier, or experiencing more extreme rainfall events. Informing conservation decisions with near‐term and seasonal ecological forecasting will be critical to harness uncertainties and lower the risk of threat management failure under global change.     

Adjusting for reporting delay in cancer incidence when combining different sets of cancer registries

Cancer registries collect cancer incidence data that can be used to calculate incidence rates in a population and track changes over time. For incidence rates to be accurate, it is critical that diagnosed cases be reported in a timely manner. Registries typically allow a fixed amount of time (e.g. two years) for diagnosed cases to be reported before releasing the initial case counts for a particular diagnosis year. Inevitably, however, additional cases are reported after the initial counts are released; these extra cases are included in subsequent releases that become more complete over time, while incidence rates based on earlier releases will underestimate the true rates. Statistical methods have been developed to estimate the distribution of reporting delay (the amount of time until a diagnosed case is reported) and to correct incidence rates for underestimation due to reporting delay. Since the observed reporting delays must be less than the length of time the registry has been collecting data, most methods estimate a truncated delay distribution. These methods can be applied to a group of registries that began collecting data in the same diagnosis year. In this paper, we extend the methods to two groups of registries that began collecting data in two different diagnosis years (so that the delay distributions are truncated at different times). We apply the proposed method to data from the National Cancer Institute's Surveillance Epidemiology and End Results (SEER) program, a consortium of U.S. cancer registries that includes nine registries with data collection beginning in 1981 and four registries with data collection beginning in 1992. We use the method to obtain delay‐adjusted incidence rates for melanoma, liver cancer, and Hodgkin lymphoma.     

When mechanism matters: Bayesian forecasting using models of ecological diffusion

Ecological diffusion is a theory that can be used to understand and forecast spatio‐temporal processes such as dispersal, invasion, and the spread of disease. Hierarchical Bayesian modelling provides a framework to make statistical inference and probabilistic forecasts, using mechanistic ecological models. To illustrate, we show how hierarchical Bayesian models of ecological diffusion can be implemented for large data sets that are distributed densely across space and time. The hierarchical Bayesian approach is used to understand and forecast the growth and geographic spread in the prevalence of chronic wasting disease in white‐tailed deer (Odocoileus virginianus). We compare statistical inference and forecasts from our hierarchical Bayesian model to phenomenological regression‐based methods that are commonly used to analyse spatial occurrence data. The mechanistic statistical model based on ecological diffusion led to important ecological insights, obviated a commonly ignored type of collinearity, and was the most accurate method for forecasting.     

Estimating absolute and relative case fatality ratios from infectious disease surveillance data

Knowing which populations are most at risk for severe outcomes from an emerging infectious disease is crucial in deciding the optimal allocation of resources during an outbreak response. The case fatality ratio (CFR) is the fraction of cases that die after contracting a disease. The relative CFR is the factor by which the case fatality in one group is greater or less than that in a second group. Incomplete reporting of the number of infected individuals, both recovered and dead, can lead to biased estimates of the CFR. We define conditions under which the CFR and the relative CFR are identifiable. Furthermore, we propose an estimator for the relative CFR that controls for time-varying reporting rates. We generalize our methods to account for elapsed time between infection and death. To demonstrate the new methodology, we use data from the 1918 influenza pandemic to estimate relative CFRs between counties in Maryland. A simulation study evaluates the performance of the methods in outbreak scenarios. An R software package makes the methods and data presented here freely available. Our work highlights the limitations and challenges associated with estimating absolute and relative CFRs in practice. However, in certain situations, the methods presented here can help identify vulnerable subpopulations early in an outbreak of an emerging pathogen such as pandemic influenza.     

Faster indicators of chikungunya incidence using Google searches

Chikungunya, a mosquito-borne disease, is a growing threat in Brazil, where over 640,000 cases have been reported since 2017. However, there are often long delays between diagnoses of chikungunya cases and their entry in the national monitoring system, leaving policymakers without the up-to-date case count statistics they need. In contrast, weekly data on Google searches for chikungunya is available with no delay. Here, we analyse whether Google search data can help improve rapid estimates of chikungunya case counts in Rio de Janeiro, Brazil. We build on a Bayesian approach suitable for data that is subject to long and varied delays, and find that including Google search data reduces both model error and uncertainty. These improvements are largest during epidemics, which are particularly important periods for policymakers. Including Google search data in chikungunya surveillance systems may therefore help policymakers respond to future epidemics more quickly.     

River Restoration in Victoria, Australia: Change is in the Wind, and None too Soon

Stream restoration has become a multibillion dollar industry worldwide, yet there are few clear success stories and the scientific basis for effective stream restoration remains uncertain. We compiled data on completed river restoration projects from four management authorities in Victoria, Australia, to examine how the available data could inform the science of restoration ecology in rivers, and thus improve future restoration efforts. We found that existing data sources are limited and much historical information has been lost through industry restructuring and poor data archiving. Examining records for 2,247 restoration projects, we found that riparian management projects were the most common, followed by bank stabilization and in‐stream habitat improvement. Only 14% of the project records indicated that some form of monitoring was carried out. It is evident that overall there is little scientific guidance and little or no monitoring and evaluation of the projects for which we had information. However, recent advances with mandatory, statewide reporting and an increased emphasis on project design and monitoring strongly suggest that the design, implementation, monitoring, and reporting of stream restoration projects have improved in recent years and will continue to do so.     

Recommended reporting items for epidemic forecasting and prediction research: The EPIFORGE 2020 guidelines

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

Simon Pollett and co-workers describe EPIFORGE, a guideline for reporting research on epidemic forecasting.     

The power of data synthesis to shape the future of the restoration community and capacity

Restoration efforts will be taking place over the next decade(s) in the largest scope and capacity ever seen. Immense commitments, goals, and budgets are set, with impactful wide‐reaching potential benefits for people and the environment. These are ambitious aims for a relatively new branch of science and practice. It is time for restoration action to scale up, the legacy of which could impact over 350 million hectares targeted for the U.N. Decade on Ecosystem Restoration. However, restoration still proceeds on a case‐by‐case, trial by error basis and restoration outcomes can be variable even under similar conditions. The ability to put each case into context—what about it worked, what did not, and why—is something that the synthesis of data across studies can facilitate. The link between data synthesis and predictive capacity is strong. There are examples of extremely ambitious and successful efforts to compile data in structured, standardized databases which have led to valuable insights across regional and global scales in other branches of science. There is opportunity and challenge in compiling, standardizing, and synthesizing restoration monitoring data to inform the future of restoration practice and science. Through global collation of restoration data, knowledge gaps can be addressed and data synthesized to advance toward a more predictive science to inform more consistent success. The interdisciplinary potential of restoration ecology sits just over the horizon of this decade. Through truly collaborative synthesis across foci within the restoration community, we have the opportunity to rapidly reach that potential and achieve extraordinary outcomes together.