The application of environmental management methods in combating dengue: a systematic review.

Among several vector control methods commonly used, environmental management is one of the control measures to mitigate dengue in such vulnerable communities. Since there is no curative treatment for dengue yet, targeted environmental and ecosystem management is increasingly relevant. Hence, this review was conducted to identify the effectiveness of environmental management intervention strategy to reduce dengue cases. We searched PUBMED, CENTRAL, SCOPUS, Web of Science, CIJE, WHO ICTRP, and ClinicalTrials.gov up to January 2021. A total of 521 articles were screened. Only 16 studies were included in this review. There were 6 studies that applied all three types of environmental management interventions (manipulation, modification and behavior), 8 studies applied two types of interventions (manipulation and behavior) and 2 studies applied one type of intervention (manipulation or behavior). All included studies reported reduction of Aedes entomological indices. The studies showed reduction in dengue cases and density of Aedes population through environmental interventions. It is recommended for the health authority to incorporate environmental management intervention in dengue control activities and enhanced the community involvement to ensure sustainability with high impact on dengue reduction.

Neighbourhood level real-time forecasting of dengue cases in tropical urban Singapore.

BACKGROUND: Dengue, a vector-borne infectious disease caused by the dengue virus, has spread through tropical and subtropical regions of the world. All four serotypes of dengue viruses are endemic in the equatorial city state of Singapore, and frequent localised outbreaks occur, sometimes leading to national epidemics. Vector control remains the primary and most effective measure for dengue control and prevention. The objective of this study is to develop a novel framework for producing a spatio-temporal dengue forecast at a neighbourhood level spatial resolution that can be routinely used by Singapore's government agencies for planning of vector control for best efficiency. METHODS: The forecasting algorithm uses a mixture of purely spatial, purely temporal and spatio-temporal data to derive dynamic risk maps for dengue transmission. LASSO-based regression was used for the prediction models and separate sub-models were constructed for each forecast window. Data were divided into training and testing sets for out-of-sample validation. Neighbourhoods were categorised as high or low risk based on the forecast number of cases within the cell. The predictive accuracy of the categorisation was measured. RESULTS: Close concordance between the projections and the eventual incidence of dengue were observed. The average Matthew's correlation coefficient for a classification of the upper risk decile (operational capacity) is similar to the predictive performance at the optimal 30% cut-off. The quality of the spatial predictive algorithm as a classifier shows areas under the curve at all forecast windows being above 0.75 and above 0.80 within the next month. CONCLUSIONS: Spatially resolved forecasts of geographically structured diseases like dengue can be obtained at a neighbourhood level in highly urban environments at a precision that is suitable for guiding control efforts. The same method can be adapted to other urban and even rural areas, with appropriate adjustment to the grid size and shape.

Construction sites as an important driver of dengue transmission: implications for disease control.

BACKGROUND: In 2013 and 2014, Singapore experienced its worst dengue outbreak known-to-date. Mosquito breeding in construction sites stood out as a probable risk factor due to its association with major dengue clusters in both years. We, therefore, investigated the contribution of construction sites to dengue transmission in Singapore, highlighting three case studies of large construction site-associated dengue clusters recorded during 2013-16. METHODS: The study included two components; a statistical analysis of cluster records from 2013 to 2016, and case studies of three biggest construction site-associated clusters. We explored the odds of construction site-associated clusters growing into major clusters and determined whether clusters seeded in construction sites demonstrated a higher tendency to expand into major clusters. DENV strains obtained from dengue patients residing in three major clusters were genotyped to determine whether the same strains expanded into the surroundings of construction sites. RESULTS: Despite less than 5% of total recorded clusters being construction site-associated, the odds of such clusters expanding into major clusters were 17.4 (2013), 9.2 (2014), 3.3 (2015) and 4.3 (2016) times higher than non-construction site clusters. Aedes premise index and average larvae count per habitat were also higher in construction sites than residential premises during the study period. The majority of cases in clusters associated with construction sites were residents living in the surroundings. Virus genotype data from three case study sites revealed a transmission link between the construction sites and the surrounding residential areas. CONCLUSIONS: Significantly high case burden and the probability of cluster expansion due to virus spill-over into surrounding areas suggested that construction sites play an important role as a driver of sustained dengue transmission. Our results emphasise that the management of construction-site associated dengue clusters should not be limited to the implicated construction sites, but be extended to the surrounding premises to prevent further transmission.

Epidemic resurgence of dengue fever in Singapore in 2013-2014: A virological and entomological perspective.

BACKGROUND: Dengue resurged in Singapore during 2013-14, causing an outbreak with unprecedented number of cases in the country. In the present study, we summarise the epidemiological, virological and entomological findings gathered through the dengue surveillance programme and highlight the drivers of the epidemic. We also describe how the surveillance system facilitated the preparedness to moderate epidemic transmission of dengue in the country. METHODS: The case surveillance was based on a mandatory notification system that requires all medical practitioners to report clinically-suspected and laboratory-confirmed cases within 24 hours. The circulating Dengue virus (DENV) populations were monitored through an island wide virus surveillance programme aimed at determining the serotypes and genotypes of circulating virus strains. Entomological surveillance included adult Aedes surveillance as well as premise checks for larval breeding. RESULTS: A switch in the dominant serotype from DENV-2 to DENV-1 in March 2013 signalled a potential spike in cases, and the alert was corroborated by an increase in average Aedes house index. The alert triggered preparedness and early response to moderate the impending outbreak. The two-year outbreak led to 22,170 cases in 2013 and 18,338 in 2014, corresponding to an incidence rate of 410.6 and 335.0 per 100,000 population, respectively. DENV-1 was the dominant serotype in 2013 (61.7 %, n = 5,071) and 2014 (79.2 %, n = 5,226), contributed largely by a newly-introduced DENV-1 genotype III strain. The percentage of houses with Ae. aegypti breeding increased significantly (p < 0.001) from 2012 (annual average of 0.07 %) to 2013 (annual average of 0.14 %), followed by a drop in 2014 (annual average of 0.10 %). Aedes breeding data further showed a wide spread distribution of Ae. aegypti in the country that corresponded with the dengue case distribution pattern in 2013 and 2014. The adult Aedes data from 34 gravitrap sentinel sites revealed that approximately 1/3 of the monitored sites remained at high risk of DENV transmission in 2013. CONCLUSIONS: The culmination of the latest epidemic is likely to be due to a number of demographic, social, virological, entomological, immunological, climatic and ecological factors that contribute to DENV transmission. A multi-pronged approach backed by the epidemiological, virological and entomological understanding paved way to moderate the case burden through an integrated vector management approach.

Spatio-temporal analysis of the main dengue vector populations in Singapore.

BACKGROUND: Despite the licensure of the world's first dengue vaccine and the current development of additional vaccine candidates, successful Aedes control remains critical to the reduction of dengue virus transmission. To date, there is still limited literature that attempts to explain the spatio-temporal population dynamics of Aedes mosquitoes within a single city, which hinders the development of more effective citywide vector control strategies. Narrowing this knowledge gap requires consistent and longitudinal measurement of Aedes abundance across the city as well as examination of relationships between variables on a much finer scale. METHODS: We utilized a high-resolution longitudinal dataset generated from Singapore's islandwide Gravitrap surveillance system over a 2-year period and built a Bayesian hierarchical model to explain the spatio-temporal dynamics of Aedes aegypti and Aedes albopictus in relation to a wide range of environmental and anthropogenic variables. We also created a baseline during our model assessment to serve as a benchmark to be compared with the model's out-of-sample prediction/forecast accuracy as measured by the mean absolute error. RESULTS: For both Aedes species, building age and nearby managed vegetation cover were found to have a significant positive association with the mean mosquito abundance, with the former being the strongest predictor. We also observed substantial evidence of a nonlinear effect of weekly maximum temperature on the Aedes abundance. Our models generally yielded modest but statistically significant reductions in the out-of-sample prediction/forecast error relative to the baseline. CONCLUSIONS: Our findings suggest that public residential estates with older buildings and more nearby managed vegetation should be prioritized for vector control inspections and community advocacy to reduce the abundance of Aedes mosquitoes and the risk of dengue transmission.

Transient transmission of Chikungunya virus in Singapore exemplifies successful mitigation of severe epidemics in a vulnerable population.

OBJECTIVES: Singapore experienced two major outbreaks of chikungunya in 2008-09 and 2013-14. Despite repeated virus introductions, fresh local outbreaks have not emerged after 2014. The present study reviews the success of chikungunya control in Singapore, despite repeated introduction of virus strains, presence of competent vectors and an immunologically naïve population. METHODS: Chikungunya virus (CHIKV) sequences (421 envelope 1 genes and 56 polyproteins) were analysed to distinguish the indigenous virus groups from 2008 to 2020. Vector surveillance data was used to incriminate the vector/s associated with local outbreaks. The population exposure to CHIKV was determined by assessing the seroprevalence status in three cohorts of sera collected in 2009 (n=2,008), 2013 (n=2,000) and 2017 (n=3,615). RESULTS: Four distinct groups of CHIKV of East, Central and South African genotype have mainly circulated since 2008, transmitted primarily by Aedes albopictus. The age weighted CHIKV IgG prevalence rates were low (1-5%) and showed a non-significant increase from 2009 to 2013, but a significant decrease in 2017. In contrast, the prevalence of CHIKV neutralising antibodies in the population increased significantly from 2009 to 2013, with no significant change in 2017, but the levels remained below 2%. CONCLUSIONS: The evidence suggested that surveillance and vector control strategies implemented were robust to avert severe epidemics, despite repeated introduction of virus strains, presence of competent vectors and an immunologically naïve population.

Hyperendemicity associated with increased dengue burden.

Over 105 million dengue infections are estimated to occur annually. Understanding the disease dynamics of dengue is often difficult due to multiple strains circulating within a population. Interactions between dengue serotype dynamics may result in complex cross-immunity dynamics at the population level and create difficulties in terms of formulating intervention strategies for the disease. In this study, a nationally representative 16-year time series with over 43 000 serotyped dengue infections was used to infer the long-run effects of between and within strain interactions and their impacts on past outbreaks. We used a novel identification strategy incorporating sign-identified Bayesian vector autoregressions, using structural impulse responses, historical decompositions and counterfactual analysis to conduct inference on dengue dynamics post-estimation. We found that on the population level: (i) across-serotype interactions on the population level were highly persistent, with a one time increase in any other serotype associated with long run decreases in the serotype of interest (range: 0.5-2.5 years) and (ii) over 38.7% of dengue cases of any serotype were associated with across-serotype interactions. The findings in this paper will substantially impact public health policy interventions with respect to dengue.

The Influence of South East Asia Forest Fires on Ambient Particulate Matter Concentrations in Singapore: An Ecological Study Using Random Forest and Vector Autoregressive Models.

Haze, due to biomass burning, is a recurring problem in Southeast Asia (SEA). Exposure to atmospheric particulate matter (PM) remains an important public health concern. In this paper, we examined the long-term seasonality of PM2.5 and PM10 in Singapore. To study the association between forest fires in SEA and air quality in Singapore, we built two machine learning models, including the random forest (RF) model and the vector autoregressive (VAR) model, using a benchmark air quality dataset containing daily PM2.5 and PM10 from 2009 to 2018. Furthermore, we incorporated weather parameters as independent variables. We observed two annual peaks, one in the middle of the year and one at the end of the year for both PM2.5 and PM10. Singapore was more affected by fires from Kalimantan compared to fires from other SEA countries. VAR models performed better than RF with Mean Absolute Percentage Error (MAPE) values being 0.8% and 6.1% lower for PM2.5 and PM10, respectively. The situation in Singapore can be reasonably anticipated with predictive models that incorporate information on forest fires and weather variations. Public communication of anticipated air quality at the national level benefits those at higher risk of experiencing poorer health due to poorer air quality.

Distribution and seasonal fluctuations of Ae. aegypti and Ae. albopictus larval and pupae in residential areas in an urban landscape.

Dengue, a vector-borne disease spread by Aedes mosquitoes, is a global threat. In the absence of an efficacious dengue vaccine, vector control is the key intervention tool in Singapore. A good understanding of vector habitats is essential to formulate operational strategies. We examined the distribution, long-term trend and seasonality of Aedes data collected during regulatory inspections in residences and public areas from 2008 to 2017. We also studied the seasonality of climate factors to understand their influence on the detection of Aedes-positive containers. The most frequently reported Aedes-positive containers were domestic containers, drains, discarded receptacles, ornamental containers, flower pot plates/trays, plants, gully traps, canvas/plastic sheet, bins, ground puddle, inspection chambers and roof tops/gutters. We found more Ae. aegypti and Ae. albopictus-positive containers per inspection in residences and public areas, respectively. The seasonality of Ae. aegypti-positive containers in residences and public areas coincided with that of mean temperature. However, the seasonality of Ae. albopictus-positive containers lagged by one month compared to that of mean temperature. Our study demonstrates the seasonal fluctuations of Aedes-positive containers in an urban environment. Understanding the distribution and seasonality of Aedes breeding helps to facilitate resource planning and community awareness to moderate dengue transmission.

A novel entomological index, Aedes aegypti Breeding Percentage, reveals the geographical spread of the dengue vector in Singapore and serves as a spatial risk indicator for dengue.

BACKGROUND: Aedes aegypti is an efficient primary vector of dengue, and has a heterogeneous distribution in Singapore. Aedes albopictus, a poor vector of dengue, is native and ubiquitous on the island. Though dengue risk follows the dispersal of Ae. aegypti, the spatial distribution of the vector is often poorly characterized. Here, based on the ubiquitous presence of Ae. albopictus, we developed a novel entomological index, Ae. aegypti Breeding Percentage (BP), to demonstrate the expansion of Ae. aegypti into new territories that redefined the dengue burden map in Singapore. We also determined the thresholds of BP that render the specific area higher risk of dengue transmission. METHODS: We performed analysis of dengue fever incidence and Aedes mosquito breeding in Singapore by utilizing island-wide dengue cases and vector surveillance data from 2003 to 2013. The percentage of Ae. aegypti breeding among the total Aedes breeding habitats (BP), and the reported number of dengue fever cases in each year were calculated for each residential grid. RESULTS: The BP of grids, for every year over the 11-year study period, had a consistent positive correlation with the annual case counts. Our findings suggest that the geographical expansion of Ae. aegypti to previously "non-dengue" areas have contributed substantially to the recent dengue fever incidence in Singapore. Our analysis further indicated that non-endemic areas in Singapore are likely to be at risk of dengue fever outbreaks beyond an Ae. aegypti BP of 20%. CONCLUSIONS: Our analyses indicate areas with increasing Ae. aegypti BP are likely to become more vulnerable to dengue outbreaks. We propose the usage of Ae. aegypti BP as a factor for spatial risk stratification of dengue fever in endemic countries. The Ae. aegypti BP could be recommended as an indicator for decision making in vector control efforts, and also be used to monitor the geographical expansion of Ae. aegypti.

Using Human Movement Data to Identify Potential Areas of Zika Transmission: Case Study of the Largest Zika Cluster in Singapore.

Singapore experienced its first Zika virus (ZIKV) cluster in August 2016. To understand the implication of human movement on disease spread, a retrospective study was conducted using aggregated and anonymized mobile phone data to examine movement from the cluster to identify areas of possible transmission. An origin⁻destination model was developed based on the movement of three groups of individuals: (i) construction workers, (ii) residents and (iii) visitors out of the cluster locality to other parts of the island. The odds ratio of ZIKV cases in a hexagon visited by an individual from the cluster, independent of the group of individuals, is 3.20 (95% CI: 2.65⁻3.87, p-value < 0.05), reflecting a higher count of ZIKV cases when there is a movement into a hexagon from the cluster locality. A comparison of independent ROC curves tested the statistical significance of the difference between the areas under the curves of the three groups of individuals. Visitors (difference in AUC = 0.119) and residents (difference in AUC = 0.124) have a significantly larger difference in area under the curve compared to the construction workers (p-value < 0.05). This study supports the proof of concept of using mobile phone data to approximate population movement, thus identifying areas at risk of disease transmission.

The effectiveness of inspections on reported mosquito larval habitats in households: A case-control study.

BACKGROUND: Dengue is an arboviral disease that imposes substantial health and economic burdens across the globe. Vector control remains a key strategy in settings where Dengvaxia (a dengue vaccine) has not been licenced due to safety concerns and where mass immunization programmes are not cost-effective. Though inspections are used as part of arboviral disease control programmes, evidence of their impact on the entomological activity in households is sparse. METHODOLOGY/PRINCIPAL FINDINGS: We analysed nationally representative household inspection data collected from Singapore over a 3-year period, to determine the effect of inspections on reported mosquito larval habitats in households. A case was a household with a positive report of a mosquito larval habitat in its most recent inspection in 2017. A control was a household that was reported free of mosquito larvae in its most recent inspection in 2017. Using multivariable logistic regression, we analysed 3,205 cases and 557,044 controls. Households averaging three inspections per annum were associated with reduced odds of mosquito larval habitat reports [Adjusted Odds Ratio (AOR): 0.49, 95% Confidence Interval (95% CI): 0.38 to 0.63]. The effect of inspections declined with decreasing inspection frequencies but remained protective at lower levels. Longer intervals (30 to 36 months) between the most recent two successive inspections were associated with increased odds of mosquito larval habitat reports (AOR: 1.28, 95% CI: 1.06 to 1.56) compared to those carried out less than 6 months apart. Mosquito larval habitat reports exhibited a dependence on spatial and household-level characteristics such as the location of the community district, housing type and housing floor level. We observed a four-fold increase in the odds of mosquito larval habitat reports in households with an immediate previous report of larval activity compared to those that did not have one (AOR: 4.52, 95% CI: 3.67 to 5.56). CONCLUSIONS/SIGNIFICANCE: Our study confirms the protective effect of inspections on reported mosquito larval habitat reporting in households. Spatial, temporal and household-level characteristics should be accounted for in prioritizing vector control resources. Alternative strategies may help address recurrent entomological activity in households.

Environmental management for dengue control: a systematic review protocol.

INTRODUCTION: Dengue is among the most important mosquito-borne diseases, with more than half of the world's population at risk of infection in dengue endemic countries. Environmental management, which includes any activities that involve environmental modification, environmental manipulation and changes to human behaviour have been used to mitigate the risk of dengue transmission. In this protocol, we will integrate the data from various sources to assess the overall effect of environmental management on the incidence of dengue and other entomological indices. METHODS AND ANALYSES: We will conduct a systematic review of intervention that assess the effect of environmental management on the incidence of dengue and/or entomological indices. We will include any studies that include intervention through environmental management for dengue control, involving environmental modification, environmental manipulation and changes to human behaviour. A comprehensive search will be performed in electronic databases PUBMED, CENTRAL, SCOPUS, Web of Science and relevant research websites such as PROPSERO, WHO ICTRP and ClinicalTrials.gov to identify studies that meet our inclusion criteria. A systematic approach to searching, screening, reviewing and data extraction will be applied based on Preferred Reporting Items for Systematic reviews and Meta-Analysis. Titles, abstract, keywords for eligibility will be examined independently by researchers. The quality of the included studies will be assessed using quality assessment tool for studies with diverse design and Cochrane risk of bias tool. The characteristics of the selected articles will be described based on the study design, types of intervention and outcomes of the study in various countries. These include the types of environmental management intervention methods and the effectiveness of the intervention in reducing dengue cases or incidence and impact on entomological indices. ETHICS AND DISSEMINATION: We will register this systematic review with the National Medical Research Register, Ministry of Health Malaysia. This protocol also had been registered with the PROSPERO. No ethical approval is necessary, as there will be no collection of primary data. The results will be disseminated though a peer-reviewed publication and conference presentation. TRIAL REGISTRATION NUMBER: CRD42018092189.

Highly Selective Transmission Success of Dengue Virus Type 1 Lineages in a Dynamic Virus Population: An Evolutionary and Fitness Perspective.

Arbovirus transmission is modulated by host, vector, virus, and environmental factors. Even though viral fitness plays a salient role in host and vector adaptation, the transmission success of individual strains in a heterogeneous population may be stochastic. Our large-scale molecular epidemiological analyses of a dengue virus type 1 population revealed that only a subset of strains (16.7%; n = 6) were able to sustain transmission, despite the population being widely dispersed, dynamic, and heterogeneous. The overall dominance was variable even among the "established" lineages, albeit sharing comparable evolutionary characteristics and replication profiles. These findings indicated that virological parameters alone were unlikely to have a profound effect on the survival of viral lineages, suggesting an important role for non-viral factors in the transmission success of lineages. Our observations, therefore, emphasize the strategic importance of a holistic understanding of vector, human host, and viral factors in the control of vector-borne diseases.

Mapping dengue risk in Singapore using Random Forest.

BACKGROUND: Singapore experiences endemic dengue, with 2013 being the largest outbreak year known to date, culminating in 22,170 cases. Given the limited resources available, and that vector control is the key approach for prevention in Singapore, it is important that public health professionals know where resources should be invested in. This study aims to stratify the spatial risk of dengue transmission in Singapore for effective deployment of resources. METHODOLOGY/PRINCIPAL FINDINGS: Random Forest was used to predict the risk rank of dengue transmission in 1km2 grids, with dengue, population, entomological and environmental data. The predicted risk ranks are categorized and mapped to four color-coded risk groups for easy operation application. The risk maps were evaluated with dengue case and cluster data. Risk maps produced by Random Forest have high accuracy. More than 80% of the observed risk ranks fell within the 80% prediction interval. The observed and predicted risk ranks were highly correlated ([Formula: see text]≥0.86, P <0.01). Furthermore, the predicted risk levels were in excellent agreement with case density, a weighted Kappa coefficient of more than 0.80 (P <0.01). Close to 90% of the dengue clusters occur in high risk areas, and the odds of cluster forming in high risk areas were higher than in low risk areas. CONCLUSIONS: This study demonstrates the potential of Random Forest and its strong predictive capability in stratifying the spatial risk of dengue transmission in Singapore. Dengue risk map produced using Random Forest has high accuracy, and is a good surveillance tool to guide vector control operations.

Dengue in Singapore from 2004 to 2016: Cyclical Epidemic Patterns Dominated by Serotypes 1 and 2.

Singapore has experienced periodic dengue epidemics despite maintaining a low Aedes house index. Each epidemic was associated with a switch in the predominant serotype. We investigated the temporal dynamics of dengue fever and dengue virus (DENV) and analyzed the epidemiological and entomological patterns of dengue in Singapore from 2004 to 2016. The case surveillance is based on a mandatory notification system that requires all medical practitioners to report clinically suspected and laboratory-confirmed cases. Circulating (DENV) serotypes are monitored through a virus surveillance program. Entomological surveillance involves inspections for larval breeding and monitoring of adults using gravitraps. Singapore experienced a similar epidemic pattern during 2004-2007 and 2013-2016. The pattern involved a 2-year DENV-1 epidemic occurring after a switch in the predominant serotype from DENV-2 to DENV-1, followed by a "lull" year. Thereafter, the predominant serotype switched back to DENV-2, tailed by a small-scale epidemic. Across the years, the highest incidence group was in the 25-44 years age group. The incidence rate of those aged ≥ 55 years was about half of that of the 15-24 years age group during DENV-1 predominant years. However, it was almost equal to the younger age group in DENV-2 predominant years. Types of Aedes aegypti breeding habitats remained similar. Dengue incidence was significantly higher in areas with high breeding percentage (BP) than areas with low BP (P < 0.05). In conclusion, the oscillation of DENV-1 and DENV-2, throughout the 13-year period, led to a cyclical epidemic pattern and older adults were more affected by DENV-2 than DENV-1.

Three-Month Real-Time Dengue Forecast Models: An Early Warning System for Outbreak Alerts and Policy Decision Support in Singapore.

BACKGROUND: With its tropical rainforest climate, rapid urbanization, and changing demography and ecology, Singapore experiences endemic dengue; the last large outbreak in 2013 culminated in 22,170 cases. In the absence of a vaccine on the market, vector control is the key approach for prevention. OBJECTIVES: We sought to forecast the evolution of dengue epidemics in Singapore to provide early warning of outbreaks and to facilitate the public health response to moderate an impending outbreak. METHODS: We developed a set of statistical models using least absolute shrinkage and selection operator (LASSO) methods to forecast the weekly incidence of dengue notifications over a 3-month time horizon. This forecasting tool used a variety of data streams and was updated weekly, including recent case data, meteorological data, vector surveillance data, and population-based national statistics. The forecasting methodology was compared with alternative approaches that have been proposed to model dengue case data (seasonal autoregressive integrated moving average and step-down linear regression) by fielding them on the 2013 dengue epidemic, the largest on record in Singapore. RESULTS: Operationally useful forecasts were obtained at a 3-month lag using the LASSO-derived models. Based on the mean average percentage error, the LASSO approach provided more accurate forecasts than the other methods we assessed. We demonstrate its utility in Singapore's dengue control program by providing a forecast of the 2013 outbreak for advance preparation of outbreak response. CONCLUSIONS: Statistical models built using machine learning methods such as LASSO have the potential to markedly improve forecasting techniques for recurrent infectious disease outbreaks such as dengue. CITATION: Shi Y, Liu X, Kok SY, Rajarethinam J, Liang S, Yap G, Chong CS, Lee KS, Tan SS, Chin CK, Lo A, Kong W, Ng LC, Cook AR. 2016. Three-month real-time dengue forecast models: an early warning system for outbreak alerts and policy decision support in Singapore. Environ Health Perspect 124:1369-1375; http://dx.doi.org/10.1289/ehp.1509981.