Economic impact of dengue in Singapore from 2010 to 2020 and the cost-effectiveness of Wolbachia interventions

Efficacy of Wolbachia-mediated sterility to reduce the incidence of dengue: a synthetic control study in Singapore

BACKGROUND

Due to the absence of available therapeutics and good vaccines, vector control solutions are needed to mitigate the spread of dengue. Matings between male Aedes aegypti mosquitoes infected with the wAlbB strain of Wolbachia and wildtype females yield non-viable eggs. We evaluated the efficacy of releasing wAlbB-infected A aegypti male mosquitoes to suppress dengue incidence.

METHODS

In this synthetic control study, we conducted large-scale field trials in Singapore involving release of wAlbB-infected A aegypti male mosquitoes for dengue control via vector population suppression, from epidemiological week (EW) 27, 2018, to EW 26, 2022. We selected two large towns (Yishun and Tampines) to adopt an expanding release strategy and two smaller towns (Bukit Batok and Choa Chu Kang) to adopt a targeted-release approach. Releases were conducted two times a week in high-rise public housing estates. All intervention and control locations practised the same baseline dengue control protocol. The main outcome was weekly dengue incidence rate caused by any dengue virus serotype. We used incidence data collected by the Singapore Ministry of Health to assess the efficacy of the interventions. To compare interventions, we used the synthetic control method to generate appropriate counterfactuals for the intervention towns using a weighted combination of 30 control towns between EW 1, 2014 and EW 26, 2022.

FINDINGS

Our study comprised an at-risk population of 607 872 individuals living in intervention sites and 3 894 544 individuals living in control sites. Interventions demonstrated up to 77·28% (121/156, 95% CI 75·81-78·58) intervention efficacy despite incomplete coverage across all towns until EW 26, 2022. Intervention efficacies increased as release coverage increased across all intervention sites. Releases led to 2242 (95% CI 2092-2391) fewer cases per 100 000 people in intervention sites during the study period. Secondary analysis showed that these intervention effects were replicated across all age groups and both sexes for intervention sites.

INTERPRETATION

Our results demonstrated the potential of Wolbachia-mediated incompatible insect technique for strengthening dengue control in tropical cities, where dengue burden is the greatest.

FUNDING

Singapore Ministry of Finance, Ministry of Sustainability, and the National Environment Agency, and the Singapore National Robotics Program.

Wolbachia-carrying Aedes mosquitoes for preventing dengue infection

BACKGROUND

Dengue is a global health problem of high significance, with 3.9 billion people at risk of infection. The geographic expansion of dengue virus (DENV) infection has resulted in increased frequency and severity of the disease, and the number of deaths has increased in recent years. Wolbachia,an intracellular bacterial endosymbiont, has been under investigation for several years as a novel dengue-control strategy. Some dengue vectors (Aedes mosquitoes) can be transinfected with specific strains of Wolbachia, which decreases their fitness (ability to survive and mate) and their ability to reproduce, inhibiting the replication of dengue. Both laboratory and field studies have demonstrated the potential effect of Wolbachia deployments on reducing dengue transmission, and modelling studies have suggested that this may be a self-sustaining strategy for dengue prevention, although long-term effects are yet to be elucidated.

OBJECTIVES

To assess the efficacy of Wolbachia-carrying Aedes speciesdeployments (specifically wMel-, wMelPop-, and wAlbB- strains of Wolbachia) for preventing dengue virus infection.

SEARCH METHODS

We searched CENTRAL, MEDLINE, Embase, four other databases, and two trial registries up to 24 January 2024.

SELECTION CRITERIA

Randomized controlled trials (RCTs), including cluster-randomized controlled trials (cRCTs), conducted in dengue endemic or epidemic-prone settings were eligible. We sought studies that investigated the impact of Wolbachia-carrying Aedes deployments on epidemiological or entomological dengue-related outcomes, utilizing either the population replacement or population suppression strategy.

DATA COLLECTION AND ANALYSIS

Two review authors independently selected eligible studies, extracted data, and assessed the risk of bias using the Cochrane RoB 2 tool. We used odds ratios (OR) with the corresponding 95% confidence intervals (CI) as the effect measure for dichotomous outcomes. For count/rate outcomes, we planned to use the rate ratio with 95% CI as the effect measure. We used adjusted measures of effect for cRCTs. We assessed the certainty of evidence using GRADE.

MAIN RESULTS

One completed cRCT met our inclusion criteria, and we identified two further ongoing cRCTs. The included trial was conducted in an urban setting in Yogyakarta, Indonesia. It utilized a nested test-negative study design, whereby all participants aged three to 45 years who presented at healthcare centres with a fever were enrolled in the study provided they had resided in the study area for the previous 10 nights. The trial showed that wMel-Wolbachia infected Ae aegypti deployments probably reduce the odds of contracting virologically confirmed dengue by 77% (OR 0.23, 95% CI 0.15 to 0.35; 1 trial, 6306 participants; moderate-certainty evidence). The cluster-level prevalence of wMel Wolbachia-carrying mosquitoes remained high over two years in the intervention arm of the trial, reported as 95.8% (interquartile range 91.5 to 97.8) across 27 months in clusters receiving wMel-Wolbachia Ae aegypti deployments, but there were no reliable comparative data for this outcome. Other primary outcomes were the incidence of virologically confirmed dengue, the prevalence of dengue ribonucleic acid in the mosquito population, and mosquito density, but there were no data for these outcomes. Additionally, there were no data on adverse events.

AUTHORS' CONCLUSIONS

The included trial demonstrates the potential significant impact of wMel-Wolbachia-carrying Ae aegypti mosquitoes on preventing dengue infection in an endemic setting, and supports evidence reported in non-randomized and uncontrolled studies. Further trials across a greater diversity of settings are required to confirm whether these findings apply to other locations and country settings, and greater reporting of acceptability and cost are important.

Precision Prediction for Dengue Fever in Singapore: A Machine Learning Approach Incorporating Meteorological Data

OBJECTIVE

This study aimed to improve dengue fever predictions in Singapore using a machine learning model that incorporates meteorological data, addressing the current methodological limitations by examining the intricate relationships between weather changes and dengue transmission.

METHOD

Using weekly dengue case and meteorological data from 2012 to 2022, the data was preprocessed and analyzed using various machine learning algorithms, including General Linear Model (GLM), Support Vector Machine (SVM), Gradient Boosting Machine (GBM), Decision Tree (DT), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost) algorithms. Performance metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R-squared (R2) were employed.

RESULTS

From 2012 to 2022, there was a total of 164,333 cases of dengue fever. Singapore witnessed a fluctuating number of dengue cases, peaking notably in 2020 and revealing a strong seasonality between March and July. An analysis of meteorological data points highlighted connections between certain climate variables and dengue fever outbreaks. The correlation analyses suggested significant associations between dengue cases and specific weather factors such as solar radiation, solar energy, and UV index. For disease predictions, the XGBoost model showed the best performance with an MAE = 89.12, RMSE = 156.07, and R2 = 0.83, identifying time as the primary factor, while 19 key predictors showed non-linear associations with dengue transmission. This underscores the significant role of environmental conditions, including cloud cover and rainfall, in dengue propagation.

CONCLUSION

In the last decade, meteorological factors have significantly influenced dengue transmission in Singapore. This research, using the XGBoost model, highlights the key predictors like time and cloud cover in understanding dengue's complex dynamics. By employing advanced algorithms, our study offers insights into dengue predictive models and the importance of careful model selection. These results can inform public health strategies, aiming to improve dengue control in Singapore and comparable regions.

Economic optimization of Wolbachia-infected Aedes aegypti release to prevent dengue

BACKGROUND

Dengue virus, primarily transmitted by the Aedes aegypti mosquito, is a major public health concern affecting ≈3.83 billion people worldwide. Recent releases of Wolbachia-transinfected Ae. aegypti in several cities worldwide have shown that it can reduce dengue transmission. However, these releases are costly, and, to date, no framework has been proposed for determining economically optimal release strategies that account for both costs associated with disease risk and releases.

RESULTS

We present a flexible stochastic dynamic programming framework for determining optimal release schedules for Wolbachia-transinfected mosquitoes that balances the cost of dengue infection with the costs of rearing and releasing transinfected mosquitoes. Using an ordinary differential equation model of Wolbachia and dengue in a hypothetical city loosely describing areas at risk of new dengue epidemics, we determined that an all-or-nothing release strategy that quickly brings Wolbachia to fixation is often the optimal solution. Based on this, we examined the optimal facility size, finding that it was inelastic with respect to the mosquito population size, with a 100% increase in population size resulting in a 50-67% increase in optimal facility size. Furthermore, we found that these results are robust to mosquito life-history parameters and are mostly determined by the mosquito population size and the fitness costs associated with Wolbachia.

CONCLUSIONS

These results reinforce that Wolbachia-transinfected mosquitoes can reduce the cost of dengue epidemics. Furthermore, they emphasize the importance of determining the size of the target population and fitness costs associated with Wolbachia before releases occur. © 2024 The Authors. Pest Management Science published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Associations between Dengue Incidence, Ecological Factors, and Anthropogenic Factors in Singapore

Singapore experiences endemic dengue. Vector control remains the primary means to reduce transmission due to the lack of available therapeutics. Resource limitations mean that vector-control tools need to be optimized, which can be achieved by studying risk factors related to disease transmission. We developed a statistical modelling framework which can account for a high-resolution and high-dimensional set of covariates to delineate spatio-temporal characteristics that are associated with dengue transmission from 2014 to 2020 in Singapore. We applied the proposed framework to two distinct datasets, stratified based on the primary type of housing within each spatial unit. Generalized additive models reveal non-linear exposure responses between a large range of ecological and anthropogenic factors as well as dengue incidence rates. At values below their mean, lesser mean total daily rainfall (Incidence rate ratio (IRR): 3.75, 95% CI: 1.00-14.05, Mean: 4.40 mm), decreased mean windspeed (IRR: 3.65, 95% CI: 1.87-7.10, Mean: 4.53 km/h), and lower building heights (IRR: 2.62, 95% CI: 1.44-4.77, Mean: 6.5 m) displayed positive associations, while higher than average annual NO2 concentrations (IRR: 0.35, 95% CI: 0.18-0.66, Mean: 13.8 ppb) were estimated to be negatively associated with dengue incidence rates. Our study provides an understanding of associations between ecological and anthropogenic characteristics with dengue transmission. These findings help us understand high-risk areas of dengue transmission, and allows for land-use planning and formulation of vector control policies.

A bacterium against the tiger: further evidence of the potential of noninundative releases of males with manipulated Wolbachia infection in reducing fertility of Aedes albopictus field populations in Italy

BACKGROUND

Incompatible insect technique (IIT) is a population suppression approach based on the release of males with manipulated Wolbachia infection inducing egg inviability in wild females. We here present results of multiple field releases of incompatible ARwP males carried out in 2019 in a 2.7-ha green area within urban Rome (Italy) to assess the effect on Aedes albopictus egg viability. Data are compared with results obtained in 2018, when the approach was tested for the first time in Europe.

RESULTS

An average of 4674 ARwP males were released weekly for 7 weeks, resulting in a mean ARwP:wild male ratio of 1.1:1 (versus 0.7:1 in 2018). Egg-viability dynamics in ovitraps significantly varied between treated and control sites, with an estimated overall reduction of 35% (versus 15% in 2018). The estimated proportion of females classified as mated with ARwP males was 41.8% and the viability rate of eggs laid by these females (9.5%) was on average significantly lower than that of females only mated with wild males (87.8%); however, high variability in fertility was observed. Values of ARwP male competitiveness were 0.36 and 0.73 based on the overall viability rate of eggs in ovitraps and on female fertility, respectively; thus, well above the conventional 0.2 threshold for an effective suppressive impact in the field.

CONCLUSIONS

Results further support the potential of IIT as a tool to contribute to Ae. albopictus control in the urban context, stressing the need for larger field trials to evaluate the cost-efficacy of the approach in temperate regions. © 2023 The Authors. Pest Management Science published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Singapore's 5 decades of dengue prevention and control-Implications for global dengue control

This paper summarises the lessons learnt in dengue epidemiology, risk factors, and prevention in Singapore over the last half a century, during which Singapore evolved from a city of 1.9 million people to a highly urban globalised city-state with a population of 5.6 million. Set in a tropical climate, urbanisation among green foliage has created ideal conditions for the proliferation of Aedes aegypti and Aedes albopictus, the mosquito vectors that transmit dengue. A vector control programme, largely for malaria, was initiated as early as 1921, but it was only in 1966 that the Vector Control Unit (VCU) was established to additionally tackle dengue haemorrhagic fever (DHF) that was first documented in the 1960s. Centred on source reduction and public education, and based on research into the bionomics and ecology of the vectors, the programme successfully reduced the Aedes House Index (HI) from 48% in 1966 to <5% in the 1970s. Further enhancement of the programme, including through legislation, suppressed the Aedes HI to around 1% from the 1990s. The current programme is characterised by 4 key features: (i) proactive inter-epidemic surveillance and control that is stepped up during outbreaks; (ii) risk-based prevention and intervention strategies based on advanced data analytics; (iii) coordinated inter-sectoral cooperation between the public, private, and people sectors; and (iv) evidence-based adoption of new tools and strategies. Dengue seroprevalence and force of infection (FOI) among residents have substantially and continuously declined over the 5 decades. This is consistent with the observation that dengue incidence has been delayed to adulthood, with severity highest among the elderly. Paradoxically, the number of reported dengue cases and outbreaks has increased since the 1990s with record-breaking epidemics. We propose that Singapore's increased vulnerability to outbreaks is due to low levels of immunity in the population, constant introduction of new viral variants, expanding urban centres, and increasing human density. The growing magnitude of reported outbreaks could also be attributed to improved diagnostics and surveillance, which at least partially explains the discord between rising trend in cases and the continuous reduction in dengue seroprevalence. Changing global and local landscapes, including climate change, increasing urbanisation and global physical connectivity are expected to make dengue control even more challenging. The adoption of new vector surveillance and control tools, such as the Gravitrap and Wolbachia technology, is important to impede the growing threat of dengue and other Aedes-borne diseases.

Wolbachia-based strategies for control of agricultural pests

Wolbachia-based incompatible insect technique (IIT) and pathogen blocking technique (PBT) have been shown to be effective at protecting humans from mosquito-borne diseases in the past decades. Population suppression based on IIT and population replacement based on PBT have become major field application strategies that have continuously been improved by the translational research on Wolbachia-transinfected mosquitoes. Similarly, Wolbachia-based approaches have been proposed for the protection of plants from agricultural pests and their associated diseases. However, a bottleneck in Wolbachia-based strategies for the control of agricultural pests is the need for methods to establish Wolbachia-transinfected insect lines. As a first step in this direction, we compare field control strategies for mosquitos with the potential strategies for agricultural pests based on Wolbachia. Our results show that there is a critical need for establishing productive insect lines and accumulating field test data.

Features of the urban environment associated with Aedes aegypti abundance in high-rise public apartments in Singapore: An environmental case-control study

Aedes aegypti abundance in residential estates is hypothesized to contribute to localised outbreaks of dengue in Singapore. Knowing the factors in the urban environment underlying high Ae. aegypti abundance could guide intervention efforts to reduce Ae. aegypti breeding and the incidence of dengue. In this study, objective data on Ae. aegypti abundance in public apartment blocks estimated by Singapore's nationally representative Gravitrap surveillance system was obtained from the National Environmental Agency. Low and high abundance status public apartment blocks were classified based on the Gravitrap Aegypti Index, corresponding to the lowest and highest quartiles respectively. An environmental case-control study was conducted, wherein a blinded assessment of urban features hypothesised to form breeding habitats was conducted in 50 randomly sampled public apartment blocks with low and high abundance statuses each. Logistic regression was performed to identify features that correlated with abundance status. A multivariable logistic model was created to determine key urban features found in corridors and void decks which were predictive of the Ae. aegypti abundance status of the public apartment block. At a statistical level of significance of 0.20, the presence of gully traps [Odds Ratio (OR): 1.34, 95% Confidence Interval (CI): 1.10, 1.66], age of the public apartment block [OR: 2.23, 95% CI: 1.48, 3.60], housing price [OR: 0.33, 95% CI: 0.16, 0.61] and corridor cleanliness [OR: 0.67, 95% CI: 0.40, 1.07] were identified as important predictors of abundance status. To reduce Ae. aegypti abundance around public apartment blocks and potential onward dengue transmission, gully traps could be remodelled or replaced by other drainage types. Routine inspections of Ae. aegypti breeding should be targeted at older and low-income neighbourhoods. Campaigns for cleaner corridors should be promoted.

A pan-serotype antiviral to prevent and treat dengue: A journey from discovery to clinical development driven by public-private partnerships

While progress has been made in fighting diseases disproportionally affecting underserved populations, unmet medical needs persist for many neglected tropical diseases. The World Health Organization has encouraged strong public-private partnerships to address this issue and several public and private organizations have set an example in the past showing a strong commitment to combat these diseases. Pharmaceutical companies are contributing in different ways to address the imbalance in research efforts. With this review, we exemplify the role of a public-private partnership in research and development by the journey of our dengue antiviral molecule that is now in early clinical development. We detail the different steps of drug development and outline the contribution of each partner to this process. Years of intensive collaboration resulted in the identification of two antiviral compounds, JNJ-A07 and JNJ-1802, the latter of which has advanced to clinical development.

Perspectives of vector management in the control and elimination of vector-borne zoonoses

The complex transmission profiles of vector-borne zoonoses (VZB) and vector-borne infections with animal reservoirs (VBIAR) complicate efforts to break the transmission circuit of these infections. To control and eliminate VZB and VBIAR, insecticide application may not be conducted easily in all circumstances, particularly for infections with sylvatic transmission cycle. As a result, alternative approaches have been considered in the vector management against these infections. In this review, we highlighted differences among the environmental, chemical, and biological control approaches in vector management, from the perspectives of VZB and VBIAR. Concerns and knowledge gaps pertaining to the available control approaches were discussed to better understand the prospects of integrating these vector control approaches to synergistically break the transmission of VZB and VBIAR in humans, in line with the integrated vector management (IVM) developed by the World Health Organization (WHO) since 2004.

Assessing the efficacy of male Wolbachia-infected mosquito deployments to reduce dengue incidence in Singapore: study protocol for a cluster-randomized controlled trial

BACKGROUND

Dengue is a severe environmental public health challenge in tropical and subtropical regions. In Singapore, decreasing seroprevalence and herd immunity due to successful vector control has paradoxically led to increased transmission potential of the dengue virus. We have previously demonstrated that incompatible insect technique coupled with sterile insect technique (IIT-SIT), which involves the release of X-ray-irradiated male Wolbachia-infected mosquitoes, reduced the Aedes aegypti population by 98% and dengue incidence by 88%. This novel vector control tool is expected to be able to complement current vector control to mitigate the increasing threat of dengue on a larger scale. We propose a multi-site protocol to study the efficacy of IIT-SIT at reducing dengue incidence.

METHODS/DESIGN

The study is designed as a parallel, two-arm, non-blinded cluster-randomized (CR) controlled trial to be conducted in high-rise public housing estates in Singapore, an equatorial city-state. The aim is to determine whether large-scale deployment of male Wolbachia-infected Ae. aegypti mosquitoes can significantly reduce dengue incidence in intervention clusters. We will use the CR design, with the study area comprising 15 clusters with a total area of 10.9 km², covering approximately 722,204 residents in 1713 apartment blocks. Eight clusters will be randomly selected to receive the intervention, while the other seven will serve as non-intervention clusters. Intervention efficacy will be estimated through two primary endpoints: (1) odds ratio of Wolbachia exposure distribution (i.e., probability of living in an intervention cluster) among laboratory-confirmed reported dengue cases compared to test-negative controls and (2) laboratory-confirmed reported dengue counts normalized by population size in intervention versus non-intervention clusters.

DISCUSSION

This study will provide evidence from a multi-site, randomized controlled trial for the efficacy of IIT-SIT in reducing dengue incidence. The trial will provide valuable information to estimate intervention efficacy for this novel vector control approach and guide plans for integration into national vector control programs in dengue-endemic settings.

TRIAL REGISTRATION

ClinicalTrials.gov, identifier: NCT05505682 . Registered on 16 August 2022. Retrospectively registered.

A mass rearing cost calculator for the control of Culex quinquefasciatus in Hawai'i using the incompatible insect technique

BACKGROUND

Hawai'i's native forest avifauna is experiencing drastic declines due to climate change-induced increases in temperature encroaching on their upper-elevation montane rainforest refugia. Higher temperatures support greater avian malaria infection rates due to greater densities of its primary vector, the southern house mosquito Culex quinquefasciatus, and enhance development of the avian malaria parasite Plasmodium relictum. Here we propose the use of the incompatible insect technique (IIT) or the combined IIT/sterile insect technique (SIT) for the landscape-scale (i.e., area-wide) control of Cx. quinquefasciatus, and have developed a calculator to estimate the costs of IIT and IIT/SIT applications at various sites in Hawai'i.

METHODS

The overall cost of the infrastructure, personnel, and space necessary to produce incompatible adult males for release is calculated in a unit of ~ 1 million culicid larvae/week. We assessed the rearing costs and need for effective control at various elevations in Hawai'i using a 10:1 overflooding ratio at each elevation. The calculator uses a rate describing the number of culicids needed to control wild-type mosquitoes at each site/elevation, in relation to the number of larval rearing units. This rate is a constant from which other costs are quantified. With minor modifications, the calculator described here can be applied to other areas, mosquito species, and similar techniques. To test the robustness of our calculator, the Kaua'i-specific culicid IIT/SIT infrastructure costs were also compared to costs from Singapore, Mexico, and China using the yearly cost of control per hectare, and purchasing power parity between sites for the cost of 1000 IIT/SIT males.

RESULTS

As a proof of concept, we have used the calculator to estimate rearing infrastructure costs for an application of IIT in the Alaka'i Wilderness Reserve on the island of Kaua'i. Our analysis estimated an initial investment of at least ~ $1.16M with subsequent yearly costs of approximately $376K. Projections of rearing costs for control at lower elevations are ~ 100 times greater than in upper elevation forest bird refugia. These results are relatively comparable to those real-world cost estimates developed for IIT/SIT culicid male production in other countries when inflation and purchasing power parity are considered. We also present supplemental examples of infrastructure costs needed to control Cx. quinquefasciatus in the home range of 'i'iwi Drepanis coccinea, and the yellow fever vector Aedes aegypti.

CONCLUSIONS

Our cost calculator can be used to effectively estimate the mass rearing cost of an IIT/SIT program. Therefore, the linear relationship of rearing infrastructure to costs used in this calculator is useful for developing a conservative cost estimate for IIT/SIT culicid mass rearing infrastructure. These mass rearing cost estimates vary based on the density of the targeted organism at the application site.

Strategies to Mitigate Establishment Using the Wolbachia Incompatible Insect Technique

The Incompatible Insect Technique (IIT) strategy involves the release of male mosquitoes infected with the bacterium Wolbachia. Regular releases of male Wolbachia-infected mosquitoes can lead to the suppression of mosquito populations, thereby reducing the risk of transmission of vector-borne diseases such as dengue. However, due to imperfect sex-sorting under IIT, fertile Wolbachia-infected female mosquitoes may potentially be unintentionally released into the environment, which may result in replacement and failure to suppress the mosquito populations. As such, mitigating Wolbachia establishment requires a combination of IIT with other strategies. We introduced a simple compartmental model to simulate ex-ante mosquito population dynamics subjected to a Wolbachia-IIT programme. In silico, we explored the risk of replacement, and strategies that could mitigate the establishment of the released Wolbachia strain in the mosquito population. Our results suggest that mitigation may be achieved through the application of a sterile insect technique. Our simulations indicate that these interventions do not override the intended wild type suppression of the IIT approach. These findings will inform policy makers of possible ways to mitigate the potential establishment of Wolbachia using the IIT population control strategy.

The Diagnosis of Dengue in Patients Presenting With Acute Febrile Illness Using Supervised Machine Learning and Impact of Seasonality

Background

Symptomatic dengue infection can result in a life-threatening shock syndrome and timely diagnosis is essential. Point-of-care tests for non-structural protein 1 and IgM are used widely but performance can be limited. We developed a supervised machine learning model to predict whether patients with acute febrile illnesses had a diagnosis of dengue or other febrile illnesses (OFI). The impact of seasonality on model performance over time was examined.

Methods

We analysed data from a prospective observational clinical study in Vietnam. Enrolled patients presented with an acute febrile illness of <72 h duration. A gradient boosting model (XGBoost) was used to predict final diagnosis using age, sex, haematocrit, platelet, white cell, and lymphocyte count collected on enrolment. Data was randomly split 80/20% into a training and hold-out set, respectively, with the latter not used in model development. Cross-validation and hold out set testing was used, with performance over time evaluated through a rolling window approach.

Results

We included 8,100 patients recruited between 16th October 2010 and 10th December 2014. In total 2,240 (27.7%) patients were diagnosed with dengue infection. The optimised model from training data had an overall median area under the receiver operator curve (AUROC) of 0.86 (interquartile range 0.84-0.86), specificity of 0.92, sensitivity of 0.56, positive predictive value of 0.73, negative predictive value (NPV) of 0.84, and Brier score of 0.13 in predicting the final diagnosis, with similar performances in hold-out set testing (AUROC of 0.86). Model performances varied significantly over time as a function of seasonality and other factors. Incorporation of a dynamic threshold which continuously learns from recent cases resulted in a more consistent performance throughout the year (NPV >90%).

Conclusion

Supervised machine learning models are able to discriminate between dengue and OFI diagnoses in patients presenting with an early undifferentiated febrile illness. These models could be of clinical utility in supporting healthcare decision-making and provide passive surveillance across dengue endemic regions. Effects of seasonality and changing disease prevalence must however be taken into account-this is of significant importance given unpredictable effects of human-induced climate change and the impact on health.