Sensitivity of large dengue epidemics in Ecuador to long-lead predictions of El Niño

Petrova, Desislava; Lowe, Rachel; Stewart-Ibarra, Anna M.; Ballester, Joan; Koopman, Siem Jan; Rodó, Xavier

doi:10.1016/j.cliser.2019.02.003

Cited by 13 publications

(15 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Delayed effects of meteorological factors on dengue were accounted for by incorporating a 3-month moving average of temperature, precipitation, specific humidity, and DTR centred at lag of 1 month (i.e., each value is given the average of lag 0 to a lag of 2 months). Sea surface temperature anomalies in the Niño region 3.4 were lagged 0 to 3 months based on previous research [ 20 , 33 , 70 ] and exploratory analyses. No delayed effects were considered for wind speed based on previous studies, indicating a highest effect of wind at a lag of 0 months [ 71 ].…”

Section: Methodsmentioning

confidence: 99%

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

et al. 2021

Self Cite

View full text Add to dashboard Cite

Background With 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 findings We 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. Conclusions This 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.

show abstract

Section: Methodsmentioning

confidence: 99%

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

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…Delayed effects of meteorological factors on dengue were accounted for by incorporating the moving average of temperature, precipitation, specific humidity, and diurnal temperature range (DTR) all of them lagged zero to two months, and sea surface temperature anomalies in the Nino region 3.4 lagged three months based on previous research (e.g. ( Petrova et al, 2019; Colón-González et al, 2018a; Lowe et al, 2018 ) and exploratory analyses. No delayed effects were considered for wind speed.…”

Section: Methodsmentioning

confidence: 99%

Probabilistic seasonal dengue forecasting in Vietnam using superensembles

Colón‐González

Bastos

Hofmann

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Timely information is key for decision-making. The ability to predict dengue transmission ahead of time would significantly benefit planners and decision-makers. Dengue is climate-sensitive. Monitoring climate variability could provide advance warning about dengue risk. Multiple dengue early warning systems have been proposed. Often, these systems are based on deterministic models that have limitations for quantifying the probability that a public health event may occur. We introduce an operational seasonal dengue forecasting system where Earth observations and seasonal climate forecasts are used to drive a superensemble of probabilistic dengue models to predict dengue risk up to six months ahead. We demonstrate that the system has skill and relative economic value at multiple forecast horizons, seasons, and locations. The superensemble generated, on average, more accurate forecasts than those obtained from the models used to create it. We argue our system provides a useful tool for the development and deployment of targeted vector control interventions, and a more efficient allocation of resources in Vietnam.

show abstract

“…These studies have also used measures of local climate conditions, such as temperature, humidity and precipitation, to understand variations in dengue incidence (Huang et al ., 2015; Duarte et al ., 2019; Sippy et al ., 2019; Zhang et al ., 2019). In a series of previous studies (Stewart‐Ibarra and Lowe, 2013; Lowe et al ., 2017; Petrova et al ., 2019), we demonstrated that climate information and in particular local seasonal climate and ENSO forecasts can be used to improve the prediction of dengue outbreaks in southern coastal Ecuador in El Oro Province, and more importantly, to extend the lead time of such predictions to several seasons in advance. Both temperature and rainfall are known to affect the physiology of the dengue vectors Aedes aegypti and Aedes albopictus mosquitoes in terms of their larval development and replication rates (Mordecai et al ., 2017).…”

Section: Introductionmentioning

confidence: 93%

“…Generally, the climatological precipitation and temperature rates are enhanced during El Niño years (Petrova et al . (2019); Figure 1).…”

Section: Introductionmentioning

confidence: 96%

“…The abundance or scarcity of rainfall can increase larval mosquito habitats depending on local access to piped water and local storage practices (Stewart Ibarra et al ., 2013; Lowe et al ., 2018). ENSO is known to affect climate patterns through atmospheric teleconnections (Ropelewski and Halpert, 1987; Kiladis and Diaz, 1989; Rodó et al ., 2006; Sarachik and Cane, 2010), and southern coastal Ecuador is typically associated with heavy rainfall during and after El Niño events (Larkin and Harrison, 2002; Rossel and Cadier, 2009; Petrova et al ., 2019). Moreover, due to the proximity of El Oro Province to the equatorial Pacific area where ENSO occurs, its impact on local temperature is also well‐understood with increased temperature during and after the warm events (Aceituno, 1988; Bendix and Bendix, 2006; Santos, 2006; Rossel and Cadier, 2009; Moran‐Tejeda et al ., 2016).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The 2018–2019 weak El Niño: Predicting the risk of a dengue outbreak in Machala, Ecuador

Petrova

Rodó

Sippy

et al. 2020

Intl Journal of Climatology

Self Cite

View full text Add to dashboard Cite

Between October 2018 ‐ May 2019, sea surface temperature conditions in the central‐eastern tropical Pacific indicated a mild El Niño event. In May 2019, the global El Niño Southern Oscillation (ENSO) forecast consensus was that these generally weak warm patterns will persist at least until the end of the northern hemisphere summer. El Niño and its impact on local climatic conditions in southern coastal Ecuador influence the inter‐annual transmission of dengue fever in the region. In this study, we use an ENSO model to issue forecasts of El Niño for the year 2019, which are then used to predict local climate variables, precipitation and minimum temperature, in the city of Machala, Ecuador. All these forecasts are incorporated in a dengue transmission model, specifically developed and tested for this area, to produce out‐of‐sample predictions of dengue risk. Predictions are issued at the beginning of January 2019 for the whole year, thus providing the longest forecast lead time of 12 months. Preliminary results indicate that the mild and ongoing El Niño event did not provide the optimum climate conditions for dengue transmission, with the model predicting a very low probability of a dengue outbreak during the typical peak season in Machala in 2019. This is contrary to 2016, when a large El Niño event resulted in excess rainfall and warmer temperatures in the region, and a dengue outbreak occurred 3 months earlier than expected. This event was successfully predicted using a similar prediction framework to the one applied here. With the present study, we continue our efforts to build and test a climate service tool to issue early warnings of dengue outbreaks in the region.

show abstract

Sensitivity of large dengue epidemics in Ecuador to long-lead predictions of El Niño

Cited by 13 publications

References 47 publications

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

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

Probabilistic seasonal dengue forecasting in Vietnam using superensembles

The 2018–2019 weak El Niño: Predicting the risk of a dengue outbreak in Machala, Ecuador

Contact Info

Product

Resources

About