Alert threshold algorithms and malaria epidemic detection

Optimal environmental testing frequency for outbreak surveillance

Public health surveillance for pathogens presents an optimization problem: we require enough sampling to identify intervention-triggering shifts in pathogen epidemiology, such as new introductions or sudden increases in prevalence, but not so much that costs due to surveillance itself outweigh those from pathogen-associated illness. To determine this optimal sampling frequency, we developed a general mathematical model for the introduction of a new pathogen that, once introduced, increases in prevalence exponentially. Given the relative cost of infection vs. sampling, we derived equations for the expected combined cost per unit time of disease burden and surveillance for a specified sampling frequency, and thus the sampling frequency for which the expected total cost per unit time is lowest.

Utilizing a novel high-resolution malaria dataset for climate-informed predictions with a deep learning transformer model

Climatic factors influence malaria transmission via the effect on the Anopheles vector and Plasmodium parasite. Modelling and understanding the complex effects that climate has on malaria incidence can enable important early warning capabilities. Deep learning applications across fields are proving valuable, however the field of epidemiological forecasting is still in its infancy with a lack of applied deep learning studies for malaria in southern Africa which leverage quality datasets. Using a novel high resolution malaria incidence dataset containing 23 years of daily data from 1998 to 2021, a statistical model and XGBOOST machine learning model were compared to a deep learning Transformer model by assessing the accuracy of their numerical predictions. A novel loss function, used to account for the variable nature of the data yielded performance around + 20% compared to the standard MSE loss. When numerical predictions were converted to alert thresholds to mimic use in a real-world setting, the Transformer's performance of 80% according to AUROC was 20-40% higher than the statistical and XGBOOST models and it had the highest overall accuracy of 98%. The Transformer performed consistently with increased accuracy as more climate variables were used, indicating further potential for this prediction framework to predict malaria incidence at a daily level using climate data for southern Africa.

Geo-epidemiology of malaria incidence in the Vhembe District to guide targeted elimination strategies, South-Africa, 2015-2018: a local resurgence

In South Africa, the population at risk of malaria is 10% (around six million inhabitants) and concern only three provinces of which Limpopo Province is the most affected, particularly in Vhembe District. As the elimination approaches, a finer scale analysis is needed to accelerate the results. Therefore, in the process of refining local malaria control and elimination strategies, the aim of this study was to identify and describe malaria incidence patterns at the locality scale in the Vhembe District, Limpopo Province, South Africa. The study area comprised 474 localities in Vhembe District for which smoothed malaria incidence curve were fitted with functional data method based on their weekly observed malaria incidence from July 2015 to June 2018. Then, hierarchical clustering algorithm was carried out considering different distances to classify the 474 smoothed malaria incidence curves. Thereafter, validity indices were used to determine the number of malaria incidence patterns. The cumulative malaria incidence of the study area was 4.1 cases/1000 person-years. Four distinct patterns of malaria incidence were identified: high, intermediate, low and very low with varying characteristics. Malaria incidence increased across transmission seasons and patterns. The localities in the two highest incidence patterns were mainly located around farms, and along the rivers. Some unusual malaria phenomena in Vhembe District were also highlighted as resurgence. Four distinct malaria incidence patterns were found in Vhembe District with varying characteristics. Findings show also unusual malaria phenomena in Vhembe District that hinder malaria elimination in South Africa. Assessing the factors associated with these unusual malaria phenome would be helpful on building innovative strategies that lead South Africa on malaria elimination.

Comparing malaria early detection methods in a declining transmission setting in northwestern Ethiopia

Background

Despite remarkable progress in the reduction of malaria incidence, this disease remains a public health threat to a significant portion of the world’s population. Surveillance, combined with early detection algorithms, can be an effective intervention strategy to inform timely public health responses to potential outbreaks. Our main objective was to compare the potential for detecting malaria outbreaks by selected event detection methods.

Methods

We used historical surveillance data with weekly counts of confirmed Plasmodium falciparum (including mixed) cases from the Amhara region of Ethiopia, where there was a resurgence of malaria in 2019 following several years of declining cases. We evaluated three methods for early detection of the 2019 malaria events: 1) the Centers for Disease Prevention and Control (CDC) Early Aberration Reporting System (EARS), 2) methods based on weekly statistical thresholds, including the WHO and Cullen methods, and 3) the Farrington methods.

Results

All of the methods evaluated performed better than a naïve random alarm generator. We also found distinct trade-offs between the percent of events detected and the percent of true positive alarms. CDC EARS and weekly statistical threshold methods had high event sensitivities (80–100% CDC; 57–100% weekly statistical) and low to moderate alarm specificities (25–40% CDC; 16–61% weekly statistical). Farrington variants had a wide range of scores (20–100% sensitivities; 16–100% specificities) and could achieve various balances between sensitivity and specificity.

Conclusions

Of the methods tested, we found that the Farrington improved method was most effective at maximizing both the percent of events detected and true positive alarms for our dataset (> 70% sensitivity and > 70% specificity). This method uses statistical models to establish thresholds while controlling for seasonality and multi-year trends, and we suggest that it and other model-based approaches should be considered more broadly for malaria early detection.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-021-10850-5.

Application of Functional Data Analysis to Identify Patterns of Malaria Incidence, to Guide Targeted Control Strategies

We introduce an approach based on functional data analysis to identify patterns of malaria incidence to guide effective targeting of malaria control in a seasonal transmission area. Using functional data method, a smooth function (functional data or curve) was fitted from the time series of observed malaria incidence for each of 575 villages in west-central Senegal from 2008 to 2012. These 575 smooth functions were classified using hierarchical clustering (Ward’s method), and several different dissimilarity measures. Validity indices were used to determine the number of distinct temporal patterns of malaria incidence. Epidemiological indicators characterizing the resulting malaria incidence patterns were determined from the velocity and acceleration of their incidences over time. We identified three distinct patterns of malaria incidence: high-, intermediate-, and low-incidence patterns in respectively 2% (12/575), 17% (97/575), and 81% (466/575) of villages. Epidemiological indicators characterizing the fluctuations in malaria incidence showed that seasonal outbreaks started later, and ended earlier, in the low-incidence pattern. Functional data analysis can be used to identify patterns of malaria incidence, by considering their temporal dynamics. Epidemiological indicators derived from their velocities and accelerations, may guide to target control measures according to patterns.

Climatic changes and their role in emergence and re-emergence of diseases

Global warming and the associated climate changes are predictable. They are enhanced by burning of fossil fuels and the emission of huge amounts of CO₂ gas which resulted in greenhouse effect. It is expected that the average global temperature will increase with 2-5 °C in the next decades. As a result, the earth will exhibit marked climatic changes characterized by extremer weather events in the coming decades, such as the increase in temperature, rainfall, summertime, droughts, more frequent and stronger tornadoes and hurricanes. Epidemiological disease cycle includes host, pathogen and in certain cases intermediate host/vector. A complex mixture of various environmental conditions (e.g. temperature and humidity) determines the suitable habitat/ecological niche for every vector host. The availability of suitable vectors is a precondition for the emergence of vector-borne pathogens. Climate changes and global warming will have catastrophic effects on human, animal and environmental ecosystems. Pathogens, especially neglected tropical disease agents, are expected to emerge and re-emerge in several countries including Europe and North America. The lives of millions of people especially in developing countries will be at risk in direct and indirect ways. In the present review, the role of climate changes in the spread of infectious agents and their vectors is discussed. Examples of the major emerging viral, bacterial and parasitic diseases are also summarized.

Identifying exceptional malaria occurrences in the absence of historical data in South Sudan: a method validation

Background

Detecting unusual malaria events that may require an operational intervention is challenging, especially in endemic contexts with continuous transmission such as South Sudan. Médecins Sans Frontières (MSF) utilises the classic average plus standard deviation (AV+SD) method for malaria surveillance. This and other available approaches, however, rely on antecedent data, which are often missing.

Objective

To investigate whether a method using linear regression (LR) over only 8 weeks of retrospective data could be an alternative to AV+SD.

Design

In the absence of complete historical malaria data from South Sudan, data from weekly influenza reports from 19 Norwegian counties (2006-2015) were used as a testing data set to compare the performance of the LR and the AV+SD methods. The moving epidemic method was used as the gold standard. Subsequently, the LR method was applied in a case study on malaria occurrence in MSF facilities in South Sudan (2010-2016) to identify malaria events that required a MSF response.

Results

For the Norwegian influenza data, LR and AV+SD methods did not perform differently (P > 0.05). For the South Sudanese malaria data, the LR method identified historical periods when an operational response was mounted.

Conclusion

The LR method seems a plausible alternative to the AV+SD method in situations where retrospective data are missing.

Malaria's association with climatic variables and an epidemic early warning system using historical data from Gezira State, Sudan

Malaria is a major public health problem in Sudan. Climatic variability is the main risk factor for seasonal and secular patterns of P. falciparum malaria transmission in Gezira state. The purposes of this study is to (1) develop thresholds for action in a malaria epidemic early warning system using three traditional statistical methods including the mean number of malaria cases + 2 standard deviations (SD), percentiles over the median (medium + upper third quartile), and the cumulative sum over prior 10 years (C-SUM) and (2) explore to what extent the climate variability affects malaria transmission. Pearson's correlation coefficient for malaria incidence and rainfall, maximum temperature, relative humidity, and the Blue Nile River level was statistically significant (p < 0.05). However, there was an insignificant correlation between the number of malaria cases and the minimum temperature. Furthermore, the number of cases in 2015 was significantly higher than expected. An evaluation and comparison of the statistical methods for the early detection of malaria showed that there was a considerable variation in the number of cases exceeding an epidemic alert threshold.

Devising a method towards development of early warning tool for detection of malaria outbreak

Background & objectives

Uncertainty often arises in differentiating seasonal variation from outbreaks of malaria. The present study was aimed to generalize the theoretical structure of sine curve for detecting an outbreak so that a tool for early warning of malaria may be developed.

Methods

A 'case/mean-ratio scale' system was devised for labelling the outbreak in respect of two diverse districts of Assam and Rajasthan. A curve-based method of analysis was developed for determining outbreak and using the properties of sine curve. It could be used as an early warning tool for Plasmodium falciparum malaria outbreaks.

Result

In the present method of analysis, the critical C_max(peak value of sine curve) value of seasonally adjusted curve for P. falciparum malaria outbreak was 2.3 for Karbi Anglong and 2.2 for Jaisalmer districts. On case/mean-ratio scale, the C_max value of malaria curve between C_maxand 3.5, the outbreak could be labelled as minor while >3.5 may be labelled as major. In epidemic years, with mean of case/mean ratio of ≥1.00 and root mean square (RMS) ≥1.504 of case/mean ratio, outbreaks can be predicted 1-2 months in advance.

Interpretation & conclusions

The present study showed that in P. falciparum cases in Karbi Anglong (Assam) and Jaisalmer (Rajasthan) districts, the rise in C_maxvalue of curve was always followed by rise in average/RMS or both and hence could be used as an early warning tool. The present method provides better detection of outbreaks than the conventional method of mean plus two standard deviation (mean+2 SD). The identified tools are simple and may be adopted for preparedness of malaria outbreaks.

Application of Artificial Neural Networks for Dengue Fever Outbreak Predictions in the Northwest Coast of Yucatan, Mexico and San Juan, Puerto Rico

Modelling dengue fever in endemic areas is important to mitigate and improve vector-borne disease control to reduce outbreaks. This study applied artificial neural networks (ANNs) to predict dengue fever outbreak occurrences in San Juan, Puerto Rico (USA), and in several coastal municipalities of the state of Yucatan, Mexico, based on specific thresholds. The models were trained with 19 years of dengue fever data for Puerto Rico and six years for Mexico. Environmental and demographic data included in the predictive models were sea surface temperature (SST), precipitation, air temperature (i.e., minimum, maximum, and average), humidity, previous dengue cases, and population size. Two models were applied for each study area. One predicted dengue incidence rates based on population at risk (i.e., numbers of people younger than 24 years), and the other on the size of the vulnerable population (i.e., number of people younger than five years and older than 65 years). The predictive power was above 70% for all four model runs. The ANNs were able to successfully model dengue fever outbreak occurrences in both study areas. The variables with the most influence on predicting dengue fever outbreak occurrences for San Juan, Puerto Rico, included population size, previous dengue cases, maximum air temperature, and date. In Yucatan, Mexico, the most important variables were population size, previous dengue cases, minimum air temperature, and date. These models have predictive skills and should help dengue fever mitigation and management to aid specific population segments in the Caribbean region and around the Gulf of Mexico.

Health information system strengthening and malaria elimination in Papua New Guinea

Background

The objective of the study was to describe an m-health initiative to strengthen malaria surveillance in a 184-health facility, multi-province, project aimed at strengthening the National Health Information System (NHIS) in a country with fragmented malaria surveillance, striving towards enhanced control, pre-elimination.

Methods

A remote-loading mobile application and secure online platform for health professionals was created to interface with the new system (eNHIS). A case-based malaria testing register was developed and integrated geo-coded households, villages and health facilities. A malaria programme management dashboard was created, with village-level malaria mapping tools, and statistical algorithms to identify malaria outbreaks.

Results

Since its inception in 2015, 160,750 malaria testing records, including village of residence, have been reported to the eNHIS. These case-based, geo-coded malaria data are 100% complete, with a median data entry delay of 9 days from the date of testing. The system maps malaria to the village level in near real-time as well as the availability of treatment and diagnostics to health facility level. Data aggregation, analysis, outbreak detection, and reporting are automated.

Conclusions

The study demonstrates that using mobile technologies and GIS in the capture and reporting of NHIS data in Papua New Guinea provides timely, high quality, geo-coded, case-based malaria data required for malaria elimination. The health systems strengthening approach of integrating malaria information management into the eNHIS optimizes sustainability and provides enormous flexibility to cater for future malaria programme needs.

Analysing trends and forecasting malaria epidemics in Madagascar using a sentinel surveillance network: a web-based application

Background

The use of a malaria early warning system (MEWS) to trigger prompt public health interventions is a key step in adding value to the epidemiological data routinely collected by sentinel surveillance systems.

Methods

This study describes a system using various epidemic thresholds and a forecasting component with the support of new technologies to improve the performance of a sentinel MEWS. Malaria-related data from 21 sentinel sites collected by Short Message Service are automatically analysed to detect malaria trends and malaria outbreak alerts with automated feedback reports.

Results

Roll Back Malaria partners can, through a user-friendly web-based tool, visualize potential outbreaks and generate a forecasting model. The system already demonstrated its ability to detect malaria outbreaks in Madagascar in 2014.

Conclusion

This approach aims to maximize the usefulness of a sentinel surveillance system to predict and detect epidemics in limited-resource environments.

Time Series Analysis of Meteorological Factors Influencing Malaria in South Eastern Iran

BACKGROUND

The Malaria Early Warning System is defined as the use of prognostic variables for predicting the occurrence of malaria epidemics several months in advance. The principal objective of this study was to provide a malaria prediction model by using meteorological variables and historical malaria morbidity data for malaria-endemic areas in south eastern Iran.

METHODS

A total of 2002 locally transmitted microscopically confirmed malaria cases, which occurred in the Minab district of Hormozgan Province in Iran over a period of 6 years from March 2003 to March 2009, were analysed. Meteorological variables (the rainfall, temperature, and relative humidity in this district) were also assessed. Monthly and weekly autocorrelation functions, partial autocorrelation functions, and cross-correlation graphs were examined to explore the relationship between the historical morbidity data and meteorological variables and the number of cases of malaria. Having used univariate auto-regressive integrated moving average or transfer function models, significant predictors among the meteorological variables were selected to predict the number of monthly and weekly malaria cases. Ljung-Box statistics and stationary R-squared were used for model diagnosis and model fit, respectively.

RESULTS

The weekly model had a better fit (R(2)= 0.863) than the monthly model (R(2)= 0.424). However, the Ljung-Box statistic was significant for the weekly model. In addition to autocorrelations, meteorological variables were not significant, except for different orders of maximum and minimum temperatures in the monthly model.

CONCLUSIONS

Time-series models can be used to predict malaria incidence with acceptable accuracy in a malaria early-warning system. The applicability of using routine meteorological data in statistical models is seriously limited.

Dengue disease outbreak definitions are implicitly variable

•

With appropriate and timely control, disease outbreak burden can be minimised.

•

Many different case data-based statistical methods are used to trigger outbreak response.

•

Here we show that these methods are inconsistent and incomparable.

•

This may hinder the effectiveness of outbreak response.

•

Clear quantitative definitions of an outbreak are a prerequisite for effective outbreak early warning and response.

Infectious diseases rarely exhibit simple dynamics. Outbreaks (defined as excess cases beyond response capabilities) have the potential to cause a disproportionately high burden due to overwhelming health care systems. The recommendations of international policy guidelines and research agendas are based on a perceived standardised definition of an outbreak characterised by a prolonged, high-caseload, extra-seasonal surge. In this analysis we apply multiple candidate outbreak definitions to reported dengue case data from Brazil to test this assumption. The methods identify highly heterogeneous outbreak characteristics in terms of frequency, duration and case burden. All definitions identify outbreaks with characteristics that vary over time and space. Further, definitions differ in their timeliness of outbreak onset, and thus may be more or less suitable for early intervention. This raises concerns about the application of current outbreak guidelines for early warning/identification systems. It is clear that quantitatively defining the characteristics of an outbreak is an essential prerequisite for effective reactive response. More work is needed so that definitions of disease outbreaks can take into account the baseline capacities of treatment, surveillance and control. This is essential if outbreak guidelines are to be effective and generalisable across a range of epidemiologically different settings.

Monitoring data quality in syndromic surveillance: learnings from a resource limited setting

Background:

India is in the process of integrating all disease surveillance systems with the support of a World Bank funded program called the Integrated Disease Surveillance System. In this context the objective of the study was to evaluate the components of the Orissa Multi Disease Surveillance System.

Materials and Methods:

Multistage sampling was carried out, starting with four districts, followed by sequentially sampling two blocks; and in each block, two sectors and two health sub-centers were selected, all based on the best and worst performances. Two study instruments were developed for data validation, for assessing the components of the surveillance and diagnostic algorithm. The Organizational Ethics Group reviewed and approved the study.

Results:

In all 178 study subjects participated in the survey. The case definition of suspected meningitis in disease surveillance was found to be difficult, with only 29.94%, who could be correctly identified. Syndromic diagnosis following the diagnostic algorithm was difficult for suspected malaria (28.1%), ‘unusual syndrome’ (28.1%), and simple diarrhea (62%). Only 17% could correctly answer questions on follow-up cases, but only 50% prioritized diseases. Our study showed that 54% cross-checked the data before compilation. Many (22%) faltered on timeliness even during emergencies. The constraints identified were logistics (56%) and telecommunication (41%). The reason for participation in surveillance was job responsibility (34.83%).

Conclusions:

Most of the deficiencies arose from human errors when carrying out day-to-day processes of surveillance activities, hence, should be improved by retraining. Enhanced laboratory support and electronic transmission would improve data quality and timeliness. Validity of some of the case definitions need to be rechecked. Training Programs should focus on motivating the surveillance personnel.

Critical review of research literature on climate-driven malaria epidemics in sub-Saharan Africa

OBJECTIVES

To obtain a better understanding of existing research evidence towards the development of climate-driven malaria early warning systems (MEWS) through critical review of published literature in order to identify challenges and opportunities for future research.

STUDY DESIGN

Literature review.

METHODS

A comprehensive search of English literature published between 1990 and 2009 was conducted using the electronic bibliographic database, PubMed. Only studies that explored the associations between environmental and meteorological covariates, El Nino Southern Oscillation (ENSO) and malaria as the basis for developing, testing or implementing MEWS were considered.

RESULTS

In total, 35 relevant studies revealed that the development of functional climate-based MEWS remains a challenge, partly due to the complex web of causality and partly due to the use of imprecise malaria data, spatially and temporally varying covariate data, and different analytical approaches with divergent underlying assumptions. Nevertheless, high resolution spatial and temporal data, innovative analytical tools, and new and automated approaches for early warning and the development of operational MEWS.

CONCLUSIONS

Future research should exploit these opportunities and incorporate the various aspects of MEWS for functional epidemic forecasting systems to be realized.

Epidemiology of malaria in the municipality of Cruzeiro do Sul, State of Acre, Brazil, in 2010: uses of a control chart at the local level

INTRODUCTION: This study describes the uses of a control chart in the malaria surveillance at the local level, signaling whether there is a need to intensify or adapt control measures. METHODS: The districts of Cruzeiro do Sul (n=14), State of Acre, Brazil, were classified into three groups: I) those with an incidence lower than expected; II) those with an incidence within the expected range; and III) those with an epidemic. RESULTS: Thirteen of the fourteen districts had outbreaks of malaria at some point in 2010, and six districts showed persistent malaria epidemic throughout the year. CONCLUSIONS: The control chart may help the malaria control at the local level.

Defining and detecting malaria epidemics in south-east Iran

BACKGROUND

A lack of consensus on how to define malaria epidemics has impeded the evaluation of early detection systems. This study aimed to develop local definitions of malaria epidemics in a known malarious area of Iran, and to use that definition to evaluate the validity of several epidemic alert thresholds.

METHODS

Epidemic definition variables generated from surveillance data were plotted against weekly malaria counts to assess which most accurately labelled aberrations. Various alert thresholds were then generated from weekly counts or log counts. Finally, the best epidemic definition was used to calculate and compare sensitivities, specificities, detection delays, and areas under ROC curves of the alert thresholds.

RESULTS

The best epidemic definition used a minimum duration of four weeks and week-specific and overall smoothed geometric means plus 1.0 standard deviation. It defined 13 epidemics. A modified C-SUM alert of untransformed weekly counts using a threshold of mean+0.25 SD had the highest combined sensitivity and specificity. Untransformed C-SUM alerts also had the highest area under the ROC curve.

CONCLUSIONS

Defining local malaria epidemics using objective criteria facilitated the evaluation of alert thresholds. This approach needs further study to refine epidemic definitions and prospectively evaluate epidemic alerts.

Epidemic of Plasmodium falciparum malaria involving substandard antimalarial drugs, Pakistan, 2003

Because of instability in eastern Afghanistan, new refugees crossed into the federally administered tribal areas of northwestern Pakistan in 2002. In 2003, we investigated an epidemic of Plasmodium falciparum malaria in 1 of the camps. Incidence was 100.4 cases/1,000 person-years; in other nearby camps it was only 2.1/1,000 person-years. Anopheline mosquitoes were found despite an earlier spray campaign. Documented clinical failures at the basic health unit prompted a drug resistance survey of locally manufactured sulfadoxine-pyrimethamine used for routine treatment. The in vivo failure rate was 28.5%. PCR analysis of the P. falciparum dihydrofolate reductase and dihyropteroate synthase genes showed no mutations associated with clinical failure. However, chemical analysis of the drug showed that it was substandard. As global incidence decreases and epidemics become more of a threat, enhanced quality assurance of control interventions is essential.

Beyond traditional surveillance: applying syndromic surveillance to developing settings--opportunities and challenges

BACKGROUND

All countries need effective disease surveillance systems for early detection of outbreaks. The revised International Health Regulations [IHR], which entered into force for all 194 World Health Organization member states in 2007, have expanded traditional infectious disease notification to include surveillance for public health events of potential international importance, even if the causative agent is not yet known. However, there are no clearly established guidelines for how countries should conduct this surveillance, which types of emerging disease syndromes should be reported, nor any means for enforcement.

DISCUSSION

The commonly established concept of syndromic surveillance in developed regions encompasses the use of pre-diagnostic information in a near real time fashion for further investigation for public health action. Syndromic surveillance is widely used in North America and Europe, and is typically thought of as a highly complex, technology driven automated tool for early detection of outbreaks. Nonetheless, low technology applications of syndromic surveillance are being used worldwide to augment traditional surveillance.

SUMMARY

In this paper, we review examples of these novel applications in the detection of vector-borne diseases, foodborne illness, and sexually transmitted infections. We hope to demonstrate that syndromic surveillance in its basic version is a feasible and effective tool for surveillance in developing countries and may facilitate compliance with the new IHR guidelines.

Epochal changes in the association between malaria epidemics and El Niño in Sri Lanka

BACKGROUND

El Niño events were suggested as a potential predictor for malaria epidemics in Sri Lanka based on the coincidence of nine out of 16 epidemics with El Niño events from 1870 to 1945. Here the potential for the use of El Niño predictions to anticipate epidemics was examined using enhanced climatic and epidemiological data from 1870 to 2000.

METHODS

The epidemics start years were identified by the National Malaria Control Programme and verified against epidemiological records for consistency. Monthly average rainfall climatologies were estimated for epidemic and non-epidemic years; as well El Niño, Neutral and La Niña climatic phases. The relationship between El Niño indices and epidemics was examined to identify 'epochs' of consistent association. The statistical significance of the association between El Niño and epidemics for different epochs was characterized. The changes in the rainfall-El Niño relationships over the decade were examined using running windowed correlations. The anomalies in rainfall climatology during El Niño events for different epochs were compared.

RESULTS

The relationship between El Niño and epidemics from 1870 to 1927 was confirmed. The anomalies in monthly average rainfall during El Niño events resembled the anomalies in monthly average rainfall during epidemics during this period. However, the relationship between El Niño and epidemics broke down from 1928 to 1980. Of the three epidemics in these six decades, only one coincided with an El Niño. Not only did this relationship breakdown but epidemics were more likely to occur in periods with a La Niña tendency. After 1980, three of four epidemics coincided with El Niño.

CONCLUSION

The breakdown of the association between El Niño and epidemics after 1928 is likely due to an epochal change in the El Niño-rainfall relationship in Sri Lanka around the 1930's. It is unlikely that this breakdown is due to the insecticide spraying programme that began in 1945 since the breakdown started in 1928. Nor does it explain the occurrence of epidemics during La Niña phase from 1928 to 1980. Although there has been renewed coincidence with El Niño after 1980, this record is too short for establishing a reliable relationship.

Climate change and infectious diseases in North America: the road ahead

Global climate change is inevitable--the combustion of fossil fuels has resulted in a buildup of greenhouse gases within the atmosphere, causing unprecedented changes to the earth's climate. The Fourth Assessment Report of the Intergovernmental Panel on Climate Change suggests that North America will experience marked changes in weather patterns in coming decades, including warmer temperatures and increased rainfall, summertime droughts and extreme weather events (e.g., tornadoes and hurricanes). Although these events may have direct consequences for health (e.g., injuries and displacement of populations due to thermal stress), they are also likely to cause important changes in the incidence and distribution of infectious diseases, including vector-borne and zoonotic diseases, water-and food-borne diseases and diseases with environmental reservoirs (e.g., endemic fungal diseases). Changes in weather patterns and ecosystems, and health consequences of climate change will probably be most severe in far northern regions (e.g., the Arctic). We provide an overview of the expected nature and direction of such changes, which pose current and future challenges to health care providers and public health agencies.

Response to malaria epidemics in Africa

Actions should focus on early recognition of abnormal transmission and rapid deployment of mass drug administration.

Malaria epidemics affect nonimmune populations in many highland and semi-arid areas of Africa. Effective prevention of these epidemics is challenging, particularly in the highlands, where predictive accuracy of indicators is not sufficiently high to allow decisions involving expensive measures such as indoor residual spraying of insecticides. Advances in geographic information systems have proved useful in stratification of areas to guide selective targeting of interventions, including barrier application of insecticides in transmission foci to prevent spread of infection. Because rainfall is associated with epidemics in semi-arid areas, early warning methods based on seasonal climate predictions have been proposed. For most areas, response measures should focus on early recognition of anomalies and rapid mass drug administration. Vector control measures are useful if abnormal transmission is highly likely and if they can be selectively implemented at the early stages of an outbreak.

Forecasting malaria incidence based on monthly case reports and environmental factors in Karuzi, Burundi, 1997-2003

Background

The objective of this work was to develop a model to predict malaria incidence in an area of unstable transmission by studying the association between environmental variables and disease dynamics.

Methods

The study was carried out in Karuzi, a province in the Burundi highlands, using time series of monthly notifications of malaria cases from local health facilities, data from rain and temperature records, and the normalized difference vegetation index (NDVI). Using autoregressive integrated moving average (ARIMA) methodology, a model showing the relation between monthly notifications of malaria cases and the environmental variables was developed.

Results

The best forecasting model (R²_adj = 82%, p < 0.0001 and 93% forecasting accuracy in the range ± 4 cases per 100 inhabitants) included the NDVI, mean maximum temperature, rainfall and number of malaria cases in the preceding month.

Conclusion

This model is a simple and useful tool for producing reasonably reliable forecasts of the malaria incidence rate in the study area.

Approaches to the evaluation of outbreak detection methods

Background

An increasing number of methods are being developed for the early detection of infectious disease outbreaks which could be naturally occurring or as a result of bioterrorism; however, no standardised framework for examining the usefulness of various outbreak detection methods exists. To promote comparability between studies, it is essential that standardised methods are developed for the evaluation of outbreak detection methods.

Methods

This analysis aims to review approaches used to evaluate outbreak detection methods and provide a conceptual framework upon which recommendations for standardised evaluation methods can be based. We reviewed the recently published literature for reports which evaluated methods for the detection of infectious disease outbreaks in public health surveillance data. Evaluation methods identified in the recent literature were categorised according to the presence of common features to provide a conceptual basis within which to understand current approaches to evaluation.

Results

There was considerable variation in the approaches used for the evaluation of methods for the detection of outbreaks in public health surveillance data, and appeared to be no single approach of choice. Four main approaches were used to evaluate performance, and these were labelled the Descriptive, Derived, Epidemiological and Simulation approaches. Based on the approaches identified, we propose a basic framework for evaluation and recommend the use of multiple approaches to evaluation to enable a comprehensive and contextualised description of outbreak detection performance.

Conclusion

The varied nature of performance evaluation demonstrated in this review supports the need for further development of evaluation methods to improve comparability between studies. Our findings indicate that no single approach can fulfil all evaluation requirements. We propose that the cornerstone approaches to evaluation identified provide key contributions to support internal and external validity and comparability of study findings, and suggest these be incorporated into future recommendations for performance assessment.

Malaria

Malaria is the most important parasitic infection in people, accounting for more than 1 million deaths a year. Malaria has become a priority for the international health community and is now the focus of several new initiatives. Prevention and treatment of malaria could be greatly improved with existing methods if increased financial and labour resources were available. However, new approaches for prevention and treatment are needed. Several new drugs are under development, which are likely to be used in combinations to slow the spread of resistance, but the high cost of treatments would make sustainability difficult. Insecticide-treated bed-nets provide a simple but effective means of preventing malaria, especially with the development of longlasting nets in which insecticide is incorporated into the net fibres. One malaria vaccine, RTS,S/AS02, has shown promise in endemic areas and will shortly enter further trials. Other vaccines are being studied in clinical trials, but it will probably be at least 10 years before a malaria vaccine is ready for widespread use.

Weather-based prediction of Plasmodium falciparum malaria in epidemic-prone regions of Ethiopia II. Weather-based prediction systems perform comparably to early detection systems in identifying times for interventions

BACKGROUND

Timely and accurate information about the onset of malaria epidemics is essential for effective control activities in epidemic-prone regions. Early warning methods that provide earlier alerts (usually by the use of weather variables) may permit control measures to interrupt transmission earlier in the epidemic, perhaps at the expense of some level of accuracy.

METHODS

Expected case numbers were modeled using a Poisson regression with lagged weather factors in a 4th-degree polynomial distributed lag model. For each week, the numbers of malaria cases were predicted using coefficients obtained using all years except that for which the prediction was being made. The effectiveness of alerts generated by the prediction system was compared against that of alerts based on observed cases. The usefulness of the prediction system was evaluated in cold and hot districts.

RESULTS

The system predicts the overall pattern of cases well, yet underestimates the height of the largest peaks. Relative to alerts triggered by observed cases, the alerts triggered by the predicted number of cases performed slightly worse, within 5% of the detection system. The prediction-based alerts were able to prevent 10-25% more cases at a given sensitivity in cold districts than in hot ones.

CONCLUSIONS

The prediction of malaria cases using lagged weather performed well in identifying periods of increased malaria cases. Weather-derived predictions identified epidemics with reasonable accuracy and better timeliness than early detection systems; therefore, the prediction of malarial epidemics using weather is a plausible alternative to early detection systems.

Weather-based prediction of Plasmodium falciparum malaria in epidemic-prone regions of Ethiopia I. Patterns of lagged weather effects reflect biological mechanisms

Background

Malaria epidemics due to Plasmodium falciparum are reported frequently in the East African highlands with high case fatality rates. There have been formal attempts to predict epidemics by the use of climatic variables that are predictors of transmission potential. However, little consensus has emerged about the relative importance and predictive value of different factors. Understanding the reasons for variation is crucial to determining specific and important indicators for epidemic prediction. The impact of temperature on the duration of a mosquito's life cycle and the sporogonic phase of the parasite could explain the inconsistent findings.

Methods

Daily average number of cases was modeled using a robust Poisson regression with rainfall, minimum temperature and maximum temperatures as explanatory variables in a polynomial distributed lag model in 10 districts of Ethiopia. To improve reliability and generalizability within similar climatic conditions, we grouped the districts into two climatic zones, hot and cold.

Results

In cold districts, rainfall was associated with a delayed increase in malaria cases, while the association in the hot districts occurred at relatively shorter lags. In cold districts, minimum temperature was associated with malaria cases with a delayed effect. In hot districts, the effect of minimum temperature was non-significant at most lags, and much of its contribution was relatively immediate.

Conclusions

The interaction between climatic factors and their biological influence on mosquito and parasite life cycle is a key factor in the association between weather and malaria. These factors should be considered in the development of malaria early warning system.