Community-based early warning systems involve community driven collection and analysis of information that enable warning messages to help a community to react to a hazard and reduce the resulting loss or harm. Most early warning systems are designed at the national or global level. Local communities’ capacity to predict weather conditions using indigenous knowledge has been demonstrated in studies focusing on climate change and agriculture in some African countries. This review was motivated by successes made in non-disease specific community-based early warning systems with a view to identify opportunities for developing similar systems for malaria. This article reviewed the existing community-based early warning systems documented in literature. The types of disasters that are addressed by these systems and the methodologies utilised in the development of the systems were identified. The review showed that most of the documented community-based early warning systems focus on natural disasters such as floods, drought, and landslides. Community-based early warning systems for human diseases are very few, even though such systems exist at national and regional and global levels. There is a clear gap in terms of community-based malaria early warning systems. The methodologies for the development of the community-based early warning systems reviewed mainly derive from the four elements of early warning systems; namely risk knowledge, monitoring, warning communication and response capability. The review indicated the need for the development of community based early warning systems for human diseases.
Early warning systems are an essential component of the Community-Based Disaster Risk Management (Centre for International Studies and Cooperation
Zimbabwe, like other countries of the Southern African Development Community (SADC), is committed to the Abuja targets for Roll Back Malaria in Africa, which includes improved response to epidemics. To meet the targets, countries are expected to detect 60% of malaria epidemics within 2 weeks of onset and respond to 60% of the epidemics within 2 weeks of their detection (DaSilver
There has been a global commitment to the fight against malaria, as evidenced by initiatives like the Roll back Malaria, the Global Fund and the Gates Foundation. All these initiatives are foreign to most people who are affected by malaria. Some of them do not take into consideration the knowledge of local people about the disease, including ways to prevent, treat and even predict outbreaks. Diseases like malaria have existed without being considered natural disasters comparable to floods and earthquakes. Hence the capacity of the local community to predict the disease has been undermined. The top-down approach of malaria control programmes may also have contributed to the lack of community participation in the development of MEWS. Predicting malaria has been considered too complex to include the simple methods used by communities to predict diseases.
This review is based on the analysis of the existing literature on community-based early warning systems. It scrutinises the various definitions of community-based early warning systems given by various organisations and authors, and the methodologies for the development of such systems. The analysis identifies the disasters for which community-based early warning systems have been developed. The potential for developing early warning systems is explored. Early warning systems for conflict and financial crises were not considered.
This review is based on a systematic search of relevant literature on Google scholar (
Of the literature reviewed, 24 articles were on community-based early warning systems and the remaining 29 articles were on early warning systems either at regional or global level. Ten articles out of the 29 were on malaria early warning, while one was on animal diseases. Three articles among those that discussed community-based early warning systems discussed systems for human diseases. Only one out of the 3 articles explained the development of a community-based early warning system for a specific communicable disease.
Various definitions for early warning systems have been used by different authorities. According to the United Nations International Strategy for Disaster Reduction (
A complete and effective people-centred early warning system comprises of four interrelated elements: risk knowledge, monitoring and warning service, dissemination and communication and response capability (UNISDR
All the definitions identified in this review emphasise the importance of community participation in the formulation of systems and the responses triggered by them. However, the extent of the participation is not clear. Some definitions interpret community involvement as participation. Sustainability is another important aspect embedded in the definitions. Community participation creates ownership of the early warning systems and therefore enhances their sustainability. In addition, the acceptance of community-based early warning systems, just as in all early warning systems, depends on their reliability. The system should not produce false alarms or miss hazards. Most of the authors of the literature that was reviewed derived the definition of early warning systems from the UNISDR (
Early warning systems exist at different levels such as the global, national and local levels. The levels are determined by the magnitude of the problems and the capacities of the various levels to address them. Successful early warning systems have been developed for the agricultural sector and include the USAID’s Famine Early Warning System (FEWS), SADC Food Security Programme (/REWU), FAO Global Information and Early Warning System (GIEWS) on Food and Agriculture, FAO Food Insecurity and Vulnerability Information and Mapping Systems (FIVIMS) and World Food Programme (WFP) Vulnerability Analysis and Mapping (VAM). All these systems are developed at global level and communicated down to the regions and countries concerned until the warnings reach the affected populations. Nevertheless, the ordinary farmer in a village may not get some warnings unless he listens to the radio or watches television. These early warning systems are necessary for communities to understand regardless of their technological complexity. Thus, to be useful and relevant, early warning systems must be designed to suit the populations and local settings.
The WHO has promoted MEWS as a core component of epidemic risk management for many years (Cox & Abeku
The MEWS are targeted at alerting health systems. Even though communities are capable of predicting climatic events in their own areas, the existing MEWS do not take this knowledge into account. In response to the need to develop MEWS in Africa, the WHO published a framework proposing the generic concepts and potential early warning and detection indicators for use in MEWS (Thomson & Connor
Early warning systems are continually being created as new threats are being identified (Glantz
Other early warning systems that are documented include the pastoral early warning systems in the Greater Horn of Africa. According to a workshop held in 2001 on early warning systems, most of them have been designed to warn of drought-related crises even though the pastoralists are also exposed to other hazards. The early warning systems are largely funded by international donors and there is a perception that they service the information needs of these international agencies. This has serious implications on how sustainable the systems can be. When the organisations leave the respective countries, the systems are likely to collapse. It is important to note that early warning systems should be balanced in their approach. They should address the needs of the local communities, district and national governments as well as the international donors. Another lesson from the pastoralist early warning systems is that they should start at local level to the district and national level up to the international level. The movement upwards is determined by the exhaustion of the coping capacity of the preceding level.
Radice and Tekle (
Contrary to global and national early warning systems, community-based early warning systems empower communities to prepare for and confront hazards. According to the Pastoral Early Warning System document of 2006, communities are underinformed about the operation of early warning systems and underrepresented in the development of such systems. This results in early warning information that is disseminated but fails to trigger a response due to lack of community awareness and participation in response operations. In order to curb the above challenges, there is a need to develop community-based early warning systems. Community-based early warning systems are not synonymous with traditional early warning systems. We believe that traditional early warning systems or indigenous early warning systems fall under community-based systems because they are formulated and implemented by communities. The community-based early warning systems that have been identified from literature include: pastoral community-based early warning systems, flood and landslides warning systems, multi-hazard early warning systems, avian influenza warning systems and early warning system and mapping for an urban barangay. The majority of the community-based early warning systems documented in literature are for floods and landslides (Gautam & Phaiju
From the analysis of the documented community-based early warning systems, it is evident that they were initiated by outsiders who took into consideration existing traditional knowledge. Almost all of the community-based early warning systems have been initiated or advocated for by NGOs, such as CARE Philippines, Save the Children UK Ethiopia, Practical Action, Christian Aid Social Action Centre, World Vision, USAID et cetera. The existence of national and global warning systems has reduced the communities to ‘at-risk’ communities without considering their capabilities in terms of early warning. According to Save the Children Ethiopia (
Mercy Corps and Practical Action (
The IFRC (
Main characteristics of national early warning systems and community early warning systems.
Characteristics | National EWS> | Community EWS |
---|---|---|
Design | Deliberate, based on legal mandate by government or other agencies |
Flexible design on need and adapted by trial and error |
Human Resources | Technicians Specialists |
Ad hoc volunteers to individuals appointed by local leaders |
Characteristics | Formal staged warning Legislation Policies |
Ad hoc and staged warning |
Documented | Standard operating procedures MOUs Diagrammatic representations of information flow, etc. |
Informal and rarely documented |
Technology | High tech Access to telephones VHF HF radios |
Telephone to traditional or no technology used |
Trigger | Indicators, prediction, technology |
Personal local detection of a hazard or receipt of a warning from outside the community |
Warning process | Cascading or fanned in a systematic manner | Ad hoc, but may be naturally well organised and cascading/fanned |
Messages | Impersonal | Personal |
Timing | Not always the first to be received by community Produced to share with official systems at all levels |
Rapid (when message created at community level) or co when there are good linkages between all levels |
Primary needs targeted | Reduce economic and other losses | Safety, reduce stress, emotional support |
Evaluation criteria | Hazard details Lead-time provided Proportion of false warnings |
Timeliness of receipt of warning, actionable message in warning |
EWS, early warning systems; MOUs., Memoranda of understanding; VHF, Very High Frequency; HF; High Frequency.
The Red Cross and Red Crescent describe traditional early warning systems or indigenous early warning systems rather than community-based early warning system. As mentioned before, we believe that traditional early warning systems are under community-based systems. Our opinion is that for a system to be community-based, it has to be initiated at community level and operated by the communities themselves. Being community-based, does not necessarily mean that it has to be informal. A community-based early warning system can be formal and well documented. It can also use technology, depending on the affluence of the community and the setting in which the system is operated. Gonzales
Radice and Tekle (
Some documented community-based early warning systems were analysed to determine the implementing agencies, the hazards addressed, the methodologies used in their development or the characteristics of the systems and finally their strengths and weaknesses.
Methodologies or characteristics of selected community based early warning systems.
Implementing agency | Hazard | Methodology or characteristics | Strengths | Weaknesses |
---|---|---|---|---|
Save the Children Ethiopia | Drought (designed for pastoralists) | Taping of existing organisational structures and mechanisms within communities Identification of local actors, Participatory Analysis (hazard mapping, vulnerability assessment etc.) Community mobilisation and volunteerism, implementation of DRR Measures Development of food security Monitoring systems and communication dissemination plans |
Low cost Low tech Relevant to the communities Community participation Sustainable uses existing structures |
Traditional indicators not recognised |
Mercy Corps and Practical Action | Floods | Step 0 preparatory step Step I participatory situational Analysis and formulation of the early warning management committee Step II Development of Observation and monitoring system with the active participation of the community Step III development of a communication and dissemination plan Step IV preparation for response Step V follow up, review and Sustainability |
Comprehensive and easy to follow Participatory community centred Outsiders only facilitators Promotes sustainability of the system Takes into Account the existing indigenous early warning systems |
Biased towards flood disasters |
CARE Philippines | Birds flu | Step I avian influenza preparedness orientation Step II formation of five preparedness teams; (surveillance, information and education team, Communications, survey quarantine team and rapid action team); the composition tasks and function of each team are clarified Step III follow-up sessions on Early Warning System are conducted with the Preparedness and Response. teams (review signs and symptoms of avian influenza in animals and in humans to ensure early detection, introduce and fill up monitoring and incident reporting form, clarify reporting flow, rehearse key preparedness messages for public information, practice dissemination of information, updates, and corresponding emergency measures that the community will take if there are cases of avian influenza Step IV Involves agreement on the mechanics of setting up and maintaining the Early Warning System using the 4 Ws (what, where, when and who) Step V Public information activities Step VI review of the system after 6 months and incorporation of any changes that are necessary |
Comprehensive, easy to follow, has a communication and education component Capacity building on the communities Gives room for clarifications and rehearsals before implementation of the system Utilisation of existing structures |
Disease specific |
Information Technology Department De La Salle University, Manila | Natural Disasters in general | Rapid Application Development (RAD an approach in systems development that assures better and more inexpensive systems and more rapid operation by having systems developers and end users work together jointly in real time to develop systems (Hofer, George & Valacich 2002) | User involvement, Interaction with users during development, caters for Different methods of information dissemination not only disasters, system consolidates information on disaster histories Provide accurate reports |
Limited to communities with access to Internet and mobile phones communicates only to registered users Network problems SMS sending delayed due to network traffic Inherent problems with twitter and Facebook may affect the system |
Philippine Atmospheric, Geographical and astronomical Services Administration | Floods | Methodology not outlined but involves: Decision by communities to have a CBEW; consultation of community members and selection of volunteers; establishment of rainfall and water level stations; establishment of threshold values for rainfall observation; observations by volunteers; drills | Simple, inexpensive Incorporates indigenous knowledge of communities Promotes ownership thus sustainability Political commitment |
Methodology not clearly outlined, not easy to adapt Very specific to floods |
Department of Hydrology and meteorology, Nepal, Practical Action, local Government and nongovernmental organisations | Floods | Area survey, flood hazard Mapping and assessment of warning level and danger level; installation of monitoring instruments; setting up of an operation centre; development of a communication and warning dissemination system; training and capacity building cooperation with NGOs working in the area | Utilises existing communication channels Collaboration with other stakeholders working in the area Participation of vulnerable groups. Role of community was clearly outlined |
Specific to floods |
Social Action | Floods | Setting up two way radio communication system, establishment of flood watch points, installation of manual rain gauges, collaboration with existing stakeholders, training of the disaster coordinating council, establishment of linkages with scientific agencies, communication of climate forecasts to communities | Collaboration with other stake holders capacity building utilisation of community volunteers | Does not take into consideration indigenous methods, top-down approach, initiated by NGOs and brought to the people, limited community participation |
Centre Prelature of Infanta World Vision | Landslides and mudslides multi-hazard | Training, hazard identification, hazard analysis, vulnerability analysis, early warning/ community surveillance mitigation plan and implementation, preparedness and response, monitoring and evaluation, community awareness on available resources, communication system | Community participation from inception, takes into account indigenous knowledge, sustainability is addressed Identification of existing resources to address hazards bottom-up and top-down communication |
Addressing many hazards in one system, though desirable may compromise focus on some of the hazards Methodology not exclusive for early warning, general community awareness activities not clear |
Practical Action | Floods | Community awareness and Sensitisation, task groups and volunteers, flood monitoring both automatic and manual at different strategic locations, dissemination of information at all levels; institutions, individuals and networks, Response to information, preparedness for response village and district level, rescue and evacuation trainings, mock drills | Community participation, capacity building, sustainability utilises existing structures, uses available technology e.g. radio, telephone | Does not take into account existing systems and does not consider indigenous knowledge |
Oxfam | Drought | Interactive participation, needs assessment, community action plans, formation/strengthening of community groups, awareness creation | Community participation utilisation of local resources community empowerment strengthening of existing structures external support seen as complementary or gap filling |
Does not take into account existing early warning systems |
Christian Aid | Floods | Use of civil protection committees at village level, communication using megaphone and posters at strategic points, installation of weather measuring equipment, community awareness | Utilisation of existing structures Community awareness Community participation in development and utilisation of the system |
Not clear on utilisation of indigenous systems and community participation Methodology not clearly outlined, outline of implementation rather than methodology. |
USAID | Floods | Existing available warning Information collected together with planned information in future, early warning packaging, early warning needs assessment, consultation with stakeholders sensitisation of communities | Both top-down and bottom-up communication Takes into account existing structures |
Sustainability not clear |
Centre for International Studies and Cooperation, Vietnam | Floods | Empowerment approach, community participation, learning by doing, training and mentoring , cooperation and shared responsibility, participation of vulnerable groups, long-term relationship building, simple tech solutions, monitoring and evaluation | Sustainability, cultural relevance, collaboration between nongovernmental organisations Local government and villagers |
Biased towards floods |
DRR, Disater Risk Reduction.
The major strengths of all the systems analysed are that they are low cost, relevant to the communities and promote sustainability. Identified weaknesses are that some of them are not easy to adapt for disasters other than those for which they were originally formulated. Some of the systems, even though they are community-based early warning systems, do not take into account the existing traditional indicators that the communities may have. Disregarding the knowledge that exists within the communities may have implications for sustainability. The ideal would be to integrate new knowledge and existing knowledge. Technology-based systems pose problems in that many people may not have access to the technology, such that the utilisation of such systems is limited to users who do have access.
The methodology for the avian influenza early warning system by CARE Philippines and the flood early warning system by Practical Action and Mercy Corps can be adapted for the development of a community-based MEWS. The details for each system are shown in
The community-based early warning systems that are documented are mainly for natural disasters. In the reviewed literature, no community-based early warning system has been suggested for malaria. According to Chaiken (
There is generally no documentation of the initiatives in the development of community-based early warning systems for human diseases, except the one by CARE for avian influenza, (CARE Philippines
The successes in community-based early warning systems for weather-based disasters indicate a potential for the development of early warning systems for diseases that are influenced by weather. The establishment and implementation of the community-based early warning systems differ from setting to setting, but have a common conceptual framework that includes the involvement of existing organisational structures and mechanisms within communities (the identification of local actors); participatory analysis by means of hazard mapping, vulnerability and capacity assessments, community DRR planning, etc.; community mobilisation and volunteerism, implementation of DRR measures (both non-structural and structural) and the development of monitoring systems, and communication and dissemination plans.
The United Nations International Strategy for Disaster Reduction acknowledges that simple technologies, such as rainfall and river gauges, combined with equally simple rules of thumb can often enable communities to monitor threats and provide effective warnings. This shows the possibility of the utilisation of indigenous knowledge systems in the development of early warning systems especially in settings where the conventional or formal early warning systems are not reaching the affected populations. Diseases like malaria that are affected by weather conditions can give communities ample time to prepare for the hazards after prediction of the seasons.
The case studies presented show that NGOs have taken the lead in development of community-based early warning systems. The IFRC has developed guiding principles for CEWS. The IFRC simplified and slightly adapted the four core interlocking components of an early warning system published by the UNISDR. In their guidelines, they provide examples of agencies that engage in global or regional early warning monitoring. The hazards that they documented include severe weather, flooding and landslides, drought, wild land fire, earthquakes, volcanoes, tsunamis, epidemics and conflict. They warn that setting up an early warning system at any level without clear links to other disaster reduction and management efforts and entities will result in inefficient, unsustainable products and less effective impact. The setting up of CBEWS requires a considerable investment of time and resources and should not be undertaken without considering alternatives. They encourage synergy across levels, i.e. community, national, regional and/or global. The FIRC has designed guiding principles per each early warning component for community-level practice. They did not prescribe a method for the establishment of the system but rather provided principles on which CBEWS should be based. The guidelines of the IFRC recognise risk knowledge as the foundation of early warning. They emphasise that implementation agencies should accept that community priorities may not be their priorities. Early warning systems should have a monitoring mechanism that evolves as the hazards evolve. Communities need to drive their own early warning systems since passive receivers will not save lives and they can be motivated through public displays of monitoring activities. The guidelines emphasise the response to warnings and not disasters, and therefore the need to have embedded response options in annually updated contingency plans with links to funding. The response actions should be tested in order to achieve robust no-regrets response actions. In terms of warning communication, the guidelines advocate the clear delegation of responsibilities to alert or mediate the hazard. The utilisation of sophisticated warning devises should be avoided. The IFRC puts across an argument on the terminology in early warning systems with the suggestion that the term ‘community’-based early warning systems’ does not really imply community participation but may mean a system that is based at community level but implemented by other agencies. The ideal term would be ‘CEWS’. This term describes the systems more appropriately in that it gives the impression that the systems belong to the communities. The United Nations Development Programme (UNDP) and the European Commission Humanitarian Office, through their disaster preparedness programme, has compiled a review and analysis for community-based practices for reducing the risk of disasters from natural hazards affecting South East Africa and South Western Indian Ocean. The review and analysis focuses on floods caused by swelling rivers; cyclones storm surges, earthquakes, fire and other such events.
The community-based early warning systems methodologies reviewed are all based on the same principles that encompass that the methods include simplicity, be community driven, include vulnerable groups, use local capacities for warning and response and advocate ownership of the systems by the communities. These principles are actually a simplified form of the UNISDR elements of early warning systems. It is clear that there is no one method of designing early warning systems. The principles are mainly guidelines that communities and organisations can use to develop their own systems. The methodology adopted mainly depends on the setting and also the type of disaster that is addressed.
The review has revealed that the concept of early warning encompasses the issuance of the warning, the response and feedback. The definitions of early warning systems found in literature are mainly centred on the UNISDR 2006 definition. Community-based early warning systems have been developed mainly for natural disasters. There are no community-based early warning systems for malaria that have been documented in literature. The same principles used for the development of community-based early warning systems for natural disasters may be utilised in the development of community-based early warning systems for communicable diseases. This was demonstrated by CARE Philippines (
This project received financial support from the UNICEF/UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR) and the Canadian International Development Research Centre (IDRC). This study also received financial support from the College of Health Sciences of the University of KwaZulu-Natal through PhD studentship bursary awarded to Macherera M. Many thanks go to the Ministry of Health and Child Care Zimbabwe for supplying us with malaria data for Gwanda District.
The authors declare that they have no competing interests.
M.M. was responsible for structuring and writing the article and literature search. M.J.C. is the project leader and supervisor for PhD studies. He conceptualised the idea and has read and made suggestions to several versions of the manuscript.