Over the past decades, the increased demands on the landscape for food and shelter and an increased number of by-products of man's living environment have led to unparalleled changes. Some of these changes have led to an increase in the distribution of several vector-borne diseases, including malaria.
Malaria is a serious vector-borne disease affecting a greater proportion of the world's population than any other vector-transmitted disease. Over 200 million cases of malaria occur every year, and the number is increasing. Large areas of regions where malaria had been controlled are now suffering again from this significant public health problem.
The increase in malaria prevalence is determined by several factors: mosquito resistance to insecticides, parasite resistance to drugs, changes in land-use patterns, and reductions in funding and manpower dedicated to control activities. Most of the determinants are heterogeneously distributed, changing over both space and time. Factors such as topography, temperature, rainfall, land use, population movements, and degree of deforestation have a profound influence on the temporal and spatial distribution of malaria vectors and malaria.
Despite its importance, the study of environmental determinants of malaria has been hampered by the difficulties related to collecting and analyzing environmental data over large areas, and to the speed of change in the malaria epidemiological situation. Geographic information systems (GIS), remote sensing, and geographic positioning systems (GPS) are important new tools for the study and control of malaria.
GIS may be defined as a constellation of software and hardware tools capable of integrating digital images for the purpose of dealing with geographically localized data.
Large amounts of information are necessary for almost all aspects of malaria control programs. GIS offer the ability to process quantities of data beyond the capacities of manual systems. Data are stored in a structured digital format, which permits rapid retrieval and use. In addition, data may be quickly compiled into documents, using techniques such as automatic mapping and direct report printouts (Bernhardsen 1992).
Data Integration
GIS facilitate the integration of quantitative malaria determination and control data with data
obtained from maps, satellite images, and aerial photos. Frequently, socioeconomic data and
qualitative information on health facilities have a spatial basis, and can also be integrated with
malaria data from the same areas. The integration of operational and logistical data for malaria
control program planning with epidemiological data will serve to strengthen both the
epidemiological analysis and the planning and execution of control programs.
Stratification (Spatial Decision Support)
Malaria stratification aids in the development of community-based malaria control programs, by
accumulating past experiences with and solutions to different factors associated with malaria
outbreaks (Orlov 1986). Stratification can also point to the existing inequalities in resources,
allowing for a more equal and homogeneous distribution of available resources (Kadt 1993).
Stratification of malaria emerged as a strategic approach in Latin America in 1979. Since the 1980s, it has become recognized as a useful tool for attaining an objective epidemiological diagnosis, and as a basis for planning strategies of malaria prevention and control.
The ability of GIS to deal with large data sets and to incorporate satellite images increases the feasibility of studying the environmental determinants of malaria. The epidemiological mapping of high-risk areas of malaria transmission has helped national level malaria control programs to recognize those populations and geographic areas where it is possible to identify the main determinants of malaria morbidity and mortality.
One of the main functions of GIS is the selection of areas according to some specified characteristics, in a search for regularities. These characteristics may be spatial or nonspatial data from a map or database within the GIS. Data may be selected according to geographic criteria or content. Most GIS incorporate the use of Structured Query Language, which allows logic and arithmetic operations. Queries may be built to stratify areas according to inundation potential, proximity to treatment sites, availability of transport, incidence of malaria, quality of housing, and so on. Characteristics may also be combined: one could ask the GIS to find areas where the houses are of poor quality and the drainage of the terrain slow, as possible candidates for a control strategy with emphasis on drainage instead of spraying.
The stratification of areas according to their epidemiological situation is also useful to development agencies implementing agricultural projects. Many such projects require environmental and health impact assessments, which could be used as a working base by malaria control programs in the early stages of land occupation. At the same time, the projects would benefit from the control programs' previous data. It is thought that a GIS would be an adequate tool to integrate these different data sets and information.
Heuristic Modeling
Maps are heuristic models, used to communicate, interpret, and explain data. They aid in the
visualization of differences, clustering, heterogeneity, or homogeneity within data. Spatial
patterns can be perceived and correlations visualized through the use of maps. Symbols and
colours can communicate detail or the relative importance of certain features.
Malaria control program staff tend to be familiar with maps, using them for their daily activities. Maps can consequently be used to communicate ideas and explanations about the determinants of malaria and strategies of control. A map's heuristic potential can also be exploited as the medium of communication between a control program and the public.
Analysis
A number of analysis methods may be used on data acquired through a GIS. Data may be
selected according to geographic criteria or content using some form of Structured Query
Language. Geometric operations involve the computation of distance, areas, volumes, and
directions. Flow analyses calculate geometric computations to find the shortest routes between
designated points. This type of analysis is frequently important for the planning, logistics, and
operation of malaria control programs.
Statistical operations are performed primarily on attribute data, but may also be applied to some types of geometric data. Most GIS support a range of statistical operations, including sum, maxima, minima, average, frequency distribution, bidirectional comparison, standard deviation, and multivariate analysis. Pattern recognition and statistical modeling are also incorporated in some GIS (Bernhardsen 1992), allowing for the identification of spatial and/or temporal clusters of malaria occurrence. The relation between these clusters and environmental or socioeconomic characteristics can then be investigated.
GIS also open new possibilities of data analysis not limited to spatial analysis. Data generated within GIS, distances, areas, and selections based on spatial criteria, can all be used as inputs to statistical modeling.
Global Changes and Malaria
It is expected that changes in the global environment will lead to changes in malaria occurrence
patterns. These changes are currently being studied and modeled, and their impact on malaria
can now be examined. For example, researchers are looking at the impact of increased
temperatures on the spatial distribution of malaria, and on the impact of deforestation and temperature
increases on the latitudinal limits and intensity of highland malaria.
The occurrence of malaria under different land occupation strategies can be studied with the combination of remote sensing and GIS. The results could be used as guidelines for new development projects in areas receptive to malaria.
Integration with Other Information Systems
WHO technical report no. 839 (1993) recognizes that
A change from specialised information systems for malaria towards integrated health information systems is essential if the general health services are to take full responsibility for managing malaria. An integrated health information system will allow the malaria problem to be related to other health problems.Taken further, it can be seen that the main determinants of malaria are not restricted to health. Consequently, there is a need to integrate other types of data into a malaria information system. Because data are collected to fulfil different objectives and with different formats, one of the few possible strategies for their integration is the use of a geographic base. A strategy for the comprehensive integration of data for the planning and execution of health promotion activities (access to water, sanitation, housing, and so on) has been developed using a GIS as a central tool (Project Blade Runner; see Kadt 1993). A similar instrument could and should be developed by malaria control programs.
The information found within a digital image is contained in a grid constructed of spatial units called pixels. Each pixel number is related to colour intensity and brightness. The main limitations of satellite images are cloud cover (Earth Resources Satellites pass over the tropical areas early in the morning, a period when clouds and mist are most prevalent) and resolution. Even with the best resolution available (pixels < 30 m), it is not possible to see houses, to adequately classify some types of agricultural practice, or to localize some breeding sites. Some of these problems may be circumvented using satellite navigation receivers.
By using the data from the different wave bands, it is possible to identify and track environmental characteristics and changes useful to the study of malaria and other tropical diseases. Vegetation, land-use patterns, surface waters, quality and humidity of the soil, roads, built-up areas, and climatic changes may all be monitored by satellite (Lo 1986; Lillesand 1987).
Satellite Navigation System (SNS)
The satellite navigation system (SNS) was originally designed to enable a user to obtain an
instantaneous three-dimensional position, anywhere on the earth, at any time, under any weather
condition (Leick l990; Cross 1991). Clock information in satellites is emitted by radio wave.
The difference between the time when a message is sent and the time when the message is received allows for the calculation of the distance between the receiver and the satellites. Since the orbits of the satellites are known, exact positions can easily be calculated.
The SNS can be used in a number of ways to calculate absolute and relative positions with varying degrees of accuracy. The complete system consists of 24 satellites that are distributed in such a way that an adequate number of satellites is available for positioning at any given time.
When associated with a GIS, a SNS receiver is a powerful mapping tool. It can provide quick and accurate positioning of terrain features and dynamic mapping. The data received can be transferred to a computer and read by a GIS, where it is transformed into map format.
Different data sets, environmental data, and land occupation patterns from satellite imagery, coordinates of houses, and features of the terrain obtained with SNS, map data from military and road maps, data from the malaria control program, and collected socioeconomic, household microenvironment and malaria data will be integrated into the GIS. The analysis will be conducted in parallel within the GIS and with statistical software. Results from the preliminary analysis will feed back to further analysis (for example, distance from water bodies, amount of deforestation, and land use will be calculated within the GIS and used in the multivariate modelling of malaria determination).
The Information System
In the past, information systems for malaria control were designed for use by eradication
programs. The data were analyzed and used centrally with little or no feedback to the source of
the information and with no impact on local activities.
The World Health Organization soon recognized, however, that information systems are a vital element for strengthening the national and local capacities for assessing malaria situations, for selecting appropriate control measures, and for adapting activities to changes in the malaria situation. It was seen that information systems should therefore be reoriented to deal with malaria and decentralized in such a way that information is available to and used by those who need it.
Because malaria occurrence is influenced by numerous phenomena outside the habitual framework of health systems (including population movements, environmental changes, and agricultural practices), the information system should be able to incorporate and use these different types of information.
Each of these different types of data and information has a spatial basis. It was, therefore, decided that it was important that an information system to be used for malaria control should be able to deal with geographic aspects of information.
The Necessary Capabilities
It was decided that the system chosen should be able to:
Adequate Training
There exists at present a lack of trained personal. The new windowed interfaces, however, are
easier to use and will speed up the process of training. Computing skills are useful in the labour
market; consequently, staff from control programs should be willing to be trained.
GIGO (Garbage in, Garbage out)
GIS is not a tool designed to increase the quality of data. Frequently, much of the data collected
by malaria control program staff is not used. GIS use could lead to a relaxation in data collection
and consolidation. It is necessary to review all the steps in the information flow to guarantee
quality and adequacy.
Misinformation and Misinterpretation
The powerful tools of GIS can easily lead to misinformation and misinterpretation, particularly by
someone unfamiliar with their use. Ecological fallacies, problems of scales, and propagation of
error are frequent, and should be given serious consideration (Monmonier l99l). The same
problem is also true with other types of software that have helped to
advance our capacity to study and control malaria, such as spreadsheets, graphic software, and
data analysis software. GIS is not a magical solution to all the information difficulties of malaria
control, but is a powerful tool capable of transforming the way with which information is dealt.
Back to Contents -
To next section
This file was created 23 February 1996