Cover Image
close this bookC.I.S.F.A.M.: Consolidated Information System for Famine Management in Africa - Phase One Report (Centre for Research on the Epidemiology of Disasters - World Health Organisation, 1987, 33 p.)
close this folderCHAPTER 2: CISFAM: An Experimental Information System
View the document(introduction...)
View the document2.1. Background and Rationale
View the document2.2. Typology and Framework
Open this folder and view contents2.3 Overview of the Databases Examined
View the document2.4 Data Source Agencies and Negotiations


Information systems, in the context of famines are, principally, of two kinds: early warning systems and management systems. The former has had, as mentioned earlier, great success in the last two years. The latter has been, on the contrary, visibly neglected. The Organization of Emergency Operations in Africa (OEOA) has noted in its final report, that it lacked specific data to support agency activities on the field. Although the field agencies presented their data at monthly meetings, little effort was made subsequently to co-ordinate, compile or otherwise record the data routinely collected by NGOs.

While many programmes and activities are governed by political pressures, and improved information may be of little value, it is nevertheless a vital and worthwhile investment for programmes that have the will, authority and resources to manage by objectives. The availability of selected, easy-to-understand information can encourage national, international and voluntary agencies to improve programme implementations in famine relief.

The purpose of CISFAM is to provide a centralised information service for decision makers to make educated and appropriate decisions for famine interventions and policy. It would respond to requests from national, regional and international offices, governmental and non-governmental agencies, for planning, targeting and policy-making in famine management programmes. It would serve as a centralized source where limited information on sectors such as agriculture, meteorology or economy, in addition to health, would be quickly and easily available. This would eliminate the need for planners and managers to go to different specialized agencies for information on various sectors.

2.1. Background and Rationale

CISFAM was launched in March, 1986, as a joint initiative of Emergency Preparedness and Response Unit (WHO) and Centre for Research on Epidemiology of Disasters (CRED) as a initial effort to upgrade the preparedness and management concept in famine relief and recovery in Africa.


The actual project was a result of four main observations.

(i) Resource constraints were getting increasingly serious with grave implications for continued international assistance;

(ii) While the health sector is a focal point in famine relief, the crisis is essentially a multisectoral problem and requires multi-disciplinary data for effective programme planning and resource allocation;

(iii) Large quantities of data existed in specialized agencies of the larger U.N. family, and in national archives. While the international data-collections are frequently in sophisticated and technical form that are inaccessible to the uninitiated, the national ones on the other hand are non-standardized, non-computerized and dusty;

(iv) The NGOs were observed to frequently have regular data reporting systems which were not adequately processed or used either by themselves or the governments with whom they work. These agencies formed a potentially important repository of sub-national data.

Beside the principal sector of health, five additional sectors were selected as being relevant to any famine programme planning and these were: demography, agriculture, logistic and infrastructure, socio-economic, environment and meteorology.

A quick inventory of the large databases revealed that, even at the international reporting level, most of these countries had poor data, particularly in health.

The project is currently housed at the WHO Collaborating Centre for Research on Epidemiology of Disaster, (CRED), Catholic University of Louvain, Brussels, Belgium. The CISFAM project has a team of six members at CRED, with varied responsibilities in their respective areas of competence.

Its main functions are to acquire collaboration from the source agencies, collate, digest and classify the data to ensure ease of access, comprehensibility, and useability of different types of data by field planners and managers of famine and food programmes at short notice. Furthermore, it is expected once the process becomes operational, to ensure system-wide compatibility of information flows.


To provide quick and easy access to country data on multiple sectors to:

. National governments
. International and bi-lateral agencies
. Non-governmental organizations.

To conceptualize and design a ready to use information system, (including staff training), for transfer to the national health authorities or the relief and recovery management unit.

To identify and develop image, graphic and map-linked databases for quick interpretation and operational decision-making.

2.2. Typology and Framework

CISFAM is restricted to nine countries as a function of their high susceptibility to acute food shortages and famine conditions. Of these, four were chosen for sub-national data collection, while the rest remained at the aggregated national level. Other countries also threatened by famine are not included in the interest of manageability of this pilot phase. Moreover, the CISFAM project does not make any effort to collect any primary data from the field in this project.

The database is not conceived to be an exhaustive source of information on each of the represented sectors. It compiles some of the most important items in each sector which have significance in famine relief and prevention planning

The structure is defined by country blocks. Information by category, as available, is collected for each country. Any additional information generated by special survey and studies generated by other bodies is appended in the informational annexes to the country data block. These annexes are restricted to only those surveys dealing with food, nutrition and health. A graphic representation of the CISFAM structure is shown below.

Figure 1: Schematic Representation of the CISFAM Database


Several data bases were examined including both electronically maintained and data on cards, reports and other forms of hard copy. Certain CISFAM countries had better data reporting than others and the variability was significant

Statistical services of certain countries had been out of operation for several years and therefore the only viable data sources were non-governmental organizations and sample surveys and studies. Data quality also, varied widely from sector to sector.

Meteorological, ecological and climate data bases provided the best quality information in terms of reliability, coverage, accuracy and time series. The Climate System Monitoring database of the World Meteorological Organisation initiated as a response to the occurrence of significant climate anomalies over the last decades with associated adverse socio-economic effects provided potentially useful information. It carried synthesized information on climate anomalies, rainfall variability and vegetation data by small geographic areas. It has available time series data for the past 110 years on African rainfall. The data, however, is fairly technical and is not divided into the political and administrative boundaries of the countries. Similarly, the soils and temperature databases of Africa held by United Nations Environmental Programme are also very robust and reliable.

Several databases from the six different sectors were explored. A summary of the electronically registered data-sets are presented below:

Table 1:
Summary of International Computerized Database Relevant to CISFAM






Agency staff; other U.N. agency staff; External users

Food supply agricultural production, food aid; pest control; meteorology; commodity markets; and prices.



Agency staff; Other U.N. agency staff; External users

Food statistics; agricultural statistics; crops; agricultural production; imports/exports; land use; agricultural workers and machinery; fertilizers; forestry and fishery statistics.

Worldwide; Regional; National; Sub-national


Agency staff; Other U.N. agency staff; External users

Consultants; disasters; devastation; regional analyses damages compensation emergency relief; supply management; disaster prevention.



Agency Staff, Other U.N. agency staff; External users

Disasters; natural disasters; disaster prevention


Agency staff; Other U.N. agency staff; External users

Prices, commodity prices and markets.



Agency staff; Other U.N. Agency staff; External users

Economic indicators; social indicators.

Worldwide; Regional; National


Agency staff: Other U.N. agency staff; External users

Infectious diseases; transmission epidemiology


Agency staff; Other U.N. agency staff; External users

Mortality; causes of death; infectious diseases; health personnel; hospitals demographic and health statistics Morbidity

Worldwide; Regional; National


Agency staff; Other U.N. agency staff; External users

Health; legislation; environmental legislation; aging; human nutrition; food standards; pharmaceuticals; poisons occupational hygiene; health statistics

Worldwide; Regional; National; Sub-national


Agency staff; Other U.N. agency staff; External users with restrictions

Climate; meteorology; geophysics; climatology


The health sector information on the other hand, was disappointing in both coverage, quality and continuity. The physical resources, such as hospitals, dispensaries, health centres and skilled personnel enumerated in the source agencies, were the only items with regular reporting. However, a major caveat in these items are that frequently hospitals and health centres are, in fact, inoperative and therefore, the value of these numbers are questionable for certain objectives. For crisis management and long-term planning, however, the knowledge of their existence can be useful. Several sample surveys on nutritional status and incidence of nutritional deficiency diseases are available but no officially reported figures on on-going basis. Continuous data collection in this field is undertaken only by large non-governmental organisations, who, generally execute this function as an incidental by-product of their principal activities. Country statistical annuaires and card files were used for data on four of the CISFAM countries at provincial levels. Table 2 presents a summary of the CISFAM database contents listing gross categories of variables by sectors.

Table 2:
CISFAM Database Contents: Sectors files, Items Categories and Sources







Basic Health Indicators
WHO HFA Indicators
Primary causes of morbidity
Health service coverage
Health establishments and personnel
Cholera, yellow fever: incidence and deaths


Weekly Epidemiological records/WHO HFA Indicator Data Set U.N. Population Division Global Epidemiological Surveillance/WHO Country Annual Reports

Available on sub-national levels for Senegal, Mali, Niger, Mauritania

Food Agriculture

Food supply, cereal availability; cash/food crop production; livestock products prices; food intake, irrigation; means of production; land use; food aid; agricultural population


Agricultural production database; GIEWS/FAO Development Centre OECD

National; occasional sub-national; large survey data available


Soil status; animal pressure; population pressure; desertification hazard index; vegetation cover


Global Environmental Monitoring System

Sub-national; by small geographic areas with no administrative boundaries


Population estimates projections; mortality, fertility and other vital statistics


U.N. population division; World bank; World Fertility Survey International Statistical Institute

National level large survey data

Climate and meteorology

Precipitation; geographical co-ordinates; temperatures; windspeed, growing season

Numeric; raster

Climate monitoring system/WMO

Georeferenced data

Social and economic

Macro-economic indicators; education and social indicators food import/exports


OECD; World Bank Country annual reports


Desertification index; population and animal pressure



All countries

Administrative and communications

1:500,000 digitized

Institut Graphique National, Paris.

Only Senegal, Mali, Niger

Administrative and geographic


All countries

Limitations and Caveats in the Data

The confidentiality of the data returns is usually provided by law in particular, for manufacturing and production data. However, these legal provisions are applied somewhat indiscriminately in many African countries. The problem in this project was more unuseability rather than unaccessability of the data.

The limited data collected from archival records are uneven in coverage and quality in certain sectors, in particular health. The cholera and yellow fever data transferred from cards filed on the Weekly Epidemiological Reports were about the only regular incidence data to be reported to WHO by province. The data are under-reported for want of coverage and the magnitude of this under-reporting is unknown. The problem is aggravated when the data are compiled by different agencies. Estimates are different for the same variable according to the source, and on occasion delayed reporting caused inexplicable increases in incidence of diseases.

Most of the statistics from these records are either reported late or not reported at all. Obviously, the former situation is preferable to the latter, although none is desirable. An argument often advanced in favour of the late publication is that it improves the quality of statistics. But the timeliness of statistical information can be increased only at the expense of accuracy, while improvements in quality require more time and increase the cost of the information. Comparability of the statistics in Africa, both temporal and spatial, is limited by the use of non-comparable definitions, difference in coverage and the timing of the data collection. Comparability is further reduced by the details of tabulations.

Different types of conceptual problems arise in interpreting and collecting data. In agriculture, for example, under traditional African conditions, it is not always clear what is to be considered the main occupation of the holder, as persons can be occupied in different occupations at different times of the year.

In certain countries of the region, farmers frequently work away from their holdings during a large part of the year, for example, on plantations, in mines or nearby towns. The data of economically active women in the country are extremely unreliable due to the confusion arising from the definition of economic activity. The number of employees may be regarded as reliable, but a lot of confusion arise in classifying people, particularly women, into own-account or family workers. Furthermore, the fact that a man can have more than one wife who independently cultivate separate pieces of land though the land may customarily belong to the head of household create problems for both demographic and agricultural information.

The data on health, like data from other sectors, are subject to several limitations. They vary from incomplete coverage to details in the tabulations provided by various countries. There is also lack of uniformity in data collection procedures and the publication of such information. As the extent of undernotification is high and often unknown, the statistics collected are essentially useful only for the operational purposes of controlling diseases by taking immediate steps for prevention and disease surveillance, and for rough assessment of trends in the incidence of the disease.

The data also suffers from other limitations, the most important besides coverage being selectivity. Hospital statistics are, no doubt, an accurate source of diagnostic information, but they suffer from bias arising from selectivity in relation to factors such as location, type of disease, provision of health facilities, age and sex of the patient, and social and economic factors. The population served is also unknown. It is not, therefore, possible to generalize the hospital experience in respect of many diseases to community level, and these statistics cannot give a true picture of prevalence or distribution of morbid conditions in a community as a whole. However, they serve the purpose of measuring the relative distribution of diseases in the areas covered, and can be treated as valuable adjuncts to mortality statistics, suggesting priorities for provision of more medical facilities and efficient medical care.

Nevertheless, the importance of data in national disaster policies requires a reappraisal of the data needs to formulate, monitor and evaluate these policies. These demands a critical examination of the relevance, adequacy and reliability of current statistical data and of the tools used to collect and analyse these data.

2.4 Data Source Agencies and Negotiations

Since the system is not designed to create any new sources of information, efforts have been made to avoid duplication with existing systems. The project has reviewed the existing data-banks and information systems of the UN agencies and allied bodies. Efforts are made to use CISFAM as means to enhance the utilization of these databases, especially by users in developing countries. These sources of information and databases are little known and even less utilized by most of the targeted audience. A very small proportion of the potential users showed any knowledge of the UN systems databases and fewer acknowledged ever requesting or receiving any information from them.

The project has established working relations with the following agencies:

- Food and Agricultural Organisation,
- United Nations Disaster Co-ordination Office,
- World Meteorological Organisation,
- United Nations Environmental Program,
- Global Environmental Monitoring System (GEMS),
- Global Resources Information Database (GRID),
- World Food Program,
- Organisation for Economic Co-operation and Development,
- European Commission,
- World Bank,
- World Health Organisation and its internal departmental data collections,
- Centers for Disease Control, Atlanta,
- Office de la Recherche Scientifique et Technique d’Outre-Mer (ORSTOM), France, as well as: Mcins Sans Frontis, (Belgique,) Laboratoire International de Calcul et d’Intelligence Artificielle (LICIA), Paris, International Statistical Institute, The Hague.

Discussions with these agencies centred around the access of data for the demonstration module of CISFAM and the possibility of establishing co-operative linkages between CISFAM and their respective data-banks and information systems as well as for the feeding, procurement and consultancy services for CISFAM. Some agencies, moreover, expressed the view that CISFAM could actually enhance the utilization of their existing information system by becoming their “First link” with the users of their information services.

Plate 1: UNEP/GRID AFRICAN DATABASE - Desertification Hazard