GIS stands for Geographic Information System. A GIS is an automated (ie. computerised) system for the input, storage, analysis and output (viewing) of spatial information. Various software packages have been developed for the visualisation of geographically referenced data, all them resource-hungry, all of them expensive, and, for the epidemiologist, usually indispensable. In fact, running a veterinary information system without some kind of a mapping system - even a manual one - would be akin to trying to run a motor car without a speedometer or a fuel gauge.
The main advantage of GIS software is not just that the user is enabled to see how a disease is distributed geographically, but also that an animal disease can be viewed against other information - for example, rainfall maps, vegetation maps, rivers, and so on. The disease presence can then be related to other factors and more easily appreciated visually.
Using disease data within a GIS means that every observation of disease recorded in the disease database must be georeferenced (in other words, must be accompanied by latitude and longitude values or some other grid reference recognised by the software).
There are two main ways of handling this, each of which has advantages and disadvantages. One is to add georeferences to each report (ie. each questionnaire must have a place where georeferences can be filled in). The other method is to have a “master file” built into the computer database where the computer (automatically) or the input clerk (manually) can “look up” the latitude/longitude values and add them to the records.
Georeferences on the questionnaire
This can be done in a number of ways. Each field worker can be given a standard list of all georeferences for the places in his area of work, and can add to the data forms as he goes along, or he could look them up on his return to his district base and add them there. Alternatively, someone else at the district office could fill them in once the forms have been handed in, which might be preferable, as many field workers (especially those with lower educational levels) might not understand the principles involved, or might make too many errors. The district veterinarian would have to check the co-ordinates before sending the forms away for computerisation, but obviously would not be able to check everything in detail.
Getting georeferences from a computer table
This would entail having a database table containing the names and geographic co-ordinates of every place in the country that could conceivably be visited by field staff. From that point, things could happen in one of two ways:
The data entry clerk would enter the details from the lookup table into the main database each time an observation is entered. This poses some problems. Should the place name perchance not be in the lookup table, there would be no easy way to quickly ascertain the co-ordinates. The same would apply if the place name were misspelled on the questionnaire (hopefully a thorough check at district level would have obviated that possibility), or if place names had changed (as they sometimes do in traditional parts of Africa, for example). To make matters worse, one sometimes finds places in the same country and even the same area with the same names, giving rise to the question of which one is referred to on the data form.
The other possibility is to have software which scans the data entries and the lookup table systematically, and then enters the georeferences automatically after cross-matching place names in the disease data table with place names in the lookup table or master file. The same problems noted above for manual georeference entry would apply for such automatic entry.
Both concepts of georeference entry into the database - either at field level or at the computer level, are fraught with problems, and there are no easy solutions. Each country and system designer would have to cope with the issue in the way best suited to their circumstances.
Using the GIS
Nothing much will be said about use of GIS software, and the reader is advised to consult relevant literature on the subject. The prospective user must consult thoroughly before purchasing such software, bearing in mind the following points:
the software must be compatible with the computer system in use. GIS programmes are usually resource-hungry, requiring much disk space, operating memory and fast processors. It may be necessary to upgrade an older computer system in order to accommodate modern GIS programmes.
the GIS software must be compatible with the database system in use. The GIS treats different kinds of data as “layers” and essentially, disease data would be imported into the mapping system as a layer. It must first be ascertained whether the database format in use in the epidemiology unit can be “read” by the GIS; the next is to ensure that the manner in which the georeferences are stored can be read by the GIS. It may be necessary, for example, to convert degrees, minutes and seconds into decimal degrees (where, for example, 30° 30' 30" would be read as 30.5 degrees) before importing the database file into the GIS, or the GIS may do it automatically.
there must be ready maintenance for the GIS software within easy reach, and it should be easy to acquire important layers (roads, weather data, etc) locally. Usually, GIS programmes when purchased, come with a world map containing little more than country borders and the grid references for capital cities. More details have to be purchased, begged, borrowed or stolen elsewhere.
training and other support should also be available. GIS software, while having many advantages, and being important in good epidemiological practice, is notoriously difficult to use. The importance of good training cannot be over-emphasised.
Field staff must be carefully chosen and well-trained. Farmers are usually willing to part with information when they feel that field staff are intelligent, trustworthy and can offer something in return. Field staff should be able to offer advice on matters related to animal health, and should also be able to carry out “first aid” treatments should they encounter sick animals during the course of their rounds. Treatments need not necessarily be for free, but the service should be available. They should also be well-trained in questionnaire usage.
The ability to handle animals well, take blood samples competently and to interact with people are also essential prerequisites.
The same principles apply when using “outsiders” to collect information. There is a strong tendency to make use of Community Animal Health Workers or “barefoot vets” from the private sector as gatherers of information in the field. They are interviewed periodically and all disease information is transferred to questionnaires for computerisation. It should not necessarily be assumed that such people have all the training that they need. They too will have to be trained in certain basics before being used as trusted sources of information.
Basic training for livestock inspectors or community animal health workers
The kind of subjects that might be covered in such training (which must have a strong practical component) would include:
Training should not be conducted such that all subjects are dealt with in one course. The course should be divided into modules, between which the workers will return to the field and apply their knowledge in practice. The teaching should be re-inforced with frequent revisions. Evaluations should be practical rather than in the form of written examinations.
While the training above is aimed at lay persons, it should not be forgotten that veterinarians will also need training is questionnaire use.
Backing up field staff with sufficient supplies is basic to their functioning. Field staff who are undersupplied with their necessities will become frustrated and discouraged and will not function. The following issues must be considered:
regular pay and access to food and clothing supplies if the staff are on the official payroll.
transport in the field. Vehicles and motorcycles will require fuel, regular servicing and a supply of spare parts.
protective clothing. There should be a regular issue of overalls and appropriate footwear.
camping equipment should be available, especially for the rainy season.
a plentiful supply of questionnaires, carbon paper (to make duplicate copies), pens, pencils and clipboards is essential.
basic medicine kits with a few essential stock remedies, needles and syringes.
other supplies, such as serum collection tubes, labels, markers, ear-tags, vaccines, cold boxes, and so on, must be readily available at the appropriate times.
Many veterinarians in senior positions are currently not familiar with modern information systems development, and may have false expectations of information systems. They may even be suspicious of them, or simply discount them.
There are those who treat computers with contempt after having had one or two bad experiences with them, or who simply do not understand them. Others, on the other hand, may be over-enthusiastic, and expect too much. Some may think that installing a computer and some database software will create an information avalanche that will enable easy decision-making overnight. The latter group will suffer the most disillusionment when they discover that implementing an information system is a slow and painstaking process, and that the first information to come out of it is untrustworthy.
Having a computer system will not:
Having a computer system will:
Computer-analysed data form part of the information that goes into veterinary decision-making processes, but only a part. All planning must be done against a background of knowledge of staff capabilities, budgetary constraints; cultural values and traditions, and prevailing government policy. Decisions must be holistic and take all of reality into account, and not just the slice of it that comes from a computer.
Getting a system started
Leaving this section until this late in the manual has been deliberate. Having an overview of what a veterinary surveillance system will entail is essential before trying to design and implement one. Some ground has already been covered in previous pages, but some aspects still need discussion.
Agreeing on the need for systematic data collection and analysis is an essential first step, that will require intense discussion between all stakeholders - management, the epidemiologist, field staff and laboratory staff.
Designing the system, taking into account any system that already exists, is the next step. Crucial decisions must be made regarding data sources, inputs and outputs.
Database design is the next step, together with software customisation and questionnaire design. Several questionnaires may need to be designed according to information sources-visual surveillance, abattoir surveillance, and so on.
The next stage is field testing. It would be most practical to do a few pilot runs in two or three selected areas, completing questionnaires, following the information flow into the computer, and then doing some trial analyses. Shortcomings can be identified and corrected.
Then all veterinarians will have to be informed and trained in implementing the system, and they in turn will need to train their field staff.
The system is then implemented. Data flow and quality will have to be intensely and carefully monitored in order to make as many improvements as possible. It is important that both management and field staff be informed of what is happening every step of the way. If left in the dark, suspicions will develop, and enthusiasm and support will wane.
All in all, it will take one to two years from conception to implementation. Allowing for some hiccups along the way, another year or two for the system to mature enough to produce “research quality” data.
Bear in mind that no information system is ever static. It will continue to evolve and grow as needs change and users become more sophisticated.
It is not the intention here to write a textbook on veterinary epidemiology, but rather to give few pointers. From the point of view of an epidemiologist, having a strong flow of “analysable” data can be very exciting. The trick is to make good use of it. First and foremost, it must be remembered that an information system is a management tool. Using information from a system for research and retrospective studies may be exciting, but day-to-day management is the aim.
Regular preparation of month-by-month incidence graphs of the main diseases can be very informative. Doing such work for one or two years may reveal important seasonal tendencies, particularly as far as vector-borne diseases are concerned, but possibly with others, as well. This may give good indicators as the planning of official strategic controls or regular reminders to farmers.
Longitudinal studies will also be of value in monitoring progress with eradication efforts, deciding when best to enter an OIE pathway, or to change from visual surveillance for a particular disease to sero-surveillance or abattoir surveillance.
Spatial analyses using a GIS are extremely useful for monitoring disease spread, and for noting the association a disease may have with weather conditions, ecological zones or geographic features such as rivers or roads. Areas under particular threat can be noticed early in the development of a disease, enabling the more rational deployment of resources. Planning for staff distribution, transport arrangements and vaccine purchases is facilitated.
Long-term studies of disease data will also reveal cyclical tendencies, association with long-term climate changes, and so on. It may even be possible to build reliable models of some diseases which enable predictions of spread to be made with greater accuracy.
An important fact about disease information systems is the ability to link them with economic data, and make inferences regarding economic impacts of diseases and the costs and benefits of control measures. Computer software can then be harnessed for building reliable “what-if?” scenarios and allowing an intelligent choice of various control options to be made.
The preparation of this manual is testimony of FAO's commitment to assist developing countries with development of their own early warning systems. Via the EMPRES programme, FAO is involved at national, regional and global level with the development of disease early warning systems. The ultimate vision is a global network, linking member countries in an information network that will enable rapid disease reporting, and quick dissemination of information.
This network will be a part of the Global Early Warning System (GEWS) being established by FAO to cover all possible pests, diseases and natural disasters.
EMPRES is currently involved in the development of a three-tiered information system which will gather, process and disseminate information. It is essentially a computerised system, to be known as the Transboundary Animal Disease Information system (TADInfo). This will consist of three different software modules: TADInfo National, TADInfo Regional and TADInfo Global.
For countries lacking properly developed epidemiological software, TADInfo National will be available free of charge. Through the well-established TCP system, FAO will be able to assist with information system development and software installation. TADInfo National is designed to feed information upwards to TADInfo Regional, which will be installed at the level of collaborating regional organisations or projects (such as PARC, SADC or PANAFTOSA) or at regional/subregional FAO offices. Where countries already use their own internally developed software, provision can be made for feeding-in of information from these systems.
Finally, the regional TADInfo modules will feed information to the global module, located in FAO headquarters.
Basically, the functions of the different modules will be:
at national level: storage and analysis of disease information to facilitate local decision-making.
at regional level: regional early warning, regional support and co-ordination.
at global level: risk modelling, trend monitoring and global early warning.
FAO will also take the initiative of organising regional workshops for veterinary epidemiologists to share and disseminate information on disease surveillance.
At the Annual OIE General Session, held in Paris, France in May 1998, FAO was given the mandate, along with OIE and WHO, to build a global information system for disease early warning. This resolution (no. XIII of the 66th General Session) supports an earlier mandate from the 1996 World Food Summit.
The FAO is fully committed to this ideal, and will continue to work towards it via:
Software (TADInfo) development;
In-country support in the form of TCPs;
The EMPRES Livestock Website on the Internet;
and electronic mail discussion groups.
Interested CVOs and national epidemiologists should contact their nearest FAO office to enquire about the ways in which FAO can assist with the building of national and regional information systems.
Example of Report Form for use by Community Animal Workers/Animal Health Assistants
Name: District: Year: Month:
Example of reporting form (passive surveillance) for use by veterinarians and technical-level field staff
Example Disease Report Form
One form is to be completed for each focus/incident of disease reported. The questionnaire is to be completed clearly and legibly, and the shaded areas must be filled in.
|Province/Region (4-letter code)||District (6-letter code)|
|Nature of Diagnosis||Suspected||Clinical||Smear||PM||Laboratory|
(mark the correct word)
|NUMBER Cases (total affected)||female||juvenile||mixed||trad|
|NUMBER At Risk||?||all||other|
|MEASURES ADOPTED||MAIN CLIN. SIGNS||MAIN PM LESIONS||EPIDEMIOLOGY|
|Treatment||(source, rate of spread, vectors, reservoirs, sporadic, continuous etc)|
Details of reporting officer:
|Surname, initials:||Position||VET||VET AUX/PARAVET||LAY|
Alternative Specimen Report Form (passive surveillance)
|Reporting Officer||Comments and relevant background information|
|Species affected (check)||Bovine||Ovine||Caprine||Porcine||Other (specify)|
|No. Cases||No. Deaths|
|No. at Risk||No. Examined|
Categories most affected
|Age category (check)||neonate||juvenile||subadult||adult||all||unknown|
|Sex category (check)||male||female||neutered||both||unknown|
Signs and lesions observed
|Samples sent to (name lab)|
|Date of submission||Type of sample/s|
Details of Diagnosis
|Basis for diagnosis (check)||Rumour||Clinical history||Clinical signs||Blood smear||Laboratory test|
Example of reporting form for use in sero-surveillance
SERO-SURVEILLANCE REPORT FORM
First part for use by field staff Survey ID
Survey officer details
|Province/ Region||District||Locality name|
Details and history of animals sampled
|Age category||Younger than wean||Older than wean||Adults||All|
|Condition of animals||G||M||P||Condition of veld||G||M||P|
|Vaccinations past year||Diseases past year|
|No. of serum samples taken|
|Animals moved in from?||Animals moved to?|
Disease being surveyed
Second part for completion by laboratory staff
Lab. reference number: Technician: Veterinarian:
Any comments to Epidemiology Unit:
Third part for completion by Epidemiology Unit staff
Date entered Data entry clerk Checked by
Name of database table Date checked
Example of reporting form for use at abattoirs (laboratory results entered on the reverse of the form)
ABATTOIR/SLAUGHTER SLAB REPORT FORM
This form is completed when:
Where two different conditions are diagnosed in the same consignment, two separate forms must be completed.
The reverse side of the form is to be filled in when a sample is submitted to a laboratory. the original must accompany the sample to the laboratory, and a duplicate must be sent to Head Office for computerisation.
The reference number refers to any numeric series of your own choice, eg. 002/1996. These numbers must follow one another successively on successive forms, and must not be abattoir consignment number, etc...
The shaded boxes must be filled in.
(1st, 2nd & last)
|Owner of Animals||Locality/Farm Name||Province/Region||District|
|Species||Condition suspected or diagnosed||Differential Diagnosis|
|Indicate whether it was diagnosed:||Ante-Mortem||or||Post-Mortem|
|No. of animals in consignment:||No.affected||Age of affecteds (weeks, months, years, 2t, 3T, all ages)|
|Sex of affected (M = male, F = female, B = both)||Other comments:|
|If you conducted any tests, what were your findings:|
|Please mark with an X if samples to lab:|
Reverse of abattoir and field disease report form:
DETAILS FOR SPECIMEN(S) - LABORATORY
|Number & type of specimen(s):||Time collected (only for sensitive organisms):||In case of RABIES, was there any human contact:|
|If Yes, how many people affected|
|Examination(s) requested:||Owner's name on reverse of this form:|
|Owner's tel. number:|
|ID and reference number if from satellite laboratory:||If not official - does lab have permission to do extra test at owner's|
|cost: [Yes] [No]|
[Post price list]
|FOR LABORATORY USE ONLY|
|Date samples received||Lab number||Number copies required||Distribution|
|Sections||micro/path||path||chem tox||Referral centres (specify)||Add. examination decided upon|
|Is this a follow-up report||Yes||No||Another report to follow||Yes||No|
|LAB RESULT (FREE FORMAT)|
|LABORATORY COMMENT TO FIELD VET:|
|Blood smear:||Respiratory system:|
|Eggs per gram:||Central nervous system:|
|Urogenital system:||Aetiological diagnosis:|
|Circulatory system:||Differential diagnosis:|
|Construct sampling frame & determine sample size (no. of herds to be sampled)|
|Decide on issues such as stratification, sampling methodology (SRS, PPS, use of locality list vs map|
|Determine the localities from which the herds are to be drawn|
|Training of staff, planning and organising of survey|
|Randomly select herds (usually in situ)|
|Randomly select animals|
|Processing, data entry & initial analysis|
Sampling methods in randomised surveys
Conceptually, there are a number of steps involved in a disease survey. They are laid out below in a stepwise manner and will be explained in more detail later.
1. Sampling Frame
The construction of a sampling frame is the first step in the planning of a survey, and will be a reflection of “what question must this survey answer?” The sampling frame is a list of the “objects” in the sampling universe, giving the epidemiologist a catalogue or inventory from which to select the villages and herds which need to be sampled.
This is usually one of the most difficult and demanding parts of any survey exercise, as a surveillance effort will easily founder on a poorly designed sampling frame. If the sampling universe from which a sample is to be drawn is “all herds of cattle in district A”, then it essential to have either a list of all the herds, or at least a list of all the places at which herds of cattle may be found. The completeness of the frame will obviously influence the representativeness of the sample.
Sampling frames can be drawn up from farmer address lists kept by extension officers, from population census records, maps of villages/cattle posts, or, in more sophisticated areas, telephone directories and post office address lists.
In areas where such information is unavailable, or which are inhabited by pastoral nomads, it is often only possible to make use of a large-scale map: the map is drawn into quadrants, each of which is assigned a number from 1 to n, and then quadrants are selected at random. These quadrants are then physically visited on the ground. This method is obviously far less secure than a proper sampling frame.
2. Sample size determination
Sample size depends on exactly what level of disease prevalence is to be detected, and at what level of confidence (in epidemiological work, a confidence level of 95% is the norm). The following formula may be used for populations>1000:
where n=sample size, =confidence level, d=estimated number of diseased, N=population size
As is now the case in most statistical tasks, the memorisation and manual usage of complex mathematical formulae is becoming virtually redundant, and there are many simple computer software packages available (some of them free of charge) which will perform these calculations with ease.
Where a population of animals or livestock owners is extremely large, or spread over a number of different farming systems or distinct ecozones, it would be advisable to break the population into a series of smaller segments for the purpose of sampling. This process is known as stratification, and each new segment is known as a stratum. It is essential that the population within each stratum is as uniform as possible, thus ensuring more representative sampling. The various strata are treated as individual sampling universes wth their own sampling frames. While stratification effectively increases the overall number of samples taken, it also improves the overall quality of the sampling exercise. However, in order to keep costs to a minumum, it is advisable also to keep the number of strata in a single sampling exercise to a minimum - the criteria used for stratification should be such that no more than three strata are demarcated.
The magnitude of most sampling exercises is usually such that a multistage sampling strategy is adopted. It is usually impossible to create, from the very first, a complete list of all cattle herds in a region and then choose them randomly. To compile a sampling frame of this size is usually too resource-intensive.
Sampling is then undertaken in two or more stages, for example:
Draw up a list of places where herds are known to be kept (a primary sampling frame). Then choose the places to be sampled at random.
Draw up a list of herds kept only at each chosen locality (secondary sampling frame). Then choose the herds at random.
Localities can be chosen (depending on the quality of the primary sampling frame) by one of three methods: Probabilities Proportional to Size (PPS); Constant Sampling Fraction (CSF) or from a map (by grid). These methods will be dealt with in turn below.
Let us take an example where it is decided to sample a total of 15 herds. The designers determine that a minimum of 5 herds should be sampled in each locality. The sampling frame is of a high quality; ie. all the villages in the area are known, and the number of herds at each village is also known. The numbers of herds together with their villages are tabulated and cumulative totals worked out as shown below.
Three random numbers are chosen, and the villages within which random numbers fall are the ones where the sampling exercise is carried out. A detailed sampling frame is then drawn up for each of the villages, and the primary sampling units (herds) chosen by simple random sampling.
In the first stage, a number of posts would be chosen randomly, and for the second stage, a predetermined percentage of households at the selected posts would be chosen at random - eg.20% of all households at the settlement. This method is ideal for transhumant peoples, where one cannot say with certainty beforehand how many people will be at each post. Upon arrival at a settlement, the enumerator will allocate each household a number and draw a number of them (equal to 20%) from a hat.
It is assumed that the (a) the total number of animals in the area is known and that (b) at least the various posts/settlements are known. In other words, a rudimentary sampling frame containing all known inhabited settlements has been drawn up. If prior calculation shows that the total number of animals to be sampled is roughly equal to, eg.20% of the animals in the population, then 20% of each randomly sampled herd is bled. This method is not ideal, and is normally used only when there is uncertainty as to distribution of herds (eg. in situations of transhumance).
Random Geographic Co-ordinates
This method is used in areas populated by nomads, or where no sampling frame is available at all. It is the least reliable of the three methods described here. A map of the area to be surveyed is divided into squares of suitable size (5-10km blocks in real terms) and each square is allocated a number. The requisite number of squares (equivalent to the number of villages in the examples given above) are chosen using a random number table/random number generator. It is assumed that the number of herds to sampled at each point has already been decided; on the ground, this number of herds will be sampled within 5km radius of the physical centre of the square. An obvious disadvantage of this method is that travel might be undertaken to a chosen locality only to find the area uninhabited; in this case, a replacement locality is chosen at random as described above.
5. Choosing herds/animals once “on the ground”
The localities to be sampled are chosen by one of the methods described above. Actually choosing which herds will provide the candidate animals is a matter of cataloguing the herds at the sampling point and assigning each herd a number. Herds are then chosen using random numbers or literally drawing their assigned numbers from a hat.
Choosing animals to bleed is also not a major issue. It may be that the sampling frame calls for all animals in a particular age category; in which case no choice is involved. Where animals need to be chosen, those eligible for sampling can be assigned numbers and chosen randomly, or Linear Systematic sampling (LSS) can be used. With LSS, it is not necessary to number each individual animal, but the total must be known. The “sampling interval” (k) is calculated using the formula :
k = N/n where N= the total number of animals available and n = the number required in the sample.
A random number (r) falling between one and k is then chosen. The sample then consists of every kth animal. The animals are driven into a crush, and the first one sampled is r, then r+k, r+2k etc.
Control and eradication of major epidemic diseases of livestock in any area of the world requires a coordinated regional approach. Countries in a region which have lower socio-economic standards are liable to fall behind their more developed neighbours in the area of animal disease control.
The majority of the population in most developing countries are involved in smallholder agriculture. This group also represents one of the poorest sectors of society. In addition to food and draft power, livestock represent an important savings system within the village economy. Livestock diseases cause enormous losses through death and decreased production. Strengthening the veterinary services of countries is an efficient, well targeted approach to improving the livelihood of the rural poor on a national, regional and global basis.
The key to a coordinated regional disease control program is the free exchange of reliable, compatible disease information between countries, and the harmonisation of reporting and disease control procedures. Encouraging a uniform approach to disease reporting and control programs throughout a region will help in control and eradication of diseases.
The objective of integrated control is to use a vertically integrated approach to improving the collection, management and use of animal health information for disease control. It achieves this by addressing weaknesses at every level of the information chain - from livestock owners to regional organisations. The long term aim is to support the development of veterinary services to enable them to successfully and sustainably control and eradicate major diseases of livestock. Examples of this are rinderpest in Africa and FMD in South America and South East Asia.
The rationale behind this approach is based on the current difficulty experienced by poorer countries in controlling emergency animal diseases and the presence of other possibly higher priority issues for government. Priority should be given to the establishment of successful, internally funded sustainable disease control programs within each country of a region. Such programs will produce significant benefits, not only through the control of the target disease, but also through establishing the infrastructure and skill-base amongst relevant staff. These skills can then be used to tackle other diseases of importance to the international community, which may be difficult and expensive to control. Commencing a disease eradication project without providing prior experience in the successful control of livestock diseases is likely to result in an expensive failure.
Objectives directed at ensuring an emergency response capability
The following are reasonably generic objectives which need to be met if a country is to have the capability to successfully deal with emergency animal diseases. By clearly defining objectives, the necessary outputs activities required to ensure an emergency response capability can be more effectively identified.
Objective 1: Improve the collection of animal health information
Objective 2: Ensure Sustainable Laboratory Support
Objective 3: Implement Information Management System
Objective 4: Establish National and Regional Analysis and Reporting Systems
Objective 5: Control Program Formulation, Implementation and Monitoring
Outputs and Activities
In this section, the required outputs and activities are briefly summarised for each objective.
Objective 1: Improve the collection of animal health information
Improved disease reporting and specimen submission by livestock owners, village veterinary staff, district and provincial veterinary officers.
Passive surveillance gathers information through disease outbreak reporting and the submission of diagnostic specimens to veterinary laboratories. The level of under-reporting in passive surveillance systems means that the data collected is unrepresentative and unable to be used for developing control strategies, or addressing disease outbreaks. In countries where laboratory facilities are limited, the key personnel in the chain of reporting are the district officers who are responsible for submitting primary disease reports.
Training of provincial and district veterinary staff
Highly targeted training to district staff should be provided to equip them with the skills to carry out a disease outbreak investigation, recognise key transboundary diseases, restrain animals, collect specimens, collect disease history information and submit specimens to the laboratory. Training should be provided in a two-stage process, initially involving provincial staff. They should receive detailed training to develop the necessary technical skills as well as appropriate methods to pass on those skills to district staff (the “train the trainer” concept). The provincial staff can then be responsible for training district staff according to a plan which can be audited to ensure objectives are met. It is very important to provide basic training for para-veterinary staff, as they are often the only ones to have regular contact with farmers.
Provision of specimen collection kits
District staff need to be issued with basic sample collection kits (including restraint equipment, specimen collection equipment, transport containers and media, disinfectant etc), and pads of recording and laboratory submission forms.
Provision of ongoing support for district staff
After being trained, district staff need to be supported by provincial staff in carrying out field disease outbreak investigations, as they arise.
Monitoring staff activity
The activity of provincial and district staff in disease outbreak investigations, disease reporting and submission of specimens should be monitored with the assistance of an information management system of some type, preferably computerised. Provinces and districts that appear to be failing to report disease and submit specimens can then be identified, problems and constraints investigated, and any necessary further training and support provided. Information management systems need to reflect both the business of the animal health authorities and existing information flows. Examples are the systems in place in Namibia, Laos and the Philippines (for FMD only).
Establishment of specimen transport and feedback systems
Systems for the effective transport of specimens to the laboratory, feedback to province, district and villages, and the sustainable maintenance of restraint equipment and submission materials should be in place and strengthened where required. To assist laboratories with feedback to the districts and villages, simple information sheets on disease control and prevention are needed for specific common diseases.
Continuing veterinary education for provincial and district staff
Provincial staff should be invited to attend periodic ‘refresher courses’ on commonly encountered diseases or issues, run by national counterparts, to strengthen field diagnostic skills and update knowledge on locally appropriate control or prevention options. They should be provided with training materials and required to present the same information to district staff at regular provincial meetings.
Public awareness campaigns for livestock owners and traders
As these persons are the first link in early warning of TADs, it is very important to devote considerable resources to public education and awareness. Encouraging the support of livestock owners in disease reporting can be achieved through the development of appropriate public awareness and education materials in the local language and at a level consistent with local education levels.
Establish links with village-level agricultural projects
Links should be established with agricultural development projects working at the village level (e.g. NGOs) to include this message in their work and distribute educational material.
Active surveillance to collect reliable, population-based information on key diseases, and to monitor the progress of disease control/eradication campaigns is an extremely useful activity to complement passive surveillance.
Comprehensive survey techniques have been developed by various international aid projects and a variety of survey manuals, software, and training course syllabi are available. These materials should be used as a basis to implement institutionalised, sustainable surveillance systems tailored to the needs of specific countries and the disease problems they are likely to encounter.
Training of provincial and district staff in survey techniques
Existing materials and techniques can be used to provide training to provincial and district staff in basic epidemiology, survey and random selection techniques.
Implementing field disease surveillance
Training should be centred around actual field surveillance exercises to give all staff an opportunity to practice and perfect their skills.
Development of a coordinated active surveillance program
In collaboration with national staff, a coordinated program of active surveillance should be established, targeting priority diseases. This would initially aim to provide baseline measures of disease incidence and antibody prevalence with a national coverage. To ensure the sustainability of the techniques, previously trained staff should be called upon to conduct regular surveys on priority issues as part of their normal responsibilities.
Use active surveillance to support disease control programs
Surveillance activities should be heavily emphasised in priority areas to support newly established or modified disease control programs.
Training a range of personnel in reporting and data collection techniques relevant to their responsibilities is vital to ensuring that ancillary data which supports disease control is properly recorded, analysed and reported.
Train personnel in the collection of ancillary data
A range of staff should receive basic training in the use of reporting forms for the collection of information of a variety of types. These include:
Train national-level staff in the collection of socio-economic data
National-level staff should receive training in the conduct of surveys to collect socio-economic data. This should be combined with active surveillance activities, and be aimed to support priority setting and control program formulation activities.
Objective 2: Ensure sustainable laboratory support
A vital component of having the capability to manage emergency diseases is to have competent laboratory support. However, having a laboratory capability will only be fully effective where laboratory services are fully integrated into the overall surveillance and disease control programs. Where possible, training and program planning should include both field and laboratory personnel. This has been the approach in the Philippines in the FMD eradication program and has led to a greater understanding of mutual problems and more effective use of laboratory resources.
Effective laboratory support for field activities
Provision of specimens to national laboratories and international reference laboratories
Without a regular flow of specimens to these laboratories from the field, staff will not have an opportunity to perfect their skills, and little used tests may become unreliable. All field surveillance activities should feed specimens into national laboratories in a planned way which accounts for the capacity of the laboratories. Thus, the field surveillance effort must be designed so that it is fully integrated with laboratory capacity. National laboratories should also be made aware of the need to regularly send field isolates to reference laboratories for characterisation and serotyping.
Provision of diagnostic reagents to laboratories
In addition to maintaining staff skills, the sustainability of the diagnostic laboratories depends on a reliable supply of diagnostic reagents. A guaranteed supply of essential reagents and laboratory disposables must be available to support surveillance activities.
Development of systems for the sustainable local production of key diagnostic reagents
In some instances, it may be possible to produce some reagents locally within the laboratory. Others may have to be bought or imported. Reagents suitable for local production should be identified, and staff trained and systems set up for their sustainable production. The importance of standardising reagents and tests to international norms should be recognised.
Train laboratory staff in new diagnostic techniques as appropriate
Serological techniques such as ELISA have been developed to assist the diagnosis of many priority diseases. These require the provision of appropriate reagents and materials, and an ongoing access to training. For some priority diseases, additional simple and rapid tests exist. Staff should be trained in these techniques to broaden the range of diseases that can be handled at laboratories.
Objective 3: Implement information management system
The main features of a useful information management system are that it is simple to use; inexpensive and quick to develop; able to handle a wide range of information related to animal health; can be modified locally to meet changing needs; and provides specialised epidemiological analytical procedures. With modern technology and a systematic approach to the analysis of business needs, systems can be quickly developed and implemented which meet the needs of all levels of animal health personnel and which are tailored to operate within the organisational structure of the country in question. A wide range of report formats can be incorporated including automated disease mapping where base maps are available.
Efficient management of animal health information
User needs analysis and database designs
Experience suggests that databases need to be specifically designed to work within the administrative and organisational structure of a particular country. However, if a modularised system can be developed, tailoring the system to meet the needs of specific countries is made much easier. Core modules are likely to require minor modification for use in different countries, but there are likely to be country-specific needs which will require greater effort to develop.
Development and translation of users' manuals
Comprehensive users manuals should be developed and translated for each system. A core manual prepared in English could provide the basis for all manuals. This could be modified to take into account country differences, and then translated. Bilingual versions should be available in each country.
Training of staff in the use of systems
Inputs into information systems come from many different areas within the veterinary services of a country. Staff involved should all be trained in their roles and contributions to the system. A small team of national staff should be trained in the detailed operation of the system, including data entry staff, as well as national epidemiology staff involved in data analysis and reporting.
Where required, systems should be implemented in a phased fashion, running in parallel with existing systems for some time, and gradually introducing more of the modules until the whole system is up and running. Where possible, implementation should be at least at provincial and national level.
Objective 4: Establish National and Regional Analysis and Reporting Systems
Because of the ease of movement of livestock and diseases from one country to another, disease control requires a regional approach. National staff need to develop the skills to analyse data, and then be able to use the results to establish cooperative regional approaches to disease control.
Improved ability of national staff to analyse and interpret animal health information
Training of national staff in data analysis
National epidemiologists need to work alongside experienced personnel in the analysis of national disease information. On-the-job training should be provided through participating in the work. Sub-regional workshops for national staff of identified core countries should be conducted to provide more structured training in epidemiological data analysis and disease control program formulation.
English, French, and Spanish are the most common languages for animal health information systems. It is important that staff be competent in the language relevant for the region. If necessary, specific language training should be provided for staff. This will increase their ability to participate fully in regional meetings, prepare publications and reports, access the international literature and use many software programs.
Improved regional communication and coordination of disease control activities
Regional activities should be conducted in close collaboration with an international facilitator.
Establishment of regional disease outbreak database
A regional geo-referenced database should be established to facilitate collation and analysis of animal health information. Each contributing country would then have access to up-to-date information about the disease situation of their neighbours, to help prevent the cross-border spread of diseases.
Regional data analysis
Regional coordinators will be required to cooperate with national staff and regional organisations to carry out regional analysis of data, and establish an effective regional disease reporting system. Data collected through passive surveillance, active surveillance, socio-economic studies, and livestock movement records should be analysed and reported.
Country coordinators' meetings
Where regional programs are instituted, country coordinators and any counterparts need to remain in close consultation. In addition to correspondence, there should also be regular meetings, say every six months, rotating through each of the regional countries. These meetings will serve to share the experience in different countries, allow a better sharing of resources, and ensure continuing compatibility between systems.
Economic group member coordination meetings
Links should be established with formal trading group committees to institutionalise regional information sharing and disease control activities. In addition to working with formal committees, a series of technical meetings could be called to:
Staff from regional participating countries could be involved in short-term attachments or exchanges to the veterinary services in different countries in the region. Epidemiologists from the more developed countries would have an opportunity to learn of the difficulties facing their less developed neighbours and have a chance to share their expertise. Less developed country staff would have a chance to examine the operation of information systems and disease control activities in the more developed countries.
If formulated as a series of regional projects, a newsletter may be useful to be distributed to all countries in the region, focusing on practical and technical aspects of disease control, and providing an informal channel of communication. The newsletter should also be a vehicle for exchange of information on disease investigations.
Objective 5: Control Program Formulation, Implementation and Monitoring
In collaboration with expert consultants where required, national staff would use the data collected during field activities as the basis for the formulation and analysis of disease control options, and identification of preferred programs.
National disease control strategies
Subregional technical training of national epidemiologists from participating countries would include both analysis of epidemiological data and use of the results for the formulation of disease control strategies.
Formulate disease control options
In collaboration with expert consultants, national staff would formulate a range of disease control options, identifying priority diseases, and possible appropriate control measures.
Evaluate options to identify optimal strategies
Supported by disease surveillance and socio-economic data, the control options should be evaluated and optimal strategies determined.
Effective disease control activities
Support the implementation of identified priority disease control activities
Within the structure of the national veterinary services, existing control activities may be modified, or new activities implemented, and provincial and district staff would use active surveillance techniques to monitor the effectiveness of these programs.
The implementation of a program such as that described in Section 3 above would require development of a series regional of business plans and involve a reasonable number of personnel. In addition to an overall program manager, regional coordinators would be required as well as full-time in-country advisers. Regional coordinators should be veterinarians with experience in disease control in the specific region as well as project management experience. Each regional coordinator would be based in a convenient capital city within the region and would be responsible for overall project management, country coordinator support, and regional activities.
One country coordinator should be placed in each of the participating countries within a region, and be responsible for day to day running of project activities. The country coordinators should be veterinarians with an interest in epidemiology, good interpersonal and management skills, and cultural sensitivity. In particular they should have well developed training skills.
One key staff member should be identified within each participating country in a region to work as a counterpart to the country coordinator. The country coordinators' positions could be phased out after the first two years, with counterparts taking over full responsibilities.
Short term expert consultants may be required for a variety of tasks, including computer programming, development of publicity materials, economics, laboratory diagnostic techniques etc.
Implementation of a program such as that described in would require the cooperation of a number of agencies. Potential donor agencies would need to be identified during the planning phase and the program developed with their collaboration
Main Budget Items
The main items which may need to be budgeted for include:
Training and personnel costs are therefore likely to make up a significant part of the budget.
Some specific items which will need to be considered in developing budgets include:
Activities and Outputs by Administrative Level
|Regional||Participation in regional meetings|
Establishment of Standard Rules and Definitions for disease reporting and control
Forging links with appropriate regional bodies and committees
Institutionalising regional cooperation
Regional analysis of disease, and livestock movement patterns
Establishment of regional disease outbreak databases
|Subregional||Harmonised disease surveillance and reporting systems|
Sharing of disease information for improved ability to prevent cross-border movement of animals
|National||Improved passive surveillance systems|
Establishment of effective active surveillance systems
Improved understanding of priority diseases
Short-term attachments and exchanges between countries
Development of structured disease control programs for priority diseases
Support and development of laboratory capabilities
|Provincial||Training in active and passive surveillance|
Improved skills in disease outbreak investigation and response
|District||Training in active and passive surveillance|
Improved reporting of livestock demographics
Improved reporting of livestock movement
|Village||Increased awareness of the need to report and control diseases|
Project Activity Schedule
The GANNT chart below outlines how a project to deliver the required outputs might be delivered.
|Year||Year 1||Year 2||Year 3|
|Objective 1: Improve the collection of animal health information|
|Output 1.1: Passive Surveillance|
|Activity 1.1.1: Training of Provincial and district Veterinary staff|
|Activity 1.1.2: Provision of specimen Collection Kits|
|Activity 1.1.3: Provision of ongoing support for district staff|
|Activity 1.1.4: Monitoring staff activity|
|Activity 1.1.5: Establishment of specimen transport and feedback systems|
|Activity 1.1.6: Continuing veterinary education for provincial and district staff|
|Activity 1.1.7: Public awareness campaigns for livestock owners|
|Activity 1.1.8: Establish links with village level agricultural projects|
|Output 1.2: Active Surveillance|
|Activity 1.2.1: Training of provincial and district staff in survey techniques|
|Activity 1.2.2: Implementing field disease surveillance|
|Activity 1.2.3: Development of a coordinated active surveillance programs|
|Activity 1.2.4: Use active surveillance to support disease control programs|
|Output 1.3: Ancillary Data|
|Activity 1.3.1: Train veterinary staff in the collection of ancillary data|
|Output 1.4: Socio-economic Data|
|Activity 1.4.1: Train national staff in the collection of socio-economic data|
|Objective 2: Ensure Sustainable Laboratory Support|
|Output 2.1: Effective Laboratory support for field activities|
|Activity 2.1.1: Provision of specimens to national laboratories|
|Activity 2.1.2: Provision of diagnostic reagents to laboratories|
|Activity 2.1.3: Development of systems for the sustainable local production of key diagnostic reagents|
|Activity 2.1.4: Train laboratory staff in new diagnostic techniques as appropriate|
|Objective 3: Implement Information Management System|
|Output 3.1: Efficient management of animal health information|
|Activity 3.1.1: User needs analysis and database design|
|Activity 3.1.2: Development and translation of users' manuals|
|Activity 3.1.3: Training of staff in the use of the system|
|Activity 3.1.4: Phased implementation|
|Objective 4: Establish National and Regional Analysis and Reporting System|
|Output 4.1: Improved ability of national staff to analyse and interpret animal health information|
|Activity 4.1.1: Training of national staff and data analysis|
|Activity 4.1.2: English language training|
|Output 4.2: Improved regional communication and coordination of disease control activities|
|Activity 4.2.1: Establishment of regional disease outbreak database|
|Activity 4.2.2: regional data analysis|
|Activity 4.2.3: Country managers meetings|
|Activity 4.2.4: Regional member coordination meetings|
|Activity 4.2.5: Short term attachments|
|Activity 4.2.6: News letter|
|Objective 5: Control Program Formulation,|
|Implementation and Monitoring|
|Output 5.1: National disease control strategies|
|Activity 5.1.1: Training|
|Activity 5.1.2: Formulate disease control options|
|Activity 5.1.3: Evaluate options to identify optimal strategies|
|Output 5.2: Effective disease control activities|
|Activity 5.2.1: Support the implementation of identified priority disease control activities|
Anon Using dBase IV Borland, 1993
Casley, D & Lury, D Data Collection in Developing Countries ELBS(1987)
IAEA/FAO Recommended procedures for disease and serological surveillance as part of the Global Rinderpest Eradication Programme (GREP) IAEA (1994)
Last, J(Ed) A Dictionary of Epidemiology Oxford University Press (1995)
OIE Epidemiological Information Systems OIE Revue Sci et Tech March 1991
Pfeiffer, D Veterinary Epidemiology-An Introduction Massey University lecture notes (1998)
Poate, C.D. & Daplyn, P.F. Data for Agrarian Development Cambridge University Press (1993)
Schwabe, C et al Epidemiology in Veterinary Practice Lea and Febiger (1977)
Thrusfield, M Veterinary Epidemiology (2nd Ed) Blackwell Science (1995)
Wonnacott R, & Wonnacott, T Statistics - discovering its power John Wiley & Sons (1982)
FAO ANIMAL HEALTH MANUALS
Availability: October 1999
The FAO Animal Health Manuals are available through FAO Sales Agents or directly from the Sales and Marketing Group, FAO, Viale delle Terme di Caracalla, 00100 Rome, Italy; fax (+39) 06 5705 3360; e-mail: firstname.lastname@example.org
|Sales and Marketing Group Information Division, FAO|
Viale delle Terme di Caracalla, 00100 Rome, Italy
Tel.: +39 06 57051-Fax: +39 06 5705 3360
|WHERE TO PURCHASE FAO PUBLICATIONS LOCALLY|
POINTS DE VENTE DES PUBLICATIONS DE LA FAO
PUNTOS DE VENTA DE PUBLICACIONES DE LA FAO
Empresa Nacional do Disco e de Publicações, ENDIPU-U.E.E.
Rua Cirilo da Conceição Silva, № 7
C.P. № 1314-C, Luanda
Pasteur 743, 1028 Buenos Aires
World Publications S.A.
Av. Córdoba 1877, 1120 Buenos Aires
Tel./Fax: +5411 48158156
PO Box 404, Abbotsford, Vic. 3067
Tel.: 61 3 9417 5361
Fax: 61 3 9419 7154
Gerold Buch & Co.
Weihburggasse 26, 1010 Vienna
Association of Development
Agencies in Bangladesh
House No. 1/3, Block F
Lalmatia, Dhaka 1207
M.J. De Lannoy
202, avenue du Roi, B-1060 Bruxelles
Los Amigos del Libro
Av. Heroínas 311, Casilla 450
Mercado 1315, La Paz
Botsalo Books (Pty) Ltd
PO Box 1532, Gaborone
Fundação Getúlio Vargas
Praia do Botafogo 190, C.P.9052
Rio de Janeiro
Núcleo Editora da Universidade
Rua Miguel de Frias 9
220-000 Rio de Janeiro
Fundação da Universidade Federal do Paraná-FUNPAR
Rua Alfredo Bufrem 140,30o andar
Centre Africain de Diffusion et Dévelopement Social
B.P. 7317, Douala Bassa
Tél.: + 237 43 37 83
Télécopie: +237 42 77 03
5369 chemin Canotek Road, Unit 1
Ottawa, Ontario K1J 9J3
Tel.: +1 613 745 2665
Fax: +1 613 745 7660
Librería - Oficina Regional, FAO
c/o FAO, Oficina Regional para América
Latina y el Caribe (RLC)
Avda. Dag Hammarskjold, 3241
Tel.: + 56 2 33 72 314
Universitaria Textolibros Ltda.
Avda. L. Bernardo O'Higgins 1050
China National Publications
Import & Export Corporation
16 Gongti East Road, Beijing 100020
Tel.: +86 10 6506 3070
Fax: +86 10 6506 3101
Calle 72 № 13-23 Piso 3
Edificio Nueva Granada
Santafé de Bogotá
Office national des librairies populaires
B.P. 577, Brazzaville
• COSTA RICA
Librería Lehmann S.A.
Av. Central, Apartado 10011
1000 San José
Coalición Costarricense de Iniciativas de Desarrollo
Apartado 7170, 1000 San José
• CÔTE D'IVOIRE
04 B.P. 541, Abidjan 04
Tél.: +225 22 20 55
Télécopie: + 225 21 72 62
Empresa de Comercio Exterior
Obispo 461, Apartado 605, La Habana
• CZECH REPUBLIC
Artia Pegas Press Ltd
Import of Periodicals
Palác Metro, PO Box 825
Národní 25, 111 21 Praha 1
Ostergate 26 A - Post box 173
DK - 1005 Copenhagen K.
Tel.: +45 331 28570
Fax: +45 331 29387
• REPÚBLICA DOMINICANA
CUESTA - Centro del libro
Av. 27 de Febrero, esq.A. Lincoln
Centro Comercial Nacional
Apartado 1241, Santo Domingo
CEDAF - Centro para el Desarrollo
Agropecuario y Forestal, Inc.
Calle José Amado Soler, 50-Urban.
Apartado Postal, 567-2, Santo Domingo
Tel.: +001 809 544-0616/544-0634/ 565-5603
Fax: +001 809 544-4727/567-6989
Libri Mundi, Librería Internacional
Juan León Mera 851
Apartado Postal 3029, Quito
Universidad Agraria del Ecuador
Centro de Información Agraria
Av. 23 de julio, Apartado 09-01-1248
Murgeón 364 y Ulloa, Quito
The Middle East Readers' Information Centre
2 Baghat Aly Street, Appt. 24
El Masry Tower D
Tel.: +202 3413824: + 202 340388 18
Fax: +202 3419355
Fernando VI 2, 28004 Madrid
Librería de la Generalitat de Catalunya
Rambla dels Estudis 118 (Palau Moja)
Tel.: + 34 93 302 6462
Fax: +34 93 302 1299
Mundi Prensa Libros S.A.
Castello 37, 28001 Madrid
Tel.: +34 914 36 37 00
Fax: +34 915 75 39 98
Sitio Web: www.mundiprensa.com
Mundi Prensa - Barcelona
Consejo de Ciento 391
Tel.: +34 934 88 34 92
Fax: +34 934 87 76 59
PO Box 23, FIN-00371 Helsinki
Tel.: +358 9 121 4416
Fax: +358 9 121 4450
Editions A. Pedone
13, rue Soufflot, 75005 Paris
Lavoisier Tec & Doc
14, rue de Provigny
94236 Cachan Cedex
Site Web: www.lavoisier.fr
Librairie du commerce international
10, avenue d'léna
75783 Paris Cedex 16
Site Web: www.cfce.fr
10, rue Nicolas Flamand
Tél.: +33 1 4278 0578
Télécopie: + 33 1 4278 1472
Alexander Horn Internationale
Tel.: +49 6121 37 42 12
S. Toeche-Mittler GmbH
Tel.: +49 6151 336 65
Fax: +49 6151 314 043
Poppelsdorfer Allee 55
D-53115 Bonn 1
Tel.: +49 228 94 90 20
Fax: +49 228 21 74 92
SEDCO Publishing Ltd
Sedco House, Tabon Street
Off Ring Road Central, North Ridge
PO Box 2051, Accra
Readwide Bookshop Ltd
PO Box 0600 Osu, Accra
Tel.: +233 21 22 1387
Fax: +233 21 66 3347
35 Stournara Str. 10682 Athens
Tel.: +30 1 3302 980
Fax: +30 1 3648254
Guyana National Trading Corporation Ltd.
45-47 Water Street, PO Box 308
Escuela Agrícola Panamericana Librería RTAC
El Zamorano, Apartado 93, Tegucigalpa
Oficina de la Escuela Agrícola
Panamericana en Tegucigalpa
Blvd. Morazán, Apts. Glapson
Apartado 93, Tegucigalpa
PO Box 126, H-1656 Budapest
Tel.: +36 1 256 1672
Fax: +36 1 256 8727
Allied Publisher Ltd
751 Mount Road
Chennai 600 002
Tel.: +91 44 8523938/8523984
Fax: +91 44 852049
EWP Affiliated East-West Press PVT, Ltd
G-1/16, Ansari Road, Darya Gany
New Delhi 110 002
Tel.: +91 11 3264 180
Fax: +91 11 3260 358
Oxford Book and Stationery Co.
New Delhi 110001
Tel.: +91 113315310
Fax: +91 113713275
Periodical Expert Book Agency
G-56, 2nd Floor, Laxmi Nagar
Vikas Marg, Delhi 110092
Tel: + 91 11 2215045/2150534
Fax: + 91 11 2418599
2/72, Nirankari Colony, New Delhi-110009
Tel.: +91 11 725 1283
Fax: +91 11 328 13 15
24/4800, Ansari Road
Darya Ganj, New Delhi-110002
Tel.: +91 11 326 8786
The FAO Bureau, International and Regional Specialized Organizations Affairs
Ministry of Agriculture of the Islamic
Republic of Iran
Keshavarz Bld, M.O.A., 17th floor
Office of Public Work
4-5 Harcourt Road, Dublin 2
PO Box 13056, Tel Aviv 61130
Viale delle Terme di Caracalla
Tel.: +39 06 5705 2313
Fax: +39 06 5705 3360
Libreria Commissionaria Sansoni
Via Duca di Calabria 1/1
Tel.: +39 55 64 8 31
Fax: +39 55 64 12 57
Libreria Scientifica Dott. Lucio de Biasio “Aeiou”
Via Coronelli 6, 201 46 Milano
Far Eastern Booksellers
(Kyokuto Shoten Ltd)
12 Kanda-Jimbocho 2 chome
Chiyoda-ku-PO Box 72
Tel.: +81 3 3265 7531
Fax: +81 3 3265 4656
Maruzen Company Ltd
PO Box 5050
Tokyo International 100-31
Tel.: +81 3 3275 8585
Fax: +81 3 3275 0656
Text Book Centre Ltd
PO Box 47540, Nairobi
Tel.: +254 2 330 342
Fax: +254 2 22 57 79
Inter Africa Book Distribution
Kencom House, Moi Avenue
PO Box 73580, Nairobi
Tel.: +254 2 211 184
Fax: +254 2 22 3 5 70
Mezzanine 1, Loita House, Loita Street
Nairobi. PO Box 68077
Tel.: +254 2 303853
Fax : +254 2 330854
202, avenue du Roi
B-1060, Bruxelles (Belgique)
Centre d'Information et de Documentation Scientifique et Technique
Ministère de la recherche apliquée au développement
B.P. 6224, Tsimbazaza, Antananarivo
Suite 20F Northam House
55 Jalan Sultan Ahmad Shah
Tel.: +60 4 2282169
Fax: +60 4 2281758
Rue Soundiata Keita X 115
B.P. 3243, Bamako
La Librairie Internationale
70, rue T'ssoule
B.P. 302 (RP). Rabat
Tél./Télécopie: 212 7 75 01 83
Librería, Universidad Autónoma de Chapingo
Libros y Editoriales S.A.
Av. Progreso № 202-1o Piso A
Apartado. Postal 18922
Col. Escandón, 11800 México D.F.
Mundi Prensa Mexico, S.A.
Río Pánuco, 141 Col. Cuauhtémoc
C.P. 06500, México, DF
Tel.: +52 5 533 56 58
Fax: +52 5 514 67 99
Roodveldt Import b.v.
1013 HG Amsterdam
Tel.: +31 20 622 80 35
Fax: +31 20 625 54 93
Swets & Zeitlinger b.v.
PO Box 830.2160 Lisse
Heereweg 347 B.2161 CA Lisse
• NEW ZEALAND
PO Box 12418
PO Box 3627.Wellington
Tel.: +64 4 499 1551
Fax: +64 4 499 1972
Costado Este Univ. Centroamericana
Apartado Postal A-221, Managua
University Bookshop (Nigeria) Ltd
University of Ibadan. Ibadan
Mirza Book Agency
PO Box 729. Lahore 3
Editora e Impresora S.R.L.
Caballero 270 c/Mcal Estigarribia
Jirón Apurimac 375, Casilla 4937
Universidad Nacional «Pedro Ruiz Gallo»
Facultad de Agronomia, A.P. 795
International Booksource Center, Inc
1127-AAntipolo St. Barangay Valenzuela
Tel.: +632 8966501/8966505/8966507
Fax: +632 8966497
Krakowskie Przedmiescie 7
Livraria Portugal, Dias e Andrade Ltda.
Rua do Carmo. 70-74
Apartado 2681. 1200 Lisboa Codex
Select Books Pte Ltd
03-15 Tanglin Shopping Centre
19 Tanglin Road.Singapore 1024
Tel.: +65 732 1515
Fax.: +65 736 0855
• SLOVAK REPUBLIC
Institute of Scientific and Technical Information for Agriculture
Samova 9,950 10 Nitra
Tel.: +421 87 522 185
Fax: +421 87 525 275
PO Box 936, Mogadishu
• SOUTH AFRICA
David Philip Publishers (Pty) Ltd
PO Box 23408, Claremont 7735
Tel.: Cape Town +27 21 64 4136
Fax: Cape Town +27 21 64 3358
• SRI LANKA
M.D. Gunasena & Co. Ltd
217 Olcott Mawatha, PO Box 246
Palais des Nations
CH-1211 Genéve 1
Site Web: www.un.org
Van Diermen Editions Techniques
41 Lacuez. CH-1807 Blonzy
Vaco n.v. in Suriname
Domineestraat 26. PO Box 1841
Wennergren Williams AB
PO Box 1305, S-171 25 Solna
Tel.: +46 8 705 9750
Fax: +46 8 27 00 71
c/o Longus Books Import
PO Box 610, S-151 27 Södertälje
Tel.: +46 8 55 09 49 70
Fax: +46 55 01 76 10; E-mail:
Mansion 9. Rajadamnern Avenue, Bangkok
Librairie de Bon Pasteur
B.P. 1164. Lomé
Société tunisienne de diffusion
5, avenue de Carthage, Tunis
100. Yil Mahallesi
34440 Bagcilar, Istanbul
Tel.: +90 212 629 0808
Fax: +90 212 629 4689
Fountain Publishers Ltd
PO Box 488, Kampala
Tel.: +256 41 259 163
Fax: +256 41 251 160
• UNITED KINGDOM
The Stationery Office
51 Nine Elms Lanes
London SW8 5DR
Tel.: +44 171 873 9090 (orders)
+44 171 873 0011 (inquiries)
Fax: +44 171 873 8463
and through The Stationery Office
Electronic products only:
PO Box 3, Omega Road
Alton, Hampshire GU34 2PG
Tel.: +44 1420 86 884
Fax: +44 1420 89 889
Intermediate Technology Bookshop
103-105 Southampton Row
London WC1B 4HH
Tel.: +44 171 436 9761
Fax: +44 171 436 2013
Website: www.oneworld.org/itdg/ publications.html
• UNITED STATES
BERNAN Associates (ex UNIPUB)
411/F Assembly Drive
Toll-free: +1800 274 4447
Fax: +1 800 865 3450
United Nations Publications
Two UN Plaza. Room DC2-853
New York, NY 10017
Tel.: 212-963-8302 or 800-253-9646
UN Bookshop (direct sales)
The United Nation Bookshop
General Assembly Building Room 32
New York, NY 10017
Tel.: +1 212 963 7680
Fax : +1 212 963 4910
Ebsco Subscription Services
PO Box 1943
Birmingham, Al 35201-1943
Tel.: +1 205 991 6600
Fax : +1 205 991 1449
The Faxon Company Inc.
15 Southwest Park
Westwood, MA 02090
Cable: FW Faxon Wood
Librería Agropecuaria S.R.L.
Buenos Aires 335, Casilla 1755
Montevideo C.P. 11000
Fundación La Era Agrícola
Calle 31 Junín Qta coromoto 5-49
Apartado 456, Mérida
Ed.Fudeco, Apartado 254
Tel.: +58 51 538 022
Fax : +58 51 544 394
Universidad Central de Venezuela (UCV)
Libería Universitaria, C.A
Av. 3, entre Calles 29 y 30
No 29-25 Edif. EVA, Mérida
Fax: +58 74 52 0956
Tamanaco Libros Técnicos S.R.L.
Centro Comercial Ciudad Tamanaco
Nivel C-2, Caracas
Tel.: +58 2 261 3344/261 3335
Tecni-ciencia Labros S.A
Apartado Postal: 20.315.1020 Caracas
Tel.: +58 2 782 8698/781 9945
The Book Café
Fofe Avenue, Harare;
61a Fort Street, Bulawayo
Tel.: +263 4 79 31 82
Fax : +263 4 72 62 43