B.D. Perry,* P. Lessard,* R.A.I. Norval,* R. L'Eplattenier,+ T.T. Dolan,* B.E. Grandin* and A.D. Irvin++
* International Laboratory for Research on Animal Diseases
P.O. Box 30709
Nairobi, Kenya+ United Nations Environment Programme
P.O. Box 30552
Nairobi, Kenya++ Overseas Development Administration
Eland House, Stag Place
London SW1E 5DH, England
East Coast fever (ECF) is a major cause of cattle mortality and loss of productivity in the eleven African countries in which it is known to exist. However, its impact varies considerably within these countries. This is due to differing virulence of Theileria parva strains, to differences in abundance and infectivity of the vector tick, Rhipicephalus appendiculatus, to the presence of other tick vectors and to differences in susceptibility to the tick and the parasite of the cattle breeds and types present. The impact of ECF control in a given area will depend on a number of factors. These include the incidence and severity of clinical disease experienced, the abundance of cattle, the role of cattle and their products in society, the relative importance of ECF compared with other diseases and the nutritional and management constraints to livestock production that are present. Before embarking on widespread ECF control programmes, it is thus essential to identify the different circumstances of disease risk prevailing within a target area so that such programmes can be tailored accordingly to permit optimum cost-effectiveness.
SELECTION OF EPIDEMIOLOGICAL PARAMETERS
Our initial approach was to acquire, collate and represent geographically data available from secondary sources (publications, reports) on seven of the major parameters influencing ECF epidemiology in the affected countries of eastern, central and southern Africa. The parameters, listed in Table 1, were chosen on the basis of their importance as determinants of theileriosis epidemiology and of their accessibility in the form of secondary source data. They were not intended to comprise a comprehensive list of the disease determinants. This approach is an extension of studies initiated by Irvin (1987) to collate information on the distribution of R. appendiculatus and T. parva.
Table 1. Epidemiological determinants of ECF chosen for geographic representation
|
Parameter |
Quality |
Data source |
|
Host |
||
|
Cattle |
Known distribution and distribution of major indigenous breeds/types |
Inter-African Bureau of Animal Resources (IBAR) and published literature |
|
Buffalo |
Known distribution |
Published literature |
|
Restrictions to domestic host population |
||
|
Game parks |
Known distribution and perceived level of security (i.e., exclusion of livestock) |
Global Resources Information Database, United Nations Environmental Programme |
|
Tsetse fly |
Known distribution by species |
IBAR |
|
Vector |
||
|
Rhipicephalus appendiculatus |
Recorded distribution |
Published literature |
|
R. duttoni, R. nitens, R. bergeoni and R. zambeziensis |
Probable distribution |
Expert opinion |
|
Ecoclimatic suitability index (E.I.) |
Distribution by index value |
Climate: published literature, FAO, CIAT E.I.: CLIMEX (climate matching model), CSIRO |
|
Disease |
||
|
Clinical theileriosis |
Recorded distribution |
Published literature, government reports |
|
Antibodies to Theileria parva |
Recorded distribution |
Published literature government reports |
Data entry and analysis have been described by Lessard et al. (1988). In summary, data were entered into a computerized geographic information system called ARC/INFO (Environmental Systems Research Institute, Redlands, California 92373, USA). This software programme was run on a Micro Vax 3 mainframe computer at the Global Resources Information Database (GRID) offices of the United Nations Environment Programme (UNEP). Some of the data, such as available cattle distribution maps, were digitized into the programme using a Calcomp 9100 digitizer. Other data, such as climatic variables, were entered in numerical form along with their qualifying latitude and longitude.
RESULTS AND DISCUSSION
Individual maps showing the distribution of each of the disease determinants studied have been produced (Figure 1) and are currently being prepared for publication (Lessard, P., L'Eplattenier, R., Norval, R.A.I., Perry, B.D., Dolan, T.T., Croze, H., Walker, J.B. and Irvin, A.D., in preparation). In addition, overlay maps of two or more parameters are being used to evaluate interactions between determinants and to define disease-risk zones (Figure 2).
Geographic information systems (GIS) provide a valuable method of collating and displaying data on a geographic basis. With accurate data and correct selection of the key parameters, GIS offer an extremely useful tool to the disease-control planner. Although still in the early stages of development, the systems have already allowed a considerable refinement of our understanding of the distribution and potential distribution of R. appendiculatus and the interaction of this tick with livestock. Improvements can be made to the system by improving data quality and increasing the number of determinants/parameters studied.
Improving data quality
The quality of data in the list of parameters studied is not uniform, and inherent differences in data collection techniques mean that there is a limit to potential improvements in quality. Continent-wide databases on cattle populations, for example, are collected on the basis of administrative boundaries, such as district and provincial, submitted by national livestock authorities. In many cases the distribution within the administrative unit may not be uniform, although it appears uniform on the distribution map. Tick distribution data, on the other hand, are often based on site-specific studies and are not necessarily representative of entire administrative regions. Analyses of interactions between two parameters collected under different circumstances are thus limited, but serve to identify areas where further data are required. A further limit to cattle population data is that they do not differentiate cattle breeds and types or management systems. Particularly important is the inclusion of data on the distribution of exotic and grade livestock, which are at greatest risk to ECF. These data are not widely available.
Figure 1. Probable distribution of Rhipicephalus appendiculatus in Africa.
Tick collection data often represent the results of a single survey. Further surveys carried out in districts or countries neighbouring that in which an original survey was made are often conducted many years after the original study. Distribution comparisons may thus be invalidated by marked differences in climate, host availability or acaricide use between surveys. Eventually, data from a given time period will be required from selected sites to validate predicted distribution changes.
For data on all parameters, higher resolution studies at specific sites should be carried out to assess the accuracy and validity of the databases.
Increasing the number of determinants/parameters studied
The list of determinants used in this study is far from exhaustive. Further studies to identify and evaluate the key parameters for geographic representation of ECF risk are warranted. A planned addition to the list is the normalized difference vegetation index (NDVI) acquired from satellite-derived Advanced Very High Resolution Radiometer databases at UNEP and the Regional Centre for Services in Surveying, Mapping and Remote Sensing, in Nairobi. This will provide seasonal and secular data on distribution of favourable habitats for R. appendiculatus as measured on the NDVI scale.
Certain other determinants likely to be important in assessing disease risk do not lend themselves to geographic display as yet, due to inadequate surveillance. Such determinants include infection rates of R. appendiculatus with T. parva and acaricide use policies and practices.
It is planned to supplement existing determinants with socioeconomic parameters in defined regions of the continent so as to assess the ability of these parameters to serve as predictors of disease impact on a geographical basis. Socioeconomic parameters include human population density/land use intensity, land tenure, livestock production system, livestock function (such as traction, milk) and the availability of veterinary services.
REFERENCES
Irvin A.D. (1987). Monitoring patterns of distribution of Rhipicephalus appendiculatus and Theileria parva. In: Sutherst, R.W., ed. Ticks and Tick-Borne Diseases: Proceedings of an International Workshop on the Ecology of Ticks and the Epidemiology of Tick-Borne Diseases. Australian Centre for International Agricultural Research Proceedings No. 17, p. 65.
Lessard, P., Sorensen, M., Norval, R.A.I., Perry, B.D., Dolan, T.T., Burrill, A., L,Eplattenier, R., Grootenhuis, J.G. and Irvin, A.D. (1988). Geographical Information Systems in the estimation of East Coast fever risk to African livestock. Acta Veterinaria Scandinavica 84 (Supplement):234-236.
