Bovine tuberculosis in Ethiopia: A systematic review and meta-analysis

Abstract

Bovine tuberculosis (bTB) is a known endemic disease of cattle in Ethiopia; however, there is lack of a comprehensive information on the status and distribution of the disease in the country. The objectives of this systematic review and meta-analysis were to provide a pooled prevalence estimate of bTB at a national level, assess the level of in-between variance among study reports and illustrate the spatial distribution pattern in the country. Articles published on bTB from January 2000 to December, 2016 in English language were included in the review. Pubmed, CAB direct, AJOL and Web of Science were the databases used in electronic search. A total of 127 articles were retrieved from online sources, of which 56 articles were selected for data extraction based on the specified inclusion criteria. From these selected published articles, 114 animal level data were extracted for quantitative analysis. A pooled prevalence estimate of bovine tuberculosis in Ethiopia was found to be 5.8% (95% CI: 4.5, 7.5). In a multivariable meta-regression analysis, breed and production system explained 40.9% of the explainable proportion of the in-between study variance computed. The prevalence of bovine tuberculosis in Holstein-Friesians, 21.6% (95% CI: 14.7-30.7), was higher than the prevalence in local zebus 4.1 (95% CI: 3.4-4.9). Cattle kept under intensive and semi-intensive production systems had higher prevalence, 16.6% (95% CI: 12.4-21.6), of bTB than those kept in extensive livestock production system, 4.6 (95% CI: 3.4-6.2). Bovine tuberculosis is widely distributed across major livestock producing regions of Ethiopia. However, no valid data could be retrieved from Benishanul-Gumuz, Harari and Dire Dawa. Data obtained on bTB from Somali and Gambella regional states are also few and further studies are suggested in these regions. In conclusion, this review showed that bTB in cattle in Ethiopia is widespread with high prevalence in intensive and semi-intensive management systsems that keep exotic breeds and their crosses in urban and peri-urban areas. Thus, it is suggested that the design and implementation of bTB control strategies in Ethiopia should prioritize these hotspots in order to reduce the impact of the disease on the growing dairy sector.

Keywords: Bovine tuberculosis; Breed; Cattle husbandry; Ethiopia; Geographic distribution.

Figure 1. Literature selection flow diagram for systematic review of bovine tuberculosis in Ethiopia

Figure 2. Galbraith plot for assessing bovine tuberculosis study reports heterogeneity in Ethiopia

The plot is a scatter plot of standardized effect estimates against inverse standard error. In such graphical assessment points outside the confidence bounds (positioned 2 units above and below the regression line) are study reports which have major contribution to heterogeneity noted. In the absence of heterogeneity we could expect all points to lie within the confidence limits. This is one of the evidence for choosing random effect model as opposed to fixed effect model for pooled estimate made in meta-analysis.

Figure 3. Apparent (A) and logit-prevalence (B) estimate distributions for the animal-level studies of bovine tuberculosis prevalence in Ethiopia

The right skewed histogram (figure 3.A), illustrates distribution pattern of apparent prevalence reports, while figure 3.B is a normalized one, following logit-transformation of each prevalence estimate up on which all the analyses were made.

Figure 4. Distribution and mean prevalence of bovine tuberculosis in districts of Ethiopia

Most of the studies were conducted in Addis Ababa, Amhara, Oromia and Southern Nations and Nationalities Peoples regions while no valid published study was obtained from Benishangul-Gumuz, Harari and Dire Dawa regions. On the other hand, few studies were undertaken in Afar, Gambella, Somali and Tigray regions. Variable animal level prevalence of bTB were recorded in the districts of the regions ranging from 0.8% to 54.6%; the highest prevalence being reported in intensive farms in and around cities while the lowest prevalence being recorded in grazing animals in rural areas.

Figure 5. Funnel plot of logit-prevalence estimate of bovine tuberculosis in Ethiopia

The figure displays each study’s estimated effect plotted against its SE and evaluates the relationship between study results and their precision. The lack of symmetry around the inverted funnel plot illustrates the presence of potential bias due to increasing number of study reports with low prevalence and study reports based on smaller sample size.