A meta-analysis of livestock brucellosis prevalence in Indonesia.

Abstract

Indonesia has a long history of livestock brucellosis, but the overall pooled prevalence remains unclear. This study aims to determine the pooled estimated prevalence of livestock brucellosis in Indonesia using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. Five databases were searched and screened using predefined inclusion and exclusion criteria. Data from included studies were extracted and analyzed using a random effects model in R 4.3.0 for pooled estimated prevalence, subgroup meta-analysis, and meta-regression. Publication bias and sensitivity tests were conducted using Egger's test, funnel plot, trim and fill plot, and leave-one-out. The screening process identified 46 included studies, representing 47,057 samples for brucellosis testing. The pooled estimated prevalence for livestock brucellosis was 3.25% (95% CI, 1.81%-5.78%) with high heterogeneity (Q = 2130.91, p = 0, I2=98%). Subgroup meta-analysis indicated no significant difference in the prevalence of livestock brucellosis across the main islands in Indonesia (p = 0.44) and across provinces in Sulawesi Island (p = 0.83), but significant differences were found among provinces in Java (p < 0.01). The subgroup meta-analysis based on animal type showed no significant difference between cattle, small ruminants, and pig brucellosis estimated prevalence (p = 0.26). Between serological tests, no significant difference was found (p = 0.77). Meta-regression showed no significant difference in brucellosis prevalence from 1988-2023. Egger's test and funnel plot showed publication bias. Trim and fill test indicated 21 studies should be added. As most studies were conducted in Java and Sulawesi Islands, caution should be exercised in interpreting the results, emphasizing the necessity of increasing the study of brucellosis in other regions.