author(s)
Tanveer Ahmad Rather, Muneer Ahmad Dar, Mubashir Ali Rather, Raheeqa Razvi, M Maroof Shah and Showket Ahmed Ahanger

abstract
The aim of this study was to estimate the true prevalence of brucellosis in sheep population of Kashmir valley by using Bayesian theorem. Sero-diagnosis of Brucellosis by modified Rose Bengal Plate Test (mRBPT) in ovine population of Kashmir valley was conducted from April, 2014 to March 2018. A total of 13323 sera samples were collected from both farms (9129) and field (4239) and the apparent prevalence was about 20.17%. The field recorded a higher prevalence (23.44%) than the farms (18.64%). Employing the pre-established estimates of sensitivity and specificity of mRBPT, a comparatively true prevalence of the disease was established to be 21.14% by Bayesian logic. Positive predictive value (PPV) and Negative Predictive value (NPV) were recorded as 91% and 96%, respectively. The percentage of false positives and false negatives among healthy and diseased animals were 9% and 4% respectively, signifying about overestimation and underestimation of apparent prevalence by lower sensitivity of mRBPT. Before any control/surveillance strategies are thought out to do away with the economic impact and the hazards of zoonotic infection, a substitution or at least augmenting of serological techniques with molecular diagnostic procedures is a must for reducing the variation in prevalence estimates encountered in studies based on serological diagnostic procedures.