Estimation of foot-and-mouth disease virus sero-prevalence rates using novel computational approach for the susceptible bovine population in India during the period 2008–2021

Figure 1. Schematic diagram of steps undertaken in this study. A two-stage sampling design was followed for national-level surveillance of FMDV in India. At stage I villages are randomly selected from the considered states and in stage II animals are selected randomly from the selected villages (at stage I) for serological test through NSP-ELISA. After obtaining the sample-level DIVA test results data through NSP-ELISA, then such data was given input to the developed method and tool to compute various state and national-level measures for FMDV sero-surveillance.

Figure 2. FMDSeroSurv web-application server. (A) Home page of the web-application. (B) Run page of the application. Here, the user will have to upload the data in .txt or .csv file for the analysis. (C) Output page represents the state-level estimate of sero-prevalence rates and related measures. (D) Output page represents the national-level estimate of sero-prevalence rates and related measures.

Figure 3. State-wise distribution of estimated sero-prevalence rates from the year 2008 to 2021. The estimated values of sero-prevalence rates are shown for various states covered during the year 2008 (A)–2021 (N). Higher the intensity of the red color means higher NSP sero-prevalence and vice-versa. The states shown in grey color are not covered in the sero-surveillance activities. A higher sero-prevalence estimated rate indicates more natural infection/virus in circulation of FMDV and vice-versa.

Figure 4. Distribution of predicted total DIVA positive animals during the year 2008–2021. The state-wise predicted total number of DIVA positive animals are shown for the year 2008 (A) to 2021 (N). Higher the intensity of red color indicates higher number of DIVA positive animals and vice-versa. The states shown in grey color are not covered in the sero-surveillance activities.

Figure 5. National-level (India) estimation of FMDV NSP sero-prevalence. (A) Year-wise distribution of estimated sero-prevalence rates and their computed confidence intervals, standard and margin of errors are shown. X-axis represents the years. Y1-axis represents the estimated sero-prevalence rates. Y2-axis represents the estimated co-efficient of variations. Grey color indicates the 95% confidence region of the national-level sero-prevalence estimates. Red color line represents upper and lower bound of 95% confidence intervals. The blue and red color bars indicate the standard and marginal errors of the estimates respectively. The green color (cross-squares) boxes show the co-efficient of variation of the national-level sero-prevalence estimates. (B) Distribution of the field FMD outbreaks during the year 2008–2021. X-axis: Years; Y-axis: FMD outbreaks. Red color line represents the fitted regression line. (C) Bar diagram for the estimated total of DIVA positive animals over the years (2008–2021). X-axis: Years and Y-axis: Estimated DIVA positive numbers. Red color bars represent the standard errors. (D) Relation between estimated DIVA positive rates and their margin of error. X-axis:…

Figure 6. Relationship among estimated sero-prevalence rate and FMD outbreaks through structural equation modelling. (A) Performances of eight different SEMs are shown. X-axis: observations. Y-axis: values of residuals. Residual plots for the SEMs are shown in different colors. (B) Significance of current year and previous year outbreaks on estimated sero-prevalence rate is shown through SEMs. Yellow, orange and grey color shows the statistical significance of values at 10%, 5%, and 1% level of significances. (C) SEMs vs. residual standard error and error degrees of freedom plot. X-axis represents the models. Y-axis represents residual standard error. Y1 axis represents error degrees of freedom. Blue color points show the distribution of error degrees of freedom. Red color points show the distribution of residual standard errors. (D) Relationship among the SEMs. Yellow, orange and grey colors show the statistical significance of test statistic(s) at 10%, 5%, and 1% level of significances.