2.1 Search strategy
The online free databases were searched for the primary studies which reported prevalence of cysticercosis in pigs during 2000 to 2019. The databases searched were GoogleScholar, PubMed, J-GatePlus by using keywords viz. Taenia solium, Pig, Human, Cysticercosis, Cysticercus tenuicolis, Pork tapeworm, India singly or different combinations with Booleans AND / IN / OR etc. The titles, abstracts and full length papers were read thoroughly and their suitability was assessed. Primary research articles containing the data on the number of animals/carcasses screened, number of animals/ carcasses found positive for cysticercus cysts and limited to Indian cities/ towns were included in this study. Studies with data of prevalence work carried out from 2000 to 2019 were only included in this meta-analysis.
2.2 Data extraction
Studies / papers obtained through systematic search as per PRISMA guidelines (Moher et al. 2009) were then read thoroughly and requisite data was extracted and entered in MS-Excel sheet. Data from individual study necessarily included study details viz name of authors and year of publication, year of abattoir survey / sampling, place of work, method of detection, organs detected, number of carcasses/ animals detected and number of carcasses / samples found positive for harbouring Taenia solium cysts. Data were arranged systematically in excel sheet for further statistical analyses.
2.3 Study quality and characteristics
Following the predefined inclusion and exclusion criteria, study reports were assessed for suitability. The selected study must have provided the data pertaining to year of study, number of carcasses/ samples tested and positives reported, area of study and method employed. As regards method of detection slaughter house survey / necropsy studies, serological tests and molecular methods employed were taken into account. Studies without these parameters were not included in final analysis. Initial screening of studies for compliance with the objective and relevance was done and this was followed by thorough reading of reports for extraction of requisite data.
Owing to significant heterogeneity between studies, a random-effects model was adopted for this meta-analysis. The prevalence data on total sample size and number of positives was analyzed to obtain the effect size and standard error of the effect size. The estimates were then pooled with a 95% confidence interval. Cochran’s Q and Higgin’s I2 statistics were used for measuring the between-study variance. The I2 values of 25%, 50%, and 75% were considered low, moderate and high heterogeneity, respectively (Higgins and Thompson 2002). In addition, publication bias was assessed using funnel plot visualization and LFK Index respectively (Kanamori et al. 2018). To appraise the sources of heterogeneity subgroup meta-analysis was conducted by grouping variables according to regions of studies, methods of detection of cysts and sample size in the included studies. All the analyses were done with the help of MetaXL add-in (EpiGear International Pty Ltd, Queensland, Australia).