A Retrospective Meta-analysis on the Effectiveness of FOLLTROPIN-V for the Induction of Superovulation in Beef and Dairy Heifers and Cows
Anna B Nevius, FDA/CVM  Junshan Qiu, FDA CDER  *Steven V. Radecki, ASA 

Keywords: systematic review, meta-analysis, veterinary drug

Data from twenty-one research and clinical studies encompassing fifty treatment arms were included in the analysis. Variables included in the analysis were treatment arm, study, mean number of transferable embryos within a treatment arm, variation in the mean number of transferable embryos within a treatment and the number of records used in the estimation of the mean number of transferable embryos. Statistical assessments of bias included the variables percent missing, number of records used in the estimation of the mean, and the harmonized SEM. Publication bias and the heterogeneity among the treatment arms were also evaluated. The final statistical model based on these evaluations was a two-level random effects model (between study and between treatment arms) with intercept included as a random effect at the study level. A sensitivity analysis evaluating potential high risk study characteristics was also conducted. Effect size measurements were generally reported as means or least squares means. Standard effort of the mean was the measure of variation harmonization. The lower bound of the two-sided 95% confidence interval for the mean number of transferable embryos was greater than 1 for 36 of the 50 treatment arms. The estimated effect size based on the model was 4.5 transferable embryos (95% confidence interval: 3.5, 5.6). Given the lower bound of this interval was greater than the expected non-treated value of 1, it was concluded that treatment with FOLLTROPIN-V is effective in increasing the number of transferable embryos. The sensitivity analysis confirmed the validity of the chosen model.