Effect of pesticide formulation on initial residue levels in crops: a meta-analysis approach

Abstract

A meta-analysis was conducted to determine whether pesticide formulation significantly affects the initial residue levels across different crops. Data from pesticide registration trials for three vegetable crops, four fruit crops, and one cereal crop were sourced from the Agricultural Materials Evaluation System database of the Rural Development Administration. After applying inclusion and exclusion criteria, 296 trials for lettuce, 113 for perilla leaves, 232 for green onions, 94 for apples, 206 for peaches, 230 for plums, 75 for apricots, and 144 for brown rice were selected. For each crop, trials using a specific formulation were grouped, the mean residue level and standard deviation were calculated for each group, and the residues from different formulations were then compared within the same crop. The standardized mean difference (SMD) was calculated for each crop using Hedges' g, and study heterogeneity was assessed using Cochran's Q-test and Higgins' I²-statistic. The Q values ranged from 1.91 to 12.59 (p > 0.13), and I² values remained below 38 %, indicating no significant heterogeneity. A common effect model was then applied, yielding SMDs between 1.088 and 1.525 (95 % confidence interval). Funnel plots and Egger's test confirmed the absence of publication bias. In conclusion, the results showed that variations in pesticide formulation do not significantly affect initial residue levels in crops. The methodology and findings of this meta-analysis may support further investigation into the residual behavior of pesticides and the factors influencing pesticide residues in various crops.