Pest and disease management is critical to agricultural production, protecting crops from harmful pests and diseases. Threshold-based control programs, which apply pesticides only when pest and disease population densities exceed a predetermined threshold, can reduce pesticide use. However, the effectiveness of these programs is unclear and varies widely. To assess the broader impact of threshold-based control programs on agricultural arthropod pests, we conducted a meta-analysis of 126 studies, including 466 trials on 34 crops, comparing threshold-based programs with calendar-based pesticide control programs and/or untreated controls. Compared with calendar-based programs, threshold-based programs reduced pesticide application by 44% and associated costs by 40%, without affecting pest and disease control efficacy or overall crop yield. Threshold-based programs also increased beneficial insect populations and achieved similar levels of control of arthropod-borne diseases as calendar-based programs. Given the breadth and consistency of these benefits, increased political and financial support is needed to encourage the adoption of this control approach in agriculture.
Records were identified through database and other source searches, screened for relevance, assessed for eligibility, and ultimately narrowed to 126 studies, which were included in the final quantitative meta-analysis.
Not all studies reported means and variances; therefore, we calculated the mean coefficient of variation to estimate the variance of the log ratio.25 For studies with unknown standard deviations, we used Equation 4 to estimate the log ratio and Equation 5 to estimate the corresponding standard deviation. The advantage of this method is that even if the estimated standard deviation of the lnRR is missing, it can still be included in the meta-analysis by calculating the missing standard deviation using the weighted mean coefficient of variation from studies that centrally report standard deviations.
For studies with known standard deviations, the following formulas 1 and 2 are used to estimate the log ratio and the corresponding standard deviation .
For studies with unknown standard deviations, the following formulas 3 and 4 are used to estimate the log ratio and the corresponding standard deviation .
Table 1 presents the point estimates of the ratios, associated standard errors, confidence intervals, and p-values for each measure and comparison. Funnel plots were constructed to determine the presence of asymmetry for the measures in question (Supplementary Figure 1). Supplementary Figures 2–7 present the estimates for the measures in question in each study.
More details about the study design can be found in the Nature Portfolio report summary linked from this article.
Interestingly, we found virtually no significant differences in the effectiveness of threshold-based pesticide applications between specialty and conventional crops for key metrics such as pest and disease control, yield, economic benefits, and impact on beneficial insects. This result is unsurprising given that, from a biological perspective, threshold-based pesticide application programs do not significantly differ between these two crop types. Differences between conventional and specialty crops primarily stem from economic and/or regulatory factors, rather than environmental ones. These differences between crop types are more likely to influence pest and disease management practices than the biological effects of threshold-based pesticide applications. For example, specialty crops typically have a higher unit cost per hectare and therefore require more stringent quality standards, which may motivate growers to apply pesticides preventively due to concerns about less common pests and diseases. Conversely, large acreages of conventional crops make pest and disease monitoring more labor-intensive, limiting the feasibility of implementing threshold-based pesticide application programs. Thus, both systems face unique pressures that can either facilitate or hinder the implementation of threshold-based pesticide application programs. Since nearly all the studies in our meta-analysis were conducted in settings where pesticide restrictions had been lifted, it is not surprising that we observed stable threshold values across crop types.
Our analysis shows that threshold-based pesticide management programs can significantly reduce pesticide use and associated costs, but it remains unclear whether agricultural producers actually benefit from them. The studies included in our meta-analysis varied significantly in their definitions of “standard” pesticide management programs, ranging from regional practices to simplified calendar programs. Therefore, the positive results we report here may not fully reflect the actual experiences of producers. Moreover, although we documented significant cost savings due to reduced pesticide use, the initial studies generally did not consider field inspection costs. Therefore, the overall economic benefits of threshold-based management programs may be somewhat lower than the results of our analysis. However, all studies that reported field inspection costs documented reduced production costs due to reduced pesticide costs. Routine monitoring and field inspections can be challenging for busy producers and farm managers (U.S. Bureau of Labor Statistics, 2004).
Economic thresholds play a central role in the concept of integrated pest management (IPM), and researchers have long reported the positive benefits of threshold-based pesticide application programs. Our research showed that arthropod pest control is essential in most systems, as 94% of studies indicate a reduction in crop yields without pesticide application. However, prudent pesticide use is critical to promoting long-term sustainable agricultural development. We found that threshold-based application effectively controls arthropod damage without sacrificing crop yields compared to calendar-based pesticide application programs. Moreover, threshold-based application can reduce pesticide use by more than 40%. Other large-scale assessments of pesticide application patterns in French farmland and plant disease control trials have also shown that pesticide application can be reduced by 40-50 % without affecting yield. These results highlight the need for further development of new thresholds for pest management and the provision of resources to encourage their widespread use. As agricultural land use intensity increases, pesticide use will continue to threaten natural systems, including highly sensitive and valuable habitats . However, wider adoption and implementation of pesticide threshold programs can mitigate these impacts, thereby increasing the sustainability and environmental friendliness of agriculture.
Post time: Dec-04-2025