Organic fertilization selectively alters rice grain quality without a consistent yield gain: a multilevel meta-analysis in China

Reducing synthetic-fertilizer use without sacrificing rice yield or grain quality remains difficult because quality traits may respond differently. We conducted a systematic review and multilevel meta-analysis of field experiments in China. Eligible studies compared organic fertilizer alone or combined with reduced synthetic fertilizer against a synthetic-fertilizer-only control. After excluding 13 effects with ineligible comparators, the dataset contained 420 treatment-control effects from 40 studies. Outcomes comprised yield and 10 grain-quality traits. We analyzed original-direction log response ratios (lnRR) with trait-specific multilevel random-effects models. We used experiment-clustered CR2 variance estimates and Satterthwaite inference. Prespecified moderators were soil organic matter, total nitrogen, available phosphorus, and nitrogen substitution rate. We controlled the within-trait false discovery rate. The pooled yield association was -0.52% with a 95% confidence interval from -6.25% to 5.57%. This interval did not establish equivalence or assured yield maintenance. Head rice rate increased by 2.05% with a 95% confidence interval from 0.28% to 3.86% and a nominal P value of 0.025. Rapid Visco Analyzer (RVA) breakdown increased by 6.70% with a 95% confidence interval from 0.79% to 12.96% and a nominal P value of 0.030. Prediction intervals for both outcomes crossed zero. Confidence intervals for the other eight quality traits also included zero. No moderator association remained significant after false-discovery-rate control. However, the yield association with substitution rate had an adjusted P value of 0.071 and remained exploratory. Removing ineligible comparators did not reverse any pooled direction, but it changed the magnitude or heterogeneity of setback, chalkiness, and breakdown. Organic-fertilizer substitution showed trait-specific associations rather than uniform benefits. The evidence applies to the included Chinese field studies. It does not establish causal mechanisms, overall quality improvement, or assured yield maintenance.