Abstract Background Generally, low-density lipoprotein (LDL) particle size can be inferred from the LDL cholesterol concentration to total apolipoprotein B concentration ratio (LDL-C/ApoB ratio, hereinafter called LAR), which is a good predictor of cardiovascular disease. However, the predictive ability of LAR for mortality risk in the general population is still unclear. This study aimed to explore the association between LAR and cardiovascular as well as all-cause mortality among American adults. Methods The present study was a secondary analysis of existing data from the National Health and Nutrition Examination Survey (NHANES). The final analysis included 12,440 participants from 2005 to 2014. Survival differences between groups were visualized using Kaplan‒Meier curves and the log-rank test. The association of LAR with cardiovascular and all-cause mortality was evaluated using multivariate Cox regression and restricted cubic spline analysis. Age, sex, coronary artery disease, diabetes, lipid-lowering medication use and hypertriglyceridemia were analyzed in subgroup analyses. Results The median age in the study cohort was 46.0 years [interquartile range (IQR): 31.0–62.0], and 6,034 (48.5%) participants were male. During the follow-up period, there were 872 (7.0%) all-cause deaths and 150 (1.2%) cardiovascular deaths. Compared with individuals without cardiovascular events, those who experienced cardiovascular deaths had a lower LAR (1.13 vs. 1.25) (P < 0.001). The adjusted Cox regression model indicated that lower LAR was an independent risk factor for both cardiovascular [hazard ratio (HR) = 0.304, 95% confidence interval (CI): 0.114–0.812] and all-cause mortality (HR = 0.408, 95% CI: 0.270–0.617). Moreover, a significant age interaction was observed (P for interaction < 0.05), and there was a strong association between LAR and mortality among participants over 65 years of age. Further analysis showed an inverse association between LAR and both cardiovascular and all-cause mortality. Conclusions LAR can independently predict cardiovascular and all-cause mortality in the general population.