Objective: Given that previous studies have reached conflicting conclusions regarding the effects of diet (D), aerobic exercise (E) or both (DE) on blood lipid and lipoprotein concentrations in adults, the meta-analytic approach was used to address this issue. Data sources: Nine electronic databases, cross-referencing, and expert review. Study selection: Randomized controlled trials ≥ 4 weeks that included a D, E, DE and control (C) group in adults ≥ 18 years of age and in which data for one or more of the following were available: total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglycerides (TG). Data extraction: Dual data extraction by the first two authors. Statistical analysis: Random-effects models as well as mixed-effects models for between-group differences. Results: Seven-hundred and eighty-eight men and women from 6 studies were included. Non-overlapping 95% confidence intervals were observed for D and DE with respect to lowering TC, LDL-C and TG while reductions in E were limited to TG. No significant changes in HDL-C were observed. When compared to E, reductions in TC and LDL-C were greater for D and DE (p < 0.05 for all). Conclusions: Diet, especially DE, are superior to E for improving selected lipids and lipoproteins in adults.