Abstract Aims Lipoprotein(a) [Lp(a)] is elevated in 20–30% of people. This study aimed to assess the effect of statins on Lp(a) levels. Methods and results This subject-level meta-analysis includes 5256 patients (1371 on placebo and 3885 on statin) from six randomized trials, three statin-vs.-placebo trials, and three statin-vs.-statin trials, with pre- and on-treatment (4–104 weeks) Lp(a) levels. Statins included atorvastatin 10 mg/day and 80 mg/day, pravastatin 40 mg/day, rosuvastatin 40 mg/day, and pitavastatin 2 mg/day. Lipoprotein(a) levels were measured with the same validated assay. The primary analysis of Lp(a) is based on the log-transformed data. In the statin-vs.-placebo pooled analysis, the ratio of geometric means [95% confidence interval (CI)] for statin to placebo is 1.11 (1.07–1.14) (P < 0.0001), with ratio >1 indicating a higher increase in Lp(a) from baseline in statin vs. placebo. The mean percent change from baseline ranged from 8.5% to 19.6% in the statin groups and −0.4% to −2.3% in the placebo groups. In the statin-vs.-statin pooled analysis, the ratio of geometric means (95% CI) for atorvastatin to pravastatin is 1.09 (1.05–1.14) (P < 0.0001). The mean percent change from baseline ranged from 11.6% to 20.4% in the pravastatin group and 18.7% to 24.2% in the atorvastatin group. Incubation of HepG2 hepatocytes with atorvastatin showed an increase in expression of LPA mRNA and apolipoprotein(a) protein. Conclusion This meta-analysis reveals that statins significantly increase plasma Lp(a) levels. Elevations of Lp(a) post-statin therapy should be studied for effects on residual cardiovascular risk.