The article emphasizes the critical importance of accounting for cellular heterogeneity in epigenome-wide association studies (EWAS) of human disease and other quantitative traits. While DNA methylation (DNAm) profiles in peripheral blood have been widely studied, blood is a heterogeneous mixture of different cell types, each with distinct DNAm profiles. The authors use a statistical method to estimate the relative proportions of cell types from DNA methylation profiles and find strong evidence of cell composition changes across age in blood. They demonstrate that cellular composition explains a significant portion of the observed variability in DNAm and that there is high levels of confounding between age-related variability and cellular composition at the CpG level. The findings highlight the need for considering cell composition variability in EWAS based on whole blood and other heterogeneous tissue sources. The authors provide software for estimating and exploring this composition confounding for the Illumina 450k microarray, which can help improve the biological interpretation of results and reduce false positives.The article emphasizes the critical importance of accounting for cellular heterogeneity in epigenome-wide association studies (EWAS) of human disease and other quantitative traits. While DNA methylation (DNAm) profiles in peripheral blood have been widely studied, blood is a heterogeneous mixture of different cell types, each with distinct DNAm profiles. The authors use a statistical method to estimate the relative proportions of cell types from DNA methylation profiles and find strong evidence of cell composition changes across age in blood. They demonstrate that cellular composition explains a significant portion of the observed variability in DNAm and that there is high levels of confounding between age-related variability and cellular composition at the CpG level. The findings highlight the need for considering cell composition variability in EWAS based on whole blood and other heterogeneous tissue sources. The authors provide software for estimating and exploring this composition confounding for the Illumina 450k microarray, which can help improve the biological interpretation of results and reduce false positives.