Cancers of female breast, upper aero-digestive tract (UADT) (oral cavity, pharynx, larynx, oesophagus) and colorectum are causally related to alcohol consumption. Although alcohol consumption is likely to vary during life, the few studies that have explicitly measured lifetime consumption or intake over time have not been summarised. We therefore conducted a systematic review and meta-analysis.Studies were identified by searching the Medline, CINAHL (Cumulative Index to Nursing and Allied Health Literature) and Scopus databases through January 2015 using broad search criteria. Studies reporting relative risks (RR) for quantitatively defined categories of alcohol consumption over time for breast, UADT or colorectal cancer were eligible. A two-stage random-effects meta-analysis was used to estimate a dose-response relationship between alcohol intake and each cancer site. RRs were also calculated for the highest relative to the lowest intake category.Sixteen articles for breast, 16 for UADT and 7 for colorectal cancer met the eligibility criteria. We observed a weak non-linear dose-response relationship for breast cancer and positive linear dose-response relationships for UADT and colorectal cancer. The pooled RRs were 1.28 (95% confidence interval, CI: 1.07, 1.52) for breast, 2.83 (95% CI: 1.73, 4.62) for UADT, 4.84 (95% CI: 2.51, 9.32) for oral cavity and pharynx, 2.25 (95% CI: 1.49, 3.42) for larynx, 6.71 (95% CI: 4.21, 10.70) for oesophageal and 1.49 (95% CI: 1.27, 1.74) for colorectal cancer.Our findings confirm dose-dependent associations between long-term alcohol intake and breast, UADT and colorectal cancer.