Skip to main content

Table 4 Alcohol consumption frequency and adjusted risk for poor clinical outcomes, stratified sub-groups based on average alcohol consumption

From: Association between patterns of alcohol consumption (beverage type, frequency and consumption with food) and risk of adverse health outcomes: a prospective cohort study

N = 309,123. Number used for analysis N = 286,115 (92.6%) used for analysis after excluding missing values
Type of alcohol consumed All-cause mortality MACE Liver Cirrhosis Accidents/self-harm/assaults New cancer (all-cause) incidence Alcohol cancer incidence
Events = 8869 (2.9%) HR with 95% CI Events = 5246 (1.7%) HR with 95% CI Events = 838 (0.3%) HR with 95% CI Events = 16,818 (5.4%) HR with 95% CI Events = 27,543 (8.9%) HR with 95% CI Events = 6529 (2.1%) HR with 95% CI
3–4 times/week (reference) N = 103,717 (33.5%) 2598 (2.5%) 1 1603 (1.5%) 1 180 (0.2%) 1 5336 (5.1%) 1 8996 (8.7%) 1 2144 (2.1%) 1
1–2 times/week N = 115,270 (37.3%) 3084 (2.7%) 1.09 (1.03–1.16); p < 0.01 1921 (1.7%) 1.14 (1.06–1.23); p < 0.01 188 (0.2%) 1.04 (0.83–1.31); p = 0.69 6153 (5.3%) 1.04 (1.00–1.09); p = 0.04 9612 (8.3%) 1.02 (0.99–1.05); p = 0.17 2376 (2.1%) 1.01 (0.95–1.08); p = 0.73
Daily or almost daily N = 90,136 (29.2%) 3187 (3.5%) 1.06 (0.99–1.12); p = 0.052 1722 (1.9%) 0.99 (0.91–1.07); p = 0.80 470 (0.5%) 1.49 (1.21–1.82); p < 0.01 5329 (5.9%) 1.02 (0.97–1.06); p = 0.43 8935 (9.9%) 0.99 (0.96–1.02); p = 0.65 2009 (2.2%) 0.98 (0.87–1.02); p = 0.50
Global p value for heterogeneity < 0.01 < 0.01 < 0.01 < 0.01 < 0.01 0.03
  1. Legend: HR hazard ratio, CI confidence intervals, MACE major adverse cardiovascular event. Alcohol new cancers = breast, colon, rectum, larynx, liver and oesophagus. All results adjusted for age, sex, Townsend score for socio-economic deprivation (continuous), average weekly alcohol units (continuous), type of alcohol consumed, alcohol consumption pattern with/without meals, smoking habits, BMI, physical activity levels, number of long-term conditions,, self-rated health and C-reactive protein levels at baseline. MACE events model adjusted for all the above plus presence of diabetes, hypertension, systolic blood pressure and total cholesterol levels at baseline. Cirrhosis events model adjusted for all of the above plus gamma glutamyl transpeptidase levels at baseline