Distribution and dose response of laboratory markers to alcohol consumption in a general population: results of the study of health in Pomerania (SHIP)

OBJECTIVE: Biomarker distributions must be well known for use as screening tools for hazardous alcohol consumption in general populations. The aim of this study was to investigate characteristics of carbohydrate-deficient transferrin (CDT) and to compare it with gamma-glutamyltransferase (GGT) and erythrocyte mean corpuscular volume (MCV) in regard to distribution in the general population; the dose response relationship between alcohol consumption and biomarkers; and the effect of gender, age, body mass index (BMI) and smoking. METHOD: In a cross-sectional health survey in northeast Germany (1997-2001), a sample of 7,008 men and women aged 20-79 years was drawn, following stratification by age and gender. Of this sample, 4,310 subjects (2,193 women) took part in the study. Alcohol consumption was evaluated by self-reports with a beverage specific quantity-frequency method. RESULTS: Biomarker distributions differ across age and gender. The distribution of drinkers and nondrinking subjects showed considerable overlap. The association of alcohol consumption to laboratory markers is weak in the general population: strongest for GGT, followed by CDT and MCV For CDT and MCV it is weaker in women than in men. We found increasing risk of all three elevated marker values with increasing alcohol consumption and significant effects of age, gender, smoking and body mass index on the alcohol-biomarker dose response curve. CONCLUSIONS: When biomarkers are used for screening, all confounding effects have to be taken into account and adjusted normal ranges must be used. CDT shows no overall advantage over GGT. Low correlation of biomarkers with alcohol consumption, their high variability and widely spread ranges in nondrinking subjects limit the usefulness of these markers in general population settings.