Alcohol and cirrhosis: dose –response or threshold effect? Mads Kamper-Jørgensen 1, * , Morten Grønbæk 1 , Janne Tolstrup 1 , Ulrik Becker 2 1 Centre for Alcohol Research, National Institute of Public Health, Copenhagen, Denmark 2 Alcohol Unit, Hvidovre Hospital, University of Copenhagen, Hvidovre, Denmark Background/Aims: General population studies have shown a strong association between alcohol intake and death from alcoholic cirrhosis, but whether this is a dose–response or a threshold effect remains unknown, and the relation among alcohol misusers has not been studied. Methods: A cohort of 6152 alcohol misusing men and women aged 15–83 were interviewed about drinking pattern and social issues and followed for 84,257 person-years. Outcome was alcoholic cirrhosis mortality. Data was analyzed by means of Cox-regression models. Results: In this large prospective cohort study of alcohol misusers there was a 27 fold increased mortality from alcoholic cirrhosis in men and a 35 fold increased mortality from alcoholic cirrhosis in women compared to the Danish population. Number of drinks per day was not significantly associated with death from alcoholic cirrhosis, since there was no additional risk of death from alcoholic cirrhosis when exceeding an average daily number of five drinks (> 60 g/alcohol) in neither men nor women. Conclusions: The results indicate that alcohol has a threshold effect rather than a dose–response effect on mortality from alcoholic cirrhosis in alcohol misusers. q 2004 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. Keywords: Alcohol; Misuse; Cirrhosis; Mortality; Epidemiology 1. Introduction The most consistent finding from epidemiological studies on alcohol and cirrhosis in the general population has been that of an increasing risk of cirrhosis with increasing intake of alcohol [1–6]. In some studies, this relation was modified by duration of alcohol misuse [4], type of alcohol consumed [3] and frequency of alcohol intake [7,8]. Despite the elevated risk, only a minority of alcohol misusers develops alcoholic cirrhosis [9,10]. One hypothesis is that alcohol has a permissive rather than a dose-dependent effect on development of liver disease, i.e. if alcohol intake at some point in time is higher than some threshold level, the risk of developing cirrhosis is constant over time [9]. This hypothesis is best investigated in populations of alcohol misusers, but only a few small studies regarding alcohol and alcoholic cirrhosis have been carried out in men and not in women [9–11]. The purpose of this study is therefore, in a prospective study design to investigate whether average daily number of drinks, duration of alcohol misuse, predominant type of alcohol consumed and frequency of alcohol intake respectively, is associated with alcoholic cirrhosis mortality in a large cohort of alcohol misusing men and women. 2. Materials and methods 2.1. Sample The Copenhagen Alcohol Cohort consists of men and women between 15 and 83 years of age (median age ¼ 38 years) who attended Copenhagen Hospital Corporation’s outpatient clinic for alcohol misusers in the period 1977–1992. Through interview with a nurse trained as alcohol therapist, recordings of alcohol related, social issues, frequency of alcohol use and duration of alcohol misuse were obtained for each subject. These data were recorded on structured data sheets at the time of first admission to the alcohol outpatient clinic. Information on average number of drinks per day and predominant type of alcohol was only recorded in medical records at the same time as the interview was carried out. By review of these medical records, data on average number of drinks per day and type of alcohol was extracted and added to the dataset. By linkage to The Danish National Hospital Discharge Register, all hospital discharges of the 9150 subjects attending the alcohol clinic were recorded. The Danish National Discharge 0168-8278/$30.00 q 2004 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jhep.2004.03.002 Journal of Hepatology 41 (2004) 25–30 www.elsevier.com/locate/jhep Received 20 November 2003; received in revised form 6 February 2004; accepted 9 March 2004; available online 7 April 2004 * Corresponding author. Tel.: þ 45-3920-7777; fax: þ 45-3920-8010. E-mail address: mkj@niphy.dk (M. Kamper-Jørgensen).