gms | German Medical Science

Kongress Medizin und Gesellschaft 2007

17. bis 21.09.2007, Augsburg

Who drops out? Attrition in a longitudinal alcohol research telephone survey in Denmark

Meeting Abstract

Suche in Medline nach

  • Ulrike Grittner - Institute for Biometrics and Clinical Epidemiology, Charité University Medicine, Berlin, Germany
  • Kim Bloomfield - Unit for Health Promotion Research, University of Southern Denmark, Esbjerg, Denmark
  • Matthias Wicki - Swiss Institute for the Prevention of Alcohol and Drug Problems, Lausanne, Switzerland
  • Gerhard Gmel - Swiss Institute for the Prevention of Alcohol and Drug Problems, Lausanne, Switzerland

Kongress Medizin und Gesellschaft 2007. Augsburg, 17.-21.09.2007. Düsseldorf: German Medical Science GMS Publishing House; 2007. Doc07gmds122

Die elektronische Version dieses Artikels ist vollständig und ist verfügbar unter:

Veröffentlicht: 6. September 2007

© 2007 Grittner et al.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen ( Er darf vervielfältigt, verbreitet und öffentlich zugänglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.



Background: Longitudinal studies provide important information about changes with respect to specific outcome variables such as alcohol consumption and drinking pattern. But results and conclusions from such studies could be distorted due to differences between those who respond in all waves, those who drop out and those who do not respond at all. This study analyses the differences between drop outs and those who respond in all waves. In earlier alcohol research it has been shown, that especially abstainers and heavy drinkers are more likely to drop out than moderate drinkers [1], [2]. Additional factors such as gender, age, marital status and socioeconomic status could also be related to drop out.

Material / Methods: Data for Denmark come from the Nordic study of changes in drinking [3] and were collected in 4 waves in the years 2003, 2004, 2005 and 2006. In 2003 the dataset contains 1771 cases of age 16 to 69. Only 35.7% responded in all four waves. 17.8% responded in 2003 and 2004, 13,4% in 2003, 2004 and 2005, 15.9% responded only 2003. Those who dropped out in the second or third wave were asked again in the next wave and 16.9% responded again in a later wave. To identify determinants for dropping out or responding in all four waves we used discriminant analysis.

Results: Preliminary results show those who dropped out were younger. Among men the following were more likely to drop out: abstainers, those who do not live in a partnership, and those with a high volume per drinking day. Among women housewives and those with lower income were more likely to drop out.

Conclusion: Though some findings pointed to differential dropout, these effects could not explain why the recent policy changes did not affect consumption in the expected direction, i.e., higher alcohol consumption with increased availability.


Gmel G, Rehm J. Measuring alcohol consumption. Contemporary Drug Problems. 2004;31:467-540.
Lahaut, et al. Non-response bias in a sample survey on alcohol consumption. Alcohol & Alcoholism. 2002;37/3:256-60.
SoRAD (Centre for Social Research on Alcohol and Drugs). Project: “Whose drinking changes how much when liquor taxes fall? Effects of tax cuts”. Externer Link