The European Journal of Public Health Advance Access originally published online on April 15, 2009
The European Journal of Public Health 2009 19(4):383-388; doi:10.1093/eurpub/ckp044
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Alcohol-related Problems |
Per capita alcohol consumption and sickness absence in Norway
Thor Norström1 and Inger Synnøve Moan2
1 Swedish Institute for Social Research, Stockholm University, S-106 91 Stockholm, Sweden
2 Norwegian Institute for Alcohol and Drug Research – SIRUS, PB 565 Sentrum, 0105 Oslo, Norway
Correspondence: Thor Norström, Swedish Institute for Social Research, Stockholm University, S-106 91 Stockholm, Sweden, tel: +46-8-162314, fax: +46-8-154670, e-mail: totto{at}sofi.su.se
Received January 22, 2009 , accepted March 10, 2009
Background: There is only one previous study addressing the relationship between population drinking and sickness absence. That study, based on Swedish time-series data, showed a statistically significant relationship between per capita alcohol consumption and the male sickness absence rate. Estimates suggested that a 1-l increase in consumption was associated with a 13% increase in sickness absence among men. In the present study, we aim at replicating and expanding the Swedish study on the basis of data for Norway. Methods: The outcome measure comprised annual data for Norway on registered sickness absence for manual employees covering the period 1957–2001. The unemployment rate was included as a control, as this factor may be correlated with alcohol as well as sickness absence. Alcohol consumption was gauged by sales of alcohol (total and beverage specific by beer, spirits and wine) per inhabitant 15 years and above. The data were analysed using the Box–Jenkins method for time-series analysis. Results: The results suggested that a 1-l increase in total consumption was associated with a 13% increase in sickness absence among men (P < 0.05). This corresponds to an elasticity coefficient equal to 0.62. The alcohol effect was not significant for women. Unemployment was negatively associated with the outcome for men as well as for women (P < 0.05). In the beverage-specific analyses, spirits were statistically significant for men (P < 0.05), but not beer and wine. Conclusion: The present findings strengthen the conclusion from the Swedish study, that sickness absence may be added to the list of indicators of alcohol-related harm.
Keywords: alcohol consumption, Norway, sickness absence, time-series analysis, unemployment