PT  - JOURNAL ARTICLE
AU  - Diana Quirmbach, Dr
AU  - Laura Cornelsen
AU  - Susan A Jebb
AU  - Theresa Marteau
AU  - Richard Smith
TI  - Effect of increasing the price of sugar-sweetened beverages on alcoholic beverage purchases: an economic analysis of sales data
AID  - 10.1136/jech-2017-209791
DP  - 2018 Apr 01
TA  - Journal of Epidemiology and Community Health
PG  - 324--330
VI  - 72
IP  - 4
4099  - http://jech.bmj.com/content/72/4/324.short
4100  - http://jech.bmj.com/content/72/4/324.full
SO  - J Epidemiol Community Health2018 Apr 01; 72
AB  - Background Taxing soft-drinks may reduce their purchase, but assessing the impact on health demands wider consideration on alternative beverage choices. Effects on alcoholic drinks are of particular concern, as many contain similar or greater amounts of sugar than soft-drinks and have additional health harms. Changes in consumption of alcoholic drinks may reinforce or negate the intended effect of price changes for soft-drinks.Methods A partial demand model, adapted from the Almost Ideal Demand System, was applied to Kantar Worldpanel data from 31 919 households from January 2012 to December 2013, covering drink purchases for home consumption, providing ~6 million purchases aggregated into 11 groups, including three levels of soft-drink, three of other non-alcoholic drinks and five of alcoholic drinks.Results An increase in the price of high-sugar drinks leads to an increase in the purchase of lager, an increase in the price of medium-sugar drinks reduces purchases of alcoholic drinks, while an increase in the price of diet/low-sugar drinks increases purchases of beer, cider and wines. Overall, the effects of price rises are greatest in the low-income group.Conclusion Increasing the price of soft-drinks may change purchase patterns for alcohol. Increasing the price of medium-sugar drinks has the potential to have a multiplier-effect beneficial to health through reducing alcohol purchases, with the converse for increases in the price of diet-drinks. Although the reasons for such associations cannot be explained from this analysis, requiring further study, the design of fiscal interventions should now consider these wider potential outcomes.