Research design to examine the effects of alcohol marketing

Studies on the effect of alcohol advertising and marketing on alcohol consumption in young people can be divided in two main categories: experimental and observational studies. Observational studies can be categorized in three broad categories: longitudinal studies, cross-sectional studies and econometric studies (case-control studies are not used in this area). As we will explain there are ethical difficulties on experiments on this issue. Econometric studies are often measured at the aggregate level and do not predict individual behaviour. Cross-sectional studies can not establish causality. Consequently, although expensive and complex, thoroughly designed and conducted longitudinal studies are often the best choice of design to examine the effect of exposure to alcohol marketing on drinking behaviour in young people.

• Randomized and true experiments
  
  
• Econometric Studies
  
  
• Cross-sectional studies
  
  
• Longitudinal studies
  

Randomized and true experiments

Randomized controlled trials (RCTs) are the most optimal design for inferring causality, but it has not frequently been conducted in this area. It is unlikely that these randomized controlled trials will be used in the future, besides impracticalities, it may be unethical to randomise participants or communities to exposure to specific advertising and/or marketing strategies in order to evaluate potentially harmful effects [1,3].

Most experimental studies that have been published evaluated exposure to a single type of advertising and evaluated immediate effects on either attitude or liking for the advertisements or drinking behaviour. Experimental studies might underestimate the overall effects of alcohol marketing, because only a limited number of key factors could be studied at once [1]. These experimental studies meet the complexity of the advertising and marketing that people are exposed to in their daily lives, and only evaluate effects post-exposure at a single time-point, so results can not be generalised to other settings [3].

Summaries of articles:

Observational studies:

Econometric studies

Econometric studies often use time-series data (with a longitudinal character) and bring together macroeconomic data with population level data and use statistical modelling to examine relationships between exposure (eg advertising expenditure) and outcome (eg alcohol sales). Although these econometric studies have the benefit that they do not rely heavily on self-reports of respondents, an important limitation of these studies is that they examine changes within a population and they do not specifically look at drinking behaviour in young people but report aggregate alcohol consumption. The observed effect is highly dependent on the choice of factors that are used in the statistical model [3].

Duffy, M. (1989). Measuring the Contribution of Advertising to Growth in Demand: An Econometric-Accounting Framework. International Journal of Advertising, 1989, 8 (2), 95-110.

Cross-sectional studies

Cross-sectional studies examine a sample at one point in time. Cross-sectional surveys measure for example the association between exposure to alcohol advertising and drinking behaviour, but do not show whether the exposure preceded the outcome [3]. Due to the single measurement moment cross-sectional studies are vulnerable to reverse causality in which we cannot rule out that drinkers are predisposed to view and remember alcohol advertisements, rather than alcohol advertisements stimulate alcohol consumption in young people.

Summaries of articles:

Chen, M.J., Grube, J.W., Bersamin, M., Waiters, E.D. and D.B. Keefe (2005). Alcohol Advertising: What Makes It Attractive to Youth? Journal of Health Communication, 10, 553-565.

Waiters, E.D., Treno, A.J., and J. W. Grube (2001). Alcohol Advertising and youth: a focus-group analysis of what young people find appealing in alcohol advertising. Contemporary Drug Problems, 28, 695-718.

Wyllie, A., Zhang, J.F. and S. Casswell (1998). Responses to televised alcohol advertisements associated with drinking behaviour of 10-17-years-olds. Addiction, 93 (3), 361-371.

Longitudinal studies

In contrast to cross-sectional studies, longitudinal studies are designed to observe a sample over an extended period or to study changes over time. In general, longitudinal studies are often the best way to study changes over time [2]. To examine the effect of alcohol advertising and marketing on alcohol consumption in young people longitudinal studies are especially valuable because they capture what happens to individuals over time and can demonstrate whether individuals who differ in their exposure to advertising differ in their future drinking behaviour [1]. Longitudinal studies provide the highest level of evidence that is available for evaluation of advertising and marketing exposure and subsequent drinking behaviour. If such studies are well designed, conducted and analysed they can provide supportive evidence for a causal relationship between a particular exposure and an outcome [3]. Well conducted and designed longitudinal studies have a representative sample with a minimum of attrition. Besides a reliable measurement of the main predictors and outcomes, they include main confounding variables as well.

Longitudinal studies can be divided in three types of studies: trend studies, cohort studies and panel studies [2]. Trend studies are longitudinal studies that examine population changes over time. In contrast to trend studies, cohort studies are used to examine a specific subpopulation over time. The latter type of study is more suited to examine the effect of alcohol advertising on alcohol consumption in young people, since it makes it possible to examine the effect on specific age group. Prospective cohort studies define the groups of interest before the study is conducted. Although quite similar to trend and cohort studies, panel studies examine the same set of respondents at each measurement moment, which is not the case in trend and cohort studies. Most longitudinal studies on the effects of alcohol marketing, although frequently mentioned as cohort studies, are panel studies which follow the same set of respondents during several months or years.

Connolly, G.M., Casswell S, Zhang J. F, and Silva P. A, Alcohol in the mass media and drinking by adolescents: a longitudinal study. Addiction, 1994; 89(10):1255-63.

Ellickson, P.L., Collins R. L, Hambarsoomians K, and McCaffrey D. F., (2005). Does alcohol advertising promote adolescent drinking? Results from a longitudinal assessment. Addiction, 100(2), 235-246.

Fisher, L.B., Williams I., Austin B., Camargo C.A., and Colditz G.A. (2007). Predictors of Initiation of Alcohol Use Among US Adolescents. Findings from a Prospective Cohort Study. Arch Pedriatr Adolesc Med, 161, 959-966.

Henriksen L., Feighery E.C., Schleicher N.C., and Fortmann S.P. (2008). Receptivity to Alcohol Marketing Predicts Initiation of Alcohol Use. Journal of Adolescent Health, 42, 28-35.

McClure, A.C., Chin, S.D., Gibson,J. and J.D. Sargent, (2006). Ownership of Alcohol-Branded Merchandise and Initiation of Teen Drinking. American Journal of Preventive Medicine, 30 (4), 277-283.

Pasch K.E., Komro K.A., Perry C.L., Hearst M.O., Farbakhsh K. (2007).Outdoor Alcohol Advertising near schools: What does it advertise and how is it related to intentions and use of alcohol among young adolescents? J Study Alcohol Drugs), 68, 587-596.

Robinson, T.N., H.L. Chen, and J.D. Killen, (1998) Television and music video exposure and risk of adolescent alcohol use. Pediatrics, 102(5), E54.

References

[1] Anderson, P. (2007). The Impact of Alcohol Advertising: ELSA project report on the evidence to strengthen regulation to protect young people. Utrecht: National Foundation for Alcohol Prevention.

[2] Earl Babbie (2001). The Practice of Social Research. Wadsworth/Thomson Learning: Belmont, USA.

[3] Smith & Foxcroft (2007). The effect of alcohol advertising and marketing in drinking behaviour in young people: A systematic review. Derived at December 14, from: AERC
See summary for more info.