Letter to the Editor in Regard to Peacock, Bruno, and Martin (2012): “The Subjective Physiological, Psychological, and Behavioral Risk‐Taking Consequences of Alcohol and Energy Drink Co‐Ingestion”

THE CONCLUSIONS PUBLISHED in Alcoholism: Clinical and Experimental Research (2012;36:2008– 2015) by Peacock and colleagues are not justified, at least as explained in their paper. The authors conclude that alcohol mixed with energy drink, or AmED-drinking sessions, are associated with less involvement in risk behavior because individuals were observed to be less likely to engage in these behaviors during AmED sessions compared with alcoholonly drinking sessions in the past 6 months. However, AmED sessions were far less frequent than alcohol-only drinking sessions. Therefore, the relationship between type of drinking session and engagement in risk behaviors is distorted by the large difference in the total number of AmED and alcohol-only drinking sessions, resulting in spurious effect estimates. Not accounting for this difference in the analysis produced misleading results, which in turn, led to conclusions that cannot be supported by the data. A growing body of literature has observed that consumers of AmED engage in more risk-taking behavior while drinking and experience more negative alcohol-related consequences than individuals who do not consume AmED (Brache and Stockwell, 2011; O’Brien et al., 2008; Thombs et al., 2010). The recent study by Peacock and colleagues (2012) sought to discern whether (i) AmED use, or caffeinated alcohol consumption, prompts increased involvement in risk behavior during drinking sessions or (ii) AmED users represent a subpopulation of young adults who have a heightened propensity or proneness for engaging in multiple risk behaviors. These alternative hypotheses are at the center of the current debate about the need to regulate caffeinated alcohol use. Unfortunately, weaknesses of the Peakcock and colleagues study further muddle our understanding of these important issues. Others have already begun citing Peacock and colleagues (2012) as a within-participant study providing evidence of no relationship (Howland and Rohsenow, 2013) or a protective relationship (Pennay and Lubman, 2012) between AmED consumption and engagement in risk behavior. Unfortunately, due to the limitations of the study design and analysis plan, this paper, as written, cannot provide a definitive answer to that question. Any interpretation of the reported findings needs to be conditioned on severe, fundamental limitations. The study design raises several serious concerns. First, the online survey instrument has 303 items, which was reported to be completed by the participants in 10 to 30 minutes. This indicates that on average, respondents took between 2 and 6 seconds to read, comprehend, and respond to each survey item. Given such quick response times to a relatively long survey, questions need to be raised about whether participants responded indiscriminately to survey items merely for a chance to win an Apple iPad2 . Second, during these relatively rapid responses to survey items, respondents were asked to form judgments about whether their engagement in each risk behavior was due to their consumption of AmED. Serious questions can be raised about the validity of survey items that rely on the ability of respondents to correctly attribute their involvement in risk behavior to their AmED consumption (Wood et al., 1997). Such an approach may result in an overestimate of the association if respondents incorrectly attribute their risk behavior to AmED consumption. Conversely, respondent assessment of their attributions could also lead to an underestimate if they incorrectly failed to attribute their risk behavior to AmED consumption. Use of these attribution measures elicited subjective responses that contribute little to the science on caffeinated alcohol consumption. From the Department of Behavioral and Community Health (MER, DLT), University of North Texas Health Science Center, Fort Worth, Texas. Department of Biostatistics (SS), University of North Texas Health Science Center, Fort Worth, Texas. Received for publicationMarch 12, 2013; accepted April 12, 2013. Reprint requests: Matthew E. Rossheim, MPH, Department of Behavioral and Community Health, School of Public Health, 3500 Camp Bowie Blvd., UNT Health Science Center, Fort Worth, TX 76107-2699; Tel.: 941-323-0778; Fax: 817-735-2619; E-mail: matthew.rossheim@live. unthsc.edu Copyright© 2013 by the Research Society on Alcoholism.