The relationship of alcohol use to weight loss in the context of behavioral weight loss treatment

Highlights

• Most adult participants of a behavioral weight loss treatment consumed low to moderate levels of alcohol at baseline.

• Those participants who did consume alcohol at baseline had meaningfully reduced their alcohol intake by the end-of-treatment.

• End-of-treatment alcohol intake was correlated with end-of-treatment intake of non-alcohol energy.

• Behavioral impulsivity and change in alcohol intake interacted to predict weight loss.

Abstract

Despite common wisdom that reducing alcohol intake will facilitate weight loss, little research has examined whether participants in behavioral weight loss treatments actually decrease their alcohol intake, or whether reduced alcohol intake relates to weight loss outcomes in this context. This study examined the relationship of alcohol use to energy intake excluding alcohol and to weight in 283 overweight and obese adults participating in a 26-session behavioral weight loss treatment. The majority of participants consumed low to moderate levels of alcohol at baseline. Participants who consumed alcohol at baseline meaningfully reduced their alcohol intake by end-of-treatment. Alcohol use did not relate to weight at baseline or end-of-treatment when controlling for relevant demographic variables, and change in alcohol use was unrelated to weight change in the overall sample during treatment. However, end-of-treatment alcohol intake did relate to end-of-treatment energy intake excluding alcohol. In addition, behavioral impulsivity and change in alcohol intake interacted to predict weight loss, such that decreases in alcohol intake were associated with greater percent weight loss at end-of-treatment for participants with higher levels of impulsivity. Alcohol consumption may lead to overeating episodes, and highly impulsive individuals may be at risk for increased energy intake during or after episodes of drinking. Therefore, the recommendation to reduce alcohol intake in the context of behavioral weight loss treatment seems warranted, particularly for individuals with high levels of impulsivity.

Introduction

Reducing alcohol intake is a common recommendation for participants in weight loss programs, with the expectation that making such reductions will improve weight loss outcomes. For instance, government agencies (United States Department of Agriculture, 2015), health organizations (American Heart Association, 2014), and popular weight loss programs (Mayo Clinic, 2014) explicitly recommend reducing alcohol intake during weight loss attempts. Such prescriptions seem logical given that alcohol is energy-dense and that alcohol consumption can trigger impulsive behavior (Marczinski, Abroms, Van Selst, & Fillmore, 2005), with the potential to increase risk for disinhibited eating (e.g. Polivy & Herman, 1976). While there is mixed support for the hypothesis that dietary restraint (the use of cognitive strategies rather than satiety to limit energy intake; Herman & Polivy, 1980) enhances the effects of alcohol on disinhibited eating, consistent findings indicate that alcohol use is associated with increases in short-term energy intake (Caton, Nolan, & Hetherington, 2015).

Currently, available cross-sectional and longitudinal evidence regarding the general relationship between alcohol use and weight management is inconsistent (Gruchow et al., 1985, Liu et al., 1994, Sayon-Orea et al., 2011), though some recent literature suggests that only heavy or frequent alcohol consumption, rather than light-to-moderate consumption, contributes to elevated weight (Traversy & Chaput, 2015). Furthermore, the amount of alcohol consumed by individuals seeking behavioral weight loss treatment is unclear.

The degree to which individuals adopt the recommendation to reduce alcohol intake during weight loss treatment is also unknown. One study (Egger, Bolton, O'Neill, & Freeman, 1996) found that men participating in a group-based behavioral weight loss intervention were successful in losing weight, and that they had reduced their alcohol intake at the end of the treatment. On the other hand, another study (Collins, Morgan, Warren, Lubans, & Callister, 2011) found that men in an internet-based weight loss intervention were successful in losing weight, but that they had not reduced their alcohol intake at the end of the treatment. Neither study examined whether change in alcohol use was directly related to weight loss. In addition, there are no studies investigating whether women participating in behavioral weight loss treatments reduce their alcohol intake.

In addition to the lack of evidence regarding changes in alcohol use in the context of behavioral weight loss, there is no existing research evaluating whether reductions in alcohol intake relate to weight management outcomes. While research has not directly investigated relationships between change in alcohol intake and weight control over time, several experimental studies suggest that alcohol use may make weight control more challenging. For example, alcohol consumption has been shown to stimulate appetite (Yeomans & Phillips, 2002) and lead to increased food intake when consumed prior to or with a meal (Yeomans, 2010). Consumption of alcohol has also been linked to increased desire for and intake of high fat, savory foods (Schrieks et al., 2015). The recommendation to reduce alcohol intake in the context of weight-loss interventions intends to address these risks, but there is insufficient knowledge about how alcohol intake relates to weight, weight loss, or energy intake prior to or after a weight loss attempt.

It is also possible that changes in alcohol intake may differentially relate to weight outcomes across individuals. Certain individual characteristics may influence the importance of reductions in alcohol as a treatment target, and knowledge of these factors can assist in tailoring interventions. For instance, impulsivity, defined as a personality trait involving a lack of reflectiveness and planning, rapid decision making, and carelessness (Schalling, 1978), may be a relevant factor to consider in the relationship between alcohol use and weight outcomes. It is widely understood that self-report measures assess impulsivity as a stable trait, while behavioral measures assess impulsivity as a variable state; however, both types of measures tend to be positively correlated (Meule, 2013). Self-reported impulsivity (and particularly the subdomain of attentional impulsivity) is strongly related to overeating and obesity (Mobbs et al., 2010, Murphy et al., 2014), as well as to alcohol abuse (Dick et al., 2010). Research also suggests that both obese individuals (Nederkoorn, Smulders, Havermans, Roefs, & Jansen, 2006) and individuals who abuse alcohol (Verdejo-Garcia, Lawrence, & Clark, 2008) display higher levels of behavioral impulsivity. Furthermore, behavioral impulsivity is related to relapse following both obesity and alcohol treatment (Cox et al., 2002, Elfhag and Rössner, 2005). In the general population, both self-reported and behavioral impulsivity relate to unhealthy eating behaviors such as eating in response to external cues and negative emotional states and making food choices based on taste rather than health, all of which are associated with increased BMI (Jasinska et al., 2012). In this study, we choose to focus on behaviorally-measured impulsivity, as there are no studies examining alcohol use and behavioral impulsivity in the context of weight loss treatment to date.

In addition to research suggesting that impulsivity relates to risk for substance use, alcohol intake acutely increases behavioral impulsivity (Dougherty, Marsh-Richard, Hatzis, Nouvion, & Mathias, 2008). High levels of impulsivity may therefore compound the risk for alcohol-related energy intake; highly impulsive individuals may be more likely to consume alcohol, which may result in a higher risk for further impulsive behavior, such as unplanned eating. For this reason, reducing alcohol intake might be a particularly important target for highly impulsive individuals undergoing weight loss treatment to prevent these overeating events. To date, no research has evaluated the role of impulsivity in eating behaviors during periods of alcohol consumption.

The current study sought to examine how changes in alcohol use over time in behavioral weight loss treatment relate to weight loss outcomes after one year of treatment. We anticipated overall reductions in alcohol intake at end-of-treatment, consistent with program recommendations. Further, we hypothesized that alcohol intake would positively relate to energy intake excluding alcohol and to weight at both baseline and end-of-treatment time points. In addition, we expected that change in alcohol intake during treatment would be associated with change in energy intake excluding alcohol as well as change in weight, as reductions in alcohol intake should facilitate reductions in overall calorie intake. Finally, we hypothesized that there would be an interaction between baseline levels of behavioral impulsivity and change in alcohol intake in predicting weight change, such that individuals with higher levels of impulsivity who decreased their alcohol intake would see the most weight loss benefit.