Previous studies have identified a positive relationship between dietary restraint and alcohol use. However, it is unclear whether heavier drinking is associated with higher dietary restraint per se, or restraint combined with a tendency towards disinhibition. The aim of the present study was to examine alcohol use behaviours in women classified using both restraint and disinhibition scores. Forty-four young female social drinkers gave self-reported measures of their drinking behaviour, including frequency and quantity of alcohol consumed and frequency of drunkenness and binge drinking. Attentional bias for alcohol-related stimuli was also assessed using a dot probe detection task. Finally, the Temptation and Restraint Inventory was used to investigate whether preoccupation with drinking might underlie the relationship between dietary and drinking behaviours. Women classified as both highly restrained and disinhibited tended to report more episodes of drunkenness, showed an attentional bias for alcohol-related words, and had greater cognitive preoccupation with drinking compared to other dietary groups. These data suggest that a tendency towards overeating (disinhibition) combined with attempts at restriction is associated with increased alcohol use behaviours, perhaps due to a greater preoccupation with alcohol.
It is well known that women with eating disorders, especially bulimia, have more drug and alcohol problems compared with women without eating disorders (Russell, 1979). However, there is also evidence to suggest an association between drinking and eating habits in non-clinical populations of women. For example, dietary restraint is associated with higher levels of alcohol consumption by young adult women (Stewart, Angelopoulos, Baker & Boland, 2000). Furthermore, dieting frequency and severity has been reported to be positively associated with alcohol use in women students (French et al., 1995 and Krahn et al., 2005), and it has been suggested that dieting to lose weight may be a risk factor for the development of alcohol abuse (French et al., 1995 and Striegel Moore and Huydic, 1993). In support of this, a prospective study of adolescents found that frequency of dieting was predictive of future alcohol intake (Krahn, Piper, King, Olson, Kurth & Moberg, 1996). This is significant because greater alcohol intake by dieters means that they may be more susceptible to alcohol-related health problems compared with their non-dieting counterparts.
Although a positive association between dieting and alcohol has been established, it is unclear whether unhealthy drinking practices are associated with restriction of food intake versus the disinhibitory tendencies that are also often a component of dieting behaviour. This is because there has been no study to date that has measured these behaviours separately. Previous studies have either failed to use validated measures of dieting or have used the Revised Restraint Scale (RRS) (Herman & Polivy, 1980) to measure dietary restraint (Stewart et al., 2000 and Xinaris and Boland, 1990). An issue that arises here is that the RRS measures the tendency for eating to be disinhibited as well as conscious restriction of eating and so does not provide a pure measure of restraint. The aim of the present study was to examine the relationship between measures of drinking behaviour and both dietary restraint and disinhibition. Restraint was measured using the restraint scale of the Dutch Eating Behaviour Questionnaire (van Strien, Frijters, Bergers & Defares, 1986). Tendency towards disinhibition was measured using the disinhibition scale of the Three Factor Eating Questionnaire (Stunkard & Messick, 1985).
To assess drinking behaviour, we asked young women student drinkers to self-report their alcohol use according to frequency and quantity of consumption. This was done to distinguish between frequent light drinkers and infrequent binge drinkers who may be more at risk of alcohol-related health problems (Wechsler, Davenport, Dowdall, Moeykens & Castillo, 1994). In addition to these self-report measures, we took a measure of attentional bias towards alcohol-related stimuli using a visual dot probe detection task. There is evidence to suggest that the extent to which drug-related stimuli are able to capture the attention of drug users provides an index of drug use patterns (Bradley et al., 2003, Ehrman et al., 2002 and Lubman et al., 2000). For example, enhanced reaction times to dot probe stimuli replacing alcohol-related pictures has been shown in heavy but not light social drinkers (Field et al., 2004 and Townshend and Duka, 2001). According to recent theories of drug use, this is because stimuli associated with drug use acquire motivational significance and attract the attention of the drug user (e.g. Robinson & Berridge, 1993). Because previous studies have found an attentional bias in heavy alcohol users towards alcohol-related picture and word stimuli, two dot probe tasks were used, one with pictures and one with words.
Finally, we wanted to investigate the mechanisms that might underlie any relationships between dietary restraint, disinhibition, and drinking behaviour. One possibility is that alcohol-related cognitions play a role in the initiation and maintenance of drinking behaviour. In support of this, it has been shown that individual differences in emotional preoccupation with alcohol are predictive of self-reported consumption and actual laboratory consumption of alcohol in women but not in men (Marczinski, Bryant & Fillmore, 2005). Therefore, we administered the Temptation and Restraint Inventory, which includes items tapping difficulties in controlling alcohol intake, preoccupation with thoughts of drinking, and worries about controlling drinking (Collins & Lapp, 1992).
