Distress tolerance in social versus solitary college student drinkers

Highlights

• Reported differences in distress tolerance across types of college student drinkers

• Students varied by self-report but not behavioral distress tolerance measures.

• Solitary binge drinkers had more coping motives than other drinkers.

• Coping mediated the relationship between distress tolerance and alcohol problems.

Abstract

Low distress tolerance has been an inconsistent predictor of alcohol-related consequences in college students, but its relationships to depression and coping motives for alcohol have received stronger support. Research on college students who drink heavily in isolation suggests that this population is more likely to have a greater number and severity of alcohol-related problems, depression, and coping motives. Solitary heavy drinkers were therefore hypothesized to have lower distress tolerance than other drinkers. This study examined differences in self-reported and behavioral distress tolerance across two groups of university students: those who endorsed heavy solitary drinking (20.1%) versus those who endorsed other types of drinking. Students completed a self-report measure (Distress Intolerance Self-Report, or DISR) and behavioral measure of distress tolerance (Paced Auditory Serial Addition Test, or PASAT). Students who reported drinking heavily in isolation differed from other students on the DISR, F(1, 132) = 4.645, p = .033, η² = .034, but not on the PASAT, F(1, 132) = 0.056, p = .813. These students also endorsed more coping motives for alcohol. Distress tolerance did not predict drinking consequences directly, yet a mediation model linking distress tolerance to consequences through coping motives supports previous findings of distress tolerance as a distal, indirect predictor of drinking problems. The unique characteristics of solitary binge drinkers and the significance of distress tolerance as an indirect predictor of alcohol-related consequences are discussed.

Introduction

Despite years of research, college student drinking continues to present a major health problem within the United States. Full-time college students are more likely to drink and drink heavily compared to same-age peers not in college (SAMHSA, 2014), and approximately 1 in 4 students experiences consequences as a result of drinking at some point during their time in college (Wechsler, Dowdall, Maenner, Gledhill-Hoyt, & Lee, 1998). Previous research suggests that certain subsets of students may be especially at risk of serious drinking consequences, and it is therefore important to better identify such students so that more effective steps can be taken to prevent such consequences.

Young adults who report that they drink to manage negative affect may be at greater risk for problematic drinking consequences than those who drink for other reasons. Research has demonstrated that motivation to drink as a means to cope with negative affect tends to predict more problematic drinking than do enhancement and social drinking motives (Cooper, 1994, Cooper et al., 1992, Holyfield et al., 1995, Kuntsche et al., 2005, Rafnsson et al., 2006). Distress tolerance is one factor that is likely to influence coping with negative affect. It reflects a person's likelihood of enduring, attending to, and alleviating perceived emotional discomfort (Simons & Gaher, 2005), and it has predicted a variety of adverse consequences that are purportedly regulated by self-control, such as substance abuse (Grant and Chamberlain, 2014, Zvolensky and Hogan, 2013).

The ability to tolerate distress likely acts on the mechanism of negative urgency (Kaiser, Milich, Lynam, & Charnigo, 2012). Negative urgency – commonly conceptualized as a component of impulsivity – refers to the tendency to act rashly when experiencing adverse emotions such as sadness (Wray, Simons, Dvorak, & Gaher, 2012). Therefore, in a model that predicts substance abuse, negative affect is likely to be experienced as more aversive in people with low distress tolerance who then behave impulsively and engage in risky drinking behaviors due to a sense of negative urgency (Kaiser et al., 2012). Wray et al. (2012) found more support for this model than a direct relationship between distress tolerance and risky drinking behaviors. This model also helps to explain the plethora of indirect relationships that have been suggested between distress tolerance and alcohol abuse, such as evidence from Marshall-Berenz, Vujanovic, and MacPherson (2011), who found that distress tolerance partially mediated the relationship between impulsivity and coping motives among people with PTSD, and evidence from Buckner, Keough, and Schmidt (2007), who found that distress tolerance mediated the relationship between depression and problems with alcohol and cannabis.

The relationship between distress tolerance and negative affectivity is made even more intriguing when it is examined in the context of the motivation to cope with negative affect. Because distress tolerance is related to drinking to cope (Howell et al., 2010, Marshall-Berenz et al., 2011) and coping motives strongly predict alcohol-related problems in college students (Cooper et al., 1992, Kuntsche et al., 2005), the relationship between distress tolerance and alcohol-related problems may be best captured in a mediational model rather than through direct effects (Martens et al., 2008). In addition to assessing differences in distress tolerance across groups of college student drinkers, this study proposes to assess the validity of distress tolerance measures in predicting alcohol-related consequences via coping motives.

Although the majority of college students who drink do not experience many severe negative consequences, the subset of students who drink to cope appear to be at a higher risk of engaging in risky behaviors (e.g., driving while intoxicated) and experiencing negative consequences (e.g., arrests for driving while intoxicated; Wray et al., 2012). Therefore, it is important to identify the characteristics of the students who engage in this pattern of drinking. One such group may be college students who drink heavily in isolation. Previous research has demonstrated that these students are at greater risk for experiencing more and more severe drinking consequences than their social-drinking peers (Christiansen, Vik, & Jarchow, 2002). Compared to social drinkers, solitary heavy drinkers endorse more coping motives, experience more serious consequences such as increased suicidal ideation (Gonzalez, Collins, & Bradizza, 2009), and have greater difficulty regulating negative mood (Gonzales & Skewes, 2013). For these individuals, distress tolerance might exert a greater influence on their alcohol consumption and alcohol-related consequences through negative urgency and coping mechanisms. Therefore, it is likely that college students who drink heavily alone also have lower levels of distress tolerance than students who engage in other patterns of drinking. This study proposes to examine those group differences in distress tolerance.

There are many means by which researchers have been studying distress tolerance over the past decade, using either self-report questionnaires that assess one's perceived ability to withstand distress or measures that capture the behaviorally demonstrated capacity to tolerate distress (Leyro, Zvolensky, & Bernstein, 2010). While both types of measures have been linked to alcohol-related problems (Daughters et al., 2005a, Daughters et al., 2005b, Simons and Gaher, 2005), the literature is mixed on their efficacy at predicting problems directly (Howell et al., 2010). As was suggested above, distress tolerance appears to impact alcohol use and alcohol-related problems indirectly, via mechanisms such as negative urgency and coping motives. The relationship between distress tolerance and alcohol-related problems is made especially problematic by the myriad of ways in which tolerance is assessed, as well as the types of distress captured by such measures. For example, although self-report measures of distress tolerance correlate well with one another and behavioral measures of distress tolerance correlate well with one another (Anestis et al., 2012), these different modalities do not always hang together and, indeed, may be addressing different aspects of distress (e.g., perceptions of how one will respond to distress versus behavioral responses during physiological discomfort; McHugh et al., 2011). The question then becomes which of these two types of measures are better at measuring the component of distress that best predicts alcohol-related problems.

Comprehensive reviews of the utility of distress tolerance measures to predict specific psychopathology, such as alcohol abuse, are lacking (Anestis et al., 2012). Simons and Gaher (2005), Wray et al. (2012), and Buckner et al. (2007) reported evidence between low self-reported distress tolerance and more alcohol problems, but Howell et al. (2010) did not find a significant, direct relationship between low distress tolerance and a greater number of problems. With respect to behavioral measures, one study found evidence that low distress tolerance can predict subsequent dropout from a substance abuse treatment program (Daughters et al., 2005a, Daughters et al., 2005b) and another linked low distress tolerance to failed attempts to stop smoking (Brown, Lejuez, Kahler, Strong, & Zvolensky, 2005). Other studies have found evidence of an indirect relationship between behavioral measures of distress tolerance and problematic alcohol use (Daughters et al., 2009, Gorka et al., 2012). However, to our knowledge, there has not yet been evidence to suggest a direct relationship between any behavioral measure of distress tolerance and alcohol-related problems per se.

Even so, there is a robust theoretical argument for a relationship between behavioral distress tolerance and alcohol-related problems. For example, Anestis et al. (2012) found support for the predictive validity of self-report and behavioral measures of distress tolerance (e.g., the Distress Tolerance Scale and Distress Tolerance Test, respectively) for impulsive behaviors, which are important to the realm of alcohol consumption because facets of impulsivity (e.g., negative urgency) have been shown to predict maladaptive alcohol consumption and alcohol-related problems (Kaiser et al., 2012, Shin et al., 2012). The conflicting results span different types of distress tolerance measures, and the evidence suggesting that both may be useful in predicting alcohol-related consequences highlights the importance of a multimodal assessment of distress tolerance.

In an attempt to address these concerns, this study compared solitary and social drinkers across a self-report measure and a behavioral measure of distress tolerance. The first measure, the distress intolerance self-report (DISR) scale, is brief measure of distress tolerance that includes ten items from four established measures of distress and anxiety tolerance (i.e., Anxiety Sensitivity Index, Discomfort Intolerance Scale, Distress Tolerance Scale, and Frustration Discomfort Scale). In a recent study, these ten questions were shown to have the highest factor loadings for a single “distress intolerance” factor in both clinical and nonclinical samples (McHugh & Otto, 2012). This finding was important because the succinct assessment of distress tolerance is very useful given the lack of a gold standard for self-reported distress tolerance among the variety of measures available (McHugh and Otto, 2012, McHugh et al., 2011). Although the DISR has not yet been used to predict alcohol-related problems, the Discomfort Intolerance Scale (DIS) and Distress Tolerance Scale (DTS) have been independently linked to alcohol-related problems (Howell et al., 2010, Simons and Gaher, 2005). The Anxiety Sensitivity Index (ASI) and the Frustration Discomfort Scale (FDS) have not yet been used to predict alcohol-related problems, but the ASI has predicted coping motives for alcohol use (Stewart & Zeitlin, 1995), and the FDS was shown to be related to similar constructs such as “internet addiction” and problems with self-control (Harrington, 2005, Ko et al., 2008).

Although we know of no published studies that have examined the relationship between behaviorally demonstrated distress tolerance and alcohol-related problems, there are a few studies that have used behavioral distress tolerance measures to examine the relationship between distress tolerance and alcohol use. The Paced Auditory Serial Addition Test, or PASAT, has typically been the measure used in such studies. For example, Gorka et al. (2012) reported using the PASAT to predict early termination from a drug and alcohol treatment facility. These authors found that subjects with lower distress tolerance (as indicated by premature termination of the PASAT) were more likely to leave treatment early than individuals with greater distress tolerance. Daughters et al. (2009) found a relationship between early PASAT termination and greater amounts of alcohol use among Caucasian (but not ethnic minority) youths aged 9–13 (Daughters et al., 2009).

As mentioned above, the PASAT has not yet been used to predict drinking consequences in a college student population; in fact, to our knowledge there are no published studies that have attempted to link any behavioral measure of distress tolerance to alcohol-related problems per se. The aforementioned evidence suggests that the PASAT, though previously untested, is perhaps the behavioral measure of distress tolerance that has the greatest likelihood of predicting alcohol-related problems, and therefore it may be the behavioral measure that best reflects differences in distress tolerance across solitary and social college student drinkers.

In summary, college students who drink heavily in isolation seem to be at greater risk of experiencing more and more severe alcohol-related consequences. There is also literature to suggest that they have higher rates of depressive symptoms and more motives to drink to cope with negative affect. Despite the fact that distress tolerance has been linked to negative affect, coping motives, and alcohol-related problems, differences in distress tolerance have not been compared across groups of solitary heavy drinkers and other drinkers. Therefore, this study hypothesizes that distress tolerance – assessed via a self-report and a behavioral measure – will vary across groups of college student drinkers, such that individuals who drink heavily alone will have greater levels of distress tolerance than college students who endorse other drinking patterns.

A second set of analyses will examine the ability of both distress tolerance measures to predict alcohol-related consequences in college students. Although there is some evidence that both self-report and behavioral measures of distress tolerance can predict amount of alcohol use, there is less evidence for a direct relationship between such measures and alcohol-related consequences. Furthermore, much of the research seems to suggest an indirect relationship between distress tolerance and alcohol-related consequences, as previous studies have shown that low distress tolerance is related to greater incidence of depression and coping motives, which subsequently predict more drinking consequences. Therefore, specific predictions about a direct relationship between distress tolerance and alcohol-related consequences are tenuous, whereas a mediation model linking distress tolerance, coping motives, and alcohol-related problems is likely to be more robust. This set of analyses will test the hypothesis that greater coping motives will significantly mediate the relationship between low distress tolerance and alcohol-related problems.