Unhealthy Alcohol Use Associated with Obsessive-Compulsive Symptoms: The Moderating Effects of Anxiety and Depression

Abstract

Alcohol use occurs among individuals with obsessive-compulsive symptoms at a rate significantly greater than the general population. In clinical populations, obsessive-compulsive disorder (OCD) and substance use disorders (SUD) have been shown to share neurological substrates, but little is known about the mechanisms underlying substance use in individuals with OCD. Aspects of anxiety and depression frequently contribute to various SUD and are thought to play a role in the relationship between increased substance use and obsessive-compulsive symptomatology. The present research examines the moderating effects of depression and several anxiety-related constructs (anxious arousal, anxiety sensitivity, and social anxiety) on the relationship between health risk resulting from alcohol use and obsessive-compulsive symptoms in university students (n = 178). The physical concerns and social concerns subscales of the Anxiety Sensitivity Index increased the relationship between risky drinking and total OCD symptoms (as measured with the OCI-12). Additionally, general depression and social anxiety significantly increased the relationship between risky alcohol use and the obsessing dimension. All relationships were of a small to medium effect size. These findings help identify emotionally vulnerable subgroups of persons with OCD that may have greater liability for risky alcohol use.

Introduction

Obsessive-compulsive disorder (OCD) is a serious and debilitating psychiatric condition marked by intrusive and unwanted thoughts (obsessions) which are usually accompanied by ritualistic behaviors/avoidance designed to alleviate the obsessions (American Psychiatric Association, 2013). While OCD has recently been re-conceptualized to a class of disorders characterized by repetitive behaviors and/or intrusive ideas, termed the Obsessive-Compulsive and Related Disorders, the disorder is marked by significant anxiety and distress (discussed in McKay, Sookman, Neziroglu, et al, 2015). Anxiety is considered a significant liability for substance use problems (Smith & Book, 2010; Turner, Mota, Bolton, et al, 2018) yet the relationship between substance use disorders and OCD has been under examined.

Those with obsessive-compulsive symptoms have a higher rate of co-occurring substance use than the rate of substance use in the general population (Blom, Koeter, van den Brink, et al, 2011), with alcohol being the most common substance of abuse (Cuzen, Stein, Lochner, et al, 2014). The lifetime comorbidity of alcohol use disorders (AUD) in those with OCD is high with a rate of 24%, approximately five times the rate observed in the general population (5%; Kessler, Berglund, Demler, et al, 2005; Ruscio, Stein, Chiu, et al, 2010) and nearly twice the rate of groups diagnosed with other mood and anxiety disorders, dependent on gender (Torres, Prince, Bebbington, et al, 2006; Wu, Goodwin, Fuller, et al, 2010). OCD and SUD share many overlapping phenomena. Impulsivity and compulsivity are commonly observed in OCD and SUD, particularly in episodes of heightened alcohol consumption (Bø, Billieux, & Landrø, 2016; Robbins, Gillan, Smith, et al, 2012). Both populations commonly show shared deficits in executive functioning and increased levels of stress (Evans, Lewis, & Iobst, 2004; Everitt & Robbins, 2005; Koob & Volkow, 2010; Olley, Malhi, & Sachdev, 2007) as well as overlapping areas of neurofunctional dysregulation (Klugah-Brown, Zhou, Pradhan, et al, 2021).

The onset of OCD is viewed to generally precede the subsequent onset of substance use more broadly, and alcohol use specifically (Gentil, de Mathis, Torresan, et al, 2009; Kessler, et al, 2005; Mancebo, Grant, Pinto, et al, 2009), leading to the assumption that substances are used to self-medicate or cope against obsessive and co-occurring symptomatology. In contrast to this theory, it has been proposed that the relationship between OCD and SUD is enmeshed with each disorder perpetuating the other, ultimately increasing the symptoms of OCD (e.g., Back & Brady, 2008; Brady & Lydiard, 1993; Schuckit, 1996). In fact, there is a robust relation between the severity of alcohol use and the severity of obsessive-compulsive symptoms (Campos, Yoshimi, Simão, et al, 2015; Lima, Pechansky, Fleck, et al, 2005; Mancebo, et al, 2009), especially in the symptom dimension of obsessing as measured with the OCI-R (obsessions of harm, sex, or religion; Brakoulias, Starcevic, Berle, et al, 2013; Torres, Fontenelle, Shavitt, et al, 2016; Torres, et al, 2006). This comorbidity of AUD among treatment-seeking OCD patients has shown an increased risk for poor treatment outcomes, increased impairment, and distress (Angst, Gamma, Endrass, et al, 2005; Fineberg, Hengartner, Bergbaum, et al, 2013), as well as enhanced suicidality (Gentil, et al, 2009).

As previously mentioned, OCD is commonly accompanied by symptoms of anxiety and distress (discussed in McKay, et al, 2015). The inability to tolerate these states is often tied to substance use (Özdel & Ekinci, 2014). In clinical OCD populations, common comorbid conditions include major depression (56.4%), social anxiety (34.6%), generalized anxiety (34.4%), and specific phobia (31.4%; Torres, et al, 2016). Of those diagnosed with OCD, 62% have been found to have at least two co-occurring neurotic disorders, significantly higher than the comorbidity rates of groups with other mood and anxiety concerns (10%). Many of these commonly co-occurring disorders and their related predictors have previously demonstrated unique relationships with alcohol abuse. Anxiety sensitivity (AS), a cognitive risk factor for all anxiety disorders, has been significantly associated with OCD symptom severity (Calamari, Rector, Woodard, et al, 2008; DeMartini & Carey, 2011) and is predictive of the development of AUD (Schmidt, Buckner, & Keough, 2007). Social Anxiety Disorder (SAD) is highly prevalent in both AUD (Buckner, Schmidt, Lang, et al, 2008; Grant, Hasin, Blanco, et al, 2005) and OCD (Torres, et al, 2016) populations. Those with SAD are thought to use alcohol as a means of coping with distress resulting from their concerns with negative evaluation and scrutiny by others (Buckner, et al, 2008). Avoidance of others is commonly seen in OCD populations for similar reasons, mainly fear of rejection by others due to the nature of their obsessional thoughts and more overt compulsive behaviors (Newth & Rachman, 2001). Alcohol use co-occurring with OCD symptoms has recently been linked to substance use coping motives (Bakhshaie, Storch, & Zvolensky, 2021). This relationship remained significant even after controlling for the effect of anxious arousal. Research on the impact of these anxiety constructs on the relationship between obsessive-compulsive symptoms and alcohol misuse is non-existent.

In a study of conditions co-occurring with OCD, the most common comorbidity was major depression, which affected 56.4% of the population sampled (Torres, et al, 2016). Previous work suggests that AUD and depression (MDD) are often intertwined with most individuals diagnosed with either of these conditions having a history with the other disorder (Brière, Rohde, Seeley, et al, 2014). AUD and MDD also serve as predictors of each other with AUD predicting the subsequent onset of MDD and vice-versa. In those with OCD, comorbid depression enhances symptom severity (Brown, Lester, Jassi, et al, 2015) with obsessional problems being a link between the two disorders (Jones, Mair, Riemann, et al, 2018). There has been some evidence that the association of OCD with frequent or heavy drinking becomes marginal when controlling for depression (Wu, et al, 2010), but there is a lack of research on the role of depression in the relationship between OCD and substance use.

This study has two broad aims. The first is to examine alcohol misuse and its association with obsessive-compulsive symptoms. Previous research suggests an association between substance use and total OCD symptoms, but a special relationship with the specific symptom dimension of obsessing. The second aim is to examine the role of various co-occurring anxiety constructs and depression in the relationship between obsessive-compulsive symptoms and alcohol misuse in a sample of young adults. It was hypothesized that there would be significant associations between OCD symptoms, especially the symptom subtype of obsessing, and unhealthy levels of alcohol consumption. Further, it was hypothesized that depression, anxiety, and anxiety related constructs would positively moderate the relationships between obsessing symptoms and substance use.

Methods

Participants

The sample for the current study is a group of college students from a larger cross-sectional study examining the associations bridging stress/emotion and health behaviors at a large university in the southwest United States during the spring of 2019 (Bakhshaie, et al, 2021). The subset used in these analyses was composed of 178 current college students who were purposefully selected for meeting criteria for health risk from alcohol use, 76.4% of whom were female with an average age of 22.32 (SD = 4.37, ranging from 18 to 54 years old). Participants received class credit for their participation and were recruited using fliers and postings on the university’s extra credit website. Exclusion criteria included being below 18 years of age and a lack of English language proficiency. The ethnic and racial diversity of the sample was high, 44.4% Hispanic (n = 79), 26.4% non-Hispanic White (n = 47), 15.2% Asian American and Pacific Islander (n = 27), 9.0% Black (non-Hispanic; n = 16), 3.9% other (n = 7), and 1.1% American Indian/Alaskan Native (n = 2).

Measures

Obsessive-Compulsive Inventory -12 (OCI-12) (Abramovitch, Abramowitz, & McKay, 2021). The OCI-12 is a 12-item self-report scale based on the OCI-R and used to assess obsessive-compulsive symptoms. The OCI-R (Foa, Huppert, Leiberg, et al, 2002) contains 18 response items which assess symptoms of checking, hoarding, obsessing, ordering, neutralizing, and washing. Subsequent to the publication of the OCI-R, hoarding has been removed as a dimension of OCD (American Psychiatric Association, 2013) and investigations have raised concerns about the validity and reliability of the neutralizing factor (Abramovitch, et al, 2021). The OCI-12 removes the three hoarding and three neutralizing items, leaving four subscale scores consisting of checking, obsessing, ordering, and washing. Participants affirm a response pertaining to their level of distress related to these four factors during the past month (e.g., “I repeatedly check doors, windows, drawers, etc.”, “I wash my hands more often and longer than necessary”) on a Likert scale ranging from 0 (not at all) to 4 (extremely). The OCI-12, like the OCI-R, demonstrates good to excellent reliability and validity in both clinical and non-clinical contexts, including use with student populations. In the current study, both the total score and the four subscales were used. The total scale demonstrated excellent internal consistency (α = 0.92). In this study, the sub-scales demonstrated internal consistency that was acceptable to good, checking (α = 0.76), washing (α = 0.81), obsessing (α = 0.85), and ordering (α = 0.86).

Inventory of Depression and Anxiety Symptoms (IDAS) (Watson, O’Hara, Simms, et al, 2007). The IDAS contains 64 self-report items evaluating symptoms of depression and related anxiety disorders rated on a Likert scale from 1 (not at all) to 5 (extremely). The measure contains two broader scales assessing general depression and dysphoria as well as 10 specific symptoms scales (Suicidality, Lassitude, Insomnia, Appetite Loss, Appetite Gain, Ill Temper, Well-Being, Panic, Social Anxiety, and Traumatic Intrusions). In the current study the General Depression (α = 0.91), Social Anxiety (α = 0.88), and Anxious Arousal (α = 0.91) specific symptom scales were used, and all showed good to excellent internal consistency.

Anxiety Sensitivity Index (ASI-3) (Taylor, Zvolensky, Cox, et al, 2007). The ASI-3 is an 18-item measure made to assess anxiety sensitivity (fear related to arousal-related sensations). The ASI-3 is composed of three distinct factors which measure physical, cognitive, and social concerns. Replication in international populations, as well as clinical and non-clinical samples, has shown good reliability and validity along with improved psychometric properties when compared with the original ASI. In this study, the Physical Concerns (α = 0.89), Cognitive Concerns (α = 0.93), and Social Concerns (α = 0.88) subscale scores were used and showed good to excellent internal consistency.

The Alcohol, Smoking, and Substance Involvement Screening Test – Version 3.0 Modified (ASSIST V3.0r) (McNeely, Strauss, Wright, et al, 2014). The ASSIST was originally developed as a screening tool for use in primary care settings by the World Health Organization Assist Working Group and adapted over a five-year period resulting in the ASSIST V3.0. The ASSIST consists of eight items assessing the lifetime and current use of tobacco, alcohol, cocaine, amphetamine-type stimulants, sedatives, hallucinogens, inhalants, opiates, and other unspecified drugs. It also assesses the perceived consequences of and attempts to stop or cut back substance use. The ASSIST V3.0 demonstrates satisfactory construct and concurrent validity. In this study, the ASSIST score on risky alcohol intake was used, excluding the dichotomous questions pertaining to lifetime and intravenous use (range: 7 – 37; α = 0.73).

Procedures

All procedures executed in the presented study were in compliance with the regulations of the institutional review board at the university from which participants were recruited. Participants were required to complete an online form acknowledging the receipt of their informed consent prior to participating in data collection through an internet-based self-report survey. Personally identifiable information which could possibly link participants to their survey responses was not retained after data collection.

Data Analysis

Univariate analyses were conducted to describe the characteristics of the sample in its totality. Correlations were run with SPSS v25. Single-variable moderator analyses were conducted using PROCESS v3.5 (Hayes, 2017) model one (single moderator) to determine if depression and anxiety-related constructs would significantly moderate the relationship between risky alcohol intake and OCS. Moderations were then probed at the mean and ± 1SD (Aiken, West, & Reno, 1991). ASSIST risky alcohol use served as the predictor variable with the outcomes being overall OCI-12 score or the specific sub-dimension of obsessing as measured with the OCI-12. Moderators included three subscales of the IDAS (General Depression, Social Anxiety, and Anxious Arousal), and the three subscale scores of the ASI-3 (Physical, Cognitive, and Social).

Results

Descriptive data

Descriptive statistics and correlations between variables are shown in Table 1. The ASSIST risky drinking subscale ranged from 7 – 37 (M = 14.43, SD = 7.23) with scores greater than six detecting unhealthy alcohol use for women, and scores over seven indicating unhealthy use for men. Within this sample, 100% (n = 178) of participants met the criteria for risky drinking with 91% (n = 162) of the sample endorsing patterns of drinking associated with moderate levels of health risk. When assessed with the OCI-12, 45.8% (n = 80) of participants presented with scores above the cut-off point for likely OCD (15% met the criteria for severe OCD), with scores ranging from 0 – 37 (M = 11.52, SD = 9.05). Risky alcohol use was significantly associated with overall OCD symptomatology (b = 0.23, t (172) = 3.13, p < 0.01 (two-tailed test), 95% CI = [0.00, 0.44]) and each subscale as measured with the OCI-12 (Checking, b = 0.20, t (172) = 2.69, p < 0.01 (two-tailed test), 95% CI = [0.00, 0.39]; Obsessing, b = 0.25, t (172) = 3.42, p < 0.001 (two-tailed test), 95% CI = [0.01, 0.47]; Ordering, b = 0.15, t (172) = 2.00, p = 0.05 (two-tailed test), 95% CI = [0.00, 0.30]; Washing, b = 0.19, t (172) = 2.59, p = 0.01 (two-tailed test), 95% CI = [−0.00, 0.37]).

Table 1.

Bivariate Correlations

		1.	2.	3.	4.	5.	6.	7.	8.	9.	10.	11.	12.
1.	OCI 12	0.92	0.88**	0.85**	0.85**	0.87**	0.23**	0.34**	0.41**	0.33**	0.37**	0.35**	0.42**
2.	OC Checking		0.76	0.66**	0.68**	0.69**	0.20**	0.28**	0.33**	0.27**	0.25**	0.25**	0.32**
3.	OC Obsessing			0.85	0.58**	0.68**	0.25**	0.39**	0.50**	0.40**	0.49**	0.40**	0.46**
4.	OC Ordering				0.86	0.65**	0.15*	0.17*	0.22**	0.24**	0.21**	0.17*	0.31**
5.	OC Washing					0.81	0.19*	0.33**	0.37**	0.20**	0.31**	0.39**	0.33**
6.	Unhealthy Alcohol						0.73	0.28**	0.33**	0.23**	0.23**	0.26**	0.29**
7.	Physical AS							0.89	0.77**	0.59**	0.38**	0.44**	0.35**
8.	Cognitive AS								0.93	0.66**	0.48**	0.47**	0.45**
9.	Social AS									0.88	0.40**	0.32**	0.50**
10.	Depression										0.91	0.62 **	0.68**
11.	Anxious Arousal											0.91	0.66**
12.	Social Anxiety												0.88
	Mean	11.52	2.92	2.87	3.50	2.23	14.43	5.74	5.73	9.27	47.54	13.19	10.21
	SD	9.05	2.58	2.79	2.77	2.40	7.23	5.49	5.97	6.38	14.33	6.18	4.68

Note.

^*indicates p < 0.05, and

^**indicates p < 0.01.

OC symptom dimensions per Obsessive-Compulsive Inventory – 12. Unhealthy alcohol use per the Alcohol, Smoking, and Substance Involvement Screening Test. Physical, cognitive, and social anxiety sensitivity per the Anxiety Sensitivity Index – 3. Depression, anxious arousal, and social anxiety per Inventory of Depression and Anxiety Symptoms. α levels are presented, italicized, on the diagonal. The number labeling each row in the far-left column corresponds to the same variable on each subsequent column.

Moderation analyses

In the models where total OCD symptoms served as the outcome variable, the relationship between risky drinking and OCD symptoms increased as symptoms of both physical and social AS increased. In models where symptoms of the obsessing dimension served as the outcome variable, the relationship between risky drinking and obsessing symptoms increased as symptoms of both physical and social AS increased. Additionally, this relationship increased as symptoms of general depression and social anxiety increased. Please see Table 2 for moderation paths and total model figures.

Table 2.

Moderation Models

	b	SE	t	p	LLCI	ULCI
Model 1 (total OCI-12) Total Model R2 = 0.18, F(3, 165) = 12.14, p < 0.001
RD	0.12	0.10	1.27	0.21	−0.07	0.31
ASI-P	0.45	0.12	3.82	< 0.001	0.22	0.69
RD x ASI-P	0.04	0.02	2.70	< 0.01	0.01	0.07
−1 SD	−0.12	0.15	−0.76	0.45	−0.42	0.19
+1 SD	0.38	0.11	3.52	< 0.001	0.17	0.60
Model 2 (total OCI-12) Total Model R2 = 0.17, F(3, 165) = 11.30,p < 0.001
RD	0.16	0.10	1.73	0.09	−0.02	0.35
ASI-S	0.39	0.10	3.81	< 0.001	0.19	0.59
RD x ASI-S	0.03	0.01	2.33	0.02	0.01	0.06
−1 SD	−0.06	0.16	−0.37	0.71	−0.36	0.25
+1 SD	0.37	0.11	3.50	< 0.001	0.16	0.58
Model 3 (Obsessing) Total Model R2 = 0.21, F(3, 165) = 14.73, p < 0.001
RD	0.04	0.03	1.39	0.17	−0.02	0.10
ASI-P	0.16	0.04	4.59	< 0.001	0.09	0.23
RD x ASI-P	0.01	0.01	2.52	0.01	0.00	0.02
−1 SD	−0.03	0.05	−0.58	0.56	−0.18	0.06
+1 SD	0.11	0.03	3.46	< 0.001	0.05	0.18
Model 4 (Obsessing) Total Model R2 = 0.23, F(3, 165) = 16.19,p < 0.001
RD	0.05	0.03	1.68	0.09	−0.01	0.10
ASI-S	0.15	0.03	4.98	< 0.001	0.09	0.21
RD x ASI-S	0.01	0.00	2.64	< 0.01	0.00	0.02
−1 SD	−0.03	0.05	−0.59	0.56	−0.12	0.06
+1 SD	0.12	0.03	3.71	< 0.001	0.05	0.18
Model 5 (Obsessing) Total Model R2 = 0.28, F(3, 167) = 21.19, p < 0.001
RD	0.05	0.03	2.04	0.05	0.00	0.11
IDAS-GD	0.08	0.01	6.48	< 0.001	0.06	0.11
RD x IDAS-GD	0.00	0.00	1.99	0.05	0.00	0.01
−1 SD	0.02	0.04	0.42	0.67	−0.06	0.09
+1 SD	0.10	0.03	3.71	< 0.001	0.05	0.16
Model 6 (Obsessing) Total Model R2 = 0.25, F(3, 167) = 18.81,p < 0.001
RD	0.05	0.03	1.63	0.11	−0.01	0.10
IDAS-SA	0.24	0.04	5.77	< 0.001	0.16	0.32
RD x IDAS-SA	0.01	0.01	1.99	0.05	0.00	0.02
−1 SD	−0.01	0.04	−0.25	0.80	−0.10	0.08
+1 SD	0.10	0.03	2.96	< 0.01	0.03	0.16

Note. RD is risky drinking as measured with the ASSIST. ASI-P is the physical anxiety sensitivity subscale of the ASI-3. ASI-S is the social anxiety sensitivity subscale of the ASI-3. IDAS-GD is the general depression global factor as measured with the IDAS. IDAS-SA is social anxiety as measured with the IDAS. Obsessing is the level of symptoms of the OCD obsessing dimension as measured with the OCI-12. LLCI and ULCI are the lower and upper limits of the 95% confidence intervals for b. Outcome variables are denoted next to the corresponding model.

Discussion

Though other works have controlled for anxiety and depression in the relationship between obsessive-compulsive symptoms and alcohol use (e.g., Bakhshaie, et al, 2021; Wu, et al, 2010) this is the first attempt to examine the moderating role of these constructs. Like previous work (Brakoulias, et al, 2013; Torres, et al, 2016; Torres, et al, 2006), the subcategory of obsessing (unacceptable/taboo thoughts and mental compulsions) had the strongest association with risky alcohol use specifically. Though the associations were weaker than that of the specific subcategory of obsessing, in this sample total OCD symptoms and each category of OCD dimension symptoms were associated with risky drinking. It has been speculated that this association is due to a common neurological etiology (Klugah-Brown, et al, 2021), and the possibility that in OCD co-occurring with SUD both the development and persistence of each disorder may be dependent on the other (e.g., Brady & Lydiard, 1993; Schuckit, 1996). This is an important area for further investigation since alcohol may be used to cope with intrusive thoughts but may also have a deleterious long-term effect on OCD symptoms (Back & Brady, 2008).

Our results suggest that those who participate in risky alcohol use who also have higher AS experience higher levels of OCS. Alcohol use among those with OCS may be used as a bulwark against AS related cognitive distortions such as the belief that heart palpitations can lead to cardiac arrest and that public displays of anxiety symptoms will result in social rejection. This finding appears to be similar to panic disorder with agoraphobia where physical and social AS concerns are prevalent (discussed in Craske & Barlow, 2008; Westenberg & Liebowitz, 2004). In those with symptoms of panic and agoraphobia, higher levels of AS are associated with increased alcohol use and higher endorsement of beliefs that alcohol can dampen anxiety (Cox, Swinson, Shulman, et al, 1993). Individuals high in AS are more likely to use alcohol to cope with stress as AS raises the likelihood of experiencing symptoms and thought processes associated with anxiety (Stewart, Samoluk, & MacDonald, 1999). The relationship between total OCD symptoms and risky drinking was moderated by both these physical and social anxiety-related concerns as measured with the ASI-3. For both ASI subscales, the relationship was the most significant and of the strongest effect at higher scores. This relationship indicates the need to address alcohol misuse in persons who endorse higher levels of both OCD symptoms and AS. These results potentially guide future research into the creation of an integrated treatment protocol for OCD co-occuring with alcohol abuse by demonstrating that, during the maintenance phase, AS has a greater influence over levels of OCS than alcohol use. Such an effect hints at the need to test the efficacy of interoceptive exposure techniques in the treatment of this population.

In addition to the influence of AS on overall OCS, there were additional moderating relationships unique to the symptom dimension of obsessing. The relationship between risky drinking and the symptom subtype of obsessing was moderated by physically and socially related AS as well as general depression and social anxiety symptoms. For all four of these models OCS increased as the level of the moderating construct rose. The moderating effects were strongest and most significant at the highest level of symptoms of the moderating construct. Previous evidence is mixed on the connection between ASI physical and social concerns and this OCD symptom dimension (e.g., Calamari, et al, 2008; Poli, Melli, Ghisi, et al, 2017; Raines, Oglesby, Capron, et al, 2014). The added relationships with fear of social rejection and depression may be related to the nature of thoughts characteristic of the obsessing dimension. Social anxiety is highly comorbid with OCD symptoms (Torres, et al, 2016), and the sexual/religious obsessions category specifically (Assunção, Costa, de Mathis, et al, 2012). Those with these types of obsessions often have a high level of shame and guilt associated with their thoughts (Shapiro & Stewart, 2011) which can lead to fear and distress related to the intrusions themselves and the potential consequences of others becoming aware of the nature of their thoughts. Those with OCD who endorse high levels of social anxiety show a higher level of impulse-control disorders (Assunção, et al, 2012), another feature common to those with substance use issues. Our results showing these multiple known OCD comorbidities increasing the relationship between risky drinking and symptoms characteristic of this symptom subtype may be another possible explanation for the increased observance of SUD in this subpopulation of OCD.

This study is limited by several factors. First, the low prevalence of clinically significant OCD symptomatology and alcohol dependence means that the examination of obsessive-compulsive disorder and higher-level alcohol use disorders or dependence could not be performed, and thus the findings here should be considered preliminary and warrant further examination with more diverse groups, such as community and clinical samples. Second, the results are self-report in nature, with no interviewer-based clarification for alcohol use, mood states, and obsessive-compulsive symptoms. As previous research has shown that alcohol use increases with the rate of OCD symptom severity (Gentil et al., 2009), it is anticipated that additional research evaluating the moderating role of anxiety-related constructs and mood would result in a wider range of scores on probed interactions from moderator tests. Those in the present sample also endorsed low scores on all anxiety and depression related measures which precluded the interpretation of the probed moderation effects except for those one standard deviation above the mean of each construct (Hayes, 2017). In this sample, results may be affected by a female skew in gender relative to gender distribution in the general population. Previous studies have shown substance use and abuse to be more prevalent in men (National Household Survey on Drug Use and Health, 2019) while anxiety tends to be more significantly associated with alcohol use in female samples (Rohde, Lewinsohn, & Seeley, 1996). There have been conflicting results regarding the likelihood of each gender to have substance use disorders comorbid with OCS (e.g., Gentil, et al, 2009; Torres, et al, 2006; Wu, et al, 2010). Future research should evaluate the influence of gender in moderator analyses, to further refine clinical guidelines for treatment of OCD complicated by alcohol use.

Disordered alcohol use is common in those with high levels of OCS. In this examination of those who met criteria for unhealthy alcohol use, individuals who endorsed higher levels of physical AS simultaneously expressed higher levels of OCS. This was true for overall OCS and the specific symptom dimension of obsessing. Overall OCS as well as symptoms of the obsessing dimension increased along with measures of socially focused AS. Our results extend the historical association of the symptom dimension of obsessing with substance misuse. This symptom constellation’s relationship with alcohol misuse was moderated by many of the tested constructs including AS (both physical and social), depression, and social anxiety, all of which led to increased OCS. This may provide a tentative reason for the commonly observed correlation of substance use with the obsessing symptom dimension. While physical and social AS appear to be risk factors for unhealthy alcohol use in those with higher levels of OCS, the effect of other co-occurring disorders would seem to be heterogeneous, varying by symptom dimension.

Highlights.

• The relationship between unhealthy levels of alcohol use and obsessive-compulsive symptoms increased by both physical and social anxiety sensitivity

• The relationship between unhealthy alcohol use and obsessing symptoms was increased by physical and social anxiety sensitivity as well as depression and social anxiety symptoms

• Clinically, results may inform future treatment development for OCD patients with problematic alcohol use