Scientific interest exists in better understanding obsessivity, compulsivity, impulsivity, and addictive behaviors given their clinical relevance. There exists overlap and co-occurrence among these constructs, and obsessive–compulsive disorder (OCD) and obsessive–compulsive-related disorders (OCRDs), impulse control disorders (ICDs) and addictions show similarities with respect to phenomenology, pathophysiology and comorbidity (Fontenelle et al., 2011). Given similarities and new data gathered, changes in diagnostic classification, theoretical models and clinical interventions have occurred over time.

On the one hand, of the different disorders currently considered as OCRDs (e.g., body dysmorphic disorder, hoarding disorder, trichotillomania, and excoriation (skin picking) disorder) in the fifth edition of the Diagnostic and Statistical Manual (DSM-5) (APA, 2013), trichotillomania was previously classified as an ICD (APA, 1994). However, it has recently been proposed that these disorders may present with different levels of impulsivity and/or compulsivity (Fineberg et al., 2022, Fontenelle et al., 2021). Some authors have suggested categorizing OCD as an addiction, since individuals with this disorder may present with high levels of impulsivity, biased probabilistic reasoning, and risky decision-making, as occurs in the case of substance and non-substance addictions (Denys et al., 2004, Grassi et al., 2016, Holden, 2001). On the other hand, some consider conditions such as gambling disorder (GD), internet gaming disorder (IGD), compulsive sexual behavior disorder (CSBD), and compulsive buying-shopping concerns (CBSCs) as behavioral addictions given multiple similarities with substance use disorders (Grant et al., 2006, Grant et al., 2010, Kraus et al., 2016, Vereczkei et al., 2022).1 However, international diagnostic manuals such as the DSM and the the International Classification of Diseases (ICD) initially classified them as ICDs (American Psychiatric Association, 1994, World Health Organization., 2019), and only GD has subsequently come to be categorized as “Substance Related and Addictive Disorders” in the DSM-5 (APA, 2013). The ICD-11 correlate of IGD has been classified as a “Disorder due to Addictive Behaviors” (WHO, 2020). However, some studies report similarities between Substance Related and Addictive Disorders and OCD (Figee et al., 2016). The multiple theoretical proposals and dialog among researchers and clinicians highlight complexities regarding classification considerations.

Compulsivity, that may be defined as involving “the performance of repetitive and functionally impairing overt or covert behavior without adaptive function, performed in a habitual or stereotyped fashion, either according to rigid rules or as a means of avoiding perceived negative consequences” (Fineberg et al., 2014), is a transdiagnostic factor relevant across these clinical conditions (Figee et al., 2016, Fineberg et al., 2014). It has been hypothesized that these disorders show different levels of compulsivity (Kim et al., 2017), making some of them more phenotypically compulsive (such as IGD), and others, (such as GD, CSBD, and CBSCs), more phenotypically impulsive (Vats et al., 2021).

Regarding compulsivity, IGD has been associated with cognitive inflexibility (Kim et al., 2017, Klugah-Brown et al., 2021, Morris and Voon, 2016). Obsessive-compulsive (OC) tendencies have been linked to GD, with genetic underpinnings implicated (Bottesi et al., 2015, Scherrer et al., 2015). GD-related deficits have been reported in specific cognitive domains, including cognitive flexibility, attentional set-shifting, and attentional bias (Leeman and Potenza, 2012, van Timmeren et al., 2018). Gender-related differences have been reported in individuals with GD in terms of compulsivity (Mallorquí-Bagué et al., 2021). CBSCs have also been associated with OC tendencies, with individual differences noted. Specifically, some authors have suggested the existence of three profiles: compulsive-impulsive buyers, impulsive excessive buyers, and ordinary buyers (Yi, 2013). Finally, OC tendencies have been identified in individuals with CSBD. Specifically, it has been suggested that they contribute to addiction, especially in individuals who use the internet for the purpose of finding sexual partners (Levi et al., 2020). However, CSBD may be more linked to impulsivity than compulsivity (Bőthe et al., 2019, Raymond et al., 2003).

Although compulsivity may contribute importantly to behavioral addictions, few studies have examined relationships using the same measures across different addictions simultaneously. The clinical groups included in the present study with the exception of CBSCs (GD, IGD, and CSBD) were selected because (with the exception of CBSCs) they are the only three entities accepted by at least one of the international diagnostic manuals (ICD / DSM). Even though CSBD is included in the ICD-11 as an impulse control disorder and CBSCs are not included in the DSM or ICD, both CSBD and CBSCs have been described as being addictions in nature (Brand et al., 2020, Kraus et al., 2016, Stark et al., 2018), with the terminology of “other specified disorder due to addictive behaviors” being a possible ICD-11 term that may be applied. Other entities (e.g., problematic use of social media) were not significant clinical concerns at the time which data collection commenced, and few people have been seen at our clinic to date for such concerns.

The main aims of the present study were: a) to compare OC levels between patients treatment seeking for GD, CBSCs, CSBD and IGD; b) to explore correlations between OC features (OC, harm-avoidance, and persistence tendencies) and clinical features (onset and duration of the behavioral addictions, as well as substance use); c) to explore associations between sociodemographic measures and OC features; and d) to explore through a clustering procedure the existence of empirical clusters among the treatment-seeking patients based on OC features.

Based on the cumulative evidence reported in the scientific literature (Kim et al., 2017), we hypothesized that the different behavioral addictions would show different severity levels related to OC, persistence, and harm-avoidance, and that the existence of differentiated latent empirical clusters would be identified and relate to quantitative and qualitative OC classes (Bottesi et al., 2015, Scherrer et al., 2015).