Sexual addiction 25 years on: A systematic and methodological review of empirical literature and an agenda for future research

Highlights

• Research related to sexual addiction and compulsive sexual behaviors has proliferated in recent years.

• The ICD-11 has elected to include a novel diagnosis of Compulsive Sexual behavior Disorder.

• A systematic review revealed 415 empirical studies of compulsive sexual behavior in the past 25 years.

• The majority of studies made use of cross-sectional designs in non-clinical populations.

• At present, there is almost no empirical basis for the treatment of compulsive sexual behaviors.

Abstract

In 1998, Gold and Heffner authored a landmark review in Clinical Psychology Review on the topic of sexual addiction that concluded that sexual addiction, though increasingly popular in mental health settings, was largely based on speculation, with virtually no empirical basis. In the more than two decades since that review, empirical research around compulsive sexual behaviors (which subsumes prior research about sexual addiction) has flourished, ultimately culminating in the inclusion of a novel diagnosis of Compulsive Sexual Behavior Disorder in the eleventh edition of the World Health Organization's International Classification of Diseases. The present work details a systematic review of empirical research published between January 1st, 1995 and August 1st, 2020 related to compulsive sexual behaviors, with a specific focus on evaluating the methodologies of that literature. This review yielded 371 papers detailing 415 individual studies. In general, the present review finds that, although research related to compulsive sexual behaviors has proliferated, much of this work is characterized by simplistic methodological designs, a lack of theoretical integration, and an absence of quality measurement. Moreover, the present review finds a virtual absence of high-quality treatment-related research published within this time frame. Implications of these findings for both clinical practice and future research are discussed.

Introduction

A little more than two decades ago, Gold and Heffner (1998) published a pivotal paper in Clinical Psychological Review about the nature of “sexual addiction.” At the time, the concept—popularized by Patrick Carnes fifteen years earlier (Carnes, 1983)—was gaining increasing attention in clinical settings and in popular media. However, after thoroughly reviewing the available empirical literature, Gold and Heffner concluded that, “The literature on this topic [sexual addiction] consists largely of theory and conjecture based almost entirely on clinical observation rather than on research findings” (Gold & Heffner, 1998, p. 379).

Since 1998, much has changed in clinical understandings of various phenomenon that might be related to sexual addiction. Research into compulsive, addictive, impulsive, or out-of-control sexual behavior has proliferated. The fifth edition of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (hereafter: DSM-5) only narrowly excluded the diagnosis of Hypersexual Disorder, and the World Health Organization recently elected to include the diagnosis of Compulsive Sexual Behavior Disorder (hereafter: CSBD) in the forthcoming eleventh edition of the International Classifications of Diseases (hereafter: ICD-11). In short, the knowledge base regarding non-paraphilic out-of-control sexual behaviors has indeed moved substantially forward in the past two decades.¹ As such, the purpose of the present work was to conduct a systematic review of literature published since 1995 on behaviors that might fall under the greater umbrella of sexual addiction, provide an updated summary of research with a particular focus on methodology, and discuss areas of consideration for both clinical translation and research focus.

Out of control sexual behaviors have been described by physicians since the early 19th century (Rush, 1812), though descriptions of people with excessive or voracious sexual appetites can be found in various histories and mythologies that long predate such medical attention. However, it was not until the latter half of the 20th century that these ideas began to receive academic attention, with some of the earliest works on out of control sexual behaviors labeling the presentation of these phenomena “hypersexuality.” Although the early literature and case reports describing hypersexuality often lacked rigor, substance, or clarity (Orford, 1978), these works underscored the potential for hypersexuality to be of research and clinical importance and led to novel conceptualizations of problematic sexuality, especially that of “sexual addiction.”

In 1983, Patrick Carnes first introduced the concept of sexual addiction to clinical audiences in his work Out of the Shadows: Understanding Sexual Addiction, which intensified interest in excessive sexual behaviors. This work provided clinical descriptions, etiological speculations, and treatment recommendations based on Carnes's experiences in treating what he had labeled sexual addiction (Carnes, 1983). At the time, little empirical work supported his claims, and initial reactions to Carnes' work and the general notion of sexual addiction were mixed (c.f., M. Levine & Troiden, 1988). Even so, a steady trickle of research described excessive or out of control sexual behaviors followed Carnes' early writings.

In the early to mid-1990s, a number of studies began to emerge detailing various aspects of “sexual addiction,” “sexual compulsivity,” and “compulsive sexual behavior.” Much of this work was in the form of theoretical speculation (Goodman, 1992; Pincu, 1989; Sunderwirth, Milkman, & Jenks, 1996) or case reports (Coleman, 1991, Coleman, 1992). However, some empirical studies were conducted around this time (Hecker, Trepper, Wetchler, & Fontaine, 1995; Lundy, 1994), and the notion of sexual compulsivity (particularly among men who have sex with men) was the subject of various studies (Kalichman et al., 1994; Kalichman & Rompa, 1995).

Beginning in the late 1990s, research into out of control sexual behaviors began to accelerate, in part due to the advent of the internet and its subsequent uses for sexual purposes (Cooper, Putnam, Planchon, & Boies, 1999; Delmonico & Carnes, 1999). Specifically, some empirical studies emerged at this time with a clear focus on the possibility for some people to get caught in patterns of compulsive or addictive sexual behaviors facilitated by technological mediums (Cooper, Scherer, Boies, & Gordon, 1999; Delmonico, 1997; Delmonico & Carnes, 1999). Termed “cybersex,” online activities such as chat rooms, digital infidelity, and internet pornography consumption were posited as being potential expressions of out-of-control sexual behavior (Delmonico, 2002). Indeed, empirical work during the same time supported this notion, showing that some people did report that their online sexual behaviors were excessive (Cooper, Delmonico, & Burg, 2000; Delmonico & Miller, 2003).

At the time, many of these problems were attributed to what Cooper and colleagues termed the “Triple A Engine,” or the Anonymity, Affordability, and Accessibility of online sexual behaviors (Cooper, 1998; Cooper, McLoughlin, & Campbell, 2000; Cooper, Scherer, et al., 1999). According to this model, the internet presented a unique space for out-of-control sexual behaviors due to the ease at which sexual content online could be accessed (accessibility), the privacy afforded by at-home internet service (anonymity), and the relatively low cost (affordability) of online sexual behaviors in comparison to more traditional means of engaging in such behaviors (i.e., buying pornography videos or magazines; soliciting prostitutes; frequenting strip clubs). Although intuitive, this model was not without its critics, and the only empirical study to actually test the model did not find support for it (Byers, Menzies, & O'Grady, 2004). Even so, research into out-of-control sexual behaviors, including, but not limited to, online sexual activities, continued to progress.

Interest into CSBs reached a tipping point during the development of the American Psychiatric Association's DSM-5. At this time, the diagnosis of Hypersexual Disorder was proposed for inclusion in the DSM-5 (Kafka, 2010). Conceptualized as being related to both addictive disorders and sexual desire disorders, Hypersexual Disorder followed highly similar diagnostic criteria to those seen in other syndromes outlined in earlier versions of the DSM, namely Pathological Gambling, which was considered an impulse control disorder in the DSM-IV and later classified as an addictive behavior called Gambling Disorder in the DSM-5 and substance use disorders. Specifically, this proposed diagnosis required that the recipient of the diagnosis experience repetitive and impairing sexual behaviors, urges, or fantasies that consumed excessive time, were in response to negative emotions or stress, were uncontrollable despite efforts to control or reduce these symptoms, and that disregarded the well-being or safety of the self or others (Kafka, 2010). Importantly, to meet criteria for Hypersexual Disorder, sexual behaviors could not be attributable drug use or to the side effects of prescription drug use or manic episodes as these behaviors must have occurred as a result of the sexual urges themselves instead of being caused by external factors.

Field trials of Hypersexual Disorder found the disorder to be useful, reliably applied, and generally well-received by clinicians (Reid et al., 2012). However, Hypersexual Disorder was ultimately excluded from the DSM-5 due to a range of concerns related to diagnostic accuracy, moral and cultural confounds, and general skepticism from the psychiatric community (Kafka, 2014). Regardless, in the time since that decision, research into this domain has flourished, ultimately culminating in diagnostic recognition of specific cases of out-of-control sexual behaviors, as we explore below.

The landscape of research related to out-of-control sexual behaviors again changed with the proposal and subsequent inclusion of Compulsive Sexual Behavior Disorder (hereafter: CSBD) in the ICD-11. Similar to the excluded diagnosis of Hypersexual Disorder for the DSM-5, CSBD refers to persistent, repetitive engagement in sexual behaviors that results in impairment in one's life in addition to failed attempts to reduce or stop such behaviors (Kraus et al., 2018). The diagnostic criteria for this new disorder are included in Table 1.

The inclusion of CSBD in the ICD-11 has not been without its controversy, generating more commentary and critique than any other novel diagnosis considered for the ICD-11 (Fuss et al., 2019). This controversy in the public sphere mirrors the longstanding history of controversy around CSB related notions more generally, such as the use of such diagnoses to stigmatize some sexual behaviors and debates about how and when to consider certain sexual behaviors problematic. Dating back to early work on sexual addiction, attempts to diagnose excessive sexual behaviors as a discrete syndrome have been met with criticism for a variety of reasons (Levine & Troiden, 1988; Levine, 2010). Some critiques have focused on the potential of such diagnoses to stigmatize non-conforming sexual behaviors (Halpern, 2011), arguing that the choice to conceptualize certain behaviors as problematic is inherently biased against non-traditional sexual behaviors. Others have argued that the evidence in support of such disorders is lacking (Ley, Prause, & Finn, 2014; Prause, Janssen, Georgiadis, Finn, & Pfaus, 2017). Still, others have noted that application of such diagnoses may be complicated by extraneous variables such as client or therapist religiosity or client sexual orientation (Droubay & Butters, 2019; Grubbs, Kraus, Perry, Lewczuk, & Gola, 2020; Klein, Briken, Schröder, & Fuss, 2019). Despite these controversies, the recognition of this new disorder has highlighted the need for rigorous research into out-of-control sexual behaviors and has generated new interest within clinical psychology with regards to the appropriate classification of such behaviors (for example, see: Grubbs, Kraus, et al., 2020; Klein et al., 2019).

At present, CSBD has been included in the ICD-11 as an impulse control disorder. Even so, there is ongoing debate as to whether or not CSBs are rightly considered addictive, impulsive, or compulsive. The Working Group on Obsessive-Compulsive and Related Disorders proposed a conceptualization of CSBD as an impulse control disorder because individuals exhibiting this behavior repeatedly failed to resist such sexual impulses, drives, and urges which were, at some point, rewarding, despite long-term harm (Kraus et al., 2018). Building on this conceptualization, the diagnostic criteria are written in such a way that essential features of CSBD are intended to be less rigid because they rely less on arbitrary cut-offs or symptom counts, rely more on objective indicators of impairment and dysregulation, and are meant to support the exercise of clinical judgment in assigning the diagnosis (First, Reed, Hyman, & Saxena, 2015). However, one of the current challenges is drawing the clear distinction between addiction, impulsivity, and compulsivity, as these concepts overlap because impulses or urges to engage in repetitive behaviors are core features of addiction.

Strong impulses or compulsions are distinctive from addictions in that addiction is typically associated with additional features caused by the substance or behavior, including tolerance and withdrawal, whereas in impulse control disorders (e.g., kleptomania and pyromania) or compulsive disorders (e.g., obsessive compulsive disorder) the assumption is that the pathology lies primarily within the individual. Some evidence suggests that persons with CSB exhibit significantly higher levels of impulsiveness than healthy controls (Antons et al., 2019; Mechelmans et al., 2014; Miner, Raymond, Mueller, Lloyd, & Lim, 2009; Reid, Berlin, & Kingston, 2015), however, similar associations have been observed in people with other addictive disorders, such as gambling (Specker, Carlson, Christenson, & Marcotte, 1995), alcohol (Lejoyeux, Feuché, Loi, Solomon, & Adès, 1999), or cocaine (Li et al., 2008) use disorders. In short, impulsivity is a feature of most addictive disorders, which suggests that distinguishing between impulse control disorder and addiction is not a straightforward endeavor.

Further, there are some studies on patients with CSBD that show comparable levels of impulsivity to patients with addictive disorders (gambling and drug use disorders) while also finding that only half of patients in all these groups show clinically elevated scores on measures of impulsivity (Reid, Cyders, Moghaddam, & Fong, 2014). Other studies have found no difference compared to healthy controls in terms of general impulsivity, but only specific increased sensitivity for erotic cues among individuals with CSBD (Gola et al., 2017). Such specific increases of sensitivity toward one category of cues is typical for addictive behaviors and has been well described by incentive salience theory as a main factor underlying craving in addiction (Olney et al., 2018). These studies, most of which use addiction related experimental paradigms, have sought to better understand the neurobiological underpinnings of CSB. In general, these studies have found that CSB is associated with altered functioning in brain regions and networks implicated in sensitization, habituation, diminished impulse control, and reward processing in patterns like substance, gambling, and gaming addictions (Kowalewska et al., 2018; Stark et al., 2018). In addition, studies looking at dopamine replacement therapies used for individuals with Parkinson's disease have also produced similar patterns of impulse control problems that resemble CSBD (Weintraub et al., 2010). As such, there is some research which suggests that, although CSBD fits the characteristics of an impulse control disorder, it also exhibits some of the same behavior specific impulses and reward processing as an addictive disorder.

Similar critiques can be made with regards to role of compulsivity in CSB and other addictive disorders. Specifically, addictions are inherently characterized by compulsions to seek a substance or behavior that provides neurological rewards (Koob, 2017). Although compulsions may characterize other disorders as well (e.g., compulsive checking in obsessive compulsive disorder), the compulsivity associated with CSBs more closely resembles the compulsivity seen in addictive disorders. Future research on CSBD may find more evidence to support an addictive classification in which case CSBD may follow a similar trajectory as pathological gambling which has transitioned from an impulse control disorder characterized by compulsive engagement in gambling to an addictive classification renamed gambling disorder in the DSM-5.

Moving beyond the substantive debates about the correct conceptualization of CSBD, there is evidence that the majority of laypeople view CSBs, such as excessive pornography use, as addictions (Burke & MillerMacPhee, 2020; Perry, 2019; Taylor, 2019). Many people believe that sexual behaviors can be addictive (Lang and Rosenberg, 2017, Lang and Rosenberg, 2018), and when describing their own experiences with CSBs, addiction seems to be the most commonly used explanation (Cavaglion, 2008, Cavaglion, 2009; Wordecha et al., 2018). In short, public perceptions of out-of-control sexual behaviors are most often couched in terms associated with addiction, regardless of the formal distinctions still debated in the research literature. This latter difference is not surprising, as scientists are focused on finding the most parsimonious theoretical model to explain a phenomenon while members of the public are typically interested in getting help for the phenomenon regardless of how it may be classified.

There is also considerable evidence that many people may interpret their sexual behaviors as addictive, compulsive, or out-of-control, even when their behaviors do not objectively appear to be dysregulated or clinically impairing (Burke & Haltom, 2020; Grubbs & Perry, 2019). That is, several studies (particularly in the U.S.) have found that many people are likely to experience moral qualms about their sexual behaviors which, in turn, seem to promote feelings of addiction or compulsion (Griffin et al., 2016; Hook et al., 2015; Hook et al., 2015). Recently, in reference to pornography use specifically, this has been described as moral incongruence, which specifically refers to feelings that arise when one's actual pornography use behaviors are not in line with one's beliefs about pornography use (Grubbs & Perry, 2019; Grubbs, Perry, Wilt, & Reid, 2019). Importantly, this moral incongruence can appear in clinical settings among those seeking help for their self-perceived out-of-control sexual behaviors (Cantor et al., 2013; Kraus & Sweeney, 2019).