The role of behavioral inhibition, perceived parental rearing, and adverse life events in adolescents and young adults with incident obsessive-compulsive disorder

Abstract

The role of behavioral inhibition (BI), parental rearing, and adverse life events in the development of obsessive–compulsive disorder (OCD) is unresolved. We prospectively examined whether BI and perceived parental rearing predicted the subsequent first onset of OCD and whether BI moderated the association between perceived parental rearing/adverse life events and the first onset of OCD. Data come from a prospective-longitudinal study among community adolescents and young adults (aged 14–24) who were followed up over 10 years (N = 2210). OCD and adverse life events were assessed with the DSM-IV/M-CIDI. BI and parental rearing were assessed by self-report. In adjusted logistic regressions, BI (risk ratio, RR = 8.8,95% confidence interval, CI,[3.3,23.2]), both the social (RR = 2.6,95%CI[1.04,6.3]) and nonsocial fear (RR = 4.5,95%CI[2.3,8.6]) components, and paternal rejection (RR = 3.9,95%CI[1.6,9.7]) predicted OCD. Social BI moderated the association of adverse life event (RR = 11.98,95%CI[1.8,82.1]) and paternal overprotection (RR = 5.5,95%CI[2.1,14.8]) with OCD. Nonsocial fear BI moderated the association between paternal emotional warmth (RR = 0.37,95%CI[0.1,0.99]) and OCD. BI and paternal rejection were associated with an increased risk of first onset of OCD. Individuals with high social BI who experienced paternal overprotection or any adverse life events may profit from early targeted prevention. Emotional warmth may buffer the association between nonsocial fear BI and OCD.

Introduction

Obsessive-compulsive disorder (OCD) is listed among the 10 most debilitating mental and physical disorders worldwide (Murray & Lopez, 1996) and is associated with substantial psychosocial impairment (Fineberg et al., 2013) and an increased risk of subsequent mental disorders (Hofer et al., 2018). Over the last few decades, a number of studies have focused on possible risk factors that might contribute to the development of OCD. Similar to other internalizing disorders, behavioral inhibition (BI), adverse life events, and parental rearing (Brander et al., 2016, Coles et al., 2006) are of special interest from the perspective of primary prevention. From a diathesis–stress perspective, BI might be viewed as placing individuals along a continuum of vulnerability to develop OCD in the face of stress that comes with exposure to critical environmental factors such as adverse life events or dysfunctional parental rearing.

BI is a dispositional factor that is thought to be involved in the pathogenesis of OCD (Coles et al., 2006). BI has been described as a tendency to react to social and nonsocial novelty with behavior characterized by high levels of restraint, withdrawal, and avoidance (Kagan, Reznick, Clarke, Snidman, & Garcia-Coll, 1984). There is evidence for a genetic contribution to behavioral inhibition (Dilalla, Kagan, & Reznick, 1994).

Appreciating OCD as belonging to the wider spectrum of anxiety disorders (see Diagnostic and Statitical Manual of Mental Disorders (5th ed.; DSM-5; American Psychiatric Association, 2013)) the etiological pathways to anxiety suggest that the avoidance of novel situations and stimuli may undermine an individual's ability to learn that anxiety-provoking situations are not threatening or unmanageable (Chorpita and Barlow, 1998, Foa and Kozak, 1986). Thus, BI may foster avoidance behavior and prevent opportunities in which disconfirming information about the threat value of these situations/stimuli can be obtained. Previous research has identified associations between BI and anxiety disorders in general (Beesdo et al., 2010, Knappe et al., 2011, Wittchen et al., 2000). Three population-based studies suggested that there is a relationship between a behaviorally inhibited temperament in childhood and obsessive-compulsive symptoms. Muris, Meesters, and Spinder (2003) and Muris, Merckelbach, Schmidt, Gadet, and Bogie (2001) found an association between self-reported BI and obsessive-compulsive symptoms in a cross-sectional study of two samples of adolescents. In another cross-sectional study of students, Coles et al. (2006) demonstrated that retrospective self-reports of childhood BI were associated with the frequency of current obsessive-compulsive symptoms. At odds with these studies are results from two prospective population-based studies of BI and threshold OCD (Caspi et al., 1996, Rapee, 2014). In these studies, childhood BI did not predict OCD in adolescents (age 15; Rapee, 2014) or young adults (age 21; Caspi et al., 1996). Additionally, Hudson, Dodd, and Bovopoulos (2011) failed to find a link between laboratory-assessed BI and OCD in a cross-sectional study of preschoolers, age 3–4 years. However, Rapee (2014) and Hudson et al. (2011) identified only four children with OCD, and thus it is likely that their analyses were not sufficiently powered. Furthermore, Caspi et al. (1996) did not find behavioral styles at 3 years to be predictive for OCD or any anxiety disorder, including social phobia, at 21 years. To conclude, only three studies have examined whether BI increased the risk of subsequent OCD and findings were hampered by methodological limitations.

Research has provided evidence that BI is bidimensional, consisting of a social and a nonsocial fear component. It has been suggested that these components should be treated as different constructs as they have been shown to be relatively independent and to have distinct correlates (Dyson et al., 2011, Kochanska, 1991, Majdandzic and van den Boom, 2007, Mick and Telch, 1998, Neal et al., 2002, Poole et al., 2017, Rubin et al., 1997, Schofield et al., 2009). Using a student sample, Coles et al. (2006) showed associations between obsessive and compulsive symptoms and both the social and nonsocial fear components of BI. Whether these dimensions are both relevant in threshold OCD remains unclear.

Among environmental factors, parental rearing is thought to be related to anxiety disorders (e.g. Asselmann, Wittchen, Lieb, & Beesdo-Baum, 2016; Knappe et al., 2009b; Möller, Nikolic, Majdandzic, & Bögels, 2016). In a recent systematic review of environmental risk factors for OCD, Brander et al. (2016) concluded that self-reported parental overprotection and paternal rejection are associated with OCD. All of the reviewed studies, however, were flawed in that they (a) used cross-sectional data and did not consider temporal sequence of parental rearing and OCD, making it impossible to disentangle risk factors from concomitants or consequences, and (b) were based on clinical samples or students, limiting generalizability of the results.

Increasingly, experiencing adverse life events has been recognized as a possible important negative environmental factor in OCD (for a review, see Brander et al., 2016). In recent prospective analyses based on the Early Developmental Stages of Psychopathology (EDSP) Study (Asselmann, Wittchen, Lieb, Perkonigg, & Beesdo-Baum, 2017), traumatic events at baseline were associated with an increased risk of OCD. Valleni-Basile et al. (1996) showed in a prospective study that undesirable life events increased the risk of OCD in adolescents. In the Dunedin Multidisciplinary Health and Development Study, retrospectively assessed childhood stressors, such as sexual and physical abuse, number of residence changes, but not the loss of a parent, were associated with increased odds of OCD, compared to healthy controls (Grisham et al., 2011).

Consistent with a diathesis–stress model is the suggestion that BI is one manifestation of a genetic vulnerability factor, which in interaction with an environmental stressor (e.g., adverse life events or parental rearing) may play a role in the development of an anxiety disorder (see Craske, 1997; Muris & Merckelbach, 2000; Ollendick & Hirshfeld-Becker, 2002; Turner & Beidel, 1996). Little is known concerning these interaction effects in predicting OCD. Some evidence has suggested that overprotective parenting might moderate the impact of BI on the obsessive-compulsive symptoms of OCD (Coles et al., 2006), but we have not identified any study reporting on interaction effects between BI and adverse life events or parental rearing in predicting threshold OCD.

Although the reported studies provide valuable knowledge regarding the association between BI and OCD and between perceived parental rearing and OCD, we note the following limitations: Studies (a) used convenience samples such as student samples, which were not representative of the general population, (b) assessed symptoms instead of reliable OCD diagnoses, (c) used cross-sectional data, which makes it difficult to disentangle risk factors from concomitants or consequences, or (d) were not sufficiently powered. Whether the interaction between temperament and other environmental factors that have been hypothesized to be relevant to OCD, such as adverse life experiences and perceived parental rearing, predicts incident OCD has not been investigated so far.

As such, the goal of this study was to contribute to the understanding of the longitudinal association between BI/parental rearing and OCD among adolescents and young adults. Using data from the 10-year prospective population-based EDSP Study, we addressed the following questions: Is BI associated with an increased risk of OCD? Furthermore, are both the social and nonsocial fear subscales of BI associated with an increased risk of OCD? Are maternal and paternal parenting styles associated with an increased risk of OCD? And finally, given the interest in the potential interaction between temperament and environmental factors, does BI moderate the potential association between adverse life experiences/parental rearing and OCD?