Research 256 (2017) 417–422

Contents lists available at ScienceDirect

Psychiatry Research

journal homepage: www.elsevier.com/locate/psychres

Predictors of illness symptoms in patients with obsessive compulsive MARK disorder

Lillian Reumana, Ryan J. Jacobyb, Shannon M. Blakeya, Bradley C. Riemannc, Rachel C. Leonardc, ⁎ Jonathan S. Abramowitza, a Department of and Neuroscience, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA b Department of Psychiatry, Massachusetts General Hospital, Boston, MA, USA c Rogers Memorial Hospital, Oconomowoc, WI, USA

ARTICLE INFO ABSTRACT

Keywords: Illness anxiety and OCD symptoms appear to overlap in their presentation as well as in other conceptually OCD important ways (e.g., dysfunctional cognitions). Little research, however, has directly examined these putative Illness anxiety relationships. The present study examined the extent to which illness anxiety symptoms were associated with OCD symptom dimensions and relevant cognitive factors in a large treatment-seeking sample of patients with OCD. Patients completed a battery of self-report measures of OCD and health anxiety symptoms and related cognitive biases. Results from analyses indicated that illness anxiety symptoms were associated with harm obsessions and checking rituals, as well as with the tendency to overestimate threat and responsibility for harm. Illness anxiety was not associated with perfectionism. Conceptual and clinical implications of these findings are discussed.

1. Introduction HC/IAD and OCD (e.g., 15% in Bienvenu et al., 2000). OCD is hetero- geneous, with the themes of obsessions and compulsions varying from Illness anxiety entails a preoccupation with the unsubstantiated patient to patient; research (e.g., Foa and Kozak, 1995) and clinical belief that one has, or is imminently developing, a serious medical observations suggest that obsessional of illness, (e.g., HIV condition (Taylor and Asmundson, 2004). The belief may be based on a resulting from contamination), and other somatic concerns (e.g., ser- catastrophic misinterpretation of harmless bodily sensations (e.g., ious medical problems) are among the most common thematic pre- headache or pulled muscle), the presence of a benign physical condition sentations. Among patients with OCD, the prevalence of obsessions (e.g., hives), or a previous health issue that is no longer active (e.g., a focused on bodily symptoms (i.e., somatic obsessions) is as high as 34% treated heart condition). To manage illness worry, the person engages (e.g., Rasmussen and Tsuang, 1986). Such obsessions are ordinarily in unnecessary behaviors designed to provide reassurance of good accompanied by washing, checking, and reassurance-seeking rituals health, such as excessively checking one's body (including its waste that serve to reduce anxiety and provide reassurance of safety or good products), repeatedly seeking reassurance (e.g., on the internet), and health. obtaining multiple unnecessary medical evaluations. Although most Although authors have recognized overlaps between the signs and people in the general population occasionally entertain illness concerns, symptoms of illness anxiety and obsessions and compulsions in OCD in its severe form illness anxiety results in substantial personal distress (e.g., Deacon and Abramowitz, 2008; Hollander et al., 2005), IAD was and interference with work/school, relationship, and social functioning, not included among the Obsessive-Compulsive and Related Disorders in which indicates the presence of illness (IAD; formerly DSM-5 (APA, 2013). Yet, the preoccupation with fears of illness in IAD known as hypochondriasis [HC]). Prevalence rates of severe illness is comparable in form and function to obsessional thoughts in OCD in anxiety range from 0.2% to 4.5% (e.g., Bleichhardt and Hiller, 2007) that both have intrusive, repetitive, anxiety-provoking qualities and are with similar overall rates for men and women. subjectively resisted (APA, 2013). Similarly, checking and reassurance- Illness anxiety symptoms are commonly observed in individuals seeking behaviors in IAD are similar to compulsive checking rituals in with obsessive-compulsive disorder (OCD; e.g., Abramowitz et al., OCD: both are performed in response to unrealistic perceptions of 2007) and research suggests a significant rate of comorbidity between threat and reinforced by the reduction in anxiety (albeit temporary)

⁎ Correspondence to: Department of Psychology and Neuroscience, University of North Carolina at Chapel Hill, Campus Box 3270 (Davie Hall), Chapel Hill, NC 27599, USA. E-mail address: [email protected] (J.S. Abramowitz). http://dx.doi.org/10.1016/j.psychres.2017.07.012 Received 2 February 2017; Received in revised form 1 June 2017; Accepted 4 July 2017 Available online 05 July 2017 0165-1781/ © 2017 Published by Elsevier Ireland Ltd. L. Reuman et al. Psychiatry Research 256 (2017) 417–422 they frequently engender. They also prevent the natural extinction of uncertainty regarding how to best classify illness anxiety (IAD) and anxiety and thereby complete a vicious cycle that characterizes both OCD (e.g., Abramowitz and Jacoby, 2015; Cosci and Fava, 2015). De- OCD and IAD (Abramowitz and Braddock, 2008; Deacon and spite appearing as a somatoform disorder in DSM-5, some have argued Abramowitz, 2008). that IAD is more closely related to OCD than to the other somatoform In addition to these phenomenological overlaps, illness anxiety and conditions (e.g., Deacon and Abramowitz, 2008). When the phenom- OCD symptoms appear to share cognitive factors that figure promi- enological and conceptual overlaps between OCD and illness anxiety/ nently in empirically supported conceptual approaches to under- IAD are considered along with the scarcity of empirical research on this standing and treating these conditions (e.g., Abramowitz and Braddock, topic (and limitations of the existing studies), one recognizes the need 2008; Abramowitz et al., 2009). Several studies show that certain to more carefully elucidate the relationship between illness anxiety and cognitive distortions characteristic of OCD (i.e., “obsessive beliefs”) are OCD symptoms. also present in illness anxiety, including the tendency to overestimate Recently, Solem et al. (2015) examined associations between illness threat and responsibility, the need for certainty, and the need for per- anxiety and obsessive-compulsive symptoms among a large sample of fectionism (e.g., Wheaton et al., 2010a; Wheaton et al., 2010b; Deacon patients with OCD and found that nearly one third of the patients and Abramowitz, 2008; Melli et al., 2014). Patients with OCD, for ex- scored above the established cutoff for clinically significant health an- ample, overestimate the probability of becoming sick from germs (e.g., xiety symptoms. Hedman et al. (2017) also examined associations be- Wheaton et al., 2010a). Similarly, those with illness anxiety over- tween obsessive-compulsive symptoms and illness anxiety among pa- estimate the probability of serious illnesses and view their as tients with OCD. Importantly, Hedman and colleagues also examined overly catastrophic (e.g., Weck et al., 2012). The need for certainty and these associations among patients with severe illness anxiety (SHA) and the intolerance for mistakes are also observed to be a motivator of found differences between OCD and SHA patients on various symptom compulsive checking and reassurance-seeking in both illness anxiety measures (e.g., illness behaviors). A strength of Hedman and colleagues’ and OCD (e.g., Abramowitz and Braddock, 2008; Tolin et al., 2003; study is the inclusion of both patients with OCD and patients with SHA; Wheaton et al., 2010a). however, both of these aforementioned studies offered a sub-optimal Another cognitive bias that is present in both OCD and illness an- assessment of OC symptoms, as neither study included a dimensional xiety is anxiety sensitivity (AS)–the tendency to anxiety-related assessment tool of OCD. body sensations based on mistaken beliefs about their presence and Although one study to date (Fergus and Russell, 2016) has adopted meaning (Otto et al., 1998; Raines et al., 2014; Wheaton et al., 2012). a dimensional approach by examining associations between a measure Such mistaken beliefs fall into three domains: (a) physical concerns of illness anxiety and the obsessive-compulsive symptom dimensions, (e.g., When my heart races it means I am having a heart attack), (b) the researchers used a community sample to test their hypotheses. cognitive concerns (e.g., When my thoughts race, I worry I am going Fergus (2014) had previously examined associations between obsessive crazy), and (c) concerns about the social consequences of anxiety (e.g., beliefs and AS in relation to both illness anxiety and obsessive-com- If I tremble in front of others, they will think I am incompetent; Taylor, pulsive symptoms. Results revealed that AS appeared to be a shared set 1999). Wheaton and colleagues (2012) found an association between of beliefs between illness anxiety and obsessive-compulsive symptoms, the physical concerns domain of AS and certain OCD symptoms; namely whereas obsessive beliefs seemed to be more relevant to obsessive contamination and harm obsessions and checking rituals. They posited compulsive symptoms than to illness anxiety. Limitations of these stu- that individuals with these OCD symptoms misappraise physical dies (2014, 2016), however, included use of a community sample and symptoms of anxiety as indicating contamination, sickness, and harm. combining belief domains. Thus, further additional examination of Similarly, numerous studies show that illness anxiety is associated with these constructs in a clinical sample is necessary. misinterpretations of arousal-related body sensations as signs of med- Accordingly, the aim of the current study was to examine the extent ical illness (e.g., Deacon and Abramowitz, 2006; Taylor and to which illness anxiety symptoms were associated with OCD symptoms Asmundson, 2004). and relevant cognitive factors in a large sample of treatment-seeking Despite the overlaps in symptom presentation and cognitive factors, patients with OCD. To improve upon previous work, we used a di- as well as the presence of illness anxiety in many patients with OCD, mensional measure of OCD symptom severity that allowed us to de- there is a dearth of research directly examining the relationships among termine associations between illness anxiety symptoms and empirically illness anxiety and OCD symptoms. Abramowitz and colleauges (1999) established thematic OCD symptom dimensions (e.g., contamination, compared OCD patients with and without health concerns and found harm). On the basis of previous conceptual and empirical work, we that whereas the presence of illness anxiety symptoms was not asso- predicted that illness anxiety symptoms would be associated with harm ciated with the severity of OCD symptoms overall, OCD patients with and contamination obsessions and checking rituals. We also hypothe- health concerns reported more somatic and harm obsessions and sized that beliefs relating to the overestimation of threat, the need for checking rituals relative to non-health concerned OCD patients. certainty and intolerance of mistakes, as well as the tendency to fear the Surprisingly, fears of illness related to contamination were not elevated physical consequences of anxiety-related body sensations would be among the health-concerned patients. This study, however, was hin- associated with illness anxiety symptoms. dered by limitations in the assessment procedures. In another study, Abramowitz, Olatunji, and Deacon (2007) found that patients with OCD had greater illness anxiety than did those with social , specific 2. Methods , and generalized anxiety disorder. However, patients with OCD scored similarly to those with HC on measures regarding the feared 2.1. Participants likelihood of becoming ill. This study, however, used a combined sample of individuals with anxiety disorders of which only a fraction (n Participants were 132 consecutively referred treatment-seeking = 18; 11%) had OCD. patients (52.3% female; n = 69) with a primary diagnosis of OCD who The seeming overlaps between illness anxiety and OCD are of con- presented for treatment at the Obsessive-Compulsive Disorders Center ceptual and clinical consequence. Abramowitz et al. (1999), for ex- at Rogers Memorial Hospital in Oconomowoc, Wisconsin. Of the 132 ample, found that among individuals with OCD, health concerns were participants, 112 were enrolled in residential treatment and 20 in the associated with poorer insight into the senselessness of obsessions and partial hospitalization program. On average, participants were 31.29 compulsions; and poor insight is known to predict attenuated response years old (SD = 12.05; range = 18 – 64). The sample was 89.4% White, to both psychological and pharmacological treatment for OCD (e.g., 2.3% African American, 3.0% Asian, 3.6% Latino/Hispanic, and 0.7% Eisen et al., 2001). These overlaps are also of interest given continued Indian. One individual did not report their racial/ethnic background.

418 L. Reuman et al. Psychiatry Research 256 (2017) 417–422

2.2. Measures my heart beats rapidly”). Participants rate their agreement with each statement on a 0 (very little)to4(very much) scale; higher scores in- 2.2.1. Yale-Brown obsessive compulsive scale - self-report version (Y-BOCS- dicate greater AS. This widely used measure has three factors/subscales SR; Goodman et al., 1989a, 1989b) (calculated according to Taylor et al., 2007): fears of physical sensa- A measure of global OCD symptom severity, the Y-BOCS assesses tions (e.g., racing heart), fears of cognitive symptoms of anxiety (e.g., obsessions (items 1–5) and compulsions (items 6–10) on the following dizziness, racing thoughts), and concerns about the social consequences five parameters: (a) time spent, (b) interference, (c) distress, (d) re- of displaying the symptoms of anxiety (e.g., embarrassment). Reliability sistance, and (e) control. Scores on the Y-BOCS range from 0 to 40; of the ASI subscales in the present sample ranged from 0.77 to 0.80. higher scores indicate greater symptom severity. The self-report version (Y-BOCS-SR) has good reliability and validity (Steketee et al., 1996). 2.3. Procedure 2.2.2. Dimensional obsessive-compulsive scale (DOCS; Abramowitz et al., 2010) Prior to admission to an OCD treatment program, all patients The DOCS is a 20-item self-report instrument that measures the completed a telephone assessment with a trained intake staff member. severity of the four most consistently identified theme-based OCD The clinical director of the Obsessive-Compulsive Disorder Center symptom dimensions: (a) contamination obsessions and decontamina- (BCR) reviewed the results of this assessment and determined if the tion rituals, (b) harm obsessions and checking rituals, (c) symmetry/ patient was appropriate for admission. Upon admission, each patient ordering obsessions and compulsions, and (d) unacceptable (taboo) completed an initial in-person diagnostic evaluation with an experi- thoughts and neutralizing rituals. Within each symptom dimension, five enced psychiatrist. The severity rating from the self-report version of items (rated 0–4) assess the following parameters: (a) time occupied by the Yale-Brown Obsessive-Compulsive Scale (YBOCS; Goodman et al., obsessions and rituals, (b) avoidance behavior, (c) associated distress, 1989a, 1989b) was used to determine the presence of DSM-IV OCD (d) functional interference, and (e) difficulty refraining from compul- (APA, 2000). Each participant then completed a packet of self-report sions. The DOCS subscales exhibit excellent convergent validity and questionnaires, including the measures described above. As part of the reliability in clinical samples (α = 0.94–0.96; Abramowitz et al., 2010). admissions process, patients provided consent (either in person or on- Reliability of the DOCS subscales in the present sample was excellent (α line) to allow their responses to the study measures to be used for both = 0.92–0.96). clinical and research purposes. The Human Subjects Committee and the Rogers Center for Research and Training approved the consent proce- 2.2.3. Obsessive beliefs questionnaire (OBQ; Obsessive Compulsive dures and study measures. Cognitions Working Group, 2001) The OBQ is a 44-item self-report tool that assesses cognitive factors (i.e., obsessive beliefs) relevant to OCD. It contains three subscales: (a) 2.4. Statistical analyses threat overestimation and responsibility (OBQ-RT), (b) importance and control of thoughts (OBQ-ICT), and (c) need for certainty and perfection We used the following statistical approach to test our hypotheses. (OBQ-PC). The instrument has good validity, internal consistency, and First, we examined descriptive statistics and measures of central ten- test-retest reliability (Obsessive Compulsive Cognitions Working Group, dency for the study variables, and conducted independent samples t- 2001). Reliability of the OBQ subscales in the present sample was ex- tests to examine possible gender differences on the WI. Second, we cellent (α ranged from 0.93 to 0.95). computed Pearson correlation coefficients between the WI and other study variables to examine patterns of zero-order associations between 2.2.4. Beck inventory-II (BDI-II; Beck et al., 1996) illness anxiety and demographic variables, OCD symptom dimensions, The BDI-II is a 21-item self-report measure that assesses the severity cognitive factors, and the symptoms of depression and anxiety. Third, of various components of depression. Scores of 10 or less are considered we computed two linear regression analyses. The first regression ex- normal; scores of 20 or greater suggest the presence of clinical de- amined symptomatic predictors of illness anxiety (WI): OCD symptom pression. The BDI-II has excellent reliability and validity and is widely dimensions (DOCS subscales) and associated distress (BAI, BDI). The used in clinical research (Beck et al., 1996). second regression examined cognitive predictors of the WI: obsessive beliefs (OBQ subscales) and AS (ASI). 2.2.5. Beck anxiety inventory (BAI; Beck et al., 1988) The BAI assesses 21 common symptoms of clinical anxiety (e.g., indigestion, nervous) as experienced over the past month. Respondents 3. Results indicate the degree to which they have recently been bothered by each symptom during the past month ranging on a scale from 0 “Not at all” to 3.1. Descriptive statistics and preliminary analyses 3 “Severely.” The BAI assesses anxiety symptoms independently from depression symptoms. The measure exhibits good reliability and va- Means and standard deviations on all study measures appear in lidity (Beck et al., 1988). Table 1. The sample's mean Y-BOCS-SR score indicated severe OCD symptoms. The sample's mean WI score indicated mild to moderate 2.2.6. Whiteley index (WI; Pilowsky, 1967) health anxiety (e.g., Conradt et al., 2006), and 43.2% (n=57) ex- The WI is a 14-item self-report measure that assesses hallmark ceeded the cutoff for SHA (i.e., scored > 5 on the WI; Hedman et al., features of illness anxiety (e.g., “Do you think there is something ser- 2015). The mean DOCS total score was comparable to a similar sample iously wrong with your body?”). Respondents answer a series of Yes/No of treatment-seeking patients with OCD (Abramowitz et al., 2010). questions and receive one point for each affirmative answer. Total There were no gender differences on the WI, t (130) = −0.02, p = scores range from 0 to 14, and higher scores indicate higher levels of 0.99. There was also no association between age and health anxiety health anxiety. The WI is the most widely used measure of illness an- (WI), r = −0.09, p = 0.33. xiety and showed good reliability in the present sample (α = 0.85). We computed a correlation between the BDI and BAI, which often overlap, to determine whether to use both measures in the analyses that 2.2.7. Anxiety sensitivity index (ASI; Reiss et al., 1986) follow. Scores were moderately correlated (r = 0.55), indicating re- The ASI is a 16-item self-report measure of beliefs about the dan- lated but distinct constructs. Accordingly, in the analyses that follow, gerousness of arousal-related body sensations (e.g., “It scares me when both measures were considered separately.

419 L. Reuman et al. Psychiatry Research 256 (2017) 417–422

Table 1 Table 3 Means and standard deviations on all study measures. Predicting WI scores from measures of symptoms and cognitions.

Measure M (SD) Predicting WI β tp sr2 N = 132 Model 1: Symptoms Y-BOCS-SR (total) 26.67 (6.64) BDI 0.01 0.22 0.83 < 0.01 DOCS BAI 0.09 2.31 0.02 0.05 Contamination 7.70 (6.97) DOCS - Contamination 0.05 0.85 0.40 < 0.01 Responsibility for harm 6.72 (5.96) DOCS - Responsibility for harm 0.20 3.06 0.003 0.08 Unacceptable thoughts 11.02 (6.15) DOCS - Unacceptable thoughts −0.08 −1.17 0.25 0.01 Symmetry 6.05 (5.74) DOCS – Symmetry −0.04 −0.59 0.56 < 0.01 OBQ-44 Model 2: Cognitions Responsibility/Threat 63.46 (24.97) OBQ-Responsibility/Threat 0.04 2.25 0.03 0.04 Perfectionism/Certainty 68.64 (25.23) OBQ-Perfectionism/Certainty −0.01 −0.88 0.38 < 0.01 Importance/Control of Thoughts 40.63 (19.36) OBQ-Importance/Control of Thoughts 0.01 0.29 0.78 < 0.01 BDI-II 26.72 (13.28) ASI-Physical 0.23 2.75 < 0.01 0.06 BAI 21.12 (10.92) ASI-Cognitive 0.04 0.35 0.73 < 0.01 WI 4.80 (3.32) ASI-Social −0.05 −0.55 0.59 < 0.01 ASI Physical concerns 7.20 (4.92) Note. WI = Whiteley Index; BDI-II = Beck Depression Inventory-II; BAI = Beck Anxiety Cognitive concerns 4.32 (3.47) Inventory; DOCS = Dimensional Obsessive Compulsive Scale; OBQ-44 = Obsessive Social concerns 13.42 (5.88) Beliefs Questionnaire-44; ASI = Anxiety Sensitivity Index.

Note. Y-BOCS-SR = Yale-Brown Obsessive Compulsive Scale-Self report; DOCS = 3.3. Regression analyses Dimensional Obsessive Compulsive Scale; OBQ-44 = Obsessive Beliefs Questionnaire-44; BDI-II = Beck Depression Inventory-II; BAI = Beck Anxiety Inventory; WI = Whiteley Index; ASI = Anxiety Sensitivity Index. Summary statistics for the two regression models predicting WI scores appear in Table 3. In the first model, the BAI, BDI, and DOCS 3.2. Zero-order Pearson correlations subscales collectively accounted for 24.7% of the variance in WI scores, which was significant, F (6,86) = 4.70, p < 0.001. As can be seen in Table 2 presents the results of our zero-order correlational analyses. Table 3, the BAI and the DOCS harm obsessions and checking com- Given the multiple comparisons, we applied a Bonferroni correction of pulsions subscale emerged as unique significant predictors (ps ≤ p < 0.005 (0.05 / 12). As can be seen, the Whiteley Index (WI) was 0.003). weakly to moderately associated with the other study variables (rs In the second model, when the cognitions (i.e., OBQ-44 and ASI ranging from 0.12 to 0.43). Among the OCD symptom dimensions, only subscales) were entered as predictors, they collectively accounted for the DOCS harm obsessions and checking rituals subscale was significant 22.6% of the variance in WI scores, which was significant, F (6,97) = correlated with the WI. Among the obsessive belief domains, both the 4.43, p = 0.001. As can be seen in the table, the ASI Physical concerns responsibility/threat estimation and importance/control of thoughts subscale and the OBQ Responsibility/threat estimation subscale OBQ subscales were significantly correlated with the WI. The ASI emerged as significant individual predictors (ps < 0.01 − 0.03). physical concerns and social concerns subscales, but not the cognitive concerns subscale, were significantly correlated with the WI (rs ranging 4. Discussion 0.25–0.43). Finally, the BAI, but not the BDI, was significantly corre- lated with the WI. Illness anxiety symptoms are widely observed in patients with OCD and there are phenomenological and cognitive overlaps. Some authors (e.g., Deacon and Abramowitz, 2008) have proposed that illness anxiety Table 2 (and IAD) is a “variant” of OCD, yet this has not been adequately stu- Correlations between obsessive-compulsive symptoms, obsessive beliefs, died. Our first hypothesis that illness anxiety would be associated with and other symptoms with health anxiety. harm obsessions and checking rituals was supported. In fact, a robust WI association remained even after controlling for general distress and other OCD symptoms. This corroborates clinical observations that OC Symptom Dimensions (DOCS) preoccupation with illness in IAD is characterized by unwanted obses- Contamination 0.21 sion-like thoughts of harm befalling oneself (e.g., due to ) and Harm obsessions and checking 0.37a Unacceptable thoughts 0.12 frequent checking in the form of reassurance-seeking behaviors (e.g., Symmetry 0.15 unneeded diagnostic tests) that mirror compulsive rituals in OCD. Ill- OC Cognitions (OBQ-44) ness anxiety was not associated with other OCD symptoms, including a Responsibility/Threat 0.40 contamination and washing symptoms. As previous investigations have Perfectionism/Certainty 0.22 Importance/Control of Thoughts 0.26a found (Abramowitz et al., 1999), illness anxiety appears to concern Anxiety sensitivity (ASI) that are not communicable (e.g., cancer, heart disease) and Physical concerns 0.43a thus might be distinct from obsessions with germs and washing rituals. Cognitive concerns 0.25 In partial alignment with our second hypothesis, illness anxiety Social concerns 0.31a symptoms were associated with the tendency to overestimate threat Related Symptoms/Cognitions BDI-II 0.22 and responsibility for harm, but not with the need for certainty and BAI 0.35a intolerance of mistakes (i.e., perfectionism). As has been mentioned, the presence of health concerns among individuals with OCD is asso- Note. WI = Whiteley Index; Y-BOCS-SR = Yale-Brown Obsessive ciated with poorer insight into the senselessness of obsessions and Compulsive Scale-Self report; DOCS = Dimensional Obsessive Compulsive compulsions. Thus, individuals with such concerns may be more pre- Scale; OBQ-44 = Obsessive Beliefs Questionnaire-44; BDI-II = Beck Depression Inventory-II; BAI = Beck Anxiety Inventory; ASI = Anxiety occupied with certainty of health-related harm (i.e., overestimates of Sensitivity Index. threat of having or developing a serious medical illness), as opposed to a p < .005 (Bonferroni correction). the uncertainty of whether or not they are ill. Beliefs about the

420 L. Reuman et al. Psychiatry Research 256 (2017) 417–422 importance and need to control thoughts had a significant zero-order Abramowitz, J.S., Brigidi, B.D., Foa, E.B., 1999. Health concerns in patients with ob- sessive-compulsive disorder. J. Anxiety Disord. 13, 529–539. association with illness anxiety, which is consistent with the previous Abramowitz, J.S., Deacon, B.J., Olatunji, B.O., Wheaton, M.G., Berman, N.C., Losardo, D., finding that health-related intrusive thoughts are appraised as in- et al., 2010. Assessment of obsessive-compulsive symptom dimensions: development dicating responsibility for preventing harm (Freeston et al., 1994). This and evaluation of the dimensional obsessive-compulsive scale. Psychol. Assess. 22, 180–198. relationship disappeared, however, once the other predictors were ac- Abramowitz, J.S., Jacoby, R.J., 2015. Obsessive-compulsive related disorders: z critical counted for (as discussed further below). review of the new diagnostic class. Annu. Rev. Clin. Psychol. 11, 165–186. Our zero-order correlations also indicated that illness anxiety Abramowitz, J.S., Olatunji, B.O., Deacon, B.J., 2007. Health anxiety, hypochondriasis, – symptoms were associated with both the physical and social dimensions and the anxiety disorders. Behav. Ther. 38, 86 94. Abramowitz, J.S., Taylor, S., McKay, D., 2009. Obsessive-compulsive disorder. Lancet of AS. The former association is in line with previous research and 374, 491–499. clinical observations that illness anxiety is associated with the cata- American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental strophic misinterpretation of bodily sensations (e.g., thinking a tension Disorders, 4th ed. American Psychiatric Association, Washington, DC. American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental headache might be a sign of brain cancer). In addition, among people Disorders, 5th ed. American Psychiatric Association, Washington, DC. with OCD, illness anxiety symptoms also were associated with the fear Beck, A.T., Epstein, N., Brown, G., Steer, R.A., 1988. An inventory for measuring clinical that others will notice and negatively evaluate the presence of anxiety- anxiety: psychometric properties. J. Consult. Clin. Psychol. 56, 893–897. Beck, A.T., Steer, R.A., Brown, G.K., 1996. Beck Depression Inventory-II (BDI-II). related body sensations (e.g., sweating). Perhaps health anxious in- Psychological Corporation, San Antonio, TX. dividuals also harbor concerns about being perceived as ill by others. In Bienvenu, O.J., Samuels, J.F., Riddle, M.A., Hoehn-Saric, R., Liang, K.Y., Cullen, B.A., our regression model using cognitive factors to predict health anxiety, Nestadt, G., 2000. The relationship of obsessive–compulsive disorder to possible spectrum disorders: results from a family study. Biol. Psychiatry 48, 287–293. the physical concerns domain of AS and the responsibility/threat Bleichhardt, G., Hiller, W., 2007. Hypochondriasis and health anxiety in the German overestimation domain of obsessive beliefs were the only variables that population. Br. J. Health Psychol. 12, 511–523. emerged as significant unique predictors. There are various reasons as Conradt, M., Cavanagh, M., Franklin, J., Rief, W., 2006. Dimensionality of the Whiteley fi Index: assessment of hypochondriasis in an Australian sample of primary care pa- to why the cognitive dimension of AS was not a signi cant predictor of tients. J. Psychosom. Res. 60, 137–143. illness anxiety. For example, it may be that individuals with illness Cosci, F., Fava, G.A., 2015. The clinical inadequacy of the DSM-5 classification of somatic anxiety are preoccupied with somatic experiences and/or observable symptom and related disorders: an alternative trans-diagnostic model. CNS – changes (e.g., rashes) rather than loss of mental control (e.g., “going Spectr. 1 8. Deacon, B., Abramowitz, J.S., 2006. Anxiety sensitivity and its dimensions across the crazy”). anxiety disorders. J. Anxiety Disord. 20, 837–857. Our findings wield implications for understanding how illness an- Deacon, B., Abramowitz, J.S., 2008. Is hypochondriasis related to obsessive-compulsive xiety is related to OCD, at least from a cognitive perspective. In parti- disorder, , or both? An empirical evaluation. J. Cogn. Psychother. 22, 115–127. cular, they suggest that illness anxiety may be conceptualized in the Eisen, J.L., Phillips, K.A., Baer, L., Beer, D.A., Atala, K.D., Rasmussen, S.A., 1998. The context of certain OCD symptom dimensions characterized by obses- Brown assessment of beliefs scale: reliability and validity. Am. J. Psychiatry 155, sions about harm from serious illnesses that are appraised as indicating 102–108. Eisen, J.L., Rasmussen, S.A., Phillips, K.A., Price, L.H., Davidson, J., Lydiard, R.B., et al., a physical threat and that give rise to compensatory checking and re- 2001. Insight and treatment outcome in obsessive-compulsive disorder. Compr. assurance-seeking compulsions. Given that individuals with IAD usually Psychiatry 42, 494–497. present to physicians for what they perceive to be medical problems, Foa, E.B., Kozak, M.J., 1995. DSM-IV field trial: obsessive-compulsive disorder. Am. J. Psychiatry 152, 90–96. practitioners should be aware of the extent to which symptoms of ill- Fergus, T.A., 2014. Health-related dysfunctional beliefs and health anxiety: further evi- ness anxiety may present within the context of OCD, specifically the dence of cognitive specifi city. J. Clin. Psychol. 70, 248–259. responsibility for harm dimension. Accordingly, these findings support Fergus, T.A., Russell, L.H., 2016. Does cyberchondria overlap with health anxiety and obsessive-compulsive symptoms? An examination of latent structure and scale in- the need for nuanced assessment of illness anxiety in primary care terrelations. J. Anxiety Disord. 38, 88–94. settings. Freeston, M.H., Gagnon, F., Ladouceur, R., Thibodeau, N., Letarte, H., Rhéaume, J., 1994. The current study has limitations that should be considered, in- Health-related intrusive thoughts. J. Psychosom. Res. 38, 203–215. cluding its cross-sectional design and exclusive reliance on self-report Goodman, W.K., Price, L.H., Rasmussen, S.A., Mazure, C., Fleischmann, R.L., Hill, C.L., et al., 1989a. The Yale-Brown obsessive compulsive scale: i. development, use, and methodology. The former drawback prevents us from being able to reliability. Arch. Gen. Psychiatry 46, 1006–1011. draw causal inferences, and the latter raises the possibility of inflated Goodman, W.K., Price, L.H., Rasmussen, S.A., Mazure, C., Delgado, P., Heninger, G.R., associations between study measures due to method invariance. Charney, D.S., 1989b. The Yale-Brown obsessive compulsive scale: ii. validity. Arch. Gen. Psychiatry 46, 1012–1016. Further, standardized diagnostic interviews were not used to assess for Hedman, E., Lekander, M., Ljótsson, B., Lindefors, N., Rück, C., Andersson, G., Andersson, illness anxiety or other comorbidities. Updated measures of health an- E., 2015. Optimal cut-off points on the Health Anxiety Inventory, Illness Attitude xiety (e.g., the Short Health Anxiety Inventory; SHAI; Salkovskis et al., Scales and Whiteley Index to identify severe health anxiety. PLoS One 10, e0123412. Hedman, E., Ljotsson, B., Axelsson, E., Andersson, G., Ruck, C., Andersson, E., 2017. 2002) and AS (i.e., ASI-3; Taylor et al., 2007) are also available, but Health anxiety in obsessive compulsive disorder and obsessive compulsive symptoms were not used in the present study. Future investigations might include in severe health anxiety: an investigation of symptom profiles. J. Anxiety Disord. 45, multiple assessment methods and updated measures as there have been 80–86. Hollander, E., Friedberg, J.P., Wasserman, S., Yeh, C.C., Iyengar, R., 2005. The case for some reports that the WI demonstrates only adequate reliability the OCD spectrum. In: Abramowitz, J.S., Houts, A. (Eds.), Concepts and Controversies (Speckens et al., 1996). Further examination of insight into the seeming in Obsessive-Compulsive Disorder. Springer, New York, pp. 95–118. illogic of illness fears is also warranted (e.g., using measures like the Melli, G., Rustici, S., Micheli, E., Stopani, E., Carraresi, C., Bulli, F., 2014. Il ruolo della responsabilità ipertrofica nel disturbo d′ansia per la salute: un’indagine esplorativa. Brown Assessment of Beliefs Scale; Eisen et al., 1998), as heightened Psicoter. Cogn. Comport. 20, 343–355. illness anxiety in combination with poor insight may interfere with Obsessive Compulsive Cognitions Working Group, 2001. Development and initial vali- OCD treatment outcome. Additionally, the sample consisted exclusively dation of the obsessive beliefs questionnaire and the interpretation of intrusions – of patients with primary OCD; accordingly, the current sample does not Inventory. Behav. Res. Ther. 39, 987 1006. Otto, M.W., Demopulos, C.M., McLean, N.E., Pollack, M.H., Fava, M., 1998. Additional afford conclusions as to whether health anxiety is more strongly asso- findings on the association between anxiety sensitivity and hypochondriacal con- ciated with OCD than to other anxiety-related conditions (e.g., gen- cerns: examination of patients with major depression. J. Anxiety Disord. 12, 225–232. eralized anxiety disorder). Furthermore, the dimension of obsessive – fi Pilowsky, I., 1967. Dimensions of hypochondriasis. Br. J. Psychiatry 113, 89 93. beliefs that predicted health anxiety symptoms may not be speci cto Raines, A.M., Oglesby, M.E., Capron, D.W., Schmidt, N.B., 2014. Obsessive compulsive OCD (e.g., Tolin et al., 2006). disorder and anxiety sensitivity: identification of specific relations among symptom dimensions. J. Obsessive Compuls. Relat. Disord. 3, 71–76. References Rasmussen, S.A., Tsuang, M.T., 1986. Clinical characteristics and family history in DSM- III obsessive-compulsive disorder. Am. J. Psychiatry 143, 317–322. Reiss, S., Peterson, R.A., Gursky, D.M., McNally, R.J., 1986. Anxiety sensitivity, anxiety – Abramowitz, J.S., Braddock, A., 2008. Psychological Treatment of Health Anxiety and frequency, and the prediction of fearfulness. Behav. Res. Ther. 24, 1 8. Hypochondriasis: A Biopsychosocial Approach. Hogrefe, Cambridge, MA. Salkovskis, P.M., Rimes, K.A., Warwick, H.M.C., Clark, D.M., 2002. The Health Anxiety

421 L. Reuman et al. Psychiatry Research 256 (2017) 417–422

Inventory: development and validation of scales for the measurement of health an- the Anxiety Sensitivity Index-3. Psychol. Assess. 19, 176–188. xiety and hypochondriasis. Psychol. Med. 32, 843–853. Tolin, D.F., Abramowitz, J.S., Brigidi, B.D., Foa, E.B., 2003. Intolerance of uncertainty in Solem, S., Borgejordet, S., Haseth, S., Hansen, B., Haland, A., Bailey, R., 2015. Symptoms obsessive-compulsive disorder. J. Anxiety Disord. 17, 233–242. of health anxiety in obsessive-compulsive disorder: relationship with treatment out- Tolin, D.F., Worhunsky, P., Maltby, N., 2006. Are “obsessive” beliefs specific to OCD?: a come and metacognition. J. Obsessive Compuls. Relat. Disord. 5, 76–81. comparison across anxiety disorders. Behav. Res. Ther. 44, 469–480. Speckens, A.E.M., Spinhoven, P., Sloekers, P.P.A., Bolk, J.H., 1996. A validation study of Weck, F., Neng, J.M., Richtberg, S., Stangier, U., 2012. Dysfunctional beliefs about the Whitely Index, the Illness Attitude Scales, and the Somatosensory Amplification symptoms and illness in patients with hypochondriasis. Psychosomatics 53, 148–154. Scale in general medical and general practice patients. J. Psychosom. Res. 40 (1), Wheaton, M.G., Abramowitz, J.S., Berman, N.C., Riemann, B.C., Hale, L.R., 2010a. The 95–104. relationship between obsessive beliefs and symptom dimensions in obsessive-com- Steketee, G., Frost, R., Bogart, K., 1996. The Yale-Brown Obsessive Compulsive Scale: pulsive disorder. Behav. Res. Ther. 48, 949–954. Interview versus self-report. Behav. Res. Ther. 34 (8), 675–684. Wheaton, M.G., Berman, N.C., Franklin, J.C., Abramowitz, J.S., 2010b. Health anxiety: Taylor, S., 1999. Anxiety Sensitivity: Theory, Research, and Treatment of the Fear of latent structure and associations with anxiety-related psychological processes in a Anxiety. Routledge, New York, NY. student sample. J. Psychopathol. Behav. 32, 565–574. Taylor, S., Asmundson, G.J., 2004. Treating Health Anxiety: A Cognitive Behavioral Wheaton, M.G., Mahaffey, B., Timpano, K.R., Berman, N.C., Abramowitz, J.S., 2012. The Approach. Guilford Press, New York, NY. relationship between anxiety sensitivity and obsessive-compulsive symptom dimen- Taylor, S., Zvolensky, M.J., Cox, B.J., Deacon, B., Heimberg, R.G., Ledley, D.R., et al., sions. J. Behav. Ther. Exp. Psychiatry 43, 891–896. 2007. Robust dimensions of anxiety sensitivity: development and initial validation of

422