Authors: Emily E. Bernstein (1Massachusetts General Hospital; 2Harvard Medical School), Dalton Klare (1Massachusetts General Hospital), Hilary Weingarden (1Massachusetts General Hospital; 2Harvard Medical School), Jennifer L. Greenberg (1Massachusetts General Hospital; 2Harvard Medical School), Ivar Snorrason (1Massachusetts General Hospital; 2Harvard Medical School), Susanne Hoeppner (1Massachusetts General Hospital; 2Harvard Medical School), Rachel Vanderkuik (1Massachusetts General Hospital; 2Harvard Medical School), Oliver Harrison (3Koa Health), Sabine Wilhelm (1Massachusetts General Hospital; 2Harvard Medical School)
Categories: Article, body dysmorphic disorder, sleep, insomnia, cognitive behavioral therapy, smartphone
Source: Journal of affective disorders
Authors: Emily E. Bernstein, Dalton Klare, Hilary Weingarden, Jennifer L. Greenberg, Ivar Snorrason, Susanne Hoeppner, Rachel Vanderkuik, Oliver Harrison, Sabine Wilhelm
Body dysmorphic disorder (BDD) is severe, undertreated, and relatively common. Although gold-standard cognitive behavioral therapy (CBT) for BDD has strong empirical support, a significant number of patients do not respond. More work is needed to understand BDD’s etiology and modifiable barriers to treatment response. Given its high prevalence and impact on the development, maintenance, and treatment of related, frequently comorbid disorders, sleep disruption is a compelling, but not-yet studied factor.
Data were drawn from a randomized controlled trial of guided smartphone app-based CBT for BDD. Included participants were offered 12-weeks of treatment, immediately (n=40) or after a 12-week waitlist (n=37). Sleep disruption and BDD symptom severity were assessed at baseline, week-6, and week-12.
Hypotheses and analysis plan were pre-registered. Two-thirds of patients reported significant insomnia symptoms at baseline. Baseline severity of sleep disruption and BDD symptoms were not related (r=.02). Pre-treatment sleep disruption did not predict BDD symptom reduction across treatment, nor did early sleep improvements predict greater BDD symptom improvement. Early BDD symptom improvement also did not predict later improvements in sleep.
Limitations include the small sample, restricted ranges of BDD symptom severity and treatment response, and few metrics of sleep disruption.
Although insomnia was disproportionately high in this sample and both BDD symptoms and sleep improved in treatment, results suggest sleep and BDD symptoms may function largely independent of one another. More work is encouraged to replicate and better understand findings as well as potential challenges and benefits of addressing sleep in BDD.
Body dysmorphic disorder (BDD) is a severe, undertreated, yet relatively common condition impacting up to 3% of adults in the general population (Buhlmann et al., 2010). It is categorized in the DSM-5 as an Obsessive Compulsive and Related Disorder. Preoccupations with appearance concerns and related compulsive appearance-related rituals are frequently accompanied by impaired social, occupational, or other role functioning, reduced quality of life, and elevated risk of suicidal ideation, suicidal behaviors, and psychiatric comorbidities (Phillips, 2000; Phillips et al., 2005; Snorrason et al., 2020). Although therapist-delivered cognitive behavioral therapy (CBT) is the gold standard for BDD and has strong empirical support, a significant number of patients do not receive CBT let alone respond to CBT treatment if they receive it (Harrison, Fernández de la Cruz, Enander, Radua, & Mataix-Cols, 2016; Wilhelm et al., 2019). For example, even under well-controlled trials of in-person (Wilhelm et al., 2019), internet-delivered (Enander et al., 2016), and smartphone app-based (Wilhelm et al., 2022) CBT for BDD with high follow-up rates, 25% to 68% of participants did not experience symptom remission, and up to 15% to 46% did not respond to treatment (i.e., did not experience a ≥ 30% symptom reduction). Treatment response rates outside of clinical trials are even lower (Mulder et al., 2003). Thus, more work is needed to understand the etiology of BDD and modifiable barriers to treatment response.
Sleep disruption is a compelling, but not-yet studied factor in the development, maintenance, and treatment of BDD (Cox et al., 2020). Difficulty obtaining high quality, regular sleep is transdiagnostically associated with poor physical and emotional health, including depression, anxiety, and posttraumatic stress. Additionally, psychopathology is positively associated with greater odds of sleep disruption and developing sleep disorders. Although the relationship is bidirectional, worse sleep typically contributes more strongly to worse mental health than vice versa (Freeman et al., 2020; Harvey et al., 2011; Krystal, 2012). Understanding the role of sleep in the etiology and treatment of psychiatric disorders (and vice versa) is increasingly urgent as the prevalence of sleep disturbances continues to worsen in the United States and across the globe (Cox, Jessup, & Olatunji, 2018; Mandelkorn et al., 2021; Panel: et al., 2015).
Although sleep disruption has not yet been studied in BDD, in obsessive compulsive disorder (OCD), a disorder closely related to BDD, sleep disturbance has been associated with worse symptom severity across multiple studies and was found to be more prevalent overall compared to non-psychiatric samples (Cox et al., 2018, 2020; Robinson et al., 1998; Segalàs et al., 2021). An estimated half of individuals with OCD report concurrent sleep issues (Mukhopadhyay et al., 2008). Poor and delayed sleep limit treatment response in OCD, including response to cognitive and behavioral treatments (Coles et al., 2021; Cox et al., 2018; Ivarsson & Skarphedinsson, 2015) and transcranial magnetic stimulation (TMS) (Donse et al., 2017). Rates of sleep disturbance, poor sleep quality, and resultant daytime interference are similarly elevated in adults with clinically significant hoarding symptoms and positively associated with symptom severity (Mahnke et al., 2021; Nutley et al., 2022; Raines et al., 2015).
There are many possible pathways linking poor sleep to psychopathology and treatment response. Research has highlighted the negative impact of reduced sleep on emotional and cognitive control (e.g., flexibility, response inhibition) (Cox et al., 2018; Deliens et al., 2014; Nota et al., 2016). These capacities, in turn, are implicated in managing perseverative thinking and resisting compulsions (Abramovitch et al., 2013) as well as regulating strong emotions (Hendricks & Buchanan, 2016; Pruessner et al., 2020). Poor cognitive control is posited as a core neurobiological basis of BDD; in particular, weaknesses may propel the formation and maintenance of compulsive behavioral habits (Dunai et al., 2010; Grace et al., 2017; Labuschagne et al., 2013). Even more specific to the unique neurobiology of BDD, sleep disturbance impairs visual perception and processing (e.g., selective attention, distractor suppression) (Chee, 2015; Kong et al., 2011). Thus sleep disruption could exacerbate risk or severity of BDD symptoms. Moreover, cognitive control supports learning and memory (e.g., consolidation, generalization) processes, which are necessary for extinction learning and sustained treatment response (Coles et al., 2021). Thus, sleep and circadian rhythm disturbance, and primarily sleep duration, efficiency, and delayed sleep (Nota et al., 2015; Timpano et al., 2014), have been highlighted as novel targets for OCD treatment (Cox & Olatunji, 2022) and could be similarly promising for BDD. Meta-analyses have already shown that treating insomnia directly relieves symptoms of anxiety (Belleville et al., 2011) and depression (Gebara et al., 2018).
In contrast, sleep disturbance has not shown the same patterns with other disorders in the Obsessive Compulsive and Related Disorders category, such as body-focused repetitive behaviors (e.g., hair pulling, skin picking) (Cox et al., 2020). Although sleep disturbances are more prevalent in these populations than the general population as well, there is no evidence that sleep disturbance exacerbates symptom severity or interferes with treatment response in these disorders above and beyond the effects for comorbid depression and anxiety symptoms (Ricketts et al., 2017). Thus, when examining sleep in BDD, it is important to consider the role of co-occurring or general emotional distress. With eating disorders, the picture is also unclear. Although rates of sleep disruption are elevated in clinical populations with eating disorders (Tromp et al., 2016), this area requires further research to determine the directionality, causality, and treatment implications of this association (Allison et al., 2016; Christensen & Short, 2021).
The association between poor sleep quality and BDD has not been studied yet. Thus, as a first exploration of sleep disturbance in the presentation and treatment of BDD, this secondary analysis project had three primary aims. First, we aimed to examine the association between sleep disruption and BDD symptom severity in treatment-seeking adults with BDD. We predicted that the degree of sleep disruption reported would be positively correlated with BDD symptom severity at study entry (H1), given the similar relationship identified in OCD and hoarding (Cox et al., 2020; Raines et al., 2015). Second, we aimed to explore whether sleep disruption at baseline would predict BDD symptom severity change at the end of a 12-week treatment course of CBT for BDD. We hypothesized that greater pre-treatment sleep disruption would predict lesser symptom reduction (H2a), and that early improvement in sleep (baseline to mid-treatment) would predict greater post-treatment improvement in BDD symptoms (H2b), even when controlling for baseline depression. Our hypotheses for this aim were based on similar relationship identified in OCD (Coles et al., 2021; Cox et al., 2018; Ivarsson & Skarphedinsson, 2015) and in anxiety and depressive disorders (Belleville et al., 2011; Gebara et al., 2018). Third, we aimed to examine the reverse whether changes in BDD symptoms early in treatment would predict changes in sleep disruption later during treatment (mid-treatment to post-treatment). We hypothesized that early improvement in BDD symptoms could also lead to greater improvements in post-treatment sleep (H3). This is based on evidence that targeted OCD treatment can also benefit sleep (Donse et al., 2017; Ivarsson & Skarphedinsson, 2015; Nordahl et al., 2018). The relationships between hypersomnia, BDD symptoms, and treatment outcomes were examined in exploratory analyses. The aims, hypotheses, and analysis plan of this project were pre-registered on the Center for Open Science website (https://osf.io/fj8qm/).
Data were drawn from a randomized controlled trial (RCT) of guided smartphone app-based CBT for BDD (Wilhelm et al., 2022), approved by the Mass General Brigham Institutional Review Board. Of the 80 participants enrolled and randomized (immediate treatment versus waitlist) after providing written, informed consent, 77 were included in present analyses. Included participants were those who were offered and began 12-weeks of treatment, either immediately (n=40) or after a 12-week waitlist condition (n=37). Eligibility criteria included being at least 18 years old, living in the United States, and presenting with a primary DSM-5 diagnosis of BDD. Exclusion criteria included prior treatment with four or more sessions of CBT for BDD, concurrent psychotherapy, recent (within past 2 months) psychotropic medication changes, severe substance use, severe depression, acute suicidal ideation, or and lifetime mania or psychosis. Participants were instructed not to change their psychotropic medications or start any other psychosocial treatment during the trial. Standardized assessments were conducted virtually by an independent evaluator every six weeks for all participants. Time points included in the present project include baseline, mid-treatment (week 6), and post-treatment (week 12).
Perspectives BDD is a guided smartphone app-based CBT for BDD (Wilhelm et al., 2022; Wilhelm et al., 2020). It comprises seven treatment modules and the core components of CBT for BDD: psychoeducation, cognitive restructuring, exposure, ritual prevention, mindfulness and attention retraining, enhancing self-esteem and compassion through modifying core beliefs and pursuing values-based activities, and relapse prevention (Wilhelm, Phillips, & Steketee, 2013; Wilhelm, Phillips, Fama, Greenberg, & Steketee, 2011). In this trial, bachelor’s-level coaches completed standardized, pre-trial training on BDD and CBT, received a coaching manual detailing core tenets of the supportive accountability model (Mohr et al., 2011), and participated in weekly supervision with a licensed clinician with expertise in CBT for BDD. Coaches were available via asynchronous in-app secure messaging throughout the treatment to promote engagement and answer questions. Coaches also conducted two brief phone calls with participants at baseline and mid-treatment to facilitate treatment orientation and goal setting. Twelve weeks was previously determined to be an acceptable and effective duration for in-person, guided computer-delivered, and guided app-based CBT for BDD in prior trials (Enander et al., 2016; Veale et al., 2014; Wilhelm et al., 2020).
Doctoral-level independent evaluators completed the Mini International Neuropsychiatric Interview (MINI 7.02) (Sheehan et al., 1998) and the Columbia-Suicide Severity Rating Scale (C-SSRS) (Posner et al., 2008) at baseline to establish inclusion/exclusion criteria and characterize the sample. BDD symptom severity was captured via the 12-item semi-structured clinician-rated Yale-Brown Obsessive Compulsive Scale Modified for BDD (BDD-YBOCS) (Phillips et al., 1997); possible scores range from 0 to 48, with higher scores denoting more severe BDD symptoms over the past week. Depression symptom severity was measured with the 16-item Quick Inventory of Depressive Symptomatology—Self Report (QIDS-SR) (Rush et al., 2003); scores range from 0 to 48, with higher scores reflecting more severe depression symptoms over the past week.
Sleep disruption was quantified using items 1-3 on the QIDS-SR, which has been validated as a measure of global insomnia (Manber et al., 2005). Consistent with past work, we defined early insomnia (difficulty falling asleep) as a score of ≥ 2 on item 1 (“I take at least 30 minutes to fall asleep, more than half the time”), middle insomnia (difficulty staying asleep) as a score of 3 on item 2 (“I awaken more than once a night and stay awake for 20 minutes or more, more than half the time”), and late insomnia (waking up too early) as a score ≥ 1 on item 3 (“More than half the time, I awaken more than 30 minutes before I need to get up”) (Holder et al., 2019; Soehner et al., 2014). These cut-offs align with criteria for insomnia entailing ≥ 30 min of disruption more than half the time (Lichstein et al., 2003). Severity of sleep disruption was operationalized as the sum of QIDS-SR items 1-3. Hypersomnia was examined in exploratory analyses. Consistent with past work, we defined hypersomnia (sleeping too much) as a score of ≥ 1 on item 4 of the QIDS-SR (“I sleep no longer than 10 hours in a 24-hour period including naps”) (Soehner et al., 2014). Respondents may meet criteria for more than one sleep problem.
Analyses were conducted drawing data from participants’ active treatment phase (12 weeks occurring either immediately or after a 12-week waiting period, where ‘baseline’ refers to the timepoint immediately preceding the start of treatment). The relationship between pre-treatment sleep disruption and BDD symptom severity (aim 1) was assessed using zero order Pearson correlation. Prior to conducting the pre-registered analyses, we noted that pre-treatment BDD symptom severity significantly differed between the two groups with a medium to large effect, t(60.628)=2.61, p=.0115, d=0.613, suggesting that the group that received CBT after a 12-week waitlist period entered active treatment with less severe BDD symptoms than the immediate CBT group (see Supplemental Table 1). This result led to conducting analyses adjusting for treatment group as a study design variable in all regressions. We also identified three patients from the waitlist group in remission at the start of the active treatment phase (BDD-YBOCS score ≤ 16 (de la Cruz et al., 2021)) and repeated analyses excluding them from the sample.
To explore whether sleep disruption at baseline was associated with change in BDD symptoms over treatment (aim 2, H2a), we used a multiple linear regression with pre-treatment sleep disruption scores and treatment group as independent variables and change in BDD severity (baseline to post-treatment) as the dependent variable. BDD symptom change was operationalized as the change in BDD-YBOCS scores from pre-treatment to post-treatment. To test H2b, we used a multiple linear regression with early change in sleep disruption scores from baseline to mid-treatment (week 6) group predicting change in BDD-YBOCS scores from mid-treatment to post-treatment, adjusting for treatment group and baseline BDD-YBOCS scores. To explore whether change in BDD symptoms early in treatment was associated with changes in sleep disruption later in treatment (H3), we used a multiple linear regression with the change in BDD-YBOCS scores from baseline to mid-treatment (week 6) predicting change in sleep disruption scores from mid-treatment to post-treatment, adjusting for baseline sleep disruption scores and treatment group. Regressions were then repeated adjusting for baseline depression severity (QIDS-SR score, omitting sleep items). All analyses were then repeated for exploratory analyses with hypersomnia, where hypersomnia was used in place of sleep disruption.
Demographic and clinical characteristics of the sample are presented in Table 1; the majority of participants identified as women (83.1%) and white (72.7%), with an average age of 27 (SD=9.8). The proportion of the sample reporting sleep disruption and hypersomnia at pre-treatment and across treatment are presented in Table 2. Briefly, 48.1% met criteria for early insomnia, 15.6% met criteria for middle insomnia, 40.8% met criteria for late insomnia, 32.9% did not meet any criteria for insomnia subtypes, and 36.4% met criteria for hypersomnia. There were no significant differences between treatment groups in the baseline sleep disruption sum score or frequency of insomnia subtype or hypersomnia (see Supplemental Table 1).
For hypothesis 1, the relationship between sleep disruption and BDD symptom severity at baseline was not significant, r(74)=0.02, p=.868.
For hypothesis 2a, we were unable to detect an association between pre-treatment sleep disruption and BDD symptom change from baseline to post-treatment (p=.312, f^2^=.010; Table 3). Likewise, for hypothesis 2b, we were unable to detect an association between changes in sleep disruption from baseline to week 6 of treatment and changes in BDD symptom severity from week 6 to week 12 of treatment (p=.488, f^2^=−.050; Table 3). The multiple regression models for hypotheses 2a and 2b explained 10.4 and 17.7% of the variation in whole treatment BDD symptom change and BDD symptom change in the last 6 weeks of treatment, respectively. In both models, treatment group (immediate app-CBT vs. app-CBT after waitlist) was the only significant predictor, indicating that the immediate app-CBT group experienced a greater improvement in BDD symptoms throughout treatment. See Supplemental Table 2 for descriptive statistics of each treatment group’s BDD-YBOCS scores, sleep disruption sum scores, insomnia subtype and hypersomnia frequencies across treatment.
For hypothesis 3, improvement in BDD symptoms from baseline to week 6 of treatment was associated with changes in sleep disruption from week 6 to week 12 of treatment (p=.046, f^2^=.087; Table 3); however, the direction of the relationship was opposite of our prediction. The multiple regression model for hypothesis 3 explained 8.2% of the variation in sleep disruption change in the last 6 weeks of treatment. Further investigation into a correlation matrix with raw and change scores of BDD-YBOCS and sleep disruption across treatment (see Supplemental Table 3) showed that BDD symptom severity and sleep disruption actually appear to change concurrently in the early part of treatment (i.e., as symptoms of BDD severity improve, symptoms of sleep disruption also improve in the same treatment period; r=.26 for week 0 to 6 change scores). There was little evidence that change in BDD symptom severity or sleep disruption preceded the change of the other over the 6-week time-steps we used in this analysis; participants appeared to have changed more rapidly in both BDD symptom severity and sleep disruption severity early in treatment and less in the second half of treatment. Note, adjusting the multiple regression models for baseline depression severity had little to no impact on the results of the multiple regression analyses (H2a, 2b, 3; Table 3).
When replacing the sleep disruption sum score with the single hypersomnia item in exploratory analyses, the relationship between hypersomnia and BDD symptom severity at baseline was not significant r(75)=0.03, p=.827. We were unable to detect an association between pre-treatment hypersomnia and BDD symptom change from baseline to post treatment (p=.530, f^2^=.007; Table 4). We were also unable to detect an association between changes in hypersomnia from baseline to week 6 of treatment and changes in BDD symptom severity from week 6 to week 12 of treatment (p=.947, f^2^=−.036; Table 4). Lastly, we were unable to detect an association between changes in BDD symptom severity from baseline to week 6 of treatment and changes in hypersomnia from week 6 to week 12 of treatment (p=.742, f^2^=.002; Table 4).Adjusting for depression severity similarly had little to no impact on the results of the regression analyses with hypersomnia (H2a, b, 3; Table 4).
After excluding the three waitlist participants in remission at treatment entry, results held for hypotheses 1, 2a and 2b with non-significant findings. The significant effect initially found for hypothesis 3 with sleep disruption was non-significant (p=.076), potentially due to insufficient power. The direction of the relationships was maintained for all sleep disruption and hypersomnia analyses. Overall, results were consistent with the full sample.
This project explored the prevalence and impact of sleep disturbance in treatment-seeking adults with BDD. Rates of self-reported insomnia were elevated in this sample (approximately two thirds) relative to the general population and similar to adults with OCD (approximately one half) (Cox & Olatunji, 2016; Mukhopadhyay et al., 2008). However, unlike prior studies of OCD and hoarding, in this sample of individuals with moderate to severe BDD, reports of greater sleep disruption were not associated with higher BDD severity. Moreover, we were unable to detect any associations between either baseline sleep disruption or early improvements in sleep disruption with changes in BDD symptom severity during 12 weeks of CBT for BDD. Similarly, we were not able to detect any association between early changes in BDD symptom severity and later changes in sleep disruption severity. Instead, positive changes in sleep and BDD symptom severity were only modestly correlated (small effect; r=.26) early in treatment.
Although results require replication in a large, more heterogeneous sample, our initial results diverge from findings in OCD and instead are more aligned with those in body-focused repetitive behaviors (e.g., skin picking, hair pulling)—other conditions in the DSM-5 OCD and related disorders category; in these cases, sleep has not been shown to directly influence symptom severity or treatment response, but rather indirectly via its effect on anxiety and depression (Ricketts et al., 2017). Thus, CBT for insomnia (or other sleep interventions) could still benefit patients with BDD, but perhaps be most beneficial for patients also endorsing significant psychiatric comorbidities, which can worsen BDD presentation and interfere with treatment (Phillips, 1999). Targeted sleep therapies reliably improve general mental health as well as other relevant symptoms, including stress, depression, anxiety, and rumination (Scott et al., 2021). Addressing sleep disruption could thus still be clinically indicated—particularly because sleep disruption could leave patients more even vulnerable to other psychiatric concerns (e.g., depression)— but may not have an impact on BDD specific symptoms, and therefore should not be a replacement for CBT for BDD or other targeted treatment. A future trial in which individuals with BDD and sleep disruption receive CBT for insomnia (or other sleep interventions) would be informative.
The fact that sleep concerns are more common in BDD than the general population could reflect multiple, not mutually exclusive factors. First, general emotional distress or mental health problems and sleep disruption are strongly correlated; sleep disruption also serves as a diagnostic criterion for multiple emotional disorders (e.g., MDD, GAD) (Baglioni et al., 2016; Ford & Kamerow, 1989). Treatment-seeking adults with BDD are necessarily experiencing notable emotional distress. Furthermore, rates of comorbidity in this population are high (Gunstad & Phillips, 2003), consistent with the present sample; 66% of participants met criteria for another psychiatric disorder, the most common of which were social anxiety, major depression, and generalized anxiety disorder. And at start-of-treatment, 50.7% of participants endorsed at least moderately severe depression symptoms (QIDS-SR score > 10). Thus, this prevalence appears consistent with sleep disruption being a non-specific, transdiagnostic factor in mental health. Prospective research in a non-treatment seeking sample could uncover how fluctuations in general distress and sleep interact. Second, sleep disruption could be in part accounted for by compulsive behaviors or avoidance. Individuals with BDD engage in and have difficulty resisting time consuming compulsive behaviors, such as excessive grooming or mirror checking, on average 3-8 hours per day (Phillips et al., 1998). For some, behaviors can be worse in the morning (e.g., some patients wake up early to complete rituals before needing to the leave the house for work) or in the evening (e.g., without the distraction or pressure of being around others or needing to perform at a job). This may parallel findings for those with hoarding disorder for whom symptoms, namely clutter and resulting disruption to activities of daily living (e.g., sleep environment), can interrupt sleep (Raines et al., 2015). Furthermore, the distress associated with perseverating on perceived appearance flaws may also interfere with sleep (Veale & Riley, 2001). Thus, rituals characteristic of BDD may unintentionally delay, restrict, or interrupt sleep quality. Poor sleep may further exacerbate such behaviors via weakened inhibitory control (Cox et al., 2018; Nota et al., 2016) or processing distressing visual stimuli or stressors (Chee, 2015; Kong et al., 2011; Palmer & Alfano, 2017). More work is needed to disentangle these possibilities. For example, a daily diary or ecological momentary assessment (EMA) study could identify when ritualizing interferes with sleep or how poor sleep impacts next day behavior.
Limitations of this work include the small sample, restricted range of BDD symptom severity as this was a treatment-seeking group, and relatively high response rates to the treatment (i.e., reduced variability in outcomes) (Wilhelm et al., 2022). More heterogeneous presentations and treatment outcomes could reveal more effects. Additionally, given that much of treatment response—even in digital psychotherapy—often occurs during the first half of treatment, it is possible that more frequent measurement points could have revealed relationships between changes in sleep and BDD symptoms (Bisby et al., 2022). In this project, we were able to examine effects of temporal precedence; future work should be designed to more rigorously examine causality. Although the QIDS-SR items have been validated as an insomnia measure (Manber et al., 2005), it is not comprehensive. Future projects would benefit from assessing more detailed (e.g., Pittsburgh Sleep Quality Index (PSQI); (Buysse et al., 1989)) and objective sleep metrics via polysomnography and actigraphy (e.g., proportion of rapid eye movement (REM) sleep, proportion of slow wave sleep) as well as measures of other sleep disorders, sleep chronotypes, and delayed circadian rhythms (Cox & Olatunji, 2022). Some research suggests that in OCD, delayed sleep (more so than total sleep time or quality), may be most relevant for treatment response (Coles et al., 2021). For example, sleep onset latency can be measured via polysomnography (e.g., first presence of stage N2 sleep) or self-reported time needed to fall asleep (Nota et al., 2015). Being able to compare time to bed, self-reported sleep onset, and polysomnography-measured sleep onset would help home in on the nature of patients’ sleep disturbances and in turn provide more nuance when they do or not differentially relate to BDD symptoms over time and during treatment. If only time to bed were to relate to BDD, targeting daily schedules to promote an earlier bedtime could be a stronger intervention than full CBT for insomnia. Moreover, as a retrospective measure, the QIDS-SR does not allow for exploration of more short-term links between sleep quality and BDD symptoms; it remains possible that evaluation of shorter time scales with EMA would yield different results.
This is the first examination of sleep disturbance in BDD. Counter to hypotheses, although rates of sleep concerns were disproportionately high in this treatment-seeking sample—similar to adults with OCD and emotional disorders—we were not able to detect any associations with BDD symptom severity or BDD symptom change during treatment with app-CBT for BDD. This lack of evidence diverges from findings in other, frequently comorbid disorders like OCD, anxiety, and depression. Still, given the high prevalence of sleep disruption, associated distress, and known consequences for general mental and physical health, more work is encouraged to replicate and better understand these findings.