Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Mar 11.
Published in final edited form as: Behav Res Ther. 2021 Aug 18;145:103948. doi: 10.1016/j.brat.2021.103948

Outcomes of the Transdiagnostic Intervention for Sleep and Circadian Dysfunction (TranS-C) in a community setting: Unpacking comorbidity

Laurel D Sarfan a, Heather E Hilmoe b, Nicole B Gumport c, Caitlin E Gasperetti d, Garret G Zieve e, Allison G Harvey f
PMCID: PMC10927273  NIHMSID: NIHMS1965919  PMID: 34428642

Abstract

Objective:

Comorbidity and subdiagnostic symptoms are understudied for sleep and circadian problems. We evaluated 1) impairment associated with (a) number of sleep and circadian problems and (b) diagnostic threshold (full diagnosis vs. subdiagnostic symptoms), and 2) Transdiagnostic Sleep and Circadian Intervention (TranS-C) outcomes for participants with specific sleep and circadian problems.

Method:

Community participants (N=121) with serious mental illness and sleep and circadian problem(s) were randomized to receive TranS-C plus usual care (TranS-C+UC) or usual care plus delayed TranS-C (UC-DT). Overall impairment, psychiatric symptoms, and sleep and circadian dysfunction were assessed at pre-treatment, post-treatment, and 6-month follow-up.

Results:

Higher numbers of sleep and circadian problems, versus one problem, were associated with worse overall impairment, psychiatric symptoms, and sleep and circadian dysfunction (ps<0.05, ω2=0.06–0.15). Diagnostic threshold was not associated with baseline functioning (ps>0.05). TranS-C+UC versus UC-DT was associated with psychosocial and sleep and circadian improvements for specific sleep and circadian problems (insomnia, hypersomnia, parasomnias, periodic limb movement/restless leg syndrome, circadian rhythm disorders), though improvements varied by problem. TranS-C+UC outcomes were not moderated by number of sleep and circadian problems (ps>0.05).

Conclusion:

Higher numbers of sleep and circadian problems, not diagnostic threshold, were associated with greater impairment. Transdiagnostic utility of TranS-C+UC was supported.

Keywords: Sleep, circadian, transdiagnostic, comorbid, community mental health, serious mental illness

Introduction1

Comorbidity among mental illness is pervasive. More than 40% of 12-month mental illness cases in the U.S. are estimated to be comorbid, and in turn, comorbidity is associated with greater clinical severity and poorer functioning (Kessler et al., 2005). However, when developing treatments for mental illness, randomized controlled trials often exclude patients with comorbid diagnoses. Similarly, trials regularly overlook comorbid subdiagnostic symptoms, defined as core symptoms of a disorder in the absence of meeting full diagnostic criteria (Wolitzky-Taylor et al., 2014). These practices persist despite calls for dimensional approaches to mental illness (Cuthbert & Insel, 2013) and evidence that subdiagnostic symptoms are common and associated with psychosocial impairment (e.g., Balazs et al., 2013; Marangell, 2004). Promisingly though, comorbid disorders and select subdiagnostic symptoms can be effectively targeted via psychosocial treatments (Cuijpers et al., 2007; McHugh et al., 2009).

Transdiagnostic treatments, or treatments that concurrently target cognitive and behavioral processes common to different mental illness, hold promise for treating comorbid diagnoses and subdiagnostic symptoms. The transdiagnostic approach to mental illness proposes that disorders co-occur, at least in part, because they share common processes that causally contribute to the development or maintenance of symptoms (Dalgleish et al., 2020; Mansell et al., 2009). Thus, targeting these common processes may represent an efficient path for treatment (Harvey et al., 2004).

Using data from a randomized controlled trial (NCT02469233; Harvey et al., 2021), the present study focused on a transdiagnostic treatment for sleep and circadian problems—the Transdiagnostic Sleep and Circadian Intervention (TranS-C)—delivered to individuals diagnosed with a serious mental illness (SMI). Sleep and circadian problems have been proposed as a transdiagnostic mechanism of SMI (Harvey et al., 2011). Evidence to support this proposal includes that a) sleep and circadian problems—such as insomnia, hypersomnia, and evening circadian preference—predict and predate psychiatric symptoms, including depression, substance use, anxiety, and psychosis (Hertenstein et al., 2019; Kaplan et al., 2015; Kivela et al., 2018), b) common cognitive, behavioral, and biological processes maintain sleep and circadian problems as well as SMI (Harvey, 2004), and c) interventions targeting sleep and circadian problems are associated with improved psychiatric symptoms (Taylor & Pruiksma, 2014). Indeed, TranS-C, which is grounded in the Sleep Health Framework (Buysse, 2014), is associated with reductions in sleep-related problems, functional impairment, and psychiatric symptoms.

However, there are gaps in this prior research exploring comorbidity, subdiagnostic symptoms, and TranS-C outcomes. First, although there is a relatively robust literature on sleep and circadian problems that are comorbid with mental (e.g., Ford et al., 1989) or physical (e.g., Parish et al., 2009) illness, past research on comorbidities between sleep and circadian problems is limited. This small body of research has primarily focused on testing a single sleep and circadian comorbidity (e.g., insomnia and hypersomnia in a major depressive episode; Geoffroy et al., 2018), outcomes collapsing across sleep and circadian comorbidities (e.g., Harvey et al., 2021; Reeve et al., 2019), or comorbidities between common sleep symptoms assessed in isolation (e.g., Roth et al., 2006). Second, this research on sleep and circadian problems has not yet extended to subdiagnostic symptoms. Together, impairment associated with a) different numbers of sleep and circadian problems and b) levels of diagnostic threshold (e.g., full diagnosis versus subdiagnostic symptoms) of sleep and circadian problems is unknown. Elucidating these relationships is critical for clarifying the functional impact of sleep and circadian problems. Third, prior evaluation of TranS-C has collapsed across all patients, who presented with a range of comorbid sleep and circadian problems (Harvey et al., 2021). Thus, the efficacy of TranS-C for specific sleep and circadian problems and varying numbers of sleep and circadian problems has not yet been tested. This represents an important next step to help clinicians determine whether TranS-C could be an appropriate choice for a given patient’s sleep and circadian problems.

The present study sought to address these gaps in the literature. Aim 1 sought to test the relationships between 1a) number of sleep and circadian problems and baseline functioning, and 1b) diagnostic threshold of sleep and circadian problems and baseline functioning. For Aim 1a, we coded each participants’ number of sleep and circadian problems. For Aim 1b, we coded whether each participants’ sleep and circadian problems met diagnostic threshold, categorized as follows: full diagnoses only, full diagnoses plus other subdiagnostic symptoms, and subdiagnostic symptoms only. Hypothesis 1a was that, controlling for diagnostic threshold of sleep and circadian problems, higher numbers of sleep and circadian problems would be associated with poorer baseline functioning – specifically, more severe functional impairment, psychiatric symptoms, and sleep and circadian dysfunction (i.e., sleep disturbance, sleep-related daytime impairment, and sleep health). Hypothesis 1b was that, controlling for number of sleep and circadian problems, diagnostic threshold would be associated with poorer baseline functioning, such that full diagnoses would be associated with the poorest functioning, followed by full diagnoses plus other subdiagnostic symptoms, then subdiagnostic symptoms only. The latter hypothesis was based on a key premise of widely-used categorical approaches to diagnosis in research, policy, and clinical practice – namely that diagnostic threshold signifies the presence of clinically and functionally meaningful symptoms (e.g., Clark et al., 2017). To summarize, with Aim 1, we sought to disentangle the functional impact of diagnostic threshold from number of sleep and circadian problems, and thereby isolate the distinct relationships between impairment and 1a) number of sleep and circadian problems, and 1b) diagnostic threshold of sleep and circadian problems.

Aim 2 sought to evaluate the effects of TranS-C plus Usual Care (TranS-C+UC) versus Usual Care followed by Delayed Treatment with TranS-C (UC-DT) for individuals with specific sleep and circadian problems and varying numbers of sleep and circadian problems. Hypothesis 2 was that TranS-C+UC would be associated with better outcomes than UC-DT at post-treatment and 6-month follow-up on outcomes of functional impairment, general psychiatric symptoms, and sleep and circadian dysfunction for all specific sleep and circadian problems. We also hypothesized that TranS-C+UC would not be moderated by number of sleep and circadian problems.

Method

Participants and Procedures

Data from participants (N=121) in this study were collected during a randomized controlled trial funded by the National Institute of Mental Health (Harvey et al., 2016; Harvey et al., 2021; NCT02469233). Participants who met criteria for SMI were recruited from multiple sites within Alameda County Behavioral Health Care Services (ACBHCS), a community mental health center in California. SMI was operationalized according to Public Law 102–321 and previous research (Wang et al., 2002) as the presence, for 12 months, of at least one Diagnostic and Statistical Manual–5 (DSM-5; American Psychiatric Association, 2013) mental disorder leading to substantial interference with one or more major life activities.

To enhance representativeness and generalizability, inclusion and exclusion criteria were kept to a minimum. Inclusion criteria included: 1) Age 18+ years; 2) English language fluency; 3) presence of at least one DSM-5 mental disorder for 12 months; 4) having a guaranteed bed to sleep in for 3 months; 5) receiving care for SMI at ACBHCS; 6) consenting to regular communications between research team and psychiatrist and/or case manager; and 7) experiencing one or more sleep or circadian problems for at least 3 months assessed with the Sleep and Circadian Problems Interview, which is an adapted version of the Insomnia Interview Schedule (Morin, 1993).

The exclusion criteria included: 1) presence of an active and progressive physical illness or neurological degenerative disease and/or substance abuse/dependence making participation in the study infeasible; 2) current serious suicide or homicide risk (assessed by research staff, a case manager, or psychiatrist); 3) night shift work >2 nights per week in the past 3 months; 4) pregnancy or breast-feeding; 5) not able/willing to complete the pre-treatment assessments. Individuals with periodic limb movement disorder often have comorbid insomnia, poor sleep habits, and can benefit from CBT-I (Edinger et al., 1996). Hence, these individuals were included in this study. As participants’ SMI medications often need to be changed, excluding on this basis is neither feasible nor representative of clinical practice. Medication use and changes were recorded.

Participants were randomly allocated to TranS-C+UC (n=61) or 6-months of UC-DT (n=60). Randomization was stratified by age (0=under 50 years, 1=50+ years) and psychosis (0=no, 1=yes). Participants completed assessments at pre-treatment, immediately post-treatment (9–14 weeks later), and 6-month follow-up (6FU). Assessors were blind to treatment allocation. The University of California, Berkeley Committee for the Protection of Human Subjects approved the study, and informed consent was obtained from all participants. See Supplement for CONSORT diagram (Figure 1, supplement). A full description of study procedures is detailed in Harvey et al. (2021).

Treatment Conditions

TranS-C+UC

TranS-C is driven by case conceptualization. It includes four cross-cutting interventions in every session, four core modules that apply to most participants, and seven optional modules applied based on patient presentation. The average number of 50-minute sessions attended was eight. Up to 12 sessions could be provided but was rarely needed. See Table 1 for summary of treatment modules, and Harvey et al. (2021) for more details.

Table 1.

Common Sleep-Circadian Problems and TranS-C Modules

Common Sleep-Circadian Problems Treatment Module
Irregularity, difficulty winding down/waking up Core Module 1
Daytime Impairment Core Module 2
Unhelpful beliefs about sleep Core Module 3
Poor sleep-efficiency Optional Module 1
Too much time in bed Optional Module 2
Delayed or advanced phase Optional Module 3
Sleep-related worry Optional Module 4
CPAP difficulties Optional Module 5
Disruptive environments Optional Module 6
Nightmares Optional Module 7
Maintenance of behavior change Core Module 4
Open in a new tab

Note. All sessions included cross-cutting modules: functional analysis, education, motivational enhancement, and goal setting.

UC-DT

In the UC-DT condition, a case manager coordinated care within ACBHCS and referred each participant for a medication review and to various programs (e.g., health care, housing). After eight months in UC-DT, participants received eight sessions of TranS-C+UC.

Measures Included at Pre-Treatment Assessment Only

Sleep and Circadian Problems: Diagnoses and Subdiagnostic Symptoms

To determine whether participants met full diagnoses or subdiagnostic symptoms for sleep and circadian problems, the Duke Structured Interview for Sleep Disorders (DSISD; Edinger et al., 2004) was administered. The DSISD is a semi-structured interview designed to detect sleep and circadian problems according to the International Classification of Sleep Disorders (ICSD; American Academy of Sleep Medicine, 2014) and the DSM-5 criteria. During interviews with participants, the assessors ascertained whether each sleep and circadian problem was better explained by a different disorder (e.g., psychiatric, medical, or substance use disorder). If the sleep and circadian problem was better explained by a different disorder, a full diagnosis was not given for the sleep and circadian problem. The DSISD has demonstrated discriminant validity and reliability (kappa range: .71–.86; Edinger et al., 2009). For full diagnoses, the DSISD was clarified via self-report daily sleep diaries collected for the week preceding pre-treatment assessment. For periodic limb movement, full diagnosis was not determined, as this requires polysomnography (American Academy of Sleep Medicine, 2014).

To identify subdiagnostic symptoms, core symptoms from the DSM-5 were pre-determined for each sleep and circadian problem. A core symptom was defined as a symptom that was essential to diagnosis (e.g., endorsing nightmares for nightmare disorder). Note that when evaluating subdiagnostic symptoms, the diagnostic criterion of ‘unless better explained by another disorder’ was not included as a core symptom, because it is not central to any particular diagnosis. Based on prior subdiagnostic research, patients were categorized as ‘subdiagnostic symptoms only’ if they endorsed one or more core DSM-5 symptoms assessed by the DSISD but did not meet full diagnostic criteria (Laborde-Lahoz et al., 2015). For periodic limb movement, the core symptom was pre-identified from the DSISD rather than the DSM-5, as this disorder is not in the DSM-5.

Stratification and Mental Illness Diagnoses

Age was collected using a demographics form. Psychosis was assessed using the Mini International Neuropsychiatric Interview (MINI) (DSM-5, Version 7.0.0).

Measures Collected at All Assessments (Pre, Post, 6FU)

Primary outcome measures mirrored Harvey et al. (2021) and clinicaltrials.gov (NCT02469233). Note that in Harvey et al. (2021) and clinicaltrials.gov, secondary measures were included to assess psychosocial functioning more specifically. Secondary measures were not included in the present study, because the hypotheses could be tested with the primary measures. This decision to omit the secondary measures also served to limit Type I error associated with multiple testing.

The two exceptions to omitting secondary measures were that—to gain a fuller picture of transdiagnostic sleep and circadian functioning using multiple methods—we analyzed a) sleep/wake variables via the Sleep Health Composite and b) total sleep time variability via actigraphy (Buysse et al., 2006). The Sleep Health Composite incorporates sleep diary parameters and is proposed as a better measure of transdiagnostic sleep and circadian problems relative to individual sleep parameters (Dong et al., 2019). Total sleep time variability via actigraphy was selected for four reasons: 1) actigraphy served as an objective measure of sleep and circadian functioning, 2) variability (vs. average estimates) may be a better metric of sleep and circadian functioning in transdiagnostic samples, as regularizing total sleep time is likely to be a sleep goal across sleep and circadian problems, 3) increasing evidence suggests that total sleep time variability is associated with indicators of circadian dysregulation over and above mean total sleep time (e.g., Bei et al., 2017), and 4) total sleep time variability is thought to be a critical variable that drives exacerbation of various psychiatric symptoms (Bei et al. 2016; Gruber et al., 2011; Lemola et al., 2013), again aligning with the present study’s transdiagnostic approach.

Primary Measures

The Sheehan Disability Scale (SDS) assessed functional impairment (Sheehan et al., 1996). The first three items on the SDS assess impairment in work/school, social life, and home/family responsibilities on a scale from 0–10 (‘not at all’ to ‘extremely’). These items are summed and range from 0–30, with higher scores indicating greater impairment. The SDS has demonstrated adequate reliability and validity (Leon et al., 1996; Sheehan et al., 1996).

The DSM-5 Cross-Cutting Measure assessed general psychiatric symptoms (Narrow et al., 2013). Across 13 mental health domains, participants rate how bothered they were by a list of psychiatric symptoms from 0–4 (‘not at all’ to ‘nearly every day’). Total scores range from 0–52, with higher scores indicating greater psychiatric severity. The measure has demonstrated good test-retest reliability and clinical utility (Clarke & Kuhl, 2014; Narrow et al., 2013).

The 8-item PROMIS—Sleep Disturbance (PROMIS-SD) assessed sleep disturbance (e.g., restlessness, sleep quality) over the past week on a scale from 1–5 (‘not at all’ to ‘very much’; Yu et al., 2011). Scores range from 8–40. Higher scores indicate increased disturbance. This questionnaire has demonstrated adequate reliability and validity (Yu et al., 2011).

The 16-item PROMIS—Sleep Related Impairment (PROMIS-SRI) assessed sleep-related impairment during waking hours over the past week on a scale from 1–5 (‘not at all’ to ‘very much’). The PROMIS-SRI measures perceptions of alertness, sleepiness, and tiredness (Buysse et al., 2010). Raw scores range from 16–80. Higher scores indicate increased impairment. This measure has demonstrated adequate validity (Buysse et al., 2010; Yu et al., 2011). A discussion of PROMIS T-scores can be found in Harvey et al. (2021).

Sleep/Wake Variables

Sleep and circadian problems were further assessed by actigraphy (GT9X, Actigraph) and daily sleep diaries collected for seven days prior to each assessment point. With actigraphy, variability in total sleep time was assessed, as described above. The daily sleep diary variables were analyzed for the purpose of calculating the Sleep Health Composite. Specifically, the Sleep Health Composite score (Dong et al., 2019) was defined as the sum of scores on six sleep health dimensions (each dimension dichotomized as 1 = good; 0 = poor): Regularity (midpoint fluctuation via sleep diary), Satisfaction (sleep quality question on PROMIS-SD), Alertness (daytime sleepiness question on PROMIS-SRI), Timing (mean midpoint via sleep diary), Efficiency (sleep efficiency via sleep diary) and Duration (total sleep time via sleep diary). Scores range from 0–6, with higher scores indicating better sleep health. This measure is proposed to capture the complexity of sleep problems in SMI that are covered by TranS-C. Initial validity of this measure has been established (Dong et al., 2019).

Data Analysis

Data were analyzed with Stata 16.1.

Aim 1

For Aim 1, we evaluated baseline functioning associated with number of sleep and circadian problems (Aim 1a) and diagnostic threshold of sleep and circadian problems (Aim 1b) via analysis of covariance (ANCOVA). For Aim 1a, number of sleep and circadian problems was quantified by coding each participants’ number of full diagnoses and subdiagnostic symptoms of distinct sleep and circadian problems (range: 1–7 problems). For Aim 1b, diagnostic threshold was assessed using the process described above (see Measures) and represented by two dummy-coded variables (0 = full diagnoses only, 1 = full diagnoses with other subdiagnostic symptoms, 2 = subdiagnostic symptoms only). Significant ANCOVAs were followed up with planned pairwise comparisons for adjusted means using a t-statistic and the Tukey-Kramer correction to account for unbalanced data. Omega squared (ω2) is reported as an effect size for ANCOVAs and can be interpreted as the proportion of variance attributed to the independent variable of interest (Field, 2013; Salkind, 2010). Participants with six comorbidities (n = 2) and seven comorbidities (n = 2) were not included due to sample size. Listwise deletion was used for missing data (Kline, 2011).

Aim 2

Random intercept multilevel models were used to compare the effects of TranS-C+UC versus UC-DT. Because randomization was stratified for psychosis and age taken at baseline, we controlled for these variables. Data were assumed to be missing at random and estimated with restricted maximum likelihood to account for sample size. All available data were used. The fixed component of the model consisted of two dummy-coded variables for time (0=pre-treatment assessment, 1=post-treatment, 2=6FU), one dummy-coded variable for treatment condition (0=UC-DT, 1=TranS-C), two time-by-treatment interaction terms, and the variables for which we controlled (i.e., age, psychosis). The moderation models (see below) also included time-by-treatment-by-number of sleep and circadian problems interactions, with number of sleep and circadian problems scored continuously. The random component of the model consisted of a random intercept for participants and a level-1 (i.e., occasion) error term.

First, we tested whether TranS-C was associated with improved outcomes, relative to UC-DT, for individuals presenting with specific sleep and circadian problems. In other words, if an individual presented to treatment with insomnia, hypersomnia, a circadian rhythm disorder, periodic limb movement or restless leg syndrome, or a parasomnia, would TranS-C be a helpful treatment option for them? To test this question, participants were identified as having one or more full diagnosis or subdiagnostic symptoms of the following sleep and circadian problems: insomnia, hypersomnia, circadian rhythm disorders, periodic limb movement or restless leg syndrome, and parasomnias. Then separate multilevel models were evaluated for participants with a full diagnosis or subdiagnostic symptoms of each of these sleep and circadian problems. Participants were not excluded from a model based on comorbidities. For example, some participants endorsed insomnia and hypersomnia, and therefore were included in both models. Because of this overlap, which is common in ‘real life’ community settings (e.g., Hombali et al., 2019), separate models were tested for each sleep and circadian problem. This allowed us to test whether TranS-C was associated with improvements, relative to UC-DT, ‘in the presence of’ a given sleep or circadian problem. Second, we tested whether number of sleep and circadian problems moderated treatment outcomes. Specifically, we tested whether there was a significant three-way interaction between time, treatment, and number of sleep and circadian problems for each outcome.

Results

Demographic and clinical characteristics from pre-treatment assessment are in Table 2. Frequencies of sleep and circadian problems are available in Supplement Table 1. As seen in Table 2 and Supplement Figure 1, 121 participants (mean age=45.45 (13.25) years; 52.06% female, 42.98% African American or Black) began the intervention with 60 receiving UC-DT and 61 receiving TranS-C+UC. See Harvey et al. (2021) for addition details about the sample.

Table 2.

Pre-treatment Assessment Demographic and Clinical Characteristics of Participants

UC-DT (n = 60) TranS-C (n = 61)
Characteristic n % n %
Female 33 55.00 30 49.18
Ethnicity
 Hispanic or Latino 9 15.00 10 16.39
 Not Hispanic or Latino 51 85.00 50 81.97
 Missing 0 0.00 1 1.64
Race
 White 21 35.00 25 40.98
 African American/Black 26 43.33 26 42.62
 American Indian or Alaskan Native 4 6.67 4 6.56
 Asian 5 8.33 2 3.28
 Native Hawaiian/Other Pacific Islander 2 3.33 1 1.64
 Missing 2 3.33 3 4.92
Education
 High school or below 22 36.67 19 31.14
 Vocational school 2 3.34 9 14.76
 Some college or completed college 34 56.67 30 49.18
 Graduate school 2 3.34 3 4.92
Employment
 Full-time 1 1.67 1 1.64
 Part-time 6 10.00 9 14.75
 Unemployed 49 81.66 49 80.33
 Other 4 6.67 1 1.64
 Missing 0 0.00 1 1.64
MINI Diagnosis at pre-treatment (current or past)a
 Schizophrenia spectrum disorder 29 49.15 26 43.33
 Bipolar disorder 13 22.03 21 35.00
 Major depressive disorder 17 28.81 11 18.33
 Any anxiety disorder 27 45.76 30 50.00
 Obsessive compulsive disorder 13 22.03 9 15.00
 Post-traumatic stress disorder 12 20.34 6 10.00
 Substance Use Disorder 20 33.90 19 31.67
 Psychotic symptoms/features 42 71.19 39 65.00
M SD M SD
Age (in years) 45.45 13.25 47.97 11.51
Annual personal income $12,732 $15,547 $12,636 $9,850
Annual household income $24,091 $27,507 $26,537 $23,576
Open in a new tab

Note.

a

Comorbidity was common. MINI = Mini-International Neuropsychiatric Interview.

Aim 1: Baseline Functioning by Number and Diagnostic Threshold of Sleep and Circadian Problems

For Aims 1a and 1b, results of ANCOVAs are presented in Table 3, and pairwise comparisons are presented in Table 4. Means and standard deviations of baseline functioning by number of sleep and circadian problems are presented in Supplement Table 2. Means and standard deviations of baseline functioning by diagnostic threshold (full diagnoses, full diagnoses with other subdiagnostic symptoms, subdiagnostic symptoms only) of sleep and circadian problems are presented in Supplement Table 3. Graphs of these mean values with standard deviation and significance bars are presented in Supplement Figures 2–12.

Table 3.

ANCOVA Results – Baseline Functioning by Diagnostic Threshold and Number of Sleep and Circadian Problems

F-statistic p value ω2
Primary Measures
SDS
 Diagnostic threshold F(2, 109) = 1.18 0.31 0.003
 Number of problems F(4, 109) = 6.09 < 0.001 0.15
DSM5
 Diagnostic threshold F(2, 110) = 2.93 0.06 0.03
 Number of problems F(4, 110) = 2.97 0.02 0.06
PROMIS-SD
 Diagnostic threshold F(2, 110) = 1.69 0.19 0.01
 Number of problems F(4, 110) = 2.19 0.07 0.04
PROMIS-SRI
 Diagnostic threshold F(2, 110) = 0.39 0.68 0.01
 Number of problems F(4, 110) = 5.08 < 0.001 0.12
Sleep/Wake Variables
Sleep Health Composite
 Diagnostic threshold F(2, 110) = 0.34 0.71 0.01
 Number of problems F(4, 110) = 5.19 < 0.001 0.13
TST variability - actigraphy
 Diagnostic threshold F(2, 107) = 0.42 0.66 0.01
 Number of problems F(4, 107) = 0.68 0.60 0.01
Open in a new tab

Note. Omega squared (ω2) is reported as an effect size for ANCOVAs. SDS = Sheehan Disability Scale. DSM5 = DSM-5 Cross-Cutting Measure. PROMIS-SD = Patient-Reported Outcomes Measurement Information System–Sleep Disturbance. PROMIS-SRI = Patient-Reported Outcomes Measurement Information System– Sleep-Related Impairment. TST = total sleep time.

Table 4.

Planned Pairwise Comparisons for Adjusted Means Following Significant ANCOVAs

 Measures t-statistic p value
Primary Measures
SDS
 2 vs. 1 t(41) = 3.23 0.01
 3 vs. 1 t(51) = 3.99 0.001
 4 vs. 1 t(38) = 3.84 0.002
 5 vs. 1 t(26) = 4.85 < 0.001
 3 vs. 2 t(62) = 1.47 0.58
 4 vs. 2 t(49) = 1.58 0.51
 5 vs. 2 t(37) = 3.05 0.02
 4 vs. 3 t(59) = 0.32 0.98
 5 vs. 3 t(47) = 2.13 0.22
 5 vs. 4 t(34) = 1.77 0.40
DSM5
 2 vs. 1 t(41) = −0.20 0.99
 3 vs. 1 t(52) = 0.97 0.87
 4 vs. 1 t(38) = 2.15 0.21
 5 vs. 1 t(26) = 1.53 0.55
 3 vs. 2 t(63) = 1.69 0.44
 4 vs. 2 t(49) = 1.69 0.01
 5 vs. 2 t(37) = 2.18 0.20
 4 vs. 3 t(60) = 2.05 0.25
 5 vs. 3 t(48) = 1.02 0.85
 5 vs. 4 t(34) = −0.57 0.98
PROMIS-SRI
 2 vs. 1 t(41) = 2.48 0.10
 3 vs. 1 t(52) = 3.44 0.007
 4 vs. 1 t(38) = 3.66 0.004
 5 vs. 1 t(26) = 4.27 <0.001
 3 vs. 2 t(63) = 1.69 0.45
 4 vs. 2 t(49) = 2.21 0.18
 5 vs. 2 t(37) = 3.05 0.02
 4 vs. 3 t(60) = 0.84 0.92
 5 vs. 3 t(48) = 1.97 0.29
 5 vs. 4 t(34) = 1.23 0.74
Sleep/Wake Variables
Sleep Health Composite
 2 vs. 1 t(41) = −3.33 0.010
 3 vs. 1 t(52) = −3.76 0.003
 4 vs. 1 t(38) = −3.91 0.001
 5 vs. 1 t(26) = −4.38 < 0.001
 3 vs. 2 t(51) = −1.00 0.86
 4 vs. 2 t(49) = −1.55 0.53
 5 vs. 2 t(37) = −2.35 0.14
 4 vs. 3 t(60) = −0.77 0.94
 5 vs. 3 t(48) = −1.75 0.41
 5 vs. 4 t(34) = −1.08 0.82
Open in a new tab

Note. The number of problems and diagnostic threshold compared via planned pairwise t-tests is displayed in the format: x vs. y (e.g., 2 vs. 1 = 2 problems compared with 1 prolem; full vs. sub = full diagnoses only compared with subdiagnostic symptoms only). Significant p-values are bolded. Tukey-Kramer’s correction was used to account for unbalanced data. Full = participants with full diagnoses only. Full-sub = participants with full diagnoses and other subdiagnostic symptoms. Sub = participants with subdiagnostic symptoms only. SDS = Sheehan Disability Scale. DSM5 = DSM-5 Cross-Cutting Measure. PROMIS-SD = Patient-Reported Outcomes Measurement Information System–Sleep Disturbance. PROMIS-SRI = Patient-Reported Outcomes Measurement Information System– Sleep-Related Impairment. TST = total sleep time.

Aim 1a: Baseline Functioning by Number of Sleep and Circadian Problems

ANCOVAs suggested significant differences in baseline functioning by number of sleep and circadian problems, controlling for diagnostic threshold. Specifically, number of sleep and circadian problems was significantly associated with SDS, DSM-5 Cross-Cutting Measure, PROMIS-SRI, and Sleep Health Composite. For SDS, 2 or more problems vs. 1 problem and 5 problems vs. 2 problems were associated with poorer functioning. For DSM-5 Crossing-Cutting Measure, 4 problems vs. 2 problems were associated with worse psychiatric symptoms. For PROMIS-SRI, 3 or more problems vs. 1 problem and 5 problems vs. 2 problems were associated with worse sleep-related impairment. For Sleep Health Composite, 2 or more problems vs. 1 problem were associated with poorer overall sleep health.

Aim 1b: Baseline Functioning by Diagnostic Threshold

ANCOVAs suggested that there were no significant differences in baseline functioning by diagnostic threshold, accounting for number of sleep and circadian problems.

Aim 2: Outcomes for Specific Sleep and Circadian Problems and Moderation by Number of Sleep and Circadian Problems

For Aim 2, multilevel modeling results are presented in Table 5. See Supplement Table 4 for means, standard deviations, and effect sizes of outcomes at each assessment point for each specific sleep and circadian disorder. Time-by-treatment interactions, reported below, are interpretable as the difference between TranS-C+UC relative to UC-DT in mean change from pre-treatment assessment to post-treatment and from pre-treatment assessment to 6FU. Effect sizes of treatment effects for specific sleep and circadian problems are represented as ‘d’ and were calculated using mean change scores (i.e., from pre- to post-treatment and pre-treatment to 6FU) and raw standard deviations from each treatment condition, based on equation 5 in Feingold, 2009).

Table 5.

Multilevel Modeling – Outcomes for Specific Sleep and Circadian Problems

Treatment condition effect at PRE Treatment condition effect at POST Treatment condition effect at 6FU
coef. SE p coef. SE p coef. SE p
Insomnia (n = 113)
SDS −2.26 1.42 0.11 −3.47 1.52 0.02 −1.60 1.51 0.29
DSM5 Cross-Cutting Measure −0.60 1.92 0.75 −6.10 1.91 0.001 −3.84 1.90 0.04
PROMIS-SD −1.07 1.32 0.42 −5.79 1.38 <.001 −4.83 1.38 <.001
PROMIS-SRI −3.63 2.55 0.16 −9.54 2.61 <.001 −5.17 2.59 0.046
TST variability - Actigraphy −7.43 15.65 0.64 6.07 18.82 0.75 −2.37 18.80 0.90
Sleep Health Composite 0.20 0.27 0.46 0.99 0.32 0.002 0.73 0.31 0.02
Hypersomnia (n = 85)
SDS −1.06 1.67 0.53 −3.21 1.67 0.06 −0.92 1.66 0.58
DSM5 Cross-Cutting Measure 0.62 2.37 0.79 −6.08 2.28 0.008 −3.32 2.26 0.14
PROMIS-SD −0.56 1.57 0.72 −5.47 1.54 <.001 −5.08 1.52 0.001
PROMIS-SRI 0.68 3.11 0.83 −9.35 2.85 0.001 −5.54 2.82 0.050
TST variability - Actigraphy 0.52 19.80 0.98 −10.56 24.26 0.66 10.36 24.18 0.67
Sleep health composite 0.07 0.32 0.83 1.10 0.36 0.002 0.59 0.35 0.10
Circadian Rhythm Disorders (n = 41)
SDS −1.57 2.49 0.53 −2.11 2.60 0.42 −4.29 2.57 0.10
DSM5 Cross-Cutting Measure 1.10 3.26 0.74 −5.90 3.18 0.06 −3.36 3.14 0.28
PROMIS-SD 0.09 2.33 0.97 −6.36 2.36 0.007 −7.42 2.32 0.001
PROMIS-SRI −0.25 4.48 0.96 −10.37 4.41 0.02 −8.87 4.35 0.042
TST variability - Actigraphy −17.64 32.91 0.59 20.18 36.72 0.58 32.34 36.23 0.37
Sleep health composite 0.37 0.42 0.41 1.12 0.59 0.06 0.71 0.59 0.23
Periodic Limb Movement and/or Restless Leg Syndrome (n = 45)
SDS −2.33 2.39 0.33 −2.06 2.27 0.36 −3.23 2.20 0.14
DSM5 Cross-Cutting Measure −3.10 2.84 0.28 −7.92 3.02 0.009 −5.29 2.92 0.07
PROMIS-SD −1.86 2.27 0.41 −2.98 2.29 0.19 −5.05 2.22 0.02
PROMIS-SRI −4.29 4.60 0.35 −4.34 4.26 0.31 −6.90 4.12 0.09
TST variability - Actigraphy −2.65 26.03 0.92 29.12 30.84 0.35 4.40 29.82 0.88
Sleep health composite 0.29 0.44 0.51 0.30 0.54 0.59 0.88 0.53 0.10
Parasomnia (n = 57)
SDS −3.03 1.83 0.10 −2.53 1.98 0.20 −1.69 1.97 0.39
DSM5 Cross-Cutting Measure −0.93 2.48 0.71 −6.10 2.81 0.03 −3.97 2.80 0.16
PROMIS-SD 0.88 1.82 0.63 −8.16 1.88 <.001 −6.22 1.87 0.001
PROMIS-SRI −4.66 3.14 0.14 −13.48 3.18 <.001 −6.81 3.17 0.032
TST variability - Actigraphy −1.14 19.61 0.95 −5.66 25.2 0.82 −39.30 24.89 0.11
Sleep health composite −0.12 0.34 0.72 1.48 0.40 <.001 0.36 0.40 0.36
Open in a new tab

Note. Parameters presented are the time-by-treatment interactions, interpreted as the difference between TranS-C+UC relative to UC-DT in mean change from pre-treatment assessment to post-treatment and from pre-treatment assessment to 6FU. PRE = pre-treatment assessment. POST = post-treatment assessment. 6FU = 6-month follow-up assessment. SDS = Sheehan Disability Scale. DSM5 = DSM-5 Cross-Cutting Measure. PROMIS-SD = Patient-Reported Outcomes Measurement Information System–Sleep Disturbance. PROMIS-SRI = Patient-Reported Outcomes Measurement Information System–Sleep-Related Impairment. TST variability – Actigraphy = total sleep time variability measured by actigraphy.

Insomnia

For participants with full diagnoses or subdiagnostic symptoms insomnia, TranS-C+UC demonstrated reductions in SDS (b=−3.47, p=0.02, d=−0.65), DSM-5 Cross-Cutting Measure (b=−6.10, p=0.001, d=−0.66), PROMIS-SD (b=−5.79, p<0.001, d=−1.03), and PROMIS-SRI (b=−9.54, p<0.001, d=−0.85) at post-treatment. Treatment gains for TranS-C+UC were maintained through 6FU for DSM-5 Cross-Cutting Measure (b=−3.84, p=0.04, d=−0.42), PROMIS-SD (b=−4.83, p<0.001, d=−0.79), and PROMIS-SRI (b=−5.17, p=0.046, d=−0.52). For the Sleep Health Composite, TranS-C+UC exhibited improved sleep health at post-treatment (b=0.99, p=0.002, d=0.61) and to 6FU (b=0.73, p=0.02, d=0.43).

Hypersomnia

At post-treatment, for participants with full diagnoses or subdiagnostic symptoms of hypersomnia, participants in TranS-C+UC demonstrated a reduction in DSM-5 Cross-Cutting Measure (b=−6.08, p=0.008, d=−0.57), PROMIS-SD (b=−5.47, p<0.001, d=−1.00), and PROMIS-SRI (b=−9.35, p=0.001, d=−0.82). These treatment gains were maintained through 6FU for PROMIS-SD (b=−5.08, p=0.001, d=−0.98). For the Sleep Health Composite, TranS-C+UC exhibited improved sleep health at post-treatment (b=1.10, p=0.002, d=0.64) though treatment gains were not maintained through 6FU.

Circadian Rhythm Disorders

For participants with full diagnoses or subdiagnostic symptoms of circadian rhythm disorders, at post-treatment, participants in TranS-C+UC demonstrated a reduction in PROMIS-SD (b=−6.36, p=0.007, d=−0.98) and PROMIS-SRI (b=−10.37, p=0.02, d=−1.01). Treatment gains were maintained through 6FU for both measures: PROMIS-SD (b=−7.42, p=0.001, d=−1.14) and PROMIS-SRI (b=−8.87, p=0.042, d=−0.94).

Periodic Limb Movement and Restless Leg Syndrome

For participants with full diagnoses or subdiagnostic symptoms of periodic limb movement and/or restless leg syndrome, TranS-C+UC was associated with a reduction in DSM-5 Cross-Cutting Measure (b=−7.92, p=0.009, d=−1.14) at post-treatment and PROMIS-SD (b=−5.05, p=0.02, d=−0.80) at 6FU.

Parasomnias

For participants with full diagnoses or subdiagnostic symptoms of parasomnias, at post-treatment, participants in TranS-C+UC demonstrated a reduction in DSM-5 Cross-Cutting Measure (b=−6.10, p=0.03, d=−0.81), PROMIS-SD (b=−8.16, p<0.001, d=−1.51), and PROMIS-SRI (b=−13.48, p<0.001, d=−1.29). Treatment gains for TranS-C+UC were maintained through 6FU for PROMIS-SD (b=−6.22, p=0.001, d=−1.09) and PROMIS-SRI (b=−6.81, p=0.03, d=−0.64). For the Sleep Health Composite, TranS-C+UC exhibited improved sleep health at post-treatment (b=1.48, p<0.001, d=1.07), though treatment gains were not maintained through 6FU.

Moderation by Number of Sleep and Circadian Problems

Multilevel moderation results of TranS-C+UC outcomes by number of sleep and circadian problems are presented in Table 6. Number of sleep and circadian problems did not moderate effects of TranS-C+UC at post-treatment or 6FU for any outcomes (all ps > .05).

Table 6.

Multilevel Modeling Moderation 3-Way Interactions - Time by Treatment by Number of Sleep and Circadian Problems

3-Way Interaction at POST 3-Way Interaction at 6FU
coef. SE p coef. SE p
SDS 1.75 1.12 0.12 −0.27 1.11 0.81
DSM5 Cross-Cutting Measure 0.84 1.45 0.56 0.47 1.43 0.74
PROMIS-SD −0.42 1.06 0.69 −0.37 1.05 0.73
PROMIS-SRI 0.10 1.96 0.96 −0.30 1.93 0.88
TST variability - Actigraphy 0.74 15.12 0.96 8.43 14.93 0.57
Sleep Health Composite 0.29 0.24 0.23 −0.03 0.24 0.89
Open in a new tab

Note. Parameters presented are the 3-way interactions between time, treatment, and number of sleep and circadian problems. POST = post-treatment assessment. 6FU = 6-month follow-up assessment. SDS = Sheehan Disability Scale. DSM5 = DSM-5 Cross-Cutting Measure. PROMIS-SD = Patient-Reported Outcomes Measurement Information System–Sleep Disturbance. PROMIS-SRI = Patient-Reported Outcomes Measurement Information System–Sleep-Related Impairment. TST variability – Actigraphy = total sleep time variability measured by actigraphy.

Discussion

The present study evaluated comorbidities and treatment outcomes in the context of a transdiagnostic treatment for sleep and circadian problems delivered to individuals diagnosed with SMI in a community mental health center. At the outset, we highlight the high rates of comorbidity. Less than 15% of participants presented with a full diagnosis or subdiagnostic symptoms of a single sleep and circadian problem. In other words, the majority of participants (86.77%) had full diagnoses or subdiagnostic symptoms of at least two sleep and circadian problems (Supplement Table 2). These findings corroborate reports that comorbidity is the norm, rather than the exception, and underscore the need for research on comorbidity between sleep and circadian problems (Dalgleish et al., 2020; Hombali et al., 2019)

Aim 1a was to evaluate baseline functioning associated with number of sleep and circadian problems. When compared to a single sleep and circadian problem, participants with two or more problems endorsed greater overall impairment, psychiatric symptoms, and more sleep and circadian dysfunction at baseline. Notably, findings controlled for diagnostic threshold, meaning that number of problems was uniquely associated with functioning, beyond whether those problems met full diagnostic criteria.

There are two caveats worth noting for Aim 1a. First, the evidence that number of sleep and circadian problems was associated with general psychiatric symptoms was rather limited, with only one significant comparison (i.e., 4 problems vs. 2 problems). The explanation for this limited evidence is unclear. For instance, it is possible that this finding is spurious. It is also possible that different mental illnesses may have unique relationships to number of sleep and circadian problems. Future research with more specific measures of mental illness (e.g., bipolar disorder, depressive disorders) would help expand on these initial findings. Second, an exception to the pattern of findings from Aim 1a was that number of sleep and circadian problems was not associated with the actigraphy parameter of total sleep time variability. However, given the wide inclusion gates inherent to transdiagnostic research, the multi-dimensional Sleep Health Composite may be a more informative metric of sleep and circadian improvements relative to individual parameters, such as total sleep time variability (Dong et al., 2019). Indeed, individuals with two or more problems reported worse overall sleep on the Sleep Health composite than those with a single sleep and circadian problem. Together, these findings add to the growing literature that comorbidity—in this case, sleep and circadian comorbidity—is associated with more severe dysfunction (Kessler et al., 2005).

Aim 1b was to evaluate baseline functioning associated with diagnostic threshold of sleep and circadian problems (i.e., full diagnoses only, full diagnoses with other subdiagnostic symptoms, subdiagnostic symptoms only). Diagnostic threshold was not associated with poorer baseline functioning, controlling for number of sleep and circadian problems. In other words, impairment was as significant for participants with subdiagnostic symptoms as it was for participants who met full diagnostic criteria. Together, diagnostic threshold may have less of an impact on functioning than number of sleep and circadian problems. This explanation is consistent with evidence that subdiagnostic symptoms are associated with poor psychiatric outcomes (e.g., Balazs et al., 2013; Marangell, 2004). Clinically, these findings highlight the importance of evaluating and treating subdiagnostic symptoms and support calls for dimensional versus categorical approaches to mental illness (Cuthbert & Insel, 2013; Dalgleish et al., 2020).

Aim 2 was to evaluate outcomes of TranS-C+UC versus UC-DT for specific sleep and circadian problems. We also tested whether number of sleep and circadian problems moderated treatment effects. First, with respect to the latter, number of sleep and circadian problems did not moderate treatment outcomes. This finding should be considered in light of the high rates of comorbidity between sleep and circadian problems observed in the present trial and other psychiatric samples from the community (e.g., Sarfan et al., 2021; Hombali et al., 2019). These high rates of comorbidity underscore the need for transdiagnostic treatments that improve functioning for individuals with more than one sleep and circadian problem, like TranS-C+UC.

Second, across all models testing TranS-C+UC for individuals with specific sleep and circadian problems, TranS-C+UC relative to UC-DT was associated with significant improvements in psychiatric symptoms, sleep-related disturbance and daytime impairment, and overall sleep health by post-treatment and/or 6FU. Extending prior research on TranS-C+UC efficacy (e.g., Harvey et al., 2021), the results offer additional evidence that may support the use of TranS-C+UC for participants with a wide range of specific sleep and circadian problems, including insomnia, hypersomnia, parasomnias, periodic limb movement and restless leg syndrome, and circadian rhythm disorders. For example, consider a patient who presents to treatment with symptoms of insomnia, hypersomnia, parasomnias, and/or restless leg syndrome. Our results suggest that TranS-C+UCT may confer clinical benefits in the presence of each of these sets of symptoms, and therefore, may be a useful treatment for this patient. Although definitive conclusions cannot be drawn due to overlap across models of participants with specific sleep and circadian problems, these results may help clinicians determine whether TranS-C+UC could be beneficial, when faced with a particular patient’s presenting symptoms or diagnoses.

Taking a broader perspective, findings from Aim 2 replicate prior research demonstrating that sleep treatments improve both symptoms of comorbid mental health conditions and sleep and circadian functioning (Taylor & Pruiksma, 2014). Moreover, this study adds to the evidence for high rates of comorbidity (Kessler et al, 2005) and growing recognition of ‘massive heterogeneity’ across symptoms and comorbidities (Dalgleish et al., 2020, p. 181; Sarfan et al., 2021). Together, the results support proposals that the transdiagnostic approach may represent an efficient path to treatment in ‘real world’ practice settings, such as community mental health centers. Specifically, by targeting common processes and impairment associated with comorbid symptoms, transdiagnostic treatments may reduce training costs, ease provider burden, and expedite patient recovery (Mansell et al., 2009; McHugh et al., 2009)

There were two exceptions to the pattern of TranS-C outcomes. First, for participants with circadian rhythm disorders, sleep-related disruption and daytime impairment improved in TranS-C+UC relative to UC-DT, but other outcomes did not. A visual inspection of means and effect sizes for these individuals suggests an advantage for TranS-C+UC that mirrors the pattern for the other sleep and circadian problems (i.e., improvements in functional impairment, psychiatric symptoms, sleep/wake patterns, see Supplement Table 4). That said, other findings from our group suggest that the optional modules for circadian rhythm disorders were not delivered consistently for participants with circadian rhythm diagnoses (Sarfan et al., 2021), which may account for these findings. Second, for participants with periodic limb movement or restless leg syndrome, psychiatric symptoms and sleep disruption improved, but other symptoms did not. Similar to the circadian rhythm disorders, effect sizes for overall impairment, daytime sleep-related impairment, and overall sleep health were in the anticipated direction (see Supplement Table 4). However, it is possible that participants with periodic limb movement or restless leg syndrome need additional support, such as treatment modules that more specifically target their uncomfortable physical sensations. Given that few psychosocial treatments are currently recommended for periodic limb movement and restless leg syndrome (Aurora et al., 2012), this represents an exciting direction for future research.

Several limitations warrant consideration. First, our findings drew from an adequately powered parent trial (Harvey et al., 2021), and we followed multilevel modeling guidelines for sample size (McNeish & Stapleton, 2016). However, findings should be replicated and extended with larger samples. In particular, future research would benefit from larger samples of each sleep and circadian problem without needing to ‘double count’ participants across models. This will be an important challenge for comorbidity research in ‘real world’ treatment settings, given the substantial heterogeneity across combinations of sleep and circadian problems, which can be observed in Supplement Table 1 and community samples from prior research (e.g., Hombali et al., 2019). Further, we encourage readers to focus on the effect sizes of the present results, rather the values of statistical significance, given the relatively small sample size. Second, assessors were blind to treatment condition, but participants were not, which may have influenced self-report outcomes. Third, we did not use the gold standard assessment for all sleep diagnoses. Mostly notably, we did not include polysomnography, which is an objective sleep measurement required for several sleep diagnoses, including periodic limb movement disorder. Thus, the full diagnosis could not be determined for this disorder (American Academy of Sleep Medicine, 2014). Other than subdiagnostic symptoms of periodic limb movement disorder, we did not include sleep and circadian problems that require polysomnography for diagnosis. For example, we did not include assessment of obstructive sleep apnea, which may be more common among individuals with select SMI diagnoses—particularly major depressive disorder—compared to the general population (e.g., Stubbs et al., 2015). Expanding the results of the present study to sleep and circadian problems that require polysomnography for diagnosis will be an important next step for future research. Related, we used the DSM-5 criteria to identify diagnoses and symptoms of sleep and circadian problems, except periodic limb movement disorder, which is not in the DSM-5. For some sleep and circadian problems, the symptoms and assessments methods required for diagnosis by the DSM-5 differ from other diagnostic systems, such as the ICSD-3 (American Academy of Sleep Medicine, 2014). Consequently, findings may not generalize to individuals with sleep and circadian problems who were not assessed according to the DSM-5. Fourth, subdiagnostic symptoms were not determined until after completion of the study. Assessment of subdiagnostic symptoms prior to treatment might have influenced provider decision-making regarding which TranS-C modules to administer, and which in turn, could have affected treatment outcomes. Fifth, participants in this study were individuals seeking treatment in a community mental health center. The extent to which the findings generalize to other samples in an important topic for future research.

In summary, findings from the present study add to the growing literature on comorbidity by suggesting that higher numbers of sleep and circadian problems (versus a single problem) are associated with more severe overall impairment, psychiatric symptoms, and sleep and circadian dysfunction. Moreover, diagnostic threshold (e.g., full diagnosis vs. subdiagnostic symptoms) was not associated with poorer baseline functioning. Together, results suggest that the number of sleep and circadian problems—versus whether those problems meet diagnostic criteria—may be more impactful for functioning. Finally, TranS-C+UC versus UC-DT was associated with improved psychosocial functioning and sleep and circadian functioning across a range of specific sleep and circadian problems (e.g., insomnia, hypersomnia, parasomnias). Treatment effects were not moderated by number of sleep and circadian problems. Thus, TranS-C+UC may be an effective, transdiagnostic clinical choice for clinicians treating individuals with SMI and sleep and circadian problems in community mental health settings.

Supplementary Material

TranS-C Comorbidity Supplement
TranS-C Comorbidity Highlights

Acknowledgements

Trial Registration: clinicaltrials.gov identifier: NCT01828320. This study was supported by the National Institute of Mental Health (R01MH105513). The authors are grateful to the following team members for their consultation and support: Sophia Rabe-Hesketh, Vera Portnova, Firdows Mujir, and Charlie Schroeder.

Funding:

This study was funded by the National Institute of Mental Health (MH105513). This funding source had not no role in study design; collection, analysis and interpretation of data; writing the report; and the decision to submit for publication.

Declaration of Interest:

AGH has received research support from the National Institutes of Health and book royalties from American Psychological Association, Guilford Press, and Oxford University Press. The views expressed in this article do not represent those of any public entity. The other authors do not have interests to declare.

Footnotes

1

Abbreviations: ANCOVA = analysis of covariance; SMI = serious mental illness; TranS-C = Transdiagnostic Intervention for Sleep and Circadian Disorders; UC = Usual Care; UC+DT = usual care followed by delayed treatment; 6FU = 6-month follow-up.

References

  1. American Academy of Sleep Medicine. (2014). International classification of sleep disorders (3rd ed.).
  2. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.).
  3. Aurora RN, Kristo DA, Bista SR, Rowley JA, Zak RS, Casey KR, Lamm CI, Tracy SL, & Rosenberg RS (2012). The treatment of restless legs syndrome and periodic limb movement disorder in adults—an update for 2012: Practice parameters with an evidence-based systematic review and meta-analyses: An American Academy of Sleep Medicine Clinical Practice Guideline. Sleep, 35(8), 1039–1062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Balazs J, Miklosi M, Kereszteny A, Hoven CW, Carli V, Wasserman C, Apter A, Bobes J, Brunner R, Cosman D, Cotter P, Haring C, Iosue M, Kaess M, Kahn J-P, Keeley H, Marusic D, Postuvan V, Resch F..Wasserman D (2013). Adolescent subthreshold-depression and anxiety: Psychopathology, functional impairment and increased suicide risk. Journal of Child Psychology and Psychiatry, 54, 670–677. [DOI] [PubMed] [Google Scholar]
  5. Bei B, Seeman TE, Carroll JE, & Wiley JF (2017). Sleep and physiological dysregulation: A closer look at sleep intraindividual variability. SLEEP, 40(9), 1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bei B, Wiley JF, Trinder J, & Manber R (2016). Beyond the mean: A systematic review on the correlates of daily intraindividual variability in sleep/wake patterns. Sleep Medicine Reviews, 28, 108–124. [DOI] [PubMed] [Google Scholar]
  7. Buysse DJ (2010). Sleep and Psychiatric Disorders: A Revisit and Reconceptualization. The Canadian Journal of Psychiatry, 55, 401–402. [DOI] [PubMed] [Google Scholar]
  8. Buysse DJ (2014). Sleep Health: Can we define it? Does it matter? Sleep, 37(1), 9–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Buysse DJ, Ancoli-Israel S, Edinger JD, Lichstein KL, & Morin CM (2006). Recommendations for a standard research assessment of insomnia. Sleep, 29, 1155–1173. [DOI] [PubMed] [Google Scholar]
  10. Clarke DE, & Kuhl EA (2014). DSM-5 cross-cutting symptom measures: a step towards the future of psychiatric care? World Psychiatry, 13, 314–316. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Clark LA, Cuthbert B, Lewis-Fernandez R, Narrow WE, & Reed GM (2017). Three approaches to understanding and classifying mental disorders: ICD-11, DSM-5, and the National Institute of Mental Health’s Research Domain Criteria (RDoc), 18, 72–145. [DOI] [PubMed] [Google Scholar]
  12. Cuijpers P, Smit F, & van Straten A (2007). Psychological treatments of subthreshold depression: A meta-analytic review. Acta Psychiatrica Scandinavica, 115; 434–441. [DOI] [PubMed] [Google Scholar]
  13. Cuthbert BN, & Insel TR (2013). Toward the future of psychiatric diagnosis: The seven pillars of RDoC. BMC Medicine, 11, 126–134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dalgleish T, Black M, Johnston D, & Bevan A (2020). Transdiagnostic Approaches to Mental Health Problems: Current Status and Future Directions. Journal of Consulting and Clinical Psychology, 88(3), 179–195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dong L, Martinez AJ, Buysse DJ, & Harvey AG (2019). A composite measure of sleep health predicts concurrent mental and physical health outcomes in adolescents prone to eveningness. Sleep Health, 5, 166–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Edinger JD, Fins AI, Sullivan RJ, Marsh GR, Dailey DS, & Young M (1996). Comparison of cognitive-behavioral therapy and clonazepam for treating periodic limb movement disorder. Sleep: Journal of Sleep Research & Sleep Medicine, 19, 442–444. [PubMed] [Google Scholar]
  17. Edinger J, Kirby A, Lineberger M, Loiselle M, Wohlgemuth W, & Means M (2004). The Duke structured interview for sleep disorders. Durham: University Medical Center. [Google Scholar]
  18. Edinger JD, Wyatt JK, Olsen MK, Stechuchak KM, Carney CE, Chiang A, Krystal AD, Lineberger MD, Means MK, & Radtke RA (2009). Reliability and validity of the Duke Structured Interview for sleep disorders for insomnia screening. Sleep, A265. [Google Scholar]
  19. Field A (2013). Discovering statistics using IBM SPSS statistics (4th ed.). Sage. [Google Scholar]
  20. Ford DE, & Kamerow DB (1989). Epidemiologic study of sleep disturbance and psychiatric disorders: An opportunity for prevention? JAMA, 262, 1479–1484. [DOI] [PubMed] [Google Scholar]
  21. Geoffroy PA, Hoertel N, Etain B, Bellivier F, Delorme R, Limosin F, & Peyre H (2018). Insomnia and hypersomnia in major depressive episode: Prevalence, sociodemographic characteristics and psychiatric comorbidity in a population-based study. Journal of Affective Disorders, 226, 132–141. [DOI] [PubMed] [Google Scholar]
  22. Gruber J, Milkowitz DJ, Harvey AG, Frank E, Kupfer D, Thase ME, Sachs GS, & Ketter TA (2011). Sleep matters: Sleep functioning and course of illness over bipolar disorder. Journal of Affective Disorders, 134, 416–420. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Harvey AG, Dong L, Hein K, Yu SH, Martinez A, Gumport NB, … Buysse DJ (2021). A randomized controlled trial of the Transdiagnostic Intervention for Sleep and Circadian Dysfunction (TranS-C) to improve serious mental illness outcomes in a community setting. Journal of Consulting and Clinical Psychology, 89, 537–550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Harvey AG, Hein K, Dong L, Smith FL, Lisman M, Yu S, Rabe-Hesketh S, & Buysse DJ (2016). A transdiagnostic sleep and circadian treatment to improve severe mental illness outcomes in a community setting: Study protocol for a randomized controlled trial. Trials, 17, 606–617. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Harvey AG, Watkins ER, Mansell W, & Shafran R (2004). Cognitive behavioural processes across psychological disorders: A transdiagnostic approach to research and treatment. Oxford University Press. [Google Scholar]
  26. Hertenstein E, Feige B, Gmeiner T, Kienzler C, Spiegelhalder K, Johann A, Jansson-Fröjmark M, Palagini L, Rücker G, Riemann D, Baglioni C (2019). Insomnia as a predictor of mental disorders: A systematic review and meta-analysis. Sleep Medicine Reviews, 43, 96–105. [DOI] [PubMed] [Google Scholar]
  27. Hombali A, Seow E, Yuan Q, Chang SHS, Satghare P, Kumar S, Verma WK, Mok YM, Chong s. A., & Subramaniam M (2019). Prevalence and correlates of sleep disorder symptoms in psychiatric disorders. Psychiatry Research, 279, 116–122. [DOI] [PubMed] [Google Scholar]
  28. Kaplan KA, McGlinchey EL, Soehner A, Gershon A, Talbot LS, Eidelman P, Gruber J, & Harvey AG (2015). Hypersomnia subtypes, sleep and relapse in bipolar disorder. Psychological Medicine, 45, 1751–1763. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kessler RC, Chiu WT, Delmer O, & Walters EE (2005). Prevalence, severity, and comorbidity of twelve-month DSM-IV disorders in the National Comorbidity Survey Replication (NCS-R). Archives of General Psychiatry, 62, 617–627. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kivela L, Papadopoulos MR, & Antypa N (2018). Chronotype and psychiatric disorders. Current Sleep Medicine Reports, 4, 94–103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kline RB (2011). Principles and practice of structural equation modeling. The Guilford Press. [Google Scholar]
  32. Laborde-Lahoz P, El-Gabalawy R, Kinley J, Kirwin PD, Sareen J, & Pietrzak RH (2015). Subsyndromal depression among older adults in the USA: Prevalence, comorbidity, and risk for new-onset psychiatric disorders in late life. International Journal of Geriatric Psychiatry, 30, 677–685. [DOI] [PubMed] [Google Scholar]
  33. Lemola S, Ledermann T, & Friedman EM (2013). Variability of sleep duration is related to subjective sleep quality and subjective well-being: An actigraphy study. PLoS ONE, 8(8), e71292. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Leon AC, Olfson M, Portera L, Farber L, & Sheehan DV (1997). Assessing Psychiatric Impairment in Primary Care with the Sheehan Disability Scale. The International Journal of Psychiatry in Medicine, 27(2), 93–105. [DOI] [PubMed] [Google Scholar]
  35. Mansell W, Harvey H, Watkins E, & Shafran R (2009). Conceptual foundations of the transdiagnostic approach to CBT. Journal of Cognitive Psychotherapy: An International Quarterly, 23(1), 6–19. [Google Scholar]
  36. Marangell LB (2004). The importance of subsyndromal symptoms in bipolar disorder. Journal of Clinical Psychiatry, 65, 24–27. [PubMed] [Google Scholar]
  37. McHugh RK, Murray HW, & Barlow DH (2009). Balancing fidelity and adaptation in the dissemination of empirically-supported treatments: The promise of transdiagnostic interventions. Behaviour Research and Therapy, 47, 946–953. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. McNeish DM, & Stapleton LM (2016). The effect of small sample size on two-level model estimates: A review and illustration. Educational Psychology Review, 28, 295–314. [Google Scholar]
  39. Morin CM (1993). Insomnia: Psychological assessment and management. Guilford Press. [Google Scholar]
  40. Narrow WE, Clarke DE, Kuramoto SJ, Kraemer HC, Kupfer DJ, Greiner L, & Regier DA (2013). DSM-5 field trials in the United States and Canada, Part III: Development and reliability testing of a cross-cutting symptom assessment for DSM-5. American Journal of Psychiatry, 170(1), 71–82. [DOI] [PubMed] [Google Scholar]
  41. Parish JM (2009). Sleep-related problems in common medical conditions. Chest, 135, 563–572. [DOI] [PubMed] [Google Scholar]
  42. Reeve S, Sheaves B, & Freeman D (2019). Sleep disorders in early psychosis: Incidence, severity, and association with clinical symptoms. Schizophrenia Bulletin, 45, 287–295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Roth T, Jaeger S, Jin R, Kalsekar A, Stang PE, & Kessler RC (2006). Sleep problems, comorbid mental disorders, and role functioning in the National Comorbidity Survey Replication (NCS-R). Biological Psychiatry, 60, 1364–1371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Salkind NJ (2010). Encyclopedia of research design (Vol. 1–2). SAGE Publications, Inc. [Google Scholar]
  45. Sarfan LD, Hilmoe HE, Gumport NB, & Harvey AG (2021). The Transdiagnostic Intervention for Sleep and Circadian Dysfunction (TranS-C) in community mental health settings under the microscope [Manuscript Under Review]. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sheehan DV, Harnett-Sheehan K, & Raj BA (1996). The measurement of disability. International Clinical Psychopharmacology, 11(Suppl 3), 89–95. [DOI] [PubMed] [Google Scholar]
  47. Stubbs B, Vancampfort D, Veronese N, Solmi M, Gaughran F, Manu P, Rosenbaum S, De Hert M, & Fornaro M (2016). The prevalence and predictors of obstructive sleep apnea in major depressive disorder, bipolar disorder and schizophrenia: A systematic review and meta-analysis. Journal of Affective Disorders, 197, 259–267. [DOI] [PubMed] [Google Scholar]
  48. Taylor DJ, & Pruiksma KE (2014). Cognitive and behavioural therapy for insomnia (CBT-I) in psychiatric populations: A systematic review. International Review of Psychiatry, 26, 205–213. [DOI] [PubMed] [Google Scholar]
  49. Wang PS, Demler O, & Kessler RC (2002). Adequacy of treatment for serious mental illness in the United States. American Journal of Public Health, 92(1), 92–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Wolitzky-Taylor K, Dour H, Zinbarg R, Mineka S, Vrshek-Schallhorn S, Epstein A, Borbova L, Griffith J, Waters A, Nazarian M, Rose R, & Craske MG (2014). Experiencing core systems of anxiety and unipolar mood disorders in late adolescence predicts disorder onset in early adulthood. Depression and Anxiety, 31, 207–213. [DOI] [PubMed] [Google Scholar]
  51. Yu L, Buysse DJ, Germain A, Moul DE, Stover A, Dodds NE, Johnston KL, & Pilkonis PA (2011). Development of Short Forms From the PROMIS™ Sleep Disturbance and Sleep-Related Impairment Item Banks. Behavioral Sleep Medicine, 10(1), 6–24. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

TranS-C Comorbidity Supplement
NIHMS1965919-supplement-TranS-C_Comorbidity_Supplement.docx (148.9KB, docx)
TranS-C Comorbidity Highlights
NIHMS1965919-supplement-TranS-C_Comorbidity_Highlights.docx (13.9KB, docx)

RESOURCES