Psychological Assessment 2014, Vol. 26, No. 2, 586-597
© 2014 American Psychological Association 1040-3590/14/$12.00 DOI: 10.1037/a0036276
Are Sluggish Cognitive Tempo and Daytime Sleepiness Distinct Constructs? Joshua M. Langberg
Stephen P. Becker
Virginia Commonwealth University
Miami University
Melissa R. Dvorsky
Aaron M. Luebbe
Virginia Commonwealth University
Miami University
Sluggish cognitive tempo (SCT) and daytime sleepiness are both common in individuals with attentiondeficit/hyperactivity disorder (ADHD). There appears to be considerable overlap between the tired and lethargic aspects of SCT and behaviors frequently exhibited by individuals with daytime sleepiness. However, no studies have examined the degree to which these constructs overlap and whether or not they are empirically distinct. In Study 1, a confirmatory factor analysis with the SCT subscale of the Barkley Adult ADHD Rating Scale-IV (BAARS-IV) and the Epworth Sleepiness Scale (ESS) was conducted in a sample of 768 college students. Results demonstrated that SCT and daytime sleepiness exhibit considerable overlap but are empirically distinct. In Study 2, we examined the relation between SCT and daytime sleepiness and also the impact of comorbid SCT and sleepiness on the functioning of 58 college students rigorously diagnosed with ADHD. Regression analyses in both Study 1 and Study 2 showed that SCT predicts daytime sleepiness above and beyond symptoms of ADHD, anxiety, and depres.sion. The 2 constructs were significantly related (r = .51 ), with the highest correlations occurring between the SCT tired and lethargic items with daytime sleepiness. College students with ADHD + SCT and daytime sleepiness were significantly more impaired than college students diagnosed with ADHD without SCT or daytime sleepiness. Together, these results fill an important gap in the literature by confirming SCT to be overlapping but empirically distinct from daytime sleepiness and demonstrating that SCT and daytime sleepiness are associated with functioning in college students with ADHD. Keywords: ADHD, attention-deficit/hyperactivity disorder, sleep, sluggish cognitive tempo, college students
Lecendreux, 2009). Not sutprisingly given these problems, individuals with ADHD often report feeling excessively sleepy during the daytime hours (Córtese et al., 2009; Owens, 2009). Daytime sleepiness is characterized by a lack of energy, persistent tiredness, and feelings of drowsiness and sluggishness (Drake et al, 2003). Daytime sleepiness is associated with increased risk for academic and behavioral impairment in samples of youth with (Langberg, Dvorsky, Marshall, & Evans, 2013) and without (Astill, Van der Heijden, Van IJzendoorn, & Van Someren, 2012; Beebe, 2011; Fallone, Acebo, Seifer, & Carskadon, 2005) ADHD. In a largely separate literature, approximately 30%-50% of youth with ADHD have been shown to exhibit slow processing, sluggishness, apathy, drowsiness, and inconsistent alertness, behaviors that have been labeled sluggish cognitive tempo (SCT). Although some SCT items emerged in the literature in the 1960s and 1970s (see Becker, Marshall, & McBumett, 2014, for a historical overview of the SCT construct), it was in the 1980s that Lahey and colleagues (e.g., Lahey et al., 1988; Neeper & Lahey, 1986) first demonstrated SCT symptoms to be distinct from the inattentive and hyperactive-impulsive symptoms characteristic of ADHD. As research examining the SCT construct began to advance, it was initially posited that a significant minority of children with ADHD, Predominantly Inattentive Type (ADHD-I) were best characterized by the presence of SCT symptoms and relative
Children and adolescents with attention-deficit/hyperactivity disorder (ADHD) frequently experience sleep disturbances, with prevalence rates for sleep problems in samples of youth with ADHD between 25% and 50% (Sung, Hiscock, Sciberras, & Efron, 2008; Yoon, Jain, & Shapiro, 2012). The relation between sleep and ADHD is not limited to childhood, with as many as 50% of adults with ADHD endorsing the presence of sleep disturbances such as insomnia (Voinescu, Szentagotai, & David, 2012). Individuals with ADHD experience a wide range of sleep disturbances, including higher rates of problems with sleep latency, bedtime resistance, night awakenings, and difficulties with morning awakenings in comparison to their peers (Córtese, Faraone, Konofal, &
This article was published Online First March 10, 2014. Joshua M. Langberg, Department of Psychology, Virginia Commonwealth University; Stephen P. Becker. Department of Psychology, Miami University; Melissa R. Dvorsky, Department of Psychology, Virginia Commonwealth University; Aaron M. Luebbe, Department of Psychology, Miami University. Correspondence concerning this article should he addressed to Joshua M. Langberg, Department of Psychology, Virginia Commonwealth University, 806 West Franklin Street, P.O. Box 842018, Richmond, VA 23284-2018. E-mail: [email protected] 586
SCT AND SLEEP absence of hyperactive/impulsive symptoms (Carlson & Mann, 2002; Lahey et al., 1988; McBumett, Pfiffner, & Frick, 2001; Milich, Balentine, & Lynam, 2001). Still, although SCT symptoms are more strongly related to ADHD symptoms of inattention than to symptoms of hyperactivity/impulsivity (Willcutt et al., 2012), a sizable number of individuals with ADHD Combined Type also display clinically significant levels of SCT symptoms (Barkley, 2012; Carlson & Mann, 2002; Willcutt et al., 2012). Similar to sleep disturbances, the presence of SCT is not unique to samples of children with ADHD, with approximately 50% of adults with elevated ADHD symptoms exhibiting clinically significant SCT (Barkley, 2012). Further, comorbid SCT symptoms are associated with increased impairment in social, academic, and executive functioning in youth with ADHD (Barkley, 2013; Becker & Langberg, 2013, 2014; Carison & Mann, 2002; Langberg, Becker, & Dvorsky, 2014; Marshall et al., 2014; McBumett et al., 2001), as well as among adults displaying elevated ADHD symptoms (Barkley, 2012). Multiple authors have noted the considerable similarities and potential overlap associated with measures of SCT and measures of daytime sleepiness (Córtese et al., 2009; Fallone et al., 2005; Mayes et al., 2009; Willoughby, Angold, & Egger, 2008). However, only one study published to date has examined the relation between these two constmcts, and no research has evaluated whether or not these two constmcts are empirically distinct. Specifically, Voinescu and colleagues (2012) examined the prevalence of sleep disturbances in a sample of adults with symptoms of ADHD. In this sample of 550 Romanian adults, participants were classified into an ADHD-likely group, a non-ADHD group, and a chronic insomnia group based on self-report on the Barkley Adult ADHD Rating Scale (BAARS-IV; Barkley, 2011a), which includes an SCT subscale along with Diagnostic and Statistical Manual of Mental Disorders, fourth edition {DSM-IV; American Psychiatric Association [APA], 2000) symptoms of ADHD. Adults in the insomnia and ADHD-likely groups had significantly higher scores on the SCT subscale in comparison to the nonADHD group. Further, in the ADHD-likely group, the correlation between the insomnia and SCT measures was significant and large {r = .66). To our knowledge, this is the only study completed to date in which the association between SCT and sleep has been reported. SCT is a multifaceted constmct, and it is likely that certain aspects of SCT are more highly related to daytime sleepiness than others. Recent measure development studies suggest that SCT consists of slow, sleepy, and daydreaming factors (Jacobson et al., 2012; Penny et al, 2009). The slow aspect of SCT includes apathy, motivation, being slow or delayed in completing tasks, and lacking initiative. Sleepy SCT behaviors include drowsiness, appearing sluggish or tired, and being underactive; daydreaming SCT behaviors include getting lost in one's own thoughts and seeming to be in one's own world. Importantiy, each of these SCT factors relates differently to the ADHD symptom dimensions. The SCT behaviors related to being apathetic and lacking motivation/initiative largely load onto a single factor along with DSM-IV inattention items (Barkley, 2013; Lee, Bums, Snell, & McBumett, 2014), whereas the sleepy/daydreamy aspects of SCT appear to be empirically distinct from DSM-IV symptoms of inattention (Becker, Luebbe, Fite, Stoppelbein, & Greening, 2014; Jacobson et al., 2012; Penny et al., 2009; Willcutt et al., 2012). These findings have implica-
587
tions for studies examining the relation between sleep and SCT. Specifically, the SCT behaviors found to be most distinct from the ADHD inattention dimension are those that are most highly characteristic of individuals with sleep disturbances (i.e., drowsy and sluggish/tired). Accordingly, it is possible that SCT drowsy/tired behaviors are actually manifestations of daytime sleepiness. Alternatively, it is possible that SCT and daytime sleepiness are distinct but related and commonly comorbid, potentially due to a shared genetic etiology. The purpose of the present study was to examine the relation between SCT and daytime sleepiness in samples of young adults with and without ADHD. Given that symptoms of anxiety and depression are frequentiy comorbid with ADHD and have been shown to relate to both sleep (Accardo, Marcus, Leonard, Shults, & Elia, 2012) and SCT (Becker & Langberg, 2013; Becker, Luebbe, et al., 2014; Gamer, Marceaux, Mmg, Patterson, & Hodgens, 2010), the relation between SCT and daytime sleepiness was examined above and beyond symptoms of ADHD, anxiety, and depression. In Study I, using a large general college student sample, confirmatory factor analysis (CFA) was used to determine whether SCT and daytime sleepiness best fit together as a single factor or as distinct constmcts. Further, the relation between SCT and sleep above and beyond ADHD, anxiety, and depression was examined. Study 2 consisted of a comprehensively diagnosed sample of college students with ADHD. In addition to examining the relation between SCT and sleep in this clinical sample, grouping analyses were conducted to explore how comorbid SCT and daytime sleepiness impact functioning. We hypothesized that SCT and daytime sleepiness would be distinct but that significant overlap would exist, especially between the sluggish/sleepy aspects of SCT and daytime sleepiness. We also predicted that SCT would be associated with daytime sleepiness above and beyond symptoms of ADHD, anxiety, and depression and that college students with comorbid SCT and daytime sleepiness would exhibit significantly more impairment than students with ADHD alone.
Study 1 Method Participants. Participants were 768 undergraduate students ages 17-34 (M = 18.76, SD = 1.15) enrolled in an introductory psychology course at a public midwestem university. Approximately two thirds of the participants were female (68%, n = 521). The majority of participants were Caucasian (91%); the remaining participants were African American (3%), Hispanic (2%), Asian/ Asian American (2%), or other/multiracial (2%). Most participants (n = 500) were in their first year of college. Procedure. This study was approved by the university Institutional Review Board (IRB). After providing informed consent, participants completed the study measures as part of a larger survey. Participants received course credit for their participation. Measures. SCT and ADHD symptoms. SCT and ADHD symptoms were assessed using the BAARS-IV (Barkley, 2011a). The BAARS-IV is a self-report measure that includes the 18 DSM-IV symptoms of ADHD (APA, 2000) and nine symptoms of SCT (e.g., easily confused; slow moving). Each item was rated using a 4-point scale (1 = never or rarely, 4 = very ofien). The four-factor stmcture of
588
LANGBERG, BECKER, DVORSKY, AND LUEBBE
this measure has been established in a nationally representative sample of adults, and the scales demonstrate satisfactory internal consistency and test-retest reliability over a 2- to 3-week period (Barkley, 2011a). Internal consistencies in the present study were SCT a = .86, ADHD Inattention a = .85, ADHD Hyperactivity a = .67, and ADHD Impulsivity a = .79. Anxious and depressive symptoms. Internalizing symptoms were measured using the Depression and Anxiety subscales of the Depression Artxiety Stress Scales-21 (DASS-21; Antony, Bieling, Cox, Enns, & Swinson, 1998). In reference to the past week, participants rated each item using a 4-point scale (1 = did not apply to me at all, 4 = applied to me very much or most of the time). The DASS-21 demonstrates acceptable internal consistency and concurrent validity with other measures of anxiety and depression (Antony et al, 1998) and has been used in studies examining ADHD symptoms in college students (e.g., Gudjonsson, Sigurdsson, Eyjolfsdottir, Smari, & Young, 2009). In the present study, internal consistencies for the Anxiety and Depression subscales were .78 and .87, respectively. Daytime sleepiness. The Epworth Sleepiness Scale (ESS; Johns, 1991) was used to measure participants' daytime sleepiness. Participants rated the likelihood of dozing or falling asleep across eight different situations commonly encountered in daily life (e.g., sitting and reading, watching TV, as a passenger in a car for an hour without a break). Each item is rated on a 4-point scale (0 = no chance of dozing, 1 = slight chance of dozing, 2 = moderate chance of dozing, 3 = high chance of dozing), and a mean score of the items was calculated as an overall index of daytime sleepiness. The ESS is a commonly used measure of daytime sleepiness (Moul, Hall, Pilkonis, & Buysee, 2004) and demonstrates adequate internal consistency and test-retest reliability in adults with and without sleep disorders (Johns, 1992). The ESS has been previously used with college students (e.g., Lund, Reider, Whiting, & Prichard, 2010). In the present study, a = .71. Data analyses. First, the latent structure of SCT and daytime sleepiness was examined in a CFA model using Mplus Version 5.1 (Muthén & Muthén, 1998-2007). The initial two-factor CFA model included the SCT and daytime sleepiness latent constructs predicting their respective indicators and included correlations among SCT and daytime sleepiness. Theory and modification indices were then used to prune the model for optimal flt. Multiple indices were used to test overall model fit, with the following indicating acceptable fit: comparative fit index (CFI) > .90, Tucker-Lewis Index (TLI) > .90, and root-mean-square error of approximation (RMSEA) < .08 (Kline, 1998). We then examined in a subsequent model whether SCT was distinct from daytime sleepiness. A chi-square difference test was used to determine whether the hypothesized two-factor solution or an alternative one-factor model was optimal, where a significant increase in chi-square in the constrained one-factor model would indicate that the two-factor model was superior. Next, hierarchical regression analyses were conducted in order to examine whether SCT symptoms predicted daytime sleepiness above and beyond ADHD and internalizing symptoms. Specifically, ADHD and internalizing symptoms were entered on Step 1 in predicting daytime sleepiness, followed by SCT symptoms on Step 2.
Results CFA. In the initial CFA measurement model, the nine SCT items and eight daytime sleepiness items were each predicted by their respective latent constructs, and the two latent constructs were allowed to correlate. The initial model flt was poor, X^(118) = 1134.43, p < .001; CFI = 0.76; TLI = 0.72; RMSEA = 0.101 (90% CI [0.100, 0.112]). Modification indices were used to prune the model, with correlated residuals considered when the items were conceptually very similar (e.g., SCT item "lethargic, more tired than others" with SCT item "underactive or have less energy than others"; sleepiness item "sitting and reading" with sleepiness item "watching TV"), Eight within-factor correlated residuals were added to the measurement model. This measurement model demonstrated acceptable fit, x^(llO) = 380.85, p < .001; CFI = 0.94; TLI = 0.92; RMSEA = 0.057 (90% CI [0.050, 0.063]). Factor loadings and the correlation between the latent SCT and daytime sleepiness factors of this optimal model are displayed in Figure 1. Next, we examined whether a one-factor model constraining SCT and daytime sleepiness to be equal fit better than the twofactor model. The one-factor model had mediocre model fit, X^(lll) = 690.65,p < .001; CFI = 0.86; TLI = 0.83; RMSEA = 0.082 (90% CI [0.077, 0.088]) and was significantly inferior to the two-factor model that did not constrain SCT and daytime sleepiness to be equal, Ax^(l) = 309.91, p < .001. These findings
60
Figure 1. Two-factor model of sluggish cognitive tempo (SCT) and daytime sleepiness in college students (Study 1). Item numbers correspond to the Barkley Adult ADHD Rating Scale-IV (Barkley, 2011a) for SCT items and the Epworth Sleepiness Scale (Johns, 1991) for daytime sleepiness items. All pathways shown are standardized estimates and significant (p < .05).
SCT AND SLEEP suggest that SCT and daytime sleepiness, although correlated, are distinct constructs in the present sample. Correlation analyses. Variable means, standard deviations, and intercorrelations are displayed in Table 1. All of the psychopathology dimensions (i.e., SCT, ADHD, anxiety, depression) were significantly correlated with each other (rs = .16-.67, ps < .001) and with daytime sleepiness (rs = .22-.40,ps < .001). SCT and ADHD inattention were the psychopathology dimensions most strongly associated with daytime sleepiness (rs = .40 and .36, respectively). SCT in relation to daytime sleepiness. Hierarchical regression analyses predicting daytime sleepiness were conducted, with ADHD and internalizing dimensions entered on Step 1 followed by SCT symptoms on Step 2. Neither age nor sex was correlated with daytime sleepiness, and so these demographic variables were not included in the regression model. All variance infiation factor (VIF) values were below 2.1 (values > 10 are typically considered problematic), and all tolerance values were above .45 (values < .10 are typically considered problematic; Cohen et al., 2003), indicating that our regression model did not suffer from problems with multicollinearity. As shown in Table 2, ADHD inattentive and anxious symptoms significantly predicted daytime sleepiness in Step I. When SCT was added to the model at Step 2, the predictive value of ADHD inattention and anxiety was reduced (though both remained significant), and SCT was the strongest predictor of daytime sleepiness.
589
sleepiness measure was only collected for 58 of the participants because of a delay in obtaining IRB approval for the Pédiatrie Daytime Sleepiness Scale (PDSS) measure (described below). In comparing the demographic characteristics of those participants for whom daytime sleepiness data were collected (n = 58) with those without daytime sleepiness data (n = 10), no differences were found for age, gender, ethnicity, year in school, parent education level, family income, and ADHD medication status (ps > .05). Similarly, no differences were found for ADHD subtype, symptoms of ADHD, SCT, anxiety, depression, gradepoint average (GPA), or functional impairment ratings (ps > .05). Participants ranged in age from 17 to 30 years (M = 19.90, SD = 2.75), and slightly over half (n = 32) were male. Forty-two participants (72%) self-identified as Caucasian; the remaining participants were either African American (n = 5), Hispanic (n = 6), or multiracial (n = 5). Approximately half the participants (n = 27) were in their first year of college, with remaining participants in their second (n = 13), third (n = 10), or fourth (n = 8) year. On the basis of procedures described below, 30 participants were diagnosed with DSM-IV ADHD-I, and 28 participants were diagnosed with ADHD-C. Thirty-five participants were taking medication for ADHD. Procedure. The study was approved by the university IRB. The inclusionary criteria were (a) attendance at the university where the research was being conducted; (b) consent for research staff to contact participants' parent/guardian for a diagnostic interview; (c) meeting full diagnostic criteria for ADHD-I or ADHD-C; and (d) not meeting criteria for a pervasive developmental disorder, bipolar disorder, or psychosis. Diagnosis was determined through administration of both Part I and Part II of the Conners' Adult ADHD Diagnostic Interview for the DSM-IV (CAADID; Epstein, Johnson, & Conners, 2000; Epstein & Kollins, 2006) separately to both the student and his or her parent/guardian. The CAADID interview assesses both current and childhood symptoms and impairment as well as age of onset and pervasiveness of symptoms across time. Part I of the interview provides a detailed patient history designed to obtain information about past mental health diagnoses, medication usage, psychiatric comorbidity, and other potential risk factors. Part 11 is the ADHD portion of the interview. Strict diagnostic inclusion criteria were adhered to in this study because of questions/debates in tbe field regarding the validity of self-report in college students with ADHD and
Study 2 Using a large sample of college students, results from Study 1 demonstrate that SCT and daytime sleepiness are correlated but distinct factors. Further, SCT was found to be strongly associated with daytime sleepiness above and beyond internalizing and ADHD symptoms. The purpose of Study 2 was to test whether the pattern of results obtained in Study 1 would be replicated in a clinical sample of college students rigorously diagnosed with ADHD. In addition, analyses in Study 2 were focused on exploring how comorbid SCT and daytime sleepiness impact the functioning of college students with ADHD.
Method Participants. Participants were 68 undergraduate students enrolled in a public university in Virginia. However, the daytime Table 1 Study 1: Means, Standard Deviations, and Bivariate Correlations Variable 1. SCT 2. ADHD Inattention 3. ADHD Hyperactivity 4. ADHD Impulsivity 5. Anxiety 6. Depression 7. Daytime sleepiness M SD
1
1.91 0.55
2
3
4
5
6
7
.67'*' —
.44*** .50*** —
.33*** .37*" .48*** —
.43"* .34*** .39"* .27*"
.46*" .37*** .27*** .16*" .63***
.40*" .36*** .26*** .23*** .26*** .22***
1.74 0.49
1.73 0.55
1.60 0.60
Note. SCT = sluggish cognitive tempo; ADHD = attention-deficit/hyperactivity disorder. * " p < .001.
1.51 0.48
1.52 0.55
1.96 0.45
LANGBERG, BECKER, DVORSKY, AND LUEBBE
590
Table 2 Study I: Multiple Regression Models Predicting Daytime Sleepiness Erom ADHD, Internalizing, and SCT Symptoms Step 1: Model summary
Step 2: Model summary
SE Variable ADHD Inattention ADHD Hyperactivity ADHD Impulsivity Anxiety Depression SCT
SE
Af(5, 759) = 28.55***.«' = .16 1.86 .34 .46 .94 .10 —
.30 .28 .23 .33 .29 —
.25 .05 .08 .13 .02 —
AF( 1,758) = 29.44**', fi^ = .19 6.25*** 1.22 1.96 2.82** .35 —
.89 .21 .39 .73 -.19 1.66
.23 .27 .23 .33 .28 .31
.12 .03 .07 .10 -.03 .26
2.59** .77 1.71 2.22' -.66 5.43***
Note. Dashes indicate that the variable was not entered in the Step 1 model. ADHD = attention-deficit/hyperactivity disorder; SCT = sluggish cognitive tempo. > < .05. * * p < . O I . ***p
© 2014 American Psychological Association 1040-3590/14/$12.00 DOI: 10.1037/a0036276
Are Sluggish Cognitive Tempo and Daytime Sleepiness Distinct Constructs? Joshua M. Langberg
Stephen P. Becker
Virginia Commonwealth University
Miami University
Melissa R. Dvorsky
Aaron M. Luebbe
Virginia Commonwealth University
Miami University
Sluggish cognitive tempo (SCT) and daytime sleepiness are both common in individuals with attentiondeficit/hyperactivity disorder (ADHD). There appears to be considerable overlap between the tired and lethargic aspects of SCT and behaviors frequently exhibited by individuals with daytime sleepiness. However, no studies have examined the degree to which these constructs overlap and whether or not they are empirically distinct. In Study 1, a confirmatory factor analysis with the SCT subscale of the Barkley Adult ADHD Rating Scale-IV (BAARS-IV) and the Epworth Sleepiness Scale (ESS) was conducted in a sample of 768 college students. Results demonstrated that SCT and daytime sleepiness exhibit considerable overlap but are empirically distinct. In Study 2, we examined the relation between SCT and daytime sleepiness and also the impact of comorbid SCT and sleepiness on the functioning of 58 college students rigorously diagnosed with ADHD. Regression analyses in both Study 1 and Study 2 showed that SCT predicts daytime sleepiness above and beyond symptoms of ADHD, anxiety, and depres.sion. The 2 constructs were significantly related (r = .51 ), with the highest correlations occurring between the SCT tired and lethargic items with daytime sleepiness. College students with ADHD + SCT and daytime sleepiness were significantly more impaired than college students diagnosed with ADHD without SCT or daytime sleepiness. Together, these results fill an important gap in the literature by confirming SCT to be overlapping but empirically distinct from daytime sleepiness and demonstrating that SCT and daytime sleepiness are associated with functioning in college students with ADHD. Keywords: ADHD, attention-deficit/hyperactivity disorder, sleep, sluggish cognitive tempo, college students
Lecendreux, 2009). Not sutprisingly given these problems, individuals with ADHD often report feeling excessively sleepy during the daytime hours (Córtese et al., 2009; Owens, 2009). Daytime sleepiness is characterized by a lack of energy, persistent tiredness, and feelings of drowsiness and sluggishness (Drake et al, 2003). Daytime sleepiness is associated with increased risk for academic and behavioral impairment in samples of youth with (Langberg, Dvorsky, Marshall, & Evans, 2013) and without (Astill, Van der Heijden, Van IJzendoorn, & Van Someren, 2012; Beebe, 2011; Fallone, Acebo, Seifer, & Carskadon, 2005) ADHD. In a largely separate literature, approximately 30%-50% of youth with ADHD have been shown to exhibit slow processing, sluggishness, apathy, drowsiness, and inconsistent alertness, behaviors that have been labeled sluggish cognitive tempo (SCT). Although some SCT items emerged in the literature in the 1960s and 1970s (see Becker, Marshall, & McBumett, 2014, for a historical overview of the SCT construct), it was in the 1980s that Lahey and colleagues (e.g., Lahey et al., 1988; Neeper & Lahey, 1986) first demonstrated SCT symptoms to be distinct from the inattentive and hyperactive-impulsive symptoms characteristic of ADHD. As research examining the SCT construct began to advance, it was initially posited that a significant minority of children with ADHD, Predominantly Inattentive Type (ADHD-I) were best characterized by the presence of SCT symptoms and relative
Children and adolescents with attention-deficit/hyperactivity disorder (ADHD) frequently experience sleep disturbances, with prevalence rates for sleep problems in samples of youth with ADHD between 25% and 50% (Sung, Hiscock, Sciberras, & Efron, 2008; Yoon, Jain, & Shapiro, 2012). The relation between sleep and ADHD is not limited to childhood, with as many as 50% of adults with ADHD endorsing the presence of sleep disturbances such as insomnia (Voinescu, Szentagotai, & David, 2012). Individuals with ADHD experience a wide range of sleep disturbances, including higher rates of problems with sleep latency, bedtime resistance, night awakenings, and difficulties with morning awakenings in comparison to their peers (Córtese, Faraone, Konofal, &
This article was published Online First March 10, 2014. Joshua M. Langberg, Department of Psychology, Virginia Commonwealth University; Stephen P. Becker. Department of Psychology, Miami University; Melissa R. Dvorsky, Department of Psychology, Virginia Commonwealth University; Aaron M. Luebbe, Department of Psychology, Miami University. Correspondence concerning this article should he addressed to Joshua M. Langberg, Department of Psychology, Virginia Commonwealth University, 806 West Franklin Street, P.O. Box 842018, Richmond, VA 23284-2018. E-mail: [email protected] 586
SCT AND SLEEP absence of hyperactive/impulsive symptoms (Carlson & Mann, 2002; Lahey et al., 1988; McBumett, Pfiffner, & Frick, 2001; Milich, Balentine, & Lynam, 2001). Still, although SCT symptoms are more strongly related to ADHD symptoms of inattention than to symptoms of hyperactivity/impulsivity (Willcutt et al., 2012), a sizable number of individuals with ADHD Combined Type also display clinically significant levels of SCT symptoms (Barkley, 2012; Carlson & Mann, 2002; Willcutt et al., 2012). Similar to sleep disturbances, the presence of SCT is not unique to samples of children with ADHD, with approximately 50% of adults with elevated ADHD symptoms exhibiting clinically significant SCT (Barkley, 2012). Further, comorbid SCT symptoms are associated with increased impairment in social, academic, and executive functioning in youth with ADHD (Barkley, 2013; Becker & Langberg, 2013, 2014; Carison & Mann, 2002; Langberg, Becker, & Dvorsky, 2014; Marshall et al., 2014; McBumett et al., 2001), as well as among adults displaying elevated ADHD symptoms (Barkley, 2012). Multiple authors have noted the considerable similarities and potential overlap associated with measures of SCT and measures of daytime sleepiness (Córtese et al., 2009; Fallone et al., 2005; Mayes et al., 2009; Willoughby, Angold, & Egger, 2008). However, only one study published to date has examined the relation between these two constmcts, and no research has evaluated whether or not these two constmcts are empirically distinct. Specifically, Voinescu and colleagues (2012) examined the prevalence of sleep disturbances in a sample of adults with symptoms of ADHD. In this sample of 550 Romanian adults, participants were classified into an ADHD-likely group, a non-ADHD group, and a chronic insomnia group based on self-report on the Barkley Adult ADHD Rating Scale (BAARS-IV; Barkley, 2011a), which includes an SCT subscale along with Diagnostic and Statistical Manual of Mental Disorders, fourth edition {DSM-IV; American Psychiatric Association [APA], 2000) symptoms of ADHD. Adults in the insomnia and ADHD-likely groups had significantly higher scores on the SCT subscale in comparison to the nonADHD group. Further, in the ADHD-likely group, the correlation between the insomnia and SCT measures was significant and large {r = .66). To our knowledge, this is the only study completed to date in which the association between SCT and sleep has been reported. SCT is a multifaceted constmct, and it is likely that certain aspects of SCT are more highly related to daytime sleepiness than others. Recent measure development studies suggest that SCT consists of slow, sleepy, and daydreaming factors (Jacobson et al., 2012; Penny et al, 2009). The slow aspect of SCT includes apathy, motivation, being slow or delayed in completing tasks, and lacking initiative. Sleepy SCT behaviors include drowsiness, appearing sluggish or tired, and being underactive; daydreaming SCT behaviors include getting lost in one's own thoughts and seeming to be in one's own world. Importantiy, each of these SCT factors relates differently to the ADHD symptom dimensions. The SCT behaviors related to being apathetic and lacking motivation/initiative largely load onto a single factor along with DSM-IV inattention items (Barkley, 2013; Lee, Bums, Snell, & McBumett, 2014), whereas the sleepy/daydreamy aspects of SCT appear to be empirically distinct from DSM-IV symptoms of inattention (Becker, Luebbe, Fite, Stoppelbein, & Greening, 2014; Jacobson et al., 2012; Penny et al., 2009; Willcutt et al., 2012). These findings have implica-
587
tions for studies examining the relation between sleep and SCT. Specifically, the SCT behaviors found to be most distinct from the ADHD inattention dimension are those that are most highly characteristic of individuals with sleep disturbances (i.e., drowsy and sluggish/tired). Accordingly, it is possible that SCT drowsy/tired behaviors are actually manifestations of daytime sleepiness. Alternatively, it is possible that SCT and daytime sleepiness are distinct but related and commonly comorbid, potentially due to a shared genetic etiology. The purpose of the present study was to examine the relation between SCT and daytime sleepiness in samples of young adults with and without ADHD. Given that symptoms of anxiety and depression are frequentiy comorbid with ADHD and have been shown to relate to both sleep (Accardo, Marcus, Leonard, Shults, & Elia, 2012) and SCT (Becker & Langberg, 2013; Becker, Luebbe, et al., 2014; Gamer, Marceaux, Mmg, Patterson, & Hodgens, 2010), the relation between SCT and daytime sleepiness was examined above and beyond symptoms of ADHD, anxiety, and depression. In Study I, using a large general college student sample, confirmatory factor analysis (CFA) was used to determine whether SCT and daytime sleepiness best fit together as a single factor or as distinct constmcts. Further, the relation between SCT and sleep above and beyond ADHD, anxiety, and depression was examined. Study 2 consisted of a comprehensively diagnosed sample of college students with ADHD. In addition to examining the relation between SCT and sleep in this clinical sample, grouping analyses were conducted to explore how comorbid SCT and daytime sleepiness impact functioning. We hypothesized that SCT and daytime sleepiness would be distinct but that significant overlap would exist, especially between the sluggish/sleepy aspects of SCT and daytime sleepiness. We also predicted that SCT would be associated with daytime sleepiness above and beyond symptoms of ADHD, anxiety, and depression and that college students with comorbid SCT and daytime sleepiness would exhibit significantly more impairment than students with ADHD alone.
Study 1 Method Participants. Participants were 768 undergraduate students ages 17-34 (M = 18.76, SD = 1.15) enrolled in an introductory psychology course at a public midwestem university. Approximately two thirds of the participants were female (68%, n = 521). The majority of participants were Caucasian (91%); the remaining participants were African American (3%), Hispanic (2%), Asian/ Asian American (2%), or other/multiracial (2%). Most participants (n = 500) were in their first year of college. Procedure. This study was approved by the university Institutional Review Board (IRB). After providing informed consent, participants completed the study measures as part of a larger survey. Participants received course credit for their participation. Measures. SCT and ADHD symptoms. SCT and ADHD symptoms were assessed using the BAARS-IV (Barkley, 2011a). The BAARS-IV is a self-report measure that includes the 18 DSM-IV symptoms of ADHD (APA, 2000) and nine symptoms of SCT (e.g., easily confused; slow moving). Each item was rated using a 4-point scale (1 = never or rarely, 4 = very ofien). The four-factor stmcture of
588
LANGBERG, BECKER, DVORSKY, AND LUEBBE
this measure has been established in a nationally representative sample of adults, and the scales demonstrate satisfactory internal consistency and test-retest reliability over a 2- to 3-week period (Barkley, 2011a). Internal consistencies in the present study were SCT a = .86, ADHD Inattention a = .85, ADHD Hyperactivity a = .67, and ADHD Impulsivity a = .79. Anxious and depressive symptoms. Internalizing symptoms were measured using the Depression and Anxiety subscales of the Depression Artxiety Stress Scales-21 (DASS-21; Antony, Bieling, Cox, Enns, & Swinson, 1998). In reference to the past week, participants rated each item using a 4-point scale (1 = did not apply to me at all, 4 = applied to me very much or most of the time). The DASS-21 demonstrates acceptable internal consistency and concurrent validity with other measures of anxiety and depression (Antony et al, 1998) and has been used in studies examining ADHD symptoms in college students (e.g., Gudjonsson, Sigurdsson, Eyjolfsdottir, Smari, & Young, 2009). In the present study, internal consistencies for the Anxiety and Depression subscales were .78 and .87, respectively. Daytime sleepiness. The Epworth Sleepiness Scale (ESS; Johns, 1991) was used to measure participants' daytime sleepiness. Participants rated the likelihood of dozing or falling asleep across eight different situations commonly encountered in daily life (e.g., sitting and reading, watching TV, as a passenger in a car for an hour without a break). Each item is rated on a 4-point scale (0 = no chance of dozing, 1 = slight chance of dozing, 2 = moderate chance of dozing, 3 = high chance of dozing), and a mean score of the items was calculated as an overall index of daytime sleepiness. The ESS is a commonly used measure of daytime sleepiness (Moul, Hall, Pilkonis, & Buysee, 2004) and demonstrates adequate internal consistency and test-retest reliability in adults with and without sleep disorders (Johns, 1992). The ESS has been previously used with college students (e.g., Lund, Reider, Whiting, & Prichard, 2010). In the present study, a = .71. Data analyses. First, the latent structure of SCT and daytime sleepiness was examined in a CFA model using Mplus Version 5.1 (Muthén & Muthén, 1998-2007). The initial two-factor CFA model included the SCT and daytime sleepiness latent constructs predicting their respective indicators and included correlations among SCT and daytime sleepiness. Theory and modification indices were then used to prune the model for optimal flt. Multiple indices were used to test overall model fit, with the following indicating acceptable fit: comparative fit index (CFI) > .90, Tucker-Lewis Index (TLI) > .90, and root-mean-square error of approximation (RMSEA) < .08 (Kline, 1998). We then examined in a subsequent model whether SCT was distinct from daytime sleepiness. A chi-square difference test was used to determine whether the hypothesized two-factor solution or an alternative one-factor model was optimal, where a significant increase in chi-square in the constrained one-factor model would indicate that the two-factor model was superior. Next, hierarchical regression analyses were conducted in order to examine whether SCT symptoms predicted daytime sleepiness above and beyond ADHD and internalizing symptoms. Specifically, ADHD and internalizing symptoms were entered on Step 1 in predicting daytime sleepiness, followed by SCT symptoms on Step 2.
Results CFA. In the initial CFA measurement model, the nine SCT items and eight daytime sleepiness items were each predicted by their respective latent constructs, and the two latent constructs were allowed to correlate. The initial model flt was poor, X^(118) = 1134.43, p < .001; CFI = 0.76; TLI = 0.72; RMSEA = 0.101 (90% CI [0.100, 0.112]). Modification indices were used to prune the model, with correlated residuals considered when the items were conceptually very similar (e.g., SCT item "lethargic, more tired than others" with SCT item "underactive or have less energy than others"; sleepiness item "sitting and reading" with sleepiness item "watching TV"), Eight within-factor correlated residuals were added to the measurement model. This measurement model demonstrated acceptable fit, x^(llO) = 380.85, p < .001; CFI = 0.94; TLI = 0.92; RMSEA = 0.057 (90% CI [0.050, 0.063]). Factor loadings and the correlation between the latent SCT and daytime sleepiness factors of this optimal model are displayed in Figure 1. Next, we examined whether a one-factor model constraining SCT and daytime sleepiness to be equal fit better than the twofactor model. The one-factor model had mediocre model fit, X^(lll) = 690.65,p < .001; CFI = 0.86; TLI = 0.83; RMSEA = 0.082 (90% CI [0.077, 0.088]) and was significantly inferior to the two-factor model that did not constrain SCT and daytime sleepiness to be equal, Ax^(l) = 309.91, p < .001. These findings
60
Figure 1. Two-factor model of sluggish cognitive tempo (SCT) and daytime sleepiness in college students (Study 1). Item numbers correspond to the Barkley Adult ADHD Rating Scale-IV (Barkley, 2011a) for SCT items and the Epworth Sleepiness Scale (Johns, 1991) for daytime sleepiness items. All pathways shown are standardized estimates and significant (p < .05).
SCT AND SLEEP suggest that SCT and daytime sleepiness, although correlated, are distinct constructs in the present sample. Correlation analyses. Variable means, standard deviations, and intercorrelations are displayed in Table 1. All of the psychopathology dimensions (i.e., SCT, ADHD, anxiety, depression) were significantly correlated with each other (rs = .16-.67, ps < .001) and with daytime sleepiness (rs = .22-.40,ps < .001). SCT and ADHD inattention were the psychopathology dimensions most strongly associated with daytime sleepiness (rs = .40 and .36, respectively). SCT in relation to daytime sleepiness. Hierarchical regression analyses predicting daytime sleepiness were conducted, with ADHD and internalizing dimensions entered on Step 1 followed by SCT symptoms on Step 2. Neither age nor sex was correlated with daytime sleepiness, and so these demographic variables were not included in the regression model. All variance infiation factor (VIF) values were below 2.1 (values > 10 are typically considered problematic), and all tolerance values were above .45 (values < .10 are typically considered problematic; Cohen et al., 2003), indicating that our regression model did not suffer from problems with multicollinearity. As shown in Table 2, ADHD inattentive and anxious symptoms significantly predicted daytime sleepiness in Step I. When SCT was added to the model at Step 2, the predictive value of ADHD inattention and anxiety was reduced (though both remained significant), and SCT was the strongest predictor of daytime sleepiness.
589
sleepiness measure was only collected for 58 of the participants because of a delay in obtaining IRB approval for the Pédiatrie Daytime Sleepiness Scale (PDSS) measure (described below). In comparing the demographic characteristics of those participants for whom daytime sleepiness data were collected (n = 58) with those without daytime sleepiness data (n = 10), no differences were found for age, gender, ethnicity, year in school, parent education level, family income, and ADHD medication status (ps > .05). Similarly, no differences were found for ADHD subtype, symptoms of ADHD, SCT, anxiety, depression, gradepoint average (GPA), or functional impairment ratings (ps > .05). Participants ranged in age from 17 to 30 years (M = 19.90, SD = 2.75), and slightly over half (n = 32) were male. Forty-two participants (72%) self-identified as Caucasian; the remaining participants were either African American (n = 5), Hispanic (n = 6), or multiracial (n = 5). Approximately half the participants (n = 27) were in their first year of college, with remaining participants in their second (n = 13), third (n = 10), or fourth (n = 8) year. On the basis of procedures described below, 30 participants were diagnosed with DSM-IV ADHD-I, and 28 participants were diagnosed with ADHD-C. Thirty-five participants were taking medication for ADHD. Procedure. The study was approved by the university IRB. The inclusionary criteria were (a) attendance at the university where the research was being conducted; (b) consent for research staff to contact participants' parent/guardian for a diagnostic interview; (c) meeting full diagnostic criteria for ADHD-I or ADHD-C; and (d) not meeting criteria for a pervasive developmental disorder, bipolar disorder, or psychosis. Diagnosis was determined through administration of both Part I and Part II of the Conners' Adult ADHD Diagnostic Interview for the DSM-IV (CAADID; Epstein, Johnson, & Conners, 2000; Epstein & Kollins, 2006) separately to both the student and his or her parent/guardian. The CAADID interview assesses both current and childhood symptoms and impairment as well as age of onset and pervasiveness of symptoms across time. Part I of the interview provides a detailed patient history designed to obtain information about past mental health diagnoses, medication usage, psychiatric comorbidity, and other potential risk factors. Part 11 is the ADHD portion of the interview. Strict diagnostic inclusion criteria were adhered to in this study because of questions/debates in tbe field regarding the validity of self-report in college students with ADHD and
Study 2 Using a large sample of college students, results from Study 1 demonstrate that SCT and daytime sleepiness are correlated but distinct factors. Further, SCT was found to be strongly associated with daytime sleepiness above and beyond internalizing and ADHD symptoms. The purpose of Study 2 was to test whether the pattern of results obtained in Study 1 would be replicated in a clinical sample of college students rigorously diagnosed with ADHD. In addition, analyses in Study 2 were focused on exploring how comorbid SCT and daytime sleepiness impact the functioning of college students with ADHD.
Method Participants. Participants were 68 undergraduate students enrolled in a public university in Virginia. However, the daytime Table 1 Study 1: Means, Standard Deviations, and Bivariate Correlations Variable 1. SCT 2. ADHD Inattention 3. ADHD Hyperactivity 4. ADHD Impulsivity 5. Anxiety 6. Depression 7. Daytime sleepiness M SD
1
1.91 0.55
2
3
4
5
6
7
.67'*' —
.44*** .50*** —
.33*** .37*" .48*** —
.43"* .34*** .39"* .27*"
.46*" .37*** .27*** .16*" .63***
.40*" .36*** .26*** .23*** .26*** .22***
1.74 0.49
1.73 0.55
1.60 0.60
Note. SCT = sluggish cognitive tempo; ADHD = attention-deficit/hyperactivity disorder. * " p < .001.
1.51 0.48
1.52 0.55
1.96 0.45
LANGBERG, BECKER, DVORSKY, AND LUEBBE
590
Table 2 Study I: Multiple Regression Models Predicting Daytime Sleepiness Erom ADHD, Internalizing, and SCT Symptoms Step 1: Model summary
Step 2: Model summary
SE Variable ADHD Inattention ADHD Hyperactivity ADHD Impulsivity Anxiety Depression SCT
SE
Af(5, 759) = 28.55***.«' = .16 1.86 .34 .46 .94 .10 —
.30 .28 .23 .33 .29 —
.25 .05 .08 .13 .02 —
AF( 1,758) = 29.44**', fi^ = .19 6.25*** 1.22 1.96 2.82** .35 —
.89 .21 .39 .73 -.19 1.66
.23 .27 .23 .33 .28 .31
.12 .03 .07 .10 -.03 .26
2.59** .77 1.71 2.22' -.66 5.43***
Note. Dashes indicate that the variable was not entered in the Step 1 model. ADHD = attention-deficit/hyperactivity disorder; SCT = sluggish cognitive tempo. > < .05. * * p < . O I . ***p
