This article was downloaded by: [New York University] On: 03 May 2015, At: 08:41 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Clinical Child & Adolescent Psychology Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/hcap20
Longitudinal Correlates of Sluggish Cognitive Tempo and ADHD-Inattention Symptom Dimensions with Spanish Children a
a
b
c
Mateu Servera , Maria del Mar Bernad , Jesus M. Carrillo , Susana Collado & G. Leonard d
Burns a
Department of Psychology, University of the Balearic Islands & Research Institute on Health Sciences (IUNICS), , b
Department of Personality, Evaluation and Clinical Psychology, Complutense University of Madrid, , c
Click for updates
Department of Physiotherapy, Occupational Therapy, Physical and Rehabilitation Medicine, Rey Juan Carlos University, , d
Department of Psychology, Washington State University, , Published online: 09 Mar 2015.
To cite this article: Mateu Servera, Maria del Mar Bernad, Jesus M. Carrillo, Susana Collado & G. Leonard Burns (2015): Longitudinal Correlates of Sluggish Cognitive Tempo and ADHD-Inattention Symptom Dimensions with Spanish Children, Journal of Clinical Child & Adolescent Psychology, DOI: 10.1080/15374416.2015.1004680 To link to this article: http://dx.doi.org/10.1080/15374416.2015.1004680
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Clinical Child & Adolescent Psychology, 0(0), 1–10, 2015 Copyright # Taylor & Francis Group, LLC ISSN: 1537-4416 print=1537-4424 online DOI: 10.1080/15374416.2015.1004680
Longitudinal Correlates of Sluggish Cognitive Tempo and ADHD-Inattention Symptom Dimensions with Spanish Children Mateu Servera and Maria del Mar Bernad
Downloaded by [New York University] at 08:41 03 May 2015
Department of Psychology, University of the Balearic Islands & Research Institute on Health Sciences (IUNICS)
Jesus M. Carrillo Department of Personality, Evaluation and Clinical Psychology, Complutense University of Madrid
Susana Collado Department of Physiotherapy, Occupational Therapy, Physical and Rehabilitation Medicine, Rey Juan Carlos University
G. Leonard Burns Department of Psychology, Washington State University
The objective was to examine the longitudinal correlates of sluggish cognitive tempo (SCT) and attention deficit=hyperactivity disorder (ADHD)-Inattention (IN) dimensions with mothers’ and fathers’ ratings of Spanish children. Mothers and fathers rated SCT, ADHD-IN, ADHD-hyperactivity=impulsivity (HI), oppositional defiant disorder (ODD), depression, academic impairment, and social impairment on 3 occasions (twice in first-grade year [6-week separation] and once in the second-grade year [12 months after the first assessment]) in Spanish children (758, 746, and 718 children at the 3 time-points with approximately 55% boys). The results showed that (a) higher levels of SCT from earlier assessments predicted higher levels of depression, academic impairment, and social impairment at Assessment 3 after controlling for ADHD-IN at earlier assessments; (b) higher levels of ADHD-IN from earlier assessments predicted higher levels of depression, academic impairment, and social impairment at Assessment 3 after controlling for SCT at earlier assessments; (c) higher levels of ADHD-IN from earlier assessments predicted higher levels of ADHD-HI and ODD at Assessment 3 after controlling for SCT from earlier assessments; and (d) higher levels of SCT from earlier assessments either showed no unique relationship with ADHD-HI and ODD or predicted lower levels of ADHD-HI and ODD at Assessment 3 after controlling for ADHD-IN from earlier assessments. Initial evidence is provided of SCT’s unique longitudinal relationships with depression and academic=social impairment and different longitudinal relationships with ADHD-HI and ODD relative to ADHD-IN, thus adding to a growing body of research underscoring the importance of SCT as distinct from ADHD-IN.
Correspondence should be addressed to G. Leonard Burns, Department of Psychology, Washington State University, P.O. Box 644820, Pullman, WA 99164-4820. E-mail: [email protected]
Sluggish cognitive tempo (SCT) is characterized by inconsistent alertness (e.g., loses train of thought, daydreams) and slow thinking=slow behavior (Lee,
Downloaded by [New York University] at 08:41 03 May 2015
2
SERVERA ET AL.
Burns, Snell, & McBurnett, 2014, Table 2). Research indicates that SCT and attention deficit=hyperactivity disorder (ADHD) inattention (IN) symptoms represent different dimensions with unique cross-sectional correlates (Barkley, 2013; Becker, Luebbe, Fite, Stoppelbein, & Greening, 2014; Bernad, Servera, Grases, Collado, & Burns, 2014; Burns, Servera, Bernad, Carrillo, & Cardo, 2013; Cortes, Servera, Becker, & Burns, 2014; Lee et al., 2014; Penny, Waschbusch, Klein, Corkum, & Eskes, 2009). For example, higher levels of SCT are associated with lower levels of externalizing problems after controlling for ADHD-IN, whereas higher levels of ADHD-IN are associated with higher levels of externalizing problems after controlling for SCT. In addition, higher levels of SCT and ADHD-IN uniquely predict higher levels of anxiety, depression, academic impairment, and social impairment. Although these cross-sectional results are promising, the measurement of predictors and outcomes at the same time is a weakness (i.e., Does SCT influence the outcomes, or do the outcomes influence SCT?). Longitudinal research where the measurement of SCT and ADHD-IN occurs prior to the outcomes would provide a stronger evaluation of SCT’s and ADHD-IN’s unique roles in children’s adjustment. At this time, only two studies have examined the correlates of SCT using a longitudinal design. With teacher ratings and a 6-month interval, Becker (2014) found that SCT uniquely predicted subsequent peer adjustment after controlling for prior adjustment, ADHD-IN and other symptom dimensions. With teachers and aides ratings and a 12-month interval, Bernad et al. (2014) found that higher levels of SCT predicted lower levels of ADHD-hyperactivity=impulsivity (HI) and oppositional defiant disorder (ODD) after controlling for ADHD-IN, whereas higher levels of ADHD-IN predicted higher levels of ADHD-HI and ODD after controlling for SCT. In addition, higher scores on SCT and ADHD-IN both uniquely predicted higher levels of academic impairment. Despite these promising findings, there is a need for additional longitudinal research. This study is the first longitudinal study to examine parent ratings of SCT and ADHD-IN in relation to children’s functioning and, further, uses both mother and father ratings. If ratings by mothers and fathers yield the same results as the two earlier longitudinal studies in school settings, then such would show for the first time that SCT has unique longitudinal effects relative to ADHD-IN on children’s adjustment in home settings with these effects being consistent across ratings by mothers and fathers.
OBJECTIVES The current study used ratings by mothers and fathers of Spanish children across three occasions—twice at
the end of the first-grade year (6-week separation) and once at the end of the second-grade year (12 months after the first assessment)—to investigate the longitudinal correlates of SCT and ADHD-IN. The first three objectives establish the construct validity of SCT at the third assessment in order to determine the unique longitudinal correlates of SCT and ADHD-IN from the first two assessments to the third assessment.
Objective 1: Convergent and Discriminant Validity of SCT Symptoms at Time 3 (T3) The first objective was to determine if the same five of eight SCT symptoms with convergent validity (substantial loadings on the SCT factor) and discriminant validity (low loadings on the ADHD-IN factor) at the two assessments in the first grade also showed convergent and discriminant validity in the second grade. These five SCT symptoms were loses train of thought, easily confused, thinking is slow, slow moving, and drowsy. The other three SCT symptoms—daydreams, alertness fluctuates, and absentminded—failed to show discriminant validity with the ADHD-IN factor in the first grade (Burns et al., 2013). However, given all eight SCT symptoms showed discriminant validity with the ADHD-IN factor with parent ratings of kindergarten through sixth-grade American children (Lee et al., 2014), the other three SCT symptoms may show improved validity with the Spanish children being 12 months older.
Objective 2: Measurement Model at T3 and across Time (T1 to T3, T2 to T3) The second objective was to evaluate the fit of an SCT, ADHD-IN, ADHD-HI, ODD toward adults [ODD-A], ODD toward peers [ODD-P], depression, and academic and social impairment eight-factor model at T3 along with the reliability of the factors (i.e., true score variance, mother–father agreement, stability of like factor correlations for the 10.5- and 12-month intervals). We also predicted that the correlation between the ADHD-IN factor and the ADHD-HI, ODD-A, and ODD-P factors would be stronger than the correlation between the SCT factor and the ADHD-HI, ODD-A, and ODD-P factors. In addition, we predicted the SCT and ADHD-IN factors would show similar correlations with depression, academic, and social impairment factors. Finally, we predicted that SCT and ADHD-IN factors from T1 and T2 would show this same pattern of correlations with ADHD-HI, ODD-A, ODD-P, depression, and academic and social impairment factors at T3.
Downloaded by [New York University] at 08:41 03 May 2015
LONGITUDINAL CORRELATES OF SLUGGISH COGNITIVE TEMPO
3
FIGURE 1 Cross-sectional and longitudinal regression models. Note: Each latent variable was defined by manifest variables (not show in path diagrams) with 12 months separating Assessments 1 and 3 and 10.5 months separating Assessments 2 and 3. Each model allowed correlations among all disturbances (not shown). SCT ¼ sluggish cognitive tempo; ADHD ¼ attention-deficit=hyperactivity disorder; IN ¼ inattention; HI ¼ hyperactivity= impulsivity; ODD ¼ oppositional defiant disorder.
Objective 3: Unique Cross-Sectional Correlates of SCT and ADHD-IN at T3 The third objective was to determine the unique cross-sectional correlates of SCT and ADHD-IN at T3. Figure 1a shows this cross-sectional structural regression model. It was predicted that higher levels of ADHD-IN would predict higher levels of ADHD-HI and ODD after controlling for SCT, whereas SCT’s relationship with ADHD-HI and ODD would be nonsignificant (or significantly negative) after controlling for ADHD-IN. It was also predicted that SCT and ADHD-IN would both uniquely predict higher levels of depression and academic and social impairment. To ensure that SCT is different from depression, we also repeated these analyses controlling for depression as well as ADHD-IN. Objective 4: Unique Longitudinal Correlates of SCT and ADHD-IN The fourth objective was to determine if SCT and ADHD-IN have unique longitudinal correlates. Figure 1b shows the two structural regression analyses (i.e., T3 regressed on T1; T3 regressed on T2). It was predicted that higher levels of ADHD-IN from the first-grade occasions (T1 and T2) would predict higher levels of ADHD-HI and ODD in the second grade after controlling for SCT, whereas SCT in the first grade would have a nonsignificant (or significantly negative) relationship with ADHD-HI and ODD in the second grade after controlling for ADHD-IN. It was also predicted that higher levels of SCT and ADHD-IN in the first grade would both uniquely predict higher levels of depression and academic and social impairment in the second grade. To again ensure the findings for SCT were independent of depression, we repeated the analyses controlling for
depression and ADHD-IN. Support for these four objectives, especially Objective 4, would uniquely add to the research that suggests SCT is distinct from ADHD-IN.
METHOD Participants and Procedures The 46 elementary schools on the island of Majorca (Spain) were invited to participate, with 43 indicating an interest. Twenty-two of 43 schools were randomly selected along with eight from Madrid (eight were asked and eight agreed). There were 1,045 first-grade children in these 30 schools at T1. A cover letter and measures were sent to mothers and fathers at T1, T2 (6 weeks after T1), and T3 (12 months after T1). At T1, 723 mothers and 603 fathers participated, with 667 mothers and 584 fathers at T2 and 604 mothers and 540 fathers at T3. There were 758 unique children (55% boys) at T1, 746 (54% boys) at T2, and 718 (54% boys) at T3. Although ethnicity was not obtained for individual children, at the school level, approximately 90% were Caucasian with 10% North African. The Institutional Review Board of the University of the Balearic Islands approved the research. Measures Child and adolescent disruptive behavior inventory. The Child and Adolescent Disruptive Behavior Inventory measures SCT (eight symptoms, Burns et al., 2013, Table 1), ADHD-IN (nine symptoms), ADHD-HI (nine symptoms), ODD-A (e.g., argues with adults, eight symptoms), ODD-P toward siblings=peers (e.g., argues with siblings=peers, eight symptoms), depression (five symptoms: seems sad, unhappy, or depressed; seems to feel
4
SERVERA ET AL. TABLE 1 Reliability, Interrater, and Stability Coefficients (Standard Errors) for Measures Stability Coefficients Reliability Coefficients
Downloaded by [New York University] at 08:41 03 May 2015
SCT ADHD-IN ADHD-HI ODD-A ODD-P Depression Academic Impairment Social Impairment
Mothersa
Fathersb
.81 .94 .93 .93 .92 .75 .92 .90
.82 .94 .93 .93 .91 .76 .92 .91
(.01) (.01) (.01) (.01) (.01) (.02) (.01) (.02)
10.5 Months Interraterc
(.02) (.01) (.01) (.01) (.01) (.02) (.01) (.02)
.77 .80 .80 .73 .67 .67 .86 .76
(.03) (.02) (.02) (.03) (.03) (.04) (.01) (.02)
12 Months
Mothersd
Fatherse
Motherf
Fathersg
.75 (.04) .80 (.02) .78 (.02) — — .72 (.04) .76 (.02) —
.76 (.03) .79 (.02) .74 (.02) — — .64 (.04) .77 (.02) —
.70 .74 .77 .68 .60 .72 .72 .52
.67 .70 .76 .62 .54 .64 .74 .55
(.04) (.02) (.02) (.03) (.03) (.04) (.02) (.03)
(.06) (.03) (.02) (.04) (.04) (.04) (.02) (.03)
Note: Confirmatory factor analytic procedures were used to calculate the reliability coefficients (i.e., the amount of true score variance in the measure), interrater factor correlations (i.e., the correlations between the same factors for mothers and fathers) and the stability coefficients (i.e., the correlations between the same factor within a source across time). SCT ¼ sluggish cognitive tempo; ADHD-IN ¼ attention-deficit=hyperactivity disorder-inattention; HI ¼ hyperactivity=impulsivity; ODD-A ¼ oppositional defiant disorder toward adults; ODD-P ¼ oppositional defiant disorder toward peers. A dash indicates the measure was not administered at the second assessment. a n ¼ 604. b n ¼ 540. c n ¼ 618. d n ¼ 733. e n ¼ 673. f n ¼ 790. g n ¼ 700.
worthless or expresses feelings of worthless; seems not to enjoy home activities anymore; seems lonely or expresses feelings of loneliness; and seems to feel hopeless about things or expresses feelings of hopelessness) and academic impairment (four items: completion of homework, reading skills, arithmetic skills, and writing skills). The ADHD and ODD symptoms were rated on a 6-point frequency of occurrence scale (i.e., occurs nearly none of the time [two or fewer times per month], seldom occurs [once per week], sometimes occurs [a few times per week], often occurs [once per day], very often occurs [several times per day], and occurs nearly all the time [many times per day]). The SCT and depression symptoms were rated on a 6-point duration of occurrence scale (nearly none of the time, seldom, sometimes, often, very often, and nearly all of the time). A 7-point scale was used for the four academic items (severe difficulty, moderate difficulty, slight difficulty, average performance for grade level, slightly above average, moderately above average, and excellent performance for grade level). The academic items were reversed keyed so higher scores represented academic impairment. Mothers and fathers rated the symptoms for home and community and were told not consider the children’s behavior at school. The instructions for the academic items were to rate the children’s current skills. Mothers and fathers were also asked to make their ratings independently. Earlier studies support the reliability and validity of the scores from the scale
(Burns et al., 2013; Lee et al., 2014). Table 1 shows the reliability values for this study. Functional impairment rating scale—children and adolescents (Barkley, 2012). The study used 7 of 15 items from Barkley’s Functional Impairment Rating Scale—Children and Adolescents to measure social impairment, that is, (a) interactions with their mother, (b) interactions with their father, (c) interactions with siblings, (d) interactions with other children, (e) interactions with other adults, (f) activities in the community, and (g) visiting others homes. These seven items were rated on a 10-point scale from 0 (not impaired) to 9 (severely impaired). (The eight at school items were not included.) Analytic Strategy All analyses treated the item ratings as ordered-categories and used the robust weighted least squares estimator (Version 7.1; Muthe´n & Muthe´n, 1998–2012). The robust weighted least squares estimator uses a pairwise deletion procedure for missing information. All the analyses took into account the children were nested within 30 schools (Type ¼ complex Mplus option.). The fit of the measurement model was evaluated with the comparative fit index (CFI; study criterion .95), Tucker–Lewis index (TLI; study criterion .95), and the root mean square error of approximation (RMSEA; study criterion .05). The Mplus model constraint procedure along with phantom
LONGITUDINAL CORRELATES OF SLUGGISH COGNITIVE TEMPO
constructs was used to determine if factor correlations differed significantly.
CFI ¼ .982, TLI ¼ .981, and RMSEA ¼ .015, 90% confidence interval [.012, .016].
RESULTS
Item-factor loadings. The five SCT symptoms showed strong loadings on the SCT factor for mothers and fathers (M ¼ .76, SD ¼ .07). Similar results occurred for the ADHD-IN (M ¼ .83, SD ¼ .08), ADHD-HI (M ¼ .84, SD ¼ .04), ODD-A (M ¼ .86, SD ¼ .03), ODD-P (M ¼ .84, SD ¼ .03), depression (M ¼ .73, SD ¼ .04), academic impairment (M ¼ .88, SD ¼ .03), and social impairment (M ¼ .84, SD ¼ .04) items.
Convergent and Discriminant Validity of SCT and ADHD-IN Symptoms
Downloaded by [New York University] at 08:41 03 May 2015
5
A two-factor model was applied to mothers’ and fathers’ ratings of the eight SCT and nine ADHD-IN symptoms to determine the convergent and discriminant validity of these symptoms at T3 (i.e., the analysis was restricted to two factors with cross-loadings allowed). For fathers’ ratings, all eight SCT symptoms showed substantial loading on the SCT factor (M ¼ .69, SD ¼ .10) in conjunction with low loadings on the ADHD-IN factor (M ¼ .09, SD ¼ .12). The ADHD-IN symptoms showed substantial loadings on the ADHD-IN factor (M ¼ .82, SD ¼ .06) and low loading on the SCT factor (M ¼ .02, SD ¼ .08). All SCT and ADHD-IN symptoms showed convergent and discriminant validity for fathers. For mothers’ ratings, the SCT symptoms daydreams, alertness fluctuates, and absentminded showed weak discriminant validity at T3. Their average loading on the SCT factor was .51 (SD ¼ .05) and their average loading on the ADHD-IN factor .37 (SD ¼ .02). These were the same three SCT symptoms that showed no discriminant validity with the ADHD-IN factor at T1 and T2 for mothers and fathers in the previous study (Burns et al., 2013). The other five SCT symptoms showed good convergent (M ¼ .66, SD ¼ .12) and discriminant validity (M ¼ .11, SD ¼ .13) for mothers at T3. All nine ADHD-IN symptoms showed good convergent (M ¼ .82, SD ¼ .09) and discriminant validity (M ¼ .01, SD ¼ .17) for mothers at T3. Operationalization of SCT Construct Although the three SCT symptoms with no discriminant validity with the ADHD-IN factor at T1 and T2 for mothers’ and fathers’ ratings showed an improvement in their discriminant validity at T3, especially for fathers, the SCT construct at T3 was defined by the same five SCT symptoms as at T1 and T2. These five SCT symptoms were loses train of thought, easily confused, drowsy, thinking is slow, and slow moving. This was necessary to meaningfully compare the cross-sectional results across assessments and the cross-sectional to the longitudinal results. Measurement Model at T3 Global fit. The fit of the multiple indicator (individual items) by multiple trait (SCT, ADHD-IN, ODD-A, ODD-P, depression, and academic and social impairment) by multiple source (mothers and fathers) measure model at T3 was excellent, v2(5081) ¼ 5,745, p < .001,
Reliability coefficients. The SCT scale scores contained 81% and 82% true score variance for mothers and fathers, respectively. The values for the other scales were higher than 90% with the exception of the depression scale with values of 75 and 76% for mothers and fathers, respectively (Table 1). Mother–father (interrater) factor correlations. The mother–father factor correlation for SCT was .77 with similar values for ADHD-IN (.80), ADHD-HI (.80), ODD-A (.73), ODD-P (.67), depression (.67), academic impairment (.86), and social impairment (.76). Stability of factor correlations. Table 1 shows the stability factor correlations for 10.5-month (T2–T3) and 12-month (T1–T3) intervals. The stability coefficients for SCT varied from .67 to .76. The values for the other factors were similar. Cross-sectional factor correlations at T3. Table 2 shows the correlations among the factors. ADHD-IN’s relationship with ADHD-HI, ODD-A, and ODD-P was significantly stronger than SCT’s relationship with these variables (ps < .002). In addition, SCT and ADHD-IN showed significant (ps < .05) relationships with depression, academic impairment, and social impairment with these relationships not being significantly different (ps > .05). Cross-Sectional Unique Effects of SCT and ADHD-IN on ADHD-HI, ODD, Depression, Academic Impairment, and Social Impairment at T3 Analytic strategy. A cross-sectional structural regression analysis was used to determine the unique effects of SCT and ADHD-IN on ADHD-HI, ODD (the two ODD scales were combined), depression, academic impairment, and social impairment at T3. Figure 1a shows this model with the partial standardized regression coefficients shown in Table 3.
6
SERVERA ET AL.
TABLE 2 Correlations Among Sluggish Cognitive Tempo, ADHD-IN, ADHD-HI, ODD-A, ODD-P, Depression, Academic Impairment, and Social Impairment Factors for Mothers (Fathers) at Time 3
SCT ADHD-IN ADHD-HI ODD-A ODD-P DEP AI SI
SCT
ADHD-IN
ADHD-HI
— .78 (.76) .48 (.49) .39 (.44) .40 (.49) .61 (.64) .54 (.58) .39 (.46)
— .68 (.66) .55 (.54) .57 (.56) .60 (.63) .48 (.57) .42 (.44)
— .61 (.60) .59 (.67) .44 (.46) .19 (.30) .40 (.36)
ODD-A
ODD-P
DEP
AI
SI
— (.69) (.48) (.25) (.51)
— .58 (.59) .20 (.26) .50 (.53)
— .34 (.42) .50 (.48)
— .12 (.21)
—
.69 .54 .18 .53
Downloaded by [New York University] at 08:41 03 May 2015
Note: All correlations were significant at p < .01. The range for the standard errors was from 0.02 to 0.05. SCT ¼ sluggish cognitive tempo; ADHD-IN ¼ attention-deficit=hyperactivity disorder-inattention; ADHD-HI ¼ hyperactivity=impulsivity; ODD-A ¼ oppositional defiant disorder toward adults; ODD-P ¼ oppositional defiant disorder toward peers; DEP ¼ depression; AI ¼ academic impairment; SI ¼ social impairment.
Unique effects of SCT and ADHD-IN on ADHD-HI and ODD. Although higher scores on ADHD-IN predicted significantly higher scores on ADHD-HI and ODD after controlling for SCT (ps < .001), SCT was no longer significantly related to ODD after controlling for ADHD-IN. Higher scores on SCT were associated with significantly lower scores on ADHD-HI after controlling ADHD-IN (p < .05) for mothers’ ratings (and unrelated for fathers’ ratings). Unique effects of SCT and ADHD-IN on depression. Across both mother and father ratings, higher scores on SCT predicted significantly greater depression after controlling for ADHD-IN with higher scores on ADHD-IN also predicting significantly greater depression after controlling for SCT (ps < .001). Unique effects of SCT and ADHD-IN on academic impairment. Across both mother and father ratings, higher scores on SCT predicted higher scores on
academic impairment after controlling for ADHD-IN (ps < .001), whereas higher scores on ADHD-IN also predicted higher scores on academic impairment after controlling for SCT (ps < .05). Unique effects of SCT and ADHD-IN on social impairment. Across both mother and father ratings, higher levels of SCT predicted significantly higher levels of social impairment after controlling for ADHD-IN with higher levels of ADHD-IN also predicting significantly higher levels of social impairment after controlling for SCT (ps < .05).
Role of Depression in SCT’s Unique Cross-Sectional Associations with ODD, ADHD-HI, Academics, and Social Impairment at T3 With depression added as a predictor along with ADHD-IN, higher scores on SCT now predicted significantly lower scores on ODD for mothers (b ¼ .26,
TABLE 3 Cross-Sectional Regression of ADHD-HI, ODD, Depression, Academic Impairment, and Social Impairment Factors on the Sluggish Cognitive Tempo and ADHD-IN Factors at Time 3 ADHD-HI Predictors SCT Mothers Fathers ADHD-IN Mothers Fathers
ODD
Outcomes DEP
AI
SI
b
SE
b
SE
b
SE
b
SE
b
SE
.15 .03ns
.06 .07
.12ns .10ns
.07 .07
.37 .38
.06 .07
.43 .36
.06 .06
.17 .29
.06 .07
.80 .68
.05 .06
.65 .47
.06 .07
.32 .34
.06 .05
.15 .30
.06 .06
.30 .22
.06 .07
Note: SCT ¼ sluggish cognitive tempo; ODD ¼ oppositional defiant disorder toward adults and peers; ADHD-IN ¼ attention-deficit=hyperactivity disorder-inattention; ADHD-HI ¼ attention-deficit hyperactivity disorder-hyperactivity=impulsivity; DEP ¼ depression; AI ¼ academic impairment; SI ¼ social impairment. p < .05. p < .001.
7
LONGITUDINAL CORRELATES OF SLUGGISH COGNITIVE TEMPO
SE ¼ 08, p < .01) with SCT’s unique relationship with maternal ratings of social impairment no longer significant (p > .05). All the other outcomes for the unique effects of SCT on ADHD-HI, ODD, academic impairment, and social impairment were unchanged when controlling for depression. Higher scores on depression uniquely predicted higher scores on ODD and social impairment for mothers and fathers (mothers: b ¼ .37, SE ¼ .04, b ¼ .39, SE ¼ .05, ps < .001; fathers: b ¼ .27, SE ¼ .08, b ¼ .28, SE ¼ .06, ps < .001).
Table 5 shows the partial standardized regression coefficients for the two intervals (i.e., T1–T3 and T2–T3). Unique longitudinal effects of SCT and ADHD-IN on ADHD-HI and ODD. Higher levels of ADHD-IN predicted significantly higher levels of ADHD-HI and ODD across the two intervals even after controlling for SCT (ps < .01). Higher levels of SCT significantly predicted lower scores on ADHD-HI across the 10.5-month interval after controlling for ADHD-IN (ps < .05). There was also tendency at least for mothers for higher SCT scores to predict lower ODD scores across the 12-month interval after controlling for ADHD-IN (p ¼ .07).
Downloaded by [New York University] at 08:41 03 May 2015
Longitudinal Factor Correlations Table 4 shows the correlations of SCT and ADHD-IN factors from T1 to T2 with ADHD-HI, ODD, depression, and academic and social impairment factors from T3. The longitudinal factor correlations showed the same pattern as the cross-sectional factor correlations. The ADHD-IN factor showed a stronger longitudinal relationship than SCT with the ADHD-HI and ODD factors (ps < .01), with the one exception being that the ADHD-IN and SCT correlations with ODD for the 12-month interval did not differ significantly (ps > .05). In addition, and consistent with the cross-sectional findings, the ADHD-IN and SCT factors showed similar longitudinal associations with depression, academic impairment, and social impairment (ps > .05).
Unique longitudinal effects of SCT and ADHD-IN on depression. Higher scores on SCT predicted marginally higher depression scores after controlling for ADHD-IN (ps < .07) with higher scores on ADHD-IN significantly predicting higher depression scores after controlling for SCT (ps < .05).
Longitudinal Structural Regression Analysis
Unique longitudinal effects of SCT and ADHD-IN on academic impairment. Higher scores on ADHD-IN predicted significantly higher levels academic impairment across the two intervals after controlling for SCT (ps < .05), whereas higher levels of SCT predicted significantly higher levels of academic impairment for three of the four coefficients across the two intervals after controlling for ADHD-IN (ps < .05).
Analytic strategy. ADHD-HI, ODD, depression, academic impairment, and social impairment factors from T3 were regressed on SCT and ADHD-IN factors from T1 to T2 to determine the unique effects of SCT and ADHD-IN across time. Figure 1b shows this model.
Unique longitudinal effects of SCT and ADHD-IN on social impairment. Higher scores on ADHD-IN predicted higher levels of social impairment after controlling for SCT across the two time intervals (ps < .05),
TABLE 4 Correlations (Standard Errors) of Sluggish Cognitive Tempo and ADHD-Inattention Factors from Times 1 and 2 with ADHD-HI, ODD, Depression, Academic Impairment, and Social Impairment Factors from Time 3 Time 1
Time 2
SCT
ADHD-IN
SCT
ADHD-IN
Time 3
Mothers
Fathers
Mothers
Fathers
Mothers
Fathers
Mothers
Fathers
ADHD-HI ODD Depression AI SI
.35 .35 .47 .45 .32
.37 .35 .47 .43 .33
.48 .40 .50 .45 .31
.51 .35 .50 .49 .31
.33 .30 .48 .43 .23
.37 .31 .52 .50 .32
.54 .42 .47 .43 .27
.56 .41 .54 .52 .35
(.04) (.04) (.05) (.04) (.03)
(.05) (.04) (.08) (.04) (.04)
(.04) (.03) (.04) (.04) (.02)
(.04) (.04) (.04) (.04) (.04)
(.04) (.03) (.04) (.05) (.03)
(.04) (.04) (.05) (.03) (.05)
(.03) (.04) (.04) (.05) (.03)
(.04) (.04) (.04) (.03) (.04)
Note: All correlations significant at p < .001. The interval from Time 1 to Time 3 was 12 months with the interval from Time 2 to Time 3 being 10.5 months. ODD ¼ oppositional defiant disorder toward adults and peers; ADHD-IN ¼ attention-deficit hyperactivity disorder–inattention; ADHD-HI ¼ attention-deficit hyperactivity disorder–hyperactivity=impulsivity; SCT ¼ sluggish cognitive tempo; AI ¼ academic impairment; SI ¼ social impairment.
8
SERVERA ET AL.
TABLE 5 Longitudinal Regression of ADHD-HI, ODD, Depression, Academic Impairment, and Social Impairment Factors on the Sluggish Cognitive Tempo and ADHD-IN Factors Outcomes ADHD-HI
Downloaded by [New York University] at 08:41 03 May 2015
b Time 1 Predictors SCT Mothers Fathers ADHD-IN Mothers Fathers Time 2 Predictors SCT Mothers Fathers ADHD-IN Mothers Fathers
ODD SE
b
DEP
AI
SI
SE
b
SE
b
SE
b
SE
.08 .08
.21 .22
.08 .13
.26 .13ns
.07 .09
.19 .22
.07 .06
.07 .08
.34 .33
.06 .10
.25 .40
.07 .09
.17 .15
.06 .06
.09 .12
.29 y .23
.09 .13
.23 .21
.10 .08
.03ns .10ns
.08 .12
.10 .11
.23 .35
.09 .08
.24 .35
.10 .08
.25 .27
.07 .11
Time 3 Outcomes (12-Month Interval) .04ns .02ns
.08 .09
.11ns .18
.07 .32 .51 .08 .22 .53 Time 3 Outcomes (10.5-Month Interval) .37 .26
.07 .11
.85 .77
.08 .11
y
.16 .09ns .55 .49
Note: SCT ¼ sluggish cognitive tempo; ODD ¼ oppositional defiant disorder toward adults and peers; ADHD-IN ¼ attention-deficit=hyperactivity disorder-inattention; ADHD-HI ¼ attention-deficit hyperactivity disorder—hyperactivity=impulsivity; DEP ¼ depression; AI ¼ academic impairment; SI ¼ social impairment. y p < .07. p < .05. p < .01. p < .001.
whereas higher levels of SCT predicted higher levels of social impairment after controlling for ADHD-IN only across the 12-month interval (ps < .05). Role of Depression in SCT’s Unique Longitudinal Associations with ODD, ADHD-HI, and Academic and Social Impairment The unique effect of SCT at T1 on ADHD-HI, ODD, and academic and social impairment at T3 was also examined after controlling for depression and ADHD-IN from T1. The unique effects of SCT remained the same with few exceptions: SCT’s unique association with fathers’ ratings of ODD was no longer significant (b ¼ .08, SE ¼ .09). In addition, neither SCT nor ADHD-IN from T1 uniquely predicted social impairment at T3 after controlling for the other as well as depression. Higher scores on depression from T1, however, predicted greater social impairment at T3 even after controlling for SCT and ADHD-IN (mothers: b ¼ .41, SE ¼ .05; fathers: b ¼ .27, SE ¼ .08, ps < .001).1
1
If ADHD-HI, ODD, depression, academic impairment, and social impairment factors at T3 were regressed on their respective factors at T1, then neither the SCT or ADHD-IN factors at T1 significantly predicted the ADHD-HI, ODD, depression, academic impairment, and social impairment factors at T3 above and beyond the ability of the T1 factors ability to predict themselves at T3.
DISCUSSION Our purpose was to determine if SCT and ADHD-IN represent different problems of attention with unique longitudinal correlates. We summarize the findings along with suggestions for future research.
SCT Symptoms—Structure, Development, and Rater Effects With mothers’ and fathers’ ratings, SCT symptoms loses train of thought, easily confused, drowsy, thinking is slow, and behavior is slow showed good convergent validity and discriminant validity in the first and second grades. The three SCT symptoms that failed to show discriminant validity with the ADHD-IN factor in the first grade for mothers and fathers (i.e., daydreams, alertness fluctuates, absentminded; Burns et al., 2013) showed stronger discriminant validity in the second grade, especially for fathers’ ratings. The reason all eight SCT symptoms showed excellent validity for fathers and only five of the eight for mothers at the end of the second grade was not obvious. It is important to note the similarities and differences between the aforementioned results for mothers and fathers in this study and results for teachers and aides for these same children from an earlier study (Bernad et al., 2014). For teachers and aides, only the SCT symptoms thinking is slow, behavior is slow, and drowsy
Downloaded by [New York University] at 08:41 03 May 2015
LONGITUDINAL CORRELATES OF SLUGGISH COGNITIVE TEMPO
showed convergent validity with the SCT factor and discriminant validity with ADHD-IN factor. These same results occurred at T1, T2, and T3 for both teachers and aides. Although the slow thinking=slow behavior characteristic of SCT occurred for mothers and fathers as well as teachers and aides for first- and second-grade children, the inconsistent alertness (e.g., loses train of thought) factor only occurred for parents with this aspect of SCT being stronger at the end of the second than first grade. The slow thinking=slow behavior features of SCT may thus show stronger discriminant validity with the ADHD-IN factor than the inconsistent alertness aspect of SCT at the younger ages with the inconsistent alertness factor also being more apparent to parents than teachers. Taken together, future studies need to evaluate the validity of the eight SCT symptoms with children and adolescents from a variety of cultures with multiple sources. It will be important to determine the validity of the inconsistent alertness and slow thinking=slow behavior aspects of SCT with ADHD-IN at different ages and if the slow thinking=slow behavior aspect is the more central component of SCT (Cortes et al., 2014; Willcutt et al., 2014, p. 32).
9
Limitations In addition to the limitations noted in the context of suggestions for future research, it is important to note the slight attrition in our sample across time (758 unique children at T1, 746 at T2, and 718 [54%] at T3). Also, even though the findings replicated across mothers and fathers, the cross-sectional and longitudinal coefficients share common rater variance (same source completed the measures); thus the coefficients may be slightly inflated. Summary This study found that SCT and ADHD-IN represent different symptom dimensions. Future research needs to determine if the slow thinking=slow behavior component or inconsistent alertness component is more central to SCT along with the development of these two components in the home and school contexts.
ACKNOWLEDGMENTS We thank Cristina Trias and Cristina Solano for their help in data collection.
Cross-Sectional and Longitudinal Correlates of SCT Even though SCT was defined by a different number of symptoms (i.e., eight in Lee et al., 2014; five in this study, and three in Bernad et al., 2014), the cross-sectional and longitudinal correlates were almost identical across these studies. Higher levels of SCT and ADHD-IN uniquely predicted higher levels anxiety=depression, depression, academic impairment, and social impairment. In addition, higher levels of ADHD-IN uniquely predicted higher levels of ADHD-HI and ODD, whereas higher levels of SCT uniquely predicted lower levels of ADHD-HI and ODD or had no unique relationship with ADHD-HI and ODD. Other studies with community or general clinical samples with slightly different measures of SCT report similar cross-sectional results (Barkley, 2013; Becker, Luebbe, et al., 2014; Penny et al., 2009). Future studies with community samples need to include more specific measures of academic and social impairment and sleep problems (Langberg, Becker, Dvorsky, & Luebbe, 2014). In addition, such studies need to include separate measures of anxiety and depression to allow a more specific examination of the SCT, anxiety, and depression relationships across time and to study mediation effects among these variables (e.g., SCT to depression to social impairment).
FUNDING A Ministry of Economy and Competitiveness grant PSI2011-23254 (Spanish Government) and a predoctoral fellowship co-financed by the European Social Fund and the Balearic Island Government (FPI=1451= 2012) supported this research.
REFERENCES Barkley, R. A. (2012). The Barkley Functional Impairment Scale–Children and Adolescents. New York, NY: Guilford. Barkley, R. A. (2013). Distinguishing sluggish cognitive tempo from ADHD in children and adolescents: Executive functioning, impairment, and comorbidity. Journal of Clinical Child & Adolescent Psychology, 42, 161–173. Becker, S. P. (2014). Sluggish cognitive tempo and peer functioning in school-aged children: A six-month longitudinal study. Psychiatry Research, 217, 72–78. Becker, S. P., Luebbe, A. M., Fite, P. J., Stoppelbein, L., & Greening, L. (2014). Sluggish cognitive tempo in psychiatrically hospitalized children: Factor structure and relations to internalizing symptoms, social problems, and observed behavioral dysregulation. Journal of Abnormal Child Psychology, 42, 49–62. Bernad, M. M., Servera, M., Grases, G., Collado, S., & Burns, G. L. (2014). A cross-sectional and longitudinal investigation of the external correlates of sluggish cognitive tempo and ADHD-INattention symptom dimensions. Journal of Abnormal Child Psychology, 42, 1225–1236.
10
SERVERA ET AL.
Downloaded by [New York University] at 08:41 03 May 2015
Burns, G. L., Servera, M., Bernad, M. M., Carrillo, J. M., & Cardo, E. (2013). Distinctions between sluggish cognitive tempo, ADHD-IN, and depression symptom dimensions in Spanish first-grade children. Journal of Clinical Child & Adolescent Psychology, 42, 796–808. Cortes, J. F., Servera, M., Becker, S. P., & Burns, G. L. (2014). External validity of ADHD inattention and sluggish cognitive tempo dimensions in Spanish Children with ADHD. Journal of Attention Disorders. Advance online publication. Langberg, J. M., Becker, S. P., Dvorsky, M. R., & Luebbe, A. M. (2014). Are sluggish cognitive tempo and daytime sleepiness distinct constructs? Psychological Assessment, 26, 586–597. Advance online publication. Lee, S., Burns, G. L., Snell, J., & McBurnett, K. (2014). Validity of the sluggish cognitive tempo symptom dimension in children: Sluggish
cognitive tempo and ADHD-Inattention as distinct symptom dimensions. Journal of Abnormal Child Psychology, 42, 7–19. Muthe´n, L. K., & Muthe´n, B. O. (1998–2012). Mplus user’s guide (7th ed.). Los Angeles, CA: Muthe´n and Muthe´n. Penny, A. M., Waschbusch, D. A., Klein, R. M., Corkum, P., & Eskes, G. (2009). Developing a measure of sluggish cognitive tempo for children: Content validity, factor structure, and reliability. Psychological Assessment, 21, 380–389. Willcutt, E. G., Chhabildas, N., Kinnear, M., DeFries, J. C., Olson, R. K., Leopold, D. R., . . . Pennington, B. F. (2014). The internal and external validity of sluggish cognitive tempo and its relation with DSM–IV ADHD. Journal of Abnormal Child Psychology, 42, 21–35.
Journal of Clinical Child & Adolescent Psychology Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/hcap20
Longitudinal Correlates of Sluggish Cognitive Tempo and ADHD-Inattention Symptom Dimensions with Spanish Children a
a
b
c
Mateu Servera , Maria del Mar Bernad , Jesus M. Carrillo , Susana Collado & G. Leonard d
Burns a
Department of Psychology, University of the Balearic Islands & Research Institute on Health Sciences (IUNICS), , b
Department of Personality, Evaluation and Clinical Psychology, Complutense University of Madrid, , c
Click for updates
Department of Physiotherapy, Occupational Therapy, Physical and Rehabilitation Medicine, Rey Juan Carlos University, , d
Department of Psychology, Washington State University, , Published online: 09 Mar 2015.
To cite this article: Mateu Servera, Maria del Mar Bernad, Jesus M. Carrillo, Susana Collado & G. Leonard Burns (2015): Longitudinal Correlates of Sluggish Cognitive Tempo and ADHD-Inattention Symptom Dimensions with Spanish Children, Journal of Clinical Child & Adolescent Psychology, DOI: 10.1080/15374416.2015.1004680 To link to this article: http://dx.doi.org/10.1080/15374416.2015.1004680
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Clinical Child & Adolescent Psychology, 0(0), 1–10, 2015 Copyright # Taylor & Francis Group, LLC ISSN: 1537-4416 print=1537-4424 online DOI: 10.1080/15374416.2015.1004680
Longitudinal Correlates of Sluggish Cognitive Tempo and ADHD-Inattention Symptom Dimensions with Spanish Children Mateu Servera and Maria del Mar Bernad
Downloaded by [New York University] at 08:41 03 May 2015
Department of Psychology, University of the Balearic Islands & Research Institute on Health Sciences (IUNICS)
Jesus M. Carrillo Department of Personality, Evaluation and Clinical Psychology, Complutense University of Madrid
Susana Collado Department of Physiotherapy, Occupational Therapy, Physical and Rehabilitation Medicine, Rey Juan Carlos University
G. Leonard Burns Department of Psychology, Washington State University
The objective was to examine the longitudinal correlates of sluggish cognitive tempo (SCT) and attention deficit=hyperactivity disorder (ADHD)-Inattention (IN) dimensions with mothers’ and fathers’ ratings of Spanish children. Mothers and fathers rated SCT, ADHD-IN, ADHD-hyperactivity=impulsivity (HI), oppositional defiant disorder (ODD), depression, academic impairment, and social impairment on 3 occasions (twice in first-grade year [6-week separation] and once in the second-grade year [12 months after the first assessment]) in Spanish children (758, 746, and 718 children at the 3 time-points with approximately 55% boys). The results showed that (a) higher levels of SCT from earlier assessments predicted higher levels of depression, academic impairment, and social impairment at Assessment 3 after controlling for ADHD-IN at earlier assessments; (b) higher levels of ADHD-IN from earlier assessments predicted higher levels of depression, academic impairment, and social impairment at Assessment 3 after controlling for SCT at earlier assessments; (c) higher levels of ADHD-IN from earlier assessments predicted higher levels of ADHD-HI and ODD at Assessment 3 after controlling for SCT from earlier assessments; and (d) higher levels of SCT from earlier assessments either showed no unique relationship with ADHD-HI and ODD or predicted lower levels of ADHD-HI and ODD at Assessment 3 after controlling for ADHD-IN from earlier assessments. Initial evidence is provided of SCT’s unique longitudinal relationships with depression and academic=social impairment and different longitudinal relationships with ADHD-HI and ODD relative to ADHD-IN, thus adding to a growing body of research underscoring the importance of SCT as distinct from ADHD-IN.
Correspondence should be addressed to G. Leonard Burns, Department of Psychology, Washington State University, P.O. Box 644820, Pullman, WA 99164-4820. E-mail: [email protected]
Sluggish cognitive tempo (SCT) is characterized by inconsistent alertness (e.g., loses train of thought, daydreams) and slow thinking=slow behavior (Lee,
Downloaded by [New York University] at 08:41 03 May 2015
2
SERVERA ET AL.
Burns, Snell, & McBurnett, 2014, Table 2). Research indicates that SCT and attention deficit=hyperactivity disorder (ADHD) inattention (IN) symptoms represent different dimensions with unique cross-sectional correlates (Barkley, 2013; Becker, Luebbe, Fite, Stoppelbein, & Greening, 2014; Bernad, Servera, Grases, Collado, & Burns, 2014; Burns, Servera, Bernad, Carrillo, & Cardo, 2013; Cortes, Servera, Becker, & Burns, 2014; Lee et al., 2014; Penny, Waschbusch, Klein, Corkum, & Eskes, 2009). For example, higher levels of SCT are associated with lower levels of externalizing problems after controlling for ADHD-IN, whereas higher levels of ADHD-IN are associated with higher levels of externalizing problems after controlling for SCT. In addition, higher levels of SCT and ADHD-IN uniquely predict higher levels of anxiety, depression, academic impairment, and social impairment. Although these cross-sectional results are promising, the measurement of predictors and outcomes at the same time is a weakness (i.e., Does SCT influence the outcomes, or do the outcomes influence SCT?). Longitudinal research where the measurement of SCT and ADHD-IN occurs prior to the outcomes would provide a stronger evaluation of SCT’s and ADHD-IN’s unique roles in children’s adjustment. At this time, only two studies have examined the correlates of SCT using a longitudinal design. With teacher ratings and a 6-month interval, Becker (2014) found that SCT uniquely predicted subsequent peer adjustment after controlling for prior adjustment, ADHD-IN and other symptom dimensions. With teachers and aides ratings and a 12-month interval, Bernad et al. (2014) found that higher levels of SCT predicted lower levels of ADHD-hyperactivity=impulsivity (HI) and oppositional defiant disorder (ODD) after controlling for ADHD-IN, whereas higher levels of ADHD-IN predicted higher levels of ADHD-HI and ODD after controlling for SCT. In addition, higher scores on SCT and ADHD-IN both uniquely predicted higher levels of academic impairment. Despite these promising findings, there is a need for additional longitudinal research. This study is the first longitudinal study to examine parent ratings of SCT and ADHD-IN in relation to children’s functioning and, further, uses both mother and father ratings. If ratings by mothers and fathers yield the same results as the two earlier longitudinal studies in school settings, then such would show for the first time that SCT has unique longitudinal effects relative to ADHD-IN on children’s adjustment in home settings with these effects being consistent across ratings by mothers and fathers.
OBJECTIVES The current study used ratings by mothers and fathers of Spanish children across three occasions—twice at
the end of the first-grade year (6-week separation) and once at the end of the second-grade year (12 months after the first assessment)—to investigate the longitudinal correlates of SCT and ADHD-IN. The first three objectives establish the construct validity of SCT at the third assessment in order to determine the unique longitudinal correlates of SCT and ADHD-IN from the first two assessments to the third assessment.
Objective 1: Convergent and Discriminant Validity of SCT Symptoms at Time 3 (T3) The first objective was to determine if the same five of eight SCT symptoms with convergent validity (substantial loadings on the SCT factor) and discriminant validity (low loadings on the ADHD-IN factor) at the two assessments in the first grade also showed convergent and discriminant validity in the second grade. These five SCT symptoms were loses train of thought, easily confused, thinking is slow, slow moving, and drowsy. The other three SCT symptoms—daydreams, alertness fluctuates, and absentminded—failed to show discriminant validity with the ADHD-IN factor in the first grade (Burns et al., 2013). However, given all eight SCT symptoms showed discriminant validity with the ADHD-IN factor with parent ratings of kindergarten through sixth-grade American children (Lee et al., 2014), the other three SCT symptoms may show improved validity with the Spanish children being 12 months older.
Objective 2: Measurement Model at T3 and across Time (T1 to T3, T2 to T3) The second objective was to evaluate the fit of an SCT, ADHD-IN, ADHD-HI, ODD toward adults [ODD-A], ODD toward peers [ODD-P], depression, and academic and social impairment eight-factor model at T3 along with the reliability of the factors (i.e., true score variance, mother–father agreement, stability of like factor correlations for the 10.5- and 12-month intervals). We also predicted that the correlation between the ADHD-IN factor and the ADHD-HI, ODD-A, and ODD-P factors would be stronger than the correlation between the SCT factor and the ADHD-HI, ODD-A, and ODD-P factors. In addition, we predicted the SCT and ADHD-IN factors would show similar correlations with depression, academic, and social impairment factors. Finally, we predicted that SCT and ADHD-IN factors from T1 and T2 would show this same pattern of correlations with ADHD-HI, ODD-A, ODD-P, depression, and academic and social impairment factors at T3.
Downloaded by [New York University] at 08:41 03 May 2015
LONGITUDINAL CORRELATES OF SLUGGISH COGNITIVE TEMPO
3
FIGURE 1 Cross-sectional and longitudinal regression models. Note: Each latent variable was defined by manifest variables (not show in path diagrams) with 12 months separating Assessments 1 and 3 and 10.5 months separating Assessments 2 and 3. Each model allowed correlations among all disturbances (not shown). SCT ¼ sluggish cognitive tempo; ADHD ¼ attention-deficit=hyperactivity disorder; IN ¼ inattention; HI ¼ hyperactivity= impulsivity; ODD ¼ oppositional defiant disorder.
Objective 3: Unique Cross-Sectional Correlates of SCT and ADHD-IN at T3 The third objective was to determine the unique cross-sectional correlates of SCT and ADHD-IN at T3. Figure 1a shows this cross-sectional structural regression model. It was predicted that higher levels of ADHD-IN would predict higher levels of ADHD-HI and ODD after controlling for SCT, whereas SCT’s relationship with ADHD-HI and ODD would be nonsignificant (or significantly negative) after controlling for ADHD-IN. It was also predicted that SCT and ADHD-IN would both uniquely predict higher levels of depression and academic and social impairment. To ensure that SCT is different from depression, we also repeated these analyses controlling for depression as well as ADHD-IN. Objective 4: Unique Longitudinal Correlates of SCT and ADHD-IN The fourth objective was to determine if SCT and ADHD-IN have unique longitudinal correlates. Figure 1b shows the two structural regression analyses (i.e., T3 regressed on T1; T3 regressed on T2). It was predicted that higher levels of ADHD-IN from the first-grade occasions (T1 and T2) would predict higher levels of ADHD-HI and ODD in the second grade after controlling for SCT, whereas SCT in the first grade would have a nonsignificant (or significantly negative) relationship with ADHD-HI and ODD in the second grade after controlling for ADHD-IN. It was also predicted that higher levels of SCT and ADHD-IN in the first grade would both uniquely predict higher levels of depression and academic and social impairment in the second grade. To again ensure the findings for SCT were independent of depression, we repeated the analyses controlling for
depression and ADHD-IN. Support for these four objectives, especially Objective 4, would uniquely add to the research that suggests SCT is distinct from ADHD-IN.
METHOD Participants and Procedures The 46 elementary schools on the island of Majorca (Spain) were invited to participate, with 43 indicating an interest. Twenty-two of 43 schools were randomly selected along with eight from Madrid (eight were asked and eight agreed). There were 1,045 first-grade children in these 30 schools at T1. A cover letter and measures were sent to mothers and fathers at T1, T2 (6 weeks after T1), and T3 (12 months after T1). At T1, 723 mothers and 603 fathers participated, with 667 mothers and 584 fathers at T2 and 604 mothers and 540 fathers at T3. There were 758 unique children (55% boys) at T1, 746 (54% boys) at T2, and 718 (54% boys) at T3. Although ethnicity was not obtained for individual children, at the school level, approximately 90% were Caucasian with 10% North African. The Institutional Review Board of the University of the Balearic Islands approved the research. Measures Child and adolescent disruptive behavior inventory. The Child and Adolescent Disruptive Behavior Inventory measures SCT (eight symptoms, Burns et al., 2013, Table 1), ADHD-IN (nine symptoms), ADHD-HI (nine symptoms), ODD-A (e.g., argues with adults, eight symptoms), ODD-P toward siblings=peers (e.g., argues with siblings=peers, eight symptoms), depression (five symptoms: seems sad, unhappy, or depressed; seems to feel
4
SERVERA ET AL. TABLE 1 Reliability, Interrater, and Stability Coefficients (Standard Errors) for Measures Stability Coefficients Reliability Coefficients
Downloaded by [New York University] at 08:41 03 May 2015
SCT ADHD-IN ADHD-HI ODD-A ODD-P Depression Academic Impairment Social Impairment
Mothersa
Fathersb
.81 .94 .93 .93 .92 .75 .92 .90
.82 .94 .93 .93 .91 .76 .92 .91
(.01) (.01) (.01) (.01) (.01) (.02) (.01) (.02)
10.5 Months Interraterc
(.02) (.01) (.01) (.01) (.01) (.02) (.01) (.02)
.77 .80 .80 .73 .67 .67 .86 .76
(.03) (.02) (.02) (.03) (.03) (.04) (.01) (.02)
12 Months
Mothersd
Fatherse
Motherf
Fathersg
.75 (.04) .80 (.02) .78 (.02) — — .72 (.04) .76 (.02) —
.76 (.03) .79 (.02) .74 (.02) — — .64 (.04) .77 (.02) —
.70 .74 .77 .68 .60 .72 .72 .52
.67 .70 .76 .62 .54 .64 .74 .55
(.04) (.02) (.02) (.03) (.03) (.04) (.02) (.03)
(.06) (.03) (.02) (.04) (.04) (.04) (.02) (.03)
Note: Confirmatory factor analytic procedures were used to calculate the reliability coefficients (i.e., the amount of true score variance in the measure), interrater factor correlations (i.e., the correlations between the same factors for mothers and fathers) and the stability coefficients (i.e., the correlations between the same factor within a source across time). SCT ¼ sluggish cognitive tempo; ADHD-IN ¼ attention-deficit=hyperactivity disorder-inattention; HI ¼ hyperactivity=impulsivity; ODD-A ¼ oppositional defiant disorder toward adults; ODD-P ¼ oppositional defiant disorder toward peers. A dash indicates the measure was not administered at the second assessment. a n ¼ 604. b n ¼ 540. c n ¼ 618. d n ¼ 733. e n ¼ 673. f n ¼ 790. g n ¼ 700.
worthless or expresses feelings of worthless; seems not to enjoy home activities anymore; seems lonely or expresses feelings of loneliness; and seems to feel hopeless about things or expresses feelings of hopelessness) and academic impairment (four items: completion of homework, reading skills, arithmetic skills, and writing skills). The ADHD and ODD symptoms were rated on a 6-point frequency of occurrence scale (i.e., occurs nearly none of the time [two or fewer times per month], seldom occurs [once per week], sometimes occurs [a few times per week], often occurs [once per day], very often occurs [several times per day], and occurs nearly all the time [many times per day]). The SCT and depression symptoms were rated on a 6-point duration of occurrence scale (nearly none of the time, seldom, sometimes, often, very often, and nearly all of the time). A 7-point scale was used for the four academic items (severe difficulty, moderate difficulty, slight difficulty, average performance for grade level, slightly above average, moderately above average, and excellent performance for grade level). The academic items were reversed keyed so higher scores represented academic impairment. Mothers and fathers rated the symptoms for home and community and were told not consider the children’s behavior at school. The instructions for the academic items were to rate the children’s current skills. Mothers and fathers were also asked to make their ratings independently. Earlier studies support the reliability and validity of the scores from the scale
(Burns et al., 2013; Lee et al., 2014). Table 1 shows the reliability values for this study. Functional impairment rating scale—children and adolescents (Barkley, 2012). The study used 7 of 15 items from Barkley’s Functional Impairment Rating Scale—Children and Adolescents to measure social impairment, that is, (a) interactions with their mother, (b) interactions with their father, (c) interactions with siblings, (d) interactions with other children, (e) interactions with other adults, (f) activities in the community, and (g) visiting others homes. These seven items were rated on a 10-point scale from 0 (not impaired) to 9 (severely impaired). (The eight at school items were not included.) Analytic Strategy All analyses treated the item ratings as ordered-categories and used the robust weighted least squares estimator (Version 7.1; Muthe´n & Muthe´n, 1998–2012). The robust weighted least squares estimator uses a pairwise deletion procedure for missing information. All the analyses took into account the children were nested within 30 schools (Type ¼ complex Mplus option.). The fit of the measurement model was evaluated with the comparative fit index (CFI; study criterion .95), Tucker–Lewis index (TLI; study criterion .95), and the root mean square error of approximation (RMSEA; study criterion .05). The Mplus model constraint procedure along with phantom
LONGITUDINAL CORRELATES OF SLUGGISH COGNITIVE TEMPO
constructs was used to determine if factor correlations differed significantly.
CFI ¼ .982, TLI ¼ .981, and RMSEA ¼ .015, 90% confidence interval [.012, .016].
RESULTS
Item-factor loadings. The five SCT symptoms showed strong loadings on the SCT factor for mothers and fathers (M ¼ .76, SD ¼ .07). Similar results occurred for the ADHD-IN (M ¼ .83, SD ¼ .08), ADHD-HI (M ¼ .84, SD ¼ .04), ODD-A (M ¼ .86, SD ¼ .03), ODD-P (M ¼ .84, SD ¼ .03), depression (M ¼ .73, SD ¼ .04), academic impairment (M ¼ .88, SD ¼ .03), and social impairment (M ¼ .84, SD ¼ .04) items.
Convergent and Discriminant Validity of SCT and ADHD-IN Symptoms
Downloaded by [New York University] at 08:41 03 May 2015
5
A two-factor model was applied to mothers’ and fathers’ ratings of the eight SCT and nine ADHD-IN symptoms to determine the convergent and discriminant validity of these symptoms at T3 (i.e., the analysis was restricted to two factors with cross-loadings allowed). For fathers’ ratings, all eight SCT symptoms showed substantial loading on the SCT factor (M ¼ .69, SD ¼ .10) in conjunction with low loadings on the ADHD-IN factor (M ¼ .09, SD ¼ .12). The ADHD-IN symptoms showed substantial loadings on the ADHD-IN factor (M ¼ .82, SD ¼ .06) and low loading on the SCT factor (M ¼ .02, SD ¼ .08). All SCT and ADHD-IN symptoms showed convergent and discriminant validity for fathers. For mothers’ ratings, the SCT symptoms daydreams, alertness fluctuates, and absentminded showed weak discriminant validity at T3. Their average loading on the SCT factor was .51 (SD ¼ .05) and their average loading on the ADHD-IN factor .37 (SD ¼ .02). These were the same three SCT symptoms that showed no discriminant validity with the ADHD-IN factor at T1 and T2 for mothers and fathers in the previous study (Burns et al., 2013). The other five SCT symptoms showed good convergent (M ¼ .66, SD ¼ .12) and discriminant validity (M ¼ .11, SD ¼ .13) for mothers at T3. All nine ADHD-IN symptoms showed good convergent (M ¼ .82, SD ¼ .09) and discriminant validity (M ¼ .01, SD ¼ .17) for mothers at T3. Operationalization of SCT Construct Although the three SCT symptoms with no discriminant validity with the ADHD-IN factor at T1 and T2 for mothers’ and fathers’ ratings showed an improvement in their discriminant validity at T3, especially for fathers, the SCT construct at T3 was defined by the same five SCT symptoms as at T1 and T2. These five SCT symptoms were loses train of thought, easily confused, drowsy, thinking is slow, and slow moving. This was necessary to meaningfully compare the cross-sectional results across assessments and the cross-sectional to the longitudinal results. Measurement Model at T3 Global fit. The fit of the multiple indicator (individual items) by multiple trait (SCT, ADHD-IN, ODD-A, ODD-P, depression, and academic and social impairment) by multiple source (mothers and fathers) measure model at T3 was excellent, v2(5081) ¼ 5,745, p < .001,
Reliability coefficients. The SCT scale scores contained 81% and 82% true score variance for mothers and fathers, respectively. The values for the other scales were higher than 90% with the exception of the depression scale with values of 75 and 76% for mothers and fathers, respectively (Table 1). Mother–father (interrater) factor correlations. The mother–father factor correlation for SCT was .77 with similar values for ADHD-IN (.80), ADHD-HI (.80), ODD-A (.73), ODD-P (.67), depression (.67), academic impairment (.86), and social impairment (.76). Stability of factor correlations. Table 1 shows the stability factor correlations for 10.5-month (T2–T3) and 12-month (T1–T3) intervals. The stability coefficients for SCT varied from .67 to .76. The values for the other factors were similar. Cross-sectional factor correlations at T3. Table 2 shows the correlations among the factors. ADHD-IN’s relationship with ADHD-HI, ODD-A, and ODD-P was significantly stronger than SCT’s relationship with these variables (ps < .002). In addition, SCT and ADHD-IN showed significant (ps < .05) relationships with depression, academic impairment, and social impairment with these relationships not being significantly different (ps > .05). Cross-Sectional Unique Effects of SCT and ADHD-IN on ADHD-HI, ODD, Depression, Academic Impairment, and Social Impairment at T3 Analytic strategy. A cross-sectional structural regression analysis was used to determine the unique effects of SCT and ADHD-IN on ADHD-HI, ODD (the two ODD scales were combined), depression, academic impairment, and social impairment at T3. Figure 1a shows this model with the partial standardized regression coefficients shown in Table 3.
6
SERVERA ET AL.
TABLE 2 Correlations Among Sluggish Cognitive Tempo, ADHD-IN, ADHD-HI, ODD-A, ODD-P, Depression, Academic Impairment, and Social Impairment Factors for Mothers (Fathers) at Time 3
SCT ADHD-IN ADHD-HI ODD-A ODD-P DEP AI SI
SCT
ADHD-IN
ADHD-HI
— .78 (.76) .48 (.49) .39 (.44) .40 (.49) .61 (.64) .54 (.58) .39 (.46)
— .68 (.66) .55 (.54) .57 (.56) .60 (.63) .48 (.57) .42 (.44)
— .61 (.60) .59 (.67) .44 (.46) .19 (.30) .40 (.36)
ODD-A
ODD-P
DEP
AI
SI
— (.69) (.48) (.25) (.51)
— .58 (.59) .20 (.26) .50 (.53)
— .34 (.42) .50 (.48)
— .12 (.21)
—
.69 .54 .18 .53
Downloaded by [New York University] at 08:41 03 May 2015
Note: All correlations were significant at p < .01. The range for the standard errors was from 0.02 to 0.05. SCT ¼ sluggish cognitive tempo; ADHD-IN ¼ attention-deficit=hyperactivity disorder-inattention; ADHD-HI ¼ hyperactivity=impulsivity; ODD-A ¼ oppositional defiant disorder toward adults; ODD-P ¼ oppositional defiant disorder toward peers; DEP ¼ depression; AI ¼ academic impairment; SI ¼ social impairment.
Unique effects of SCT and ADHD-IN on ADHD-HI and ODD. Although higher scores on ADHD-IN predicted significantly higher scores on ADHD-HI and ODD after controlling for SCT (ps < .001), SCT was no longer significantly related to ODD after controlling for ADHD-IN. Higher scores on SCT were associated with significantly lower scores on ADHD-HI after controlling ADHD-IN (p < .05) for mothers’ ratings (and unrelated for fathers’ ratings). Unique effects of SCT and ADHD-IN on depression. Across both mother and father ratings, higher scores on SCT predicted significantly greater depression after controlling for ADHD-IN with higher scores on ADHD-IN also predicting significantly greater depression after controlling for SCT (ps < .001). Unique effects of SCT and ADHD-IN on academic impairment. Across both mother and father ratings, higher scores on SCT predicted higher scores on
academic impairment after controlling for ADHD-IN (ps < .001), whereas higher scores on ADHD-IN also predicted higher scores on academic impairment after controlling for SCT (ps < .05). Unique effects of SCT and ADHD-IN on social impairment. Across both mother and father ratings, higher levels of SCT predicted significantly higher levels of social impairment after controlling for ADHD-IN with higher levels of ADHD-IN also predicting significantly higher levels of social impairment after controlling for SCT (ps < .05).
Role of Depression in SCT’s Unique Cross-Sectional Associations with ODD, ADHD-HI, Academics, and Social Impairment at T3 With depression added as a predictor along with ADHD-IN, higher scores on SCT now predicted significantly lower scores on ODD for mothers (b ¼ .26,
TABLE 3 Cross-Sectional Regression of ADHD-HI, ODD, Depression, Academic Impairment, and Social Impairment Factors on the Sluggish Cognitive Tempo and ADHD-IN Factors at Time 3 ADHD-HI Predictors SCT Mothers Fathers ADHD-IN Mothers Fathers
ODD
Outcomes DEP
AI
SI
b
SE
b
SE
b
SE
b
SE
b
SE
.15 .03ns
.06 .07
.12ns .10ns
.07 .07
.37 .38
.06 .07
.43 .36
.06 .06
.17 .29
.06 .07
.80 .68
.05 .06
.65 .47
.06 .07
.32 .34
.06 .05
.15 .30
.06 .06
.30 .22
.06 .07
Note: SCT ¼ sluggish cognitive tempo; ODD ¼ oppositional defiant disorder toward adults and peers; ADHD-IN ¼ attention-deficit=hyperactivity disorder-inattention; ADHD-HI ¼ attention-deficit hyperactivity disorder-hyperactivity=impulsivity; DEP ¼ depression; AI ¼ academic impairment; SI ¼ social impairment. p < .05. p < .001.
7
LONGITUDINAL CORRELATES OF SLUGGISH COGNITIVE TEMPO
SE ¼ 08, p < .01) with SCT’s unique relationship with maternal ratings of social impairment no longer significant (p > .05). All the other outcomes for the unique effects of SCT on ADHD-HI, ODD, academic impairment, and social impairment were unchanged when controlling for depression. Higher scores on depression uniquely predicted higher scores on ODD and social impairment for mothers and fathers (mothers: b ¼ .37, SE ¼ .04, b ¼ .39, SE ¼ .05, ps < .001; fathers: b ¼ .27, SE ¼ .08, b ¼ .28, SE ¼ .06, ps < .001).
Table 5 shows the partial standardized regression coefficients for the two intervals (i.e., T1–T3 and T2–T3). Unique longitudinal effects of SCT and ADHD-IN on ADHD-HI and ODD. Higher levels of ADHD-IN predicted significantly higher levels of ADHD-HI and ODD across the two intervals even after controlling for SCT (ps < .01). Higher levels of SCT significantly predicted lower scores on ADHD-HI across the 10.5-month interval after controlling for ADHD-IN (ps < .05). There was also tendency at least for mothers for higher SCT scores to predict lower ODD scores across the 12-month interval after controlling for ADHD-IN (p ¼ .07).
Downloaded by [New York University] at 08:41 03 May 2015
Longitudinal Factor Correlations Table 4 shows the correlations of SCT and ADHD-IN factors from T1 to T2 with ADHD-HI, ODD, depression, and academic and social impairment factors from T3. The longitudinal factor correlations showed the same pattern as the cross-sectional factor correlations. The ADHD-IN factor showed a stronger longitudinal relationship than SCT with the ADHD-HI and ODD factors (ps < .01), with the one exception being that the ADHD-IN and SCT correlations with ODD for the 12-month interval did not differ significantly (ps > .05). In addition, and consistent with the cross-sectional findings, the ADHD-IN and SCT factors showed similar longitudinal associations with depression, academic impairment, and social impairment (ps > .05).
Unique longitudinal effects of SCT and ADHD-IN on depression. Higher scores on SCT predicted marginally higher depression scores after controlling for ADHD-IN (ps < .07) with higher scores on ADHD-IN significantly predicting higher depression scores after controlling for SCT (ps < .05).
Longitudinal Structural Regression Analysis
Unique longitudinal effects of SCT and ADHD-IN on academic impairment. Higher scores on ADHD-IN predicted significantly higher levels academic impairment across the two intervals after controlling for SCT (ps < .05), whereas higher levels of SCT predicted significantly higher levels of academic impairment for three of the four coefficients across the two intervals after controlling for ADHD-IN (ps < .05).
Analytic strategy. ADHD-HI, ODD, depression, academic impairment, and social impairment factors from T3 were regressed on SCT and ADHD-IN factors from T1 to T2 to determine the unique effects of SCT and ADHD-IN across time. Figure 1b shows this model.
Unique longitudinal effects of SCT and ADHD-IN on social impairment. Higher scores on ADHD-IN predicted higher levels of social impairment after controlling for SCT across the two time intervals (ps < .05),
TABLE 4 Correlations (Standard Errors) of Sluggish Cognitive Tempo and ADHD-Inattention Factors from Times 1 and 2 with ADHD-HI, ODD, Depression, Academic Impairment, and Social Impairment Factors from Time 3 Time 1
Time 2
SCT
ADHD-IN
SCT
ADHD-IN
Time 3
Mothers
Fathers
Mothers
Fathers
Mothers
Fathers
Mothers
Fathers
ADHD-HI ODD Depression AI SI
.35 .35 .47 .45 .32
.37 .35 .47 .43 .33
.48 .40 .50 .45 .31
.51 .35 .50 .49 .31
.33 .30 .48 .43 .23
.37 .31 .52 .50 .32
.54 .42 .47 .43 .27
.56 .41 .54 .52 .35
(.04) (.04) (.05) (.04) (.03)
(.05) (.04) (.08) (.04) (.04)
(.04) (.03) (.04) (.04) (.02)
(.04) (.04) (.04) (.04) (.04)
(.04) (.03) (.04) (.05) (.03)
(.04) (.04) (.05) (.03) (.05)
(.03) (.04) (.04) (.05) (.03)
(.04) (.04) (.04) (.03) (.04)
Note: All correlations significant at p < .001. The interval from Time 1 to Time 3 was 12 months with the interval from Time 2 to Time 3 being 10.5 months. ODD ¼ oppositional defiant disorder toward adults and peers; ADHD-IN ¼ attention-deficit hyperactivity disorder–inattention; ADHD-HI ¼ attention-deficit hyperactivity disorder–hyperactivity=impulsivity; SCT ¼ sluggish cognitive tempo; AI ¼ academic impairment; SI ¼ social impairment.
8
SERVERA ET AL.
TABLE 5 Longitudinal Regression of ADHD-HI, ODD, Depression, Academic Impairment, and Social Impairment Factors on the Sluggish Cognitive Tempo and ADHD-IN Factors Outcomes ADHD-HI
Downloaded by [New York University] at 08:41 03 May 2015
b Time 1 Predictors SCT Mothers Fathers ADHD-IN Mothers Fathers Time 2 Predictors SCT Mothers Fathers ADHD-IN Mothers Fathers
ODD SE
b
DEP
AI
SI
SE
b
SE
b
SE
b
SE
.08 .08
.21 .22
.08 .13
.26 .13ns
.07 .09
.19 .22
.07 .06
.07 .08
.34 .33
.06 .10
.25 .40
.07 .09
.17 .15
.06 .06
.09 .12
.29 y .23
.09 .13
.23 .21
.10 .08
.03ns .10ns
.08 .12
.10 .11
.23 .35
.09 .08
.24 .35
.10 .08
.25 .27
.07 .11
Time 3 Outcomes (12-Month Interval) .04ns .02ns
.08 .09
.11ns .18
.07 .32 .51 .08 .22 .53 Time 3 Outcomes (10.5-Month Interval) .37 .26
.07 .11
.85 .77
.08 .11
y
.16 .09ns .55 .49
Note: SCT ¼ sluggish cognitive tempo; ODD ¼ oppositional defiant disorder toward adults and peers; ADHD-IN ¼ attention-deficit=hyperactivity disorder-inattention; ADHD-HI ¼ attention-deficit hyperactivity disorder—hyperactivity=impulsivity; DEP ¼ depression; AI ¼ academic impairment; SI ¼ social impairment. y p < .07. p < .05. p < .01. p < .001.
whereas higher levels of SCT predicted higher levels of social impairment after controlling for ADHD-IN only across the 12-month interval (ps < .05). Role of Depression in SCT’s Unique Longitudinal Associations with ODD, ADHD-HI, and Academic and Social Impairment The unique effect of SCT at T1 on ADHD-HI, ODD, and academic and social impairment at T3 was also examined after controlling for depression and ADHD-IN from T1. The unique effects of SCT remained the same with few exceptions: SCT’s unique association with fathers’ ratings of ODD was no longer significant (b ¼ .08, SE ¼ .09). In addition, neither SCT nor ADHD-IN from T1 uniquely predicted social impairment at T3 after controlling for the other as well as depression. Higher scores on depression from T1, however, predicted greater social impairment at T3 even after controlling for SCT and ADHD-IN (mothers: b ¼ .41, SE ¼ .05; fathers: b ¼ .27, SE ¼ .08, ps < .001).1
1
If ADHD-HI, ODD, depression, academic impairment, and social impairment factors at T3 were regressed on their respective factors at T1, then neither the SCT or ADHD-IN factors at T1 significantly predicted the ADHD-HI, ODD, depression, academic impairment, and social impairment factors at T3 above and beyond the ability of the T1 factors ability to predict themselves at T3.
DISCUSSION Our purpose was to determine if SCT and ADHD-IN represent different problems of attention with unique longitudinal correlates. We summarize the findings along with suggestions for future research.
SCT Symptoms—Structure, Development, and Rater Effects With mothers’ and fathers’ ratings, SCT symptoms loses train of thought, easily confused, drowsy, thinking is slow, and behavior is slow showed good convergent validity and discriminant validity in the first and second grades. The three SCT symptoms that failed to show discriminant validity with the ADHD-IN factor in the first grade for mothers and fathers (i.e., daydreams, alertness fluctuates, absentminded; Burns et al., 2013) showed stronger discriminant validity in the second grade, especially for fathers’ ratings. The reason all eight SCT symptoms showed excellent validity for fathers and only five of the eight for mothers at the end of the second grade was not obvious. It is important to note the similarities and differences between the aforementioned results for mothers and fathers in this study and results for teachers and aides for these same children from an earlier study (Bernad et al., 2014). For teachers and aides, only the SCT symptoms thinking is slow, behavior is slow, and drowsy
Downloaded by [New York University] at 08:41 03 May 2015
LONGITUDINAL CORRELATES OF SLUGGISH COGNITIVE TEMPO
showed convergent validity with the SCT factor and discriminant validity with ADHD-IN factor. These same results occurred at T1, T2, and T3 for both teachers and aides. Although the slow thinking=slow behavior characteristic of SCT occurred for mothers and fathers as well as teachers and aides for first- and second-grade children, the inconsistent alertness (e.g., loses train of thought) factor only occurred for parents with this aspect of SCT being stronger at the end of the second than first grade. The slow thinking=slow behavior features of SCT may thus show stronger discriminant validity with the ADHD-IN factor than the inconsistent alertness aspect of SCT at the younger ages with the inconsistent alertness factor also being more apparent to parents than teachers. Taken together, future studies need to evaluate the validity of the eight SCT symptoms with children and adolescents from a variety of cultures with multiple sources. It will be important to determine the validity of the inconsistent alertness and slow thinking=slow behavior aspects of SCT with ADHD-IN at different ages and if the slow thinking=slow behavior aspect is the more central component of SCT (Cortes et al., 2014; Willcutt et al., 2014, p. 32).
9
Limitations In addition to the limitations noted in the context of suggestions for future research, it is important to note the slight attrition in our sample across time (758 unique children at T1, 746 at T2, and 718 [54%] at T3). Also, even though the findings replicated across mothers and fathers, the cross-sectional and longitudinal coefficients share common rater variance (same source completed the measures); thus the coefficients may be slightly inflated. Summary This study found that SCT and ADHD-IN represent different symptom dimensions. Future research needs to determine if the slow thinking=slow behavior component or inconsistent alertness component is more central to SCT along with the development of these two components in the home and school contexts.
ACKNOWLEDGMENTS We thank Cristina Trias and Cristina Solano for their help in data collection.
Cross-Sectional and Longitudinal Correlates of SCT Even though SCT was defined by a different number of symptoms (i.e., eight in Lee et al., 2014; five in this study, and three in Bernad et al., 2014), the cross-sectional and longitudinal correlates were almost identical across these studies. Higher levels of SCT and ADHD-IN uniquely predicted higher levels anxiety=depression, depression, academic impairment, and social impairment. In addition, higher levels of ADHD-IN uniquely predicted higher levels of ADHD-HI and ODD, whereas higher levels of SCT uniquely predicted lower levels of ADHD-HI and ODD or had no unique relationship with ADHD-HI and ODD. Other studies with community or general clinical samples with slightly different measures of SCT report similar cross-sectional results (Barkley, 2013; Becker, Luebbe, et al., 2014; Penny et al., 2009). Future studies with community samples need to include more specific measures of academic and social impairment and sleep problems (Langberg, Becker, Dvorsky, & Luebbe, 2014). In addition, such studies need to include separate measures of anxiety and depression to allow a more specific examination of the SCT, anxiety, and depression relationships across time and to study mediation effects among these variables (e.g., SCT to depression to social impairment).
FUNDING A Ministry of Economy and Competitiveness grant PSI2011-23254 (Spanish Government) and a predoctoral fellowship co-financed by the European Social Fund and the Balearic Island Government (FPI=1451= 2012) supported this research.
REFERENCES Barkley, R. A. (2012). The Barkley Functional Impairment Scale–Children and Adolescents. New York, NY: Guilford. Barkley, R. A. (2013). Distinguishing sluggish cognitive tempo from ADHD in children and adolescents: Executive functioning, impairment, and comorbidity. Journal of Clinical Child & Adolescent Psychology, 42, 161–173. Becker, S. P. (2014). Sluggish cognitive tempo and peer functioning in school-aged children: A six-month longitudinal study. Psychiatry Research, 217, 72–78. Becker, S. P., Luebbe, A. M., Fite, P. J., Stoppelbein, L., & Greening, L. (2014). Sluggish cognitive tempo in psychiatrically hospitalized children: Factor structure and relations to internalizing symptoms, social problems, and observed behavioral dysregulation. Journal of Abnormal Child Psychology, 42, 49–62. Bernad, M. M., Servera, M., Grases, G., Collado, S., & Burns, G. L. (2014). A cross-sectional and longitudinal investigation of the external correlates of sluggish cognitive tempo and ADHD-INattention symptom dimensions. Journal of Abnormal Child Psychology, 42, 1225–1236.
10
SERVERA ET AL.
Downloaded by [New York University] at 08:41 03 May 2015
Burns, G. L., Servera, M., Bernad, M. M., Carrillo, J. M., & Cardo, E. (2013). Distinctions between sluggish cognitive tempo, ADHD-IN, and depression symptom dimensions in Spanish first-grade children. Journal of Clinical Child & Adolescent Psychology, 42, 796–808. Cortes, J. F., Servera, M., Becker, S. P., & Burns, G. L. (2014). External validity of ADHD inattention and sluggish cognitive tempo dimensions in Spanish Children with ADHD. Journal of Attention Disorders. Advance online publication. Langberg, J. M., Becker, S. P., Dvorsky, M. R., & Luebbe, A. M. (2014). Are sluggish cognitive tempo and daytime sleepiness distinct constructs? Psychological Assessment, 26, 586–597. Advance online publication. Lee, S., Burns, G. L., Snell, J., & McBurnett, K. (2014). Validity of the sluggish cognitive tempo symptom dimension in children: Sluggish
cognitive tempo and ADHD-Inattention as distinct symptom dimensions. Journal of Abnormal Child Psychology, 42, 7–19. Muthe´n, L. K., & Muthe´n, B. O. (1998–2012). Mplus user’s guide (7th ed.). Los Angeles, CA: Muthe´n and Muthe´n. Penny, A. M., Waschbusch, D. A., Klein, R. M., Corkum, P., & Eskes, G. (2009). Developing a measure of sluggish cognitive tempo for children: Content validity, factor structure, and reliability. Psychological Assessment, 21, 380–389. Willcutt, E. G., Chhabildas, N., Kinnear, M., DeFries, J. C., Olson, R. K., Leopold, D. R., . . . Pennington, B. F. (2014). The internal and external validity of sluggish cognitive tempo and its relation with DSM–IV ADHD. Journal of Abnormal Child Psychology, 42, 21–35.
