Cognitive disengagement syndrome (sluggish cognitive tempo) and medical conditions: A systematic review and call for future research

Abstract

Objectives:

Cognitive disengagement syndrome (CDS, previously referred to as sluggish cognitive tempo) is a unique set of symptoms distinct from ADHD inattentive symptoms that appear to be independently associated with neuropsychological and psychosocial outcomes in community and ADHD-specific samples of youth. However, our understanding of CDS in individuals with chronic or complex medical conditions is limited. The current systematic review is the first to summarize the literature on CDS prevalence rates and associations with neurocognitive and functional outcomes in youth with medical conditions, and to discuss areas of future research to guide clinical intervention.

Methods:

We conducted literature searches across four major databases and included studies assessing prevalence estimates, associations with neuropsychological and/or psychosocial functioning, or predictors of CDS in individuals with chronic or complex medical conditions.

Results:

Twenty-five studies were identified and retained. Fifteen of sixteen studies reported elevations in CDS symptoms, though findings were mixed in studies comparing mean differences to typically developing youth. Seven studies provided inconsistent evidence for CDS being associated with neuropsychological or academic functioning, with six studies demonstrating consistent effects on psychosocial functioning. Finally, nine studies identified biological and sociodemographic factors associated with CDS, though almost all await replication.

Conclusions:

CDS symptoms are significantly elevated in youth with medical conditions and appear to be particularly linked with psychosocial functioning. Future research is needed to identify prevalence of CDS across a range of medical conditions, examine associations with neuropsychological and psychosocial functioning, and examine whether CDS impacts self-management.

The past two decades have witnessed rapid growth in the study of cognitive disengagement syndrome (CDS, previously referred to as sluggish cognitive tempo), a set of symptoms characterized by excessive mind-wandering, zoning or spacing out, mental confusion and fogginess, hypoactivity, and drowsiness (for a review, see [1]). Originally, the study of CDS emerged from studies assessing the dimensionality of attention-deficit disorder (with or without hyperactivity), with initial support for a “slow tempo” or “sluggish tempo” factor (e.g., “seems to be in a world of his or her own,” “appears to be sluggish,” “seems drowsy”) on broadband rating scales (e.g., Children’s Behavior Rating Scale [CBRS], followed by the more frequently used Child Behavior Checklist [CBCL]) (for a history see [2]). Since the initial identification of this construct, studies have increasingly found CDS symptoms to be distinct from attention-deficit/hyperactivity disorder inattentive symptoms (ADHD-IN) [3]. Specifically, in a meta-analysis of more than 19,000 individuals, 16 items¹ assessing daydreaming/mind-wandering, mental confusion/fogginess, and hypoactive/sleepy behaviors provided strong discriminative validity from ADHD-IN symptoms [4]. Findings from this review initiated the development of specific rating scales (e.g., Child and Adolescent Behavior Inventory [CABI]), which demonstrated strong structural validity, reliability, and independence from ADHD-IN symptoms in community and clinical samples [3]. In response to a proliferation of research and clinical attention to this construct, an International Work Group convened and recommended a change in terminology from “sluggish cognitive tempo” to “cognitive disengagement syndrome” given the offensive nature of the term “sluggish,” families reporting negative responses to the label SCT, and to better reflect the nature of this construct [1].

The next wave of CDS research sought to determine relations with other constructs, such as anxiety, depression, or daytime sleepiness. Factor analytic studies demonstrated that CDS symptoms in children and adolescents are empirically separate from anxiety, depression, and daytime sleepiness [1]. Although separate constructs, numerous studies in community and ADHD-specific samples find CDS symptoms to remain associated with increased anxiety and depressive symptoms, social withdrawal and loneliness, suicide risk, sleep problems, and, to a less clear extent, academic problems and executive function deficits, above and beyond ADHD-IN symptoms (see [1]). Further, CDS longitudinally predicts increases in depressive symptoms [5, 6]. Although CDS is not currently recognized as a distinct mental health disorder in the DSM-5, and the optimal to way to conceptualize CDS remains unclear [1], research clearly supports CDS as a meaningful construct for understanding heterogeneous symptom presentations, co-occurring mental health symptoms, and functional impairments.

More recently, there has been an emphasis on understanding the nature and correlates of CDS beyond ADHD-specific samples. Due to symptoms comprising CDS (e.g., mental confusion, attention difficulties, and hypoactivity), which seemingly overlap with the neurocognitive sequalae of chronic or complex medical conditions [7], there is increasing interest in studying CDS as a potential marker of neurocognitive effects or as a mechanism of psychosocial risk in youth with chronic or complex medical conditions [8, 9]. Assessing CDS alongside ADHD-IN is critical given research showing CDS to be strongly associated with psychosocial risk, linked to attention and executive functioning deficits that may impact self-management of disease, and to be perhaps less responsive to traditional ADHD treatments.

Specifically, studies in samples of youth with medical conditions document substantial attention problems, internalizing symptoms, and peer difficulties [10, 11], which are specific correlates of CDS [1]. Thus, CDS symptoms may pose heightened risk for psychosocial difficulties following diagnosis and treatment recovery. Further, the negative impact of attention problems and executive function deficits on self-management and treatment adherence is well-documented [12, 13]. Given research linking CDS to deficits in sustained attention, processing speed, working memory, planning/organization, and greater task-unrelated thought (e.g., mind-wandering), even after controlling for ADHD-IN [14-18], youth with medical conditions who display CDS may experience challenges attending and managing the rapid influx of medical information and health-related responsibilities. In addition, youth with CDS may not respond equally to stimulant medication and psychosocial treatments, such as behavioral parent training and homework and organizational skills training designed for youth with ADHD [19-22]. As medication and behavioral management of ADHD is recommended for youth with chronic health conditions [23], assessing CDS symptoms may be useful to understand the heterogeneous nature of attention problems and implications for intervention.

Finally, there has been growing interest in understanding the etiology of the CDS phenotype, especially given somewhat less heritability and greater environmental influence on CDS compared to ADHD-I [24, 25]. Similar to the equifinality of etiological causes of ADHD, which include genetic underpinnings, prenatal complications, and environmental risk/injury (e.g., secondary ADHD from TBI) [25, 26], biological and medical factors may contribute to CDS. Indeed, investigators seeking to conceptualize the neurocognitive effects or consequences (e.g., fatigue, inattention, slower processing speed) emerging following diagnosis and treatment of medical conditions have called for alternative frameworks of attention dysfunction [7]. Thus, investigation of biological, medical, and/or treatment-related factors impacting the CDS construct may inform conceptualization and early prevention.

Taken together, CDS represents a distinct set of attentional problems and behaviors that may contribute to neuropsychological and psychosocial functioning, self-management of disease, and treatment response in youth with chronic or complex medical conditions. However, without a clear understanding of the current evidence for CDS in pediatric samples, recommendations for practice cannot advance. A topical review in 2013 highlighted the potential relevance of CDS as a construct that might explain attentional, psychosocial, and neuropsychological functioning in pediatric samples [27]. That review discussed four studies with preliminary evidence for elevated CDS in medical populations (e.g., epilepsy, pediatric cancer) or associated with environmental risk and injury (e.g., prenatal alcohol exposure, thyroid dysfunction). However, a systematic review was not conducted that systematically identifies and critically investigates the state of the evidence, identifies themes across studies, and highlights gaps in the extant literature in need of further investigation. The current systematic review sought to summarize existing evidence for whether CDS is 1) elevated in medical populations as compared to individuals without medical conditions, 2) associated with neuropsychological, psychosocial, and/or treatment outcomes in samples of individuals with chronic or complex medical conditions, and 3) related to early environmental risk, biological factors, and/or medical/treatment-related factors.

Methods

A systematic search was conducted to identify all relevant studies of CDS in samples of individuals with early risk or medical conditions, injury, or illnesses. Specifically, we identified studies examining CDS prevalence rates and/or associations between CDS and psychosocial/treatment outcomes. Although a SCT Work Group has recently recommended using the term CDS rather than SCT [1], our search terms focused on “sluggish cognitive tempo” as the CDS terminology was not yet published when the literature search for this systematic review was conducted. In accordance with PRISMA standards, computer searches were conducted in May 2022 in the MEDLINE (PubMed), Embase, PsychINFO, and Web of Science databases. Primary search terms included “sluggish cognitive tempo,” “slow tempo,” chronic illness OR chronic disease OR illness OR disease OR asthma OR cystic fibrosis OR sickle cell disease OR inflammatory bowel disease OR obesity OR cancer OR leukemia OR juvenile rheumatoid arthritis OR cardiovascular OR diabetes OR epilepsy OR genetic disorder OR HIV OR NICU Or neonatology OR prematurity OR intensive care OR transplant OR pain OR headache. References for included studies were examined to identify any additional relevant studies. No constraints were placed on dates of publication or inclusion criteria at this stage. Grey literature (e.g., listservs, unpublished manuscripts) were not included. An updated search was conducted in August 2022 prior to manuscript submission. A final search was conducted in July 2023 prior to submitting a revised manuscript, using both the SCT and CDS terms. The systematic review was pre-registered on PROSPERO (CRD42022328743).

After duplicates were removed, two authors independently screened the titles and abstracts for the following inclusion criteria: publication in a peer-reviewed journal; empirical study (not a review, commentary, or conference poster abstract); published in English; included a measure of CDS; sample included individuals with medical conditions, illness, injury, disease, or early environmental risk; and examined either CDS prevalence or associations between CDS and neuropsychological, psychosocial, or treatment outcomes. Given the focus of the review, studies containing samples of individuals with neurodevelopmental conditions (e.g., ADHD, autism spectrum disorder) without comorbid medical conditions, injuries, illness, or early environmental risk were excluded. Next, each of the remaining full texts were screened by the same two authors to determine inclusion, per the criteria above.

A data extraction form (available from the corresponding author) was used to collect information regarding study characteristics and key findings, with the first author coding all eligible articles and two co-authors providing double coding. Any disagreements in data extraction were addressed via discussion. Finally, a study quality rating form was developed, based on the Methods Guide for Effectiveness and Comparative Effective Reviews developed by the Agency for Healthcare Research and Quality [28] as well as the Johns Hopkins Nursing Evidence-Based Practice Evidence Level and Quality Guide [29].The study quality domains were: 1) having a priori aims/hypotheses related to CDS, 2) providing a justification of sample size for CDS-related analyses, 3) sample representativeness for intended study population, 4) quality of CDS measure, 5) quality of correlates/outcomes measures, 6) statistical control of ADHD-IN symptoms or including a separate CDS group, 7) rates of missing CDS data, 8) quality of inferences and conclusions drawn, 9) type of study, and 10) level of control related to normative or other clinical groups. See the caption of Figure 1 for the specific codes used for each of the study quality domains.

Figure 1. Summary of Quality Assessment for Included Studies.

Note. 1. A priori aim/hypothesis: whether there were aims/hypotheses specific to cognitive disengagement syndrome (CDS); “++” indicates specific aims/hypotheses related to CDS; “+” indicates CDS implied but not explicitly stated, e.g., difference in mental health; “−” indicates no aim/hypothesis related to CDS. 2. Sample size justification: whether sample size was justified; “++” indicates justified, “+” indicates justified based on outcomes other than CDS, “−” indicates unjustified. 3. Sample representativeness: representativeness of samples for the intended study population and for conclusions drawn, with “++”, “+”, and “−” indicating “good”, “fair”, and “poor”. 4. Quality of CDS measure: “++” indicates measure specifically designed for assessing CDS, “+” indicates measure not designed to comprehensively assess CDS. 5. Quality of correlates measured: “++” indicates well validated, “+” indicates not well validated. 6. Control for attention-deficit/hyperactivity disorder inattentive (ADHD-IN) symptoms or separate CDS group: “++” indicates study included a CDS group or statistical control of ADHD-IN symptoms, “+” indicates no identified CDS group or control for ADHD-IN symptoms. 7. Rates of missing in daily data: “++” indicates ≤10%, “+” indicates ≤20% and >10%, “−” indicates >20%, and “?” indicates not reported. 8. Inferences and conclusions: quality of inferences and conclusions drawn, with “++”, “+”, and “−” indicating “good”, “fair”, and “poor”. 9. Type of study: “++” indicates randomized controlled trial, “+” indicates quasi-experimental study, “−” indicates non-experimental study, “?” indicates qualitative study. 10. Level of control: “++” indicates adequate control (e.g., typically developing comparison group), “+” indicates some control (e.g., clinical comparison group), “−” indicates no control (e.g., single group clinical cases only).

Results

Article Search and Identification

Appendix 1 shows the PRISMA flow diagram. The search identified 220 unique records for title and abstract screening, with two studies identified in reference sections of retained studies. Of these 222 records, percent agreement was calculated with the two independent coders agreeing on 98.6% of records, with three records having disagreement. After discussion, two of these records proceeded to full-text screen, for a total of 26 records advanced to full-text screening. Of these, 21 met full criteria for data extraction and were retained for review. The most common reason studies were excluded was 1) they did not examine prevalence rates of CDS or associations of CDS with neuropsychological, psychosocial, or treatment outcomes or 2) did not explore relations between CDS and medical diagnoses (n = 4). One study with unclear CDS measurement was excluded after study authors were contacted and additional information could not be obtained. The updated literature search conducted in August 2022 identified two additional studies meeting eligibility criteria for inclusion. Further, a final search was conducted prior to submitting a revised manuscript to identify studies meeting criteria from August 2022 to July 2023, with an additional two studies meeting eligibility criteria. Details of the 25 included studies are provided in Table 1. Consistent with the aims of the systematic review, we present results for elevations of CDS, bivariate correlations of CDS with outcomes, group-based differences, and results from multiple regression analyses.

Table 1.

Details of the 25 Retained Studies

Article	Country	Study Design	Medical Condition, Illness, Injury, Disease, or Risk	N Age (y), range	% Male	Race/Ethnicity	Rater (s)	CDS Measure	Psychosocial or Treatment Outcome Measure	Mean-Difference and/or Bivariate Results	Unique (Multivariate) Associations
Barkley 2013	USA	Cross-Sectional	Parent report of neurological diagnoses in nationally-representative sample of children and adolescents	CDS-only: 41; M=13.2 ADHD-only: 95; M=11.4 CDS+ADHD: 61; M=12 TD: 1603; M=11.4	CDS-only: 46% ADHD-only: 64% CDS+ ADHD 66% TD: 46%	Entire Sample White: 74% Remaining 26% NR Specific race/ethnicity for each group NR "ethnic representation of the groups did not differ significantly"	PR	SCT-S	NA	None of the CDS-only group had reported diagnoses of deaf/heard of hearing. 2.3% of the CDS-only group had reported diagnoses of blind/very visually impaired. The CDS-only group had higher reported diagnosis of blind/very visually impaired compared to the comparison group. The CDS-only group did not differ from the ADHD-only or CDS + ADHD groups.	NR
Becker 2016	USA	Cross-sectional	Children referred to a pulmonary-based, Sleep Disorders Center	325 M=7.93; 6-10	62%	Non-Hispanic White: 79% Black: 11% Multiracial/Other: 6% Hispanic: 2% Asian/Asian American: 2%	PR	CBCL	ICSD-2 sleep diagnoses; CSHQ; SDIS-C; PSI-SF.	CDS was significantly correlated with less sleep duration (r=.12), greater symptoms of parasomnia (r=.21), sleep disordered breathing (r=.18), daytime sleepiness (r=.18), obstructive sleep apnea (r=.36), periodic limb movement (r=.39), delayed sleep phase (r=.12), and excessive daytime sleepiness (r=.52).	Controlling age, sex, race sleep problem severity, ADHD, anxiety, depression, and oppositional symptoms, CDS was uniquely associated with: ↑total sleep problem severity, organic-sleep problems, and daytime sleepiness. ↑ math, science, and history/social studies impairment. ↑parent-child dysfunctional interactions. Controlling for sleep problem severity: CDS was uniquely associated with ↑ depression and ADHD symptoms.
Becker 2017	USA	Cross-sectional	Psychiatrically hospitalized children	570 M=9.16; 6–12	73%	Black: 58% White: 39% Biracial: .02% Hispanic: .01% Other: .01%	PR	CBCL	TSH	Thyroid stimulating hormone was significantly correlated with CDS (r=.09)	Controlling for child age, sex, race, internalizing, externalizing, and ADHD symptoms: TSH was uniquely associated with ↑ CDS symptoms.
Birdal 2016	Turkey	Cross-sectional	Child and adolescent psychiatry outpatients diagnosed with enuresis	Children with Enuresis: 30 Control: 30 7-11	NR	NR	PR	CBCL	NR	No significant differences in CDS between children with enuresis and controls.	NR
Camprodon-Rosanas 2017	Spain	Cross-sectional	Parents/legal guardians reported on socio-demographic variables in heterogeneous sample of school-aged children (e.g., 7% to 12% were premature, had low birth weight, prenatal smoke exposure, second-hand smoke, ADHD diagnosis).	183 7-10	53%	NR	PR, TR	CBCL	NA	Children with parent-reported maternal smoking during pregnancy and second-hand smoke exposure in the home had statistically significant greater CDS symptoms.	Controlling for sex, maternal education level, paternal occupation, socioeconomic vulnerability, maternal smoking during pregnancy, second-hand smoke exposure, ADHD diagnosis, emotional problems, hyperactivity/impulsivity symptoms, peer relations problems, inattention symptoms, and dyslexia symptoms: ↑ risk of CDS symptoms in children with second-hand exposure to smoke.
Darow 2023	USA	Longitudinal (8 years)	Spina Bifida	SB: 68 M=8.34 TD: 68 M=8.49	SB: 54% TD: 54%	SB: White: 80.3% Hispanic/Latino: 6.1% Black: 4.5% Other: 4.5% Asian: 1.5% Indigenous: 1.5% Arab/Middle Eastern: 1.5% TD: White: 91.2% Asian: 4.4% Black: 1.5% Arab/Middle Eastern: 1.5% Other: 1.5%	MR, FR, TR, SR	CBCL	SPPC; SSRS; SDMQ	For youth with SB, MR CDS endorsement of symptoms ranged from 14.6% to 35.6%, while TR CDS endorsement of symptoms ranged from 28% to 47.6% For TD youth, MR CDS endorsement of symptoms ranged from 20% to 41.4%, while TR CDS endorsement of symptoms ranged from 42.4% to 62.9% At baseline, the TD youth had higher MR CDS compared to youth with SB, whereas TR CDS was higher for youth with SB.	For youth with SB: Higher growth of MR CDS predicated worse MR social skills. TR CDS at baseline predicted worse FR and TR social skills. Higher growth of MR CDS predicted lower SR behavioral autonomy. For TD youth: Higher growth of TR CDS predicted worse TR social skills. TR CDS at baseline predicted worse FR and SR social skills.
East 2018 a	Chile	Longitudinal (15 years)	Iron deficiency	1,116 (6-month old infants) M=14.6; 11-18	54%	NR	PR	CBCL	Hemoglobin levels	Infants with ID-only did not have higher Age 10 CDS (r=.06). Infants with ID anemic had higher Age 10 CDS (r=.11).	Controlling for child sex, family socioeconomic status, mothers' education level, iron supplementation status, and breastfeeding status: Infants who were Iron Deficient Anemic, compared to infants with iron supplementation, had ↑ CDS at Age 10.
East 2021 a	Chile	Longitudinal (15 years)	Iron deficiency	959 (6-month old infants, follow-up at 5, 10, and 15 years)	50%	NR	PR	CABIb (age 5), CBCL (age 10 and 16)	Hemoglobin levels	Infant ID severity was significantly correlated with greater Age 5 CDS (r=.08), Age 10 CDS (r=.10), and Age 15 CDS (r=.09).	Controlling for child age, sex, family socioeconomic status, family stress, home support, maternal education, infant temperament, infant activity level, iron supplementation, and iron status: Infant ID severity was associated with ↑ CDS at age, 5, 10, and 16.
Georgsdottir 2013	Iceland	Cross-sectional	Teenagers who had extremely low birth weight (ELBW) as infants	Teenagers with ELBW: 30 M=16.8; 14-19 Comparison: 30 M=16.5; 14-18	17%	NR	PR	CBCL	NA	Teenagers with ELBW had higher CDS compared to teenagers who were not extremely low birth weight.	NA
Graham 2013	USA	Cross-sectional	Children with and without histories of heavy prenatal alcohol exposure.	272 M=12.18; 8-16 Exposure+ADHD: 75 Exposure-only: 35 ADHD-only: 60 TD: 102	Exposure+ADHD: 63% Exposure-only: 43% ADHD-only: 72% TD: 56%	Exposure+ADHD White: 68% Hispanic: 8% Exposure-only White: 37% Hispanic: 11% ADHD-only White: 60% Hispanic: 22% TD White: 72% Hispanic: 19%	PR	SCT-S	CBCL	Youth with prenatal alcohol exposure + ADHD had higher CDS compared to youth with prenatal alcohol exposure −ADHD. Youth with prenatal alcohol exposure but −ADHD and youth with ADHD -alcohol exposure had higher CDS compared to typically developing youth. Prenatal alcohol exposure (− ADHD) was associated with elevation on CDS. In prenatal alcohol exposure + ADHD group, CDS was correlated to greater internalizing symptoms (r=.35), externalizing symptoms (r=.39), attention problems (r=.60), and higher full square IQ (r=.32). In prenatal alcohol exposure - ADHD group, CDS was correlated to greater internalizing symptoms (r=.41), externalizing symptoms (r=.43), and attention problems (r=.42).	CDS items distinguished the prenatal alcohol exposure group - ADHD from the typical developing group. CDS items distinguished the prenatal alcohol exposure and ADHD groups from the prenatal alcohol exposure + ADHD group.
Hardy 2022	USA	Cross-sectional	Children and Adolescents with Sickle Cell Disease participating in a clinical trial.	85 M=10.45; 7-16	42%		PR	K-SCT	Conners 3, BRIEF	CDS symptoms were significantly correlated with PR ADHD-IN (r=.64), ADHD-H/I (r=.46), learning problems (r=.69), executive function problems (r=.55), defiance/aggression (r=.23), and peer problems (r=.23), BRIEF metacognition (r=.66), and BRIEF behavioral regulation index (r=.47). 25% of children and adolescents had significantly elevated CDS. 71% of children and adolescents with elevated CDS, regardless of ADHD-I symptoms, had significantly elevated learning problems.	Controlling for age, genotype, stroke, parent education, family income, and ADHD symptoms: CDS was associated with ↑ learning problems and metacognition.
Loutfi 2011	Brazil	Cross-sectional	Children with idiopathic epilepsy treated in a neuropediatric ambulatory clinic.	30 M=9; 6-13	“Proportion of girls and boys was similar”	NR	PR, TR	CBCL	CBCL	In children with ADHD, 68.7% had borderline or clinical elevations on CDS vs. 14.2% in the non-ADHD group.	NA
Mahdavi 2021	USA	Cross-sectional	TBI (mild, moderate, severe) within 1 year post-injury	50 M=13.42; 6-16	58%	White: 74% Latino: 12% Other: 8% Asian: 4% Black: 2%	PR	CBCL	WISC-IV	24% of children with TBI had clinically significant elevation on CDS. Youth with mild and moderate-severe TBI did not have significant differences on CDS. CDS was correlated with higher internalizing (r=.50) and externalizing problems (r=.37). CDS was not correlated to verbal comprehension (r=.01), perceptual reasoning (r=−.12), working memory (r=−.16), or processing speed (r=−.12). Time since injury was not correlated to CDS (r=−.05).	NA
Moreira 2014	Brazil	Cross-sectional	Idiopathic epilepsy	Idiopathic epilepsy: 45 Controls: 65 6-16	NR	NR	PR	CBCL	WISC-III; APT	Youth without idiopathic epilepsy had greater CDS compared to youth with idiopathic epilepsy. No difference on CDS for youth with focal seizure and generalized seizure or for youth on and off epileptic medication. In youth with epilepsy, CDS was associated with higher IQ scores (r=.27). CDS was not related to academic performance.	NA
Olstohoorn 2022	USA	Cross-sectional	Pediatric BT survivors (at least 3 months post-treatment)	83 M=12.48; 6-18	42%	White: 58% Black: 7% Hispanic/Latino: 27% Asian: 6% Hawaiian: 1% Unknown: 1%	PR	CBCL	CBCL	30% of pediatric BT survivors had elevated CDS symptoms. Disease burden was not correlated with CDS symptoms.	Controlling for disease burden (e.g., tumor-related complications, surgical events, radiation therapy, chemotherapy): Sleep disturbances was associated with ↑ CDS symptoms.
Peng 2021	Hong Kong	Cross-sectional	Pediatric and adult survivors of ALL (average 17 years post-diagnosis)	152 M=23	52%	NR	PR, SR	CBCL, ABCL	Chronic Health Conditions: Obtained from CMS; Family functioning: CFAI; Physical activity: CUHK-PARCY; Fatigue: Peds QL MFS;	23.7% of ALL survivors had elevated CDS symptoms.	Controlling for sex, age at diagnosis, age at evaluation and cranial radiation: ALL survivors with chronic health conditions, poorer family functioning, less housing space, greater physical inactivity, more fatigue, and more academic stress had ↑ CDS. Chronic health conditions were indirectly associated with ↑ CDS via fatigue.
Peterson 2021	USA	Cross-sectional	Pediatric BT survivors (average 3 years post-diagnosis)	58 M=11.50; 4-21	40%	White: 67% Black: 19% Did not report: 10% Asian: 2% Hispanic: 2%	PR	SCT-S	NEPSY-II Inhibition subtest or DKEFS Color Word Interference subtest; Coding subtest from WISC-V or WAIS-IV; Verbal Comprehension Index from WISC-V or WAIS-IV	CDS-daydreamy (r=−.31), low initiation (r=−.30), and CDS-total (r=−.26) were each correlated with lower verbal ability. CDS symptoms were unrelated to tumor location and type of treatment (e.g., resection, radiation, chemotherapy). CDS-daydreamy (r=−.33), low initiation (r=−.30) correlated with age at diagnosis. CDS-initiation (r=.39) and CDS-total (r=−34) were significantly correlated with longer time since diagnosis.	In analyses controlling for verbal ability and ADHD-IN symptoms: CDS-daydreamy was uniquely associated with ↓ speeded inhibition performance. CDS-sleepy/sluggish and low initiation were unrelated to processing speed outcomes.
Peterson 2022	USA	Cross-sectional	Pediatric BT and ALL (average 5 to 6 years post-diagnoses)	ADHD-IN group: 149 M=12.6 BT group: 146 M=12.8 ALL group: 69 M=11.5	ADHD-IN group: 51% BT group: 74% ALL group: 36%	ADHD-IN group: White: 53% Black: 33% Hispanic: 5% Other/multi-racial: 9% BT group: White: 70% Black: 14% Hispanic: 5% Other/multi-racial: 12 ALL group: White: 75% Black: 12% Hispanic: 3% Other/multi-racial: 10%	PR	SCT-S	NA	Youth with ADHD-IN, BT, and ALL had higher CDS than the normative mean. The ALL group did not differ on CDS-daydreamy compared to the normative mean. The ADHD-IN group had higher CDS-total, sleepy/sluggish, low initiation, and daydreamy compared to the ALL and BT group. The ALL and BT group did not differ on any CDS scales except for the CDS-sleepy/sluggish score being higher in the BT than ALL group. In both the ALL and BT groups, the CDS-low initiation scale was higher than the other two CDS scales. 93% of individuals in the low CDS group had an oncology diagnosis. 53% of individuals in the medium CDS group had an ADHD-IN diagnosis, whereas 70% of individuals in the high CDS group had an ADHD-IN diagnosis.	In adjusted logistic regression model of all participants: Individuals with ALL and BT, compared to those with ADHD-IN, had significantly ↑ odds of being in the low or medium CDS class. When controlling for age, age at diagnosis and history of chemotherapy, surgery, and/or radiation, no variables were associated with CDS class membership.
Reeves 2007	USA	Cross-sectional	Pediatric survivors of ALL (1-year post-treatment)	Patient group: 80 M=12.4; 6-18 Sibling control: 19 M=12.6; 6-19	Patient: 54% Control: 53%	Patient: White: 93% Black: 6% Hispanic: 1% Control: White: 79% Other: 16% Black: 5%	PR	CBCL	CBCL; WISC-III; WIAT	Children with ALL had higher CDS compared to siblings. In the ALL group, CDS correlated with poorer WISC-III IQ (r=−.30), WIAT basic reading (r=−.26), numerical operations (r=−.29), reading comprehension (r=−.27), and math reasoning (r=−.36). CDS was unrelated to spelling.	Controlling for group status (patients vs. siblings): CDS uniquely associated with ↓ basic reading scores.
Roberts 2014	Australia	Cross-sectional	Survivors of pediatric cancer (at least 5 years post treatment)	70 M=18.25; 7-36	54%	NR	PR, SR	CBCL; YSR; ASR	NA	PR (CBCL) and SR (YSR) CDS were not different from normative mean scores. No differences were found on specific items, though mean score "confused or in a fog" were marginally significant.	NA
Saez 2019	Spain	Cross-sectional	Community/school (cross-sectional)	2,142 M=10.30; 8-13	51%	White: 90% North African: 10% (estimated based on school demographics)	FR, MR, TR	CABI	NA	NR	Controlling for child sex and ADHD-IN: CDS factor was uniquely associated with ↑ mother and father-reported history of child medical problems (asthma most common problem listed).
Simpson 2022	USA	Cross-sectional	Spina Bifida	169 M=11; 5-19 Youth with SB compared to previous unselected and clinical samples reporting descriptives statistics on the SCT-S scale	49.7%	White: 71% Black: 1% Asian: 8% Multiracial: 4% Other: 15% Not reported: 2%	PR	SCT-S	RCADS, VPAS	PR CDS was correlated with PR ADHD-I (r=.58), SR depression (r=.47), and SR anxiety (r=.34). Patient diagnosis of myelomeningocele and shunting was correlated with greater CDS (r=.21, .18). Lesion level and ambulation status was unrelated. 18% of sample were elevated in CDS, with 39% of patients with elevated CDS not meeting criteria for ADHD Inattentive Presentation. Full sample had higher CDS compared to Penny (2009) sample of school-aged children, Leopold et al. (2016) community sample of twins, and Barkley (2013) nationally representative sample of children and adolescents. Conversely, Full sample had lower CDS compared to Smith et al. (2018) sample of adolescents diagnosed with ADHD and Jacobson et al. (2018) sample of children and adolescents with psychiatric and medical diagnosis seen for neuropsychological evaluations. Myelomeningocele was correlated with higher CDS	N/A
Smith 2022	USA	Prospective (8 years)	Spina Bifida	140 M=11.40; 8-15	46%	White: 60% Hispanic/Latinx: 23% African American/Black: 12% Other: 5%	MR, FR, TR	CBCL	IQ: Full Scale IQ on the WASI; Attention: SNAP-IV, CAS Number Detection, TEA-Ch subtests; Working Memory: BRIEF, WISC-IV Digit Span Backwards; Cognitive Flexibility: BRIEF Shift subscale, DKEFS Verbal Fluency Test; Planning and Organization: BRIEF Plan/Organize and Organization of Materials, CAS Planned Connections.	Elevations in CDS: TR 9%, MR 8.3%, and FR 5.3%. TR CDS was correlated with lower attention (r=−.45), working memory (r=−.57), cognitive flexibility (r=−.32), and planning/organization (r=−.42). MR CDS was correlated with lower attention (r=−.33), working memory (r=−.40), cognitive flexibility (r=−.31), and planning/organization (r=−.44). FR CDS was correlated with lower attention (r=−.33), working memory (r=−.49), cognitive flexibility (r=−.28), and planning/organization (r=−.45).	Controlling for age and shunt status: Lower attention was associated with ↑ MR, FR, and TR CDS. Controlling for age, shunt status, and ADHD-IN symptoms: Lower working memory, cognitive flexibility, and planning/organization was associated with ↑ MR, FR, and TR CDS. Lower working memory predicted ↑ MR CDS over time. Lower cognitive flexibility predicted ↑ TR CDS over time.
Tsang 2017	Australia	Cross-sectional	FASD and Non-FASD	FASD: 21; M=8.7 Non-FASD: 87; M=8.7	FASD: 62% Non-FASD: 51%	NR	PR, TR	CBCL, TRF	NA	6% of youth with FASD had clinically significant PR CDS, whereas 6% of youth without FASD had clinically significant PR CDS. No differences on PR CDS for youth with and without FASD. 29% of youth with FASD had clinically significant TR CDS whereas 5% of youth without FASD had clinically significant TR CDS. Youth with FASD had greater TR CDS compared to youth without FASD.	After adjusting for false discovery rates and controlling for demographic factors: Youth with FASD had ↑TR CDS T-scores compared to youth without FASD.
Willard 2013	USA	Cross-sectional	Survivors of pediatric BT and ALL (around 4 year post treatment)	BT Survivors: 65; M=12 ALL Survivors: 25; M=12 Community Controls: 50; M=10.1	BT Survivors: 51% ALL Survivors: 64% Community Controls: 42%	BT Survivors: White: 89% Black: 11% ALL Survivors: White: 56% Black: 12% Biracial: 16% Other: 16% Community Controls: White: 72% Black: 16% Biracial: 10% Other: 2%	PR	CBCL	WISC-V or WAIS-IV	BT survivors had greater CDS T-scores compared to ALL survivors and community controls. No difference on CDS T-scores for ALL survivors and community controls. In BT and ALL survivors, CDS was correlated with greater attention problems (r=.71), lower working memory (r=−.28), and total IQ (r=−.22) scores, but not processing speed.	In BT survivors, shunt placement predicted CDS symptoms. Tumor location, radiation field, age at radiation, age at diagnosis, and time off-treatment were unrelated to CDS.

Note. ABC = Adult Behavior Checklist; APT = Academic Performance Test; BFIS = Barkley Functional Impairment Scale; BFIS-CA = Barkley Functional Impairment Scale-Children and Adolescents; BRIEF = Behavior Rating Inventory of Executive Functioning; CABCL = Child and Adolescent Behavior Checklist; CABI = Children's Adaptive Behavior Inventory; CAS = Cognitive Assessment System; CBCL = Child Behavior Checklist; CDS = Cognitive Disengagement Syndrome (previously referred to as sluggish cognitive tempo [SCT]); CHIP-AE = Child Health and Illness Profile-Adolescent Edition; CFAI = Chinese Family Assessment Instrument; CMS = Clinical Management System; CUHK-PARCY = Chinese University of Hong Kong: Physical Activity Rating for Children and Youth; CR = Caregiver-report; CSHQ = Children’s Sleep Habits Questionnaire; D-KEFS = Delis-Kaplan Executive Function System; ESSA = Education Stress Scale for Adolescents; FASD = Fetal Alcohol Syndrome Disorder; FR = Father-report IRS = Impairment Rating Scale; MR = Mother-report; NA = Not Applicable (if not primary purpose of study); NR = Not Reported; NEPSY-II = A Developmental Neuropsychological Assessment, 2^nd edition; PBS = Parent Behavior Scale; PedsQl MFS = Pediatric Quality of Life Inventory Multidimensional Fatigue Scale; PSI-SF = Parenting Stress Index, Short Form; SDIS-C = Sleep Disorders Inventory for Students, Child Version; PR = Parent-report; SCT= Sluggish cognitive tempo; SCT-S = Sluggish Cognitive Tempo Scale; SDMQ = Steinberg Decision-Making Questionnaire; SNAP-IV =Swanson, Nolan, and Pelham Teacher and Parent Rating Scale; SPPC = Self-Perception Profile for Children; SR = Self-report; SSRS = Social Skills Rating System; TEA-Ch = Test of Everyday Attention for Children; TR = Teacher-report; TRF = Teacher Report Form; TSH = Thyroid Stimulating Hormone; WASI = Wechsler Abbreviated Scale of Intelligence; WIAT = Wechsler Individual Achievement Test-Third Edition; WISC-R = Wechsler Intelligence Scale for Children-revised version; WISC-III = Wechsler Intelligence Scale for Children, 3^rd Edition; WISC-IV = Wechsler Intelligence Scale for Children, 4^th Edition; WJ-III = Woodcock Johnson III Tests of Achievement, Third Edition; WPPSI- R= Wechsler Preschool and Primary Scale of Intelligence; WRAT = Wide Range Achievement Test; YSR = Youth Self-Report.

^{a =}same sample used in East et al., 2018, 2021.

^{b =}CDS items from the CABI demonstrated poor internal consistency

Sample Characteristics

First, to better understand the state of the literature, we provide a brief description of sample characteristics of retained studies providing data on CDS. The following list are the number and percentage of studies with specific disease groups: childhood survivors of cancer (n = 5, 20%) [30-34], youth with spina bifida (n = 3, 12%) [8, 35, 36], children with idiopathic epilepsy (n = 2, 8%) [37, 38], children exposed to alcohol prenatally (n = 2, 8%) [39, 40], children with enuresis (n = 1, 4%) [41], youth with traumatic brain injury (TBI; n = 1, 4%) [42], pediatric patients with sickle cell disease (SCD; n = 1, 4%) [43], and teenagers who had extremely low birth weight (n = 1, 4%) [44]. Of these, two studies included samples of both pediatric and adult survivors of childhood cancer [30, 31]. Further, two studies (8%) assessed relations between CDS and parent-reported rates of child medical diagnosis [45, 46], specifically rates of blindness/visual impairment [46]. Although the second study noted that asthma was the most commonly endorsed medical problem associated with elevated CDS symptoms [45], no other information was provided on specific types of medical problems.

Twelve studies examined biological, treatment-related, and/or socio-demographic factors correlates or predictors of CDS symptoms: medical treatment-related factors (n = 5, 20%) [9, 33-35, 42], iron deficiency (n = 2, 8%) [47, 48], prenatal risk (n = 1, 4%) [49], sociodemographic characteristics (n = 1, 4%) [31], thyroid dysfunction (n = 1, 4%) [50], neuropsychological functioning (n = 1, 4%) [8], and sleep disturbances (n = 1, 4%) [51].

Finally, the majority of studies (n = 17) used the SCT (CDS) scale on the CBCL, with five studies using the Sluggish Cognitive Tempo Scale (SCT-S) [9, 34, 35, 39, 46], one study using the Kiddie-SCT scale (K-SCT) [43], one study using the SCT (CDS) scale on the Youth Self Report measure [30], and one using the CABI [45].

Prevalence and Mean Differences on CDS

Sixteen studies evaluated prevalence estimates of CDS and/or differences in rates of CDS symptoms among youth with chronic or complex medical conditions. Overall, 15 of the 16 studies documented increased prevalence rates or significant elevations in rates of CDS symptoms compared to normative scores, with only one study reporting no significant differences between survivors of childhood cancer and normative means [30]. The prevalence rates of CDS were examined according to elevations on the SCT (CDS) scale on the CBCL, with the exception of three studies [34, 35, 39], testing mean differences on the SCT-S and one study testing elevations on the parent-report K-SCT [43]. Results varied across specific conditions; specifically, assessed on the CBCL as either scores above the borderline range (T-scores ≥ 65) or above 1.5 standard deviation units from the mean, the highest prevalence rates were found among pediatric brain tumor survivors (30%) [51], children with TBI (24%) [42], and pediatric/adult survivors of acute lymphoblastic leukemia (ALL, 23.7%) [31], whereas the lowest prevalence rates were observed in youth with spina bifida (5-9% for mother, father, and teacher report) [8]. In a sample of 20 children with fetal alcohol syndrome, 6% had elevations on parent ratings of CDS whereas 29% were elevated according to teacher report [40]. Finally, Loutfi and colleagues [37] found substantial differences in parent-reported elevations in CDS among children with idiopathic epilepsy with and without ADHD; specifically, 68% of youth with epilepsy and ADHD had elevated CDS and 29% of youth with epilepsy without ADHD had elevated CDS. Conversely, in a sample of pediatric survivors of childhood cancer, Roberts and colleagues [30] did not observe significant elevations in either parent or self-reported CDS symptoms, though the small sample size (n=25 for CBCL and n=17 for Youth Self-Report version) raises concerns related to statistical power to detect significant differences. Of note, many of these studies reported significant elevations on other CBCL scales (e.g., attention problems, ADHD) [37, 40] suggesting that elevations were not specific to CDS.

Eleven of the 16 studies also compared mean differences of CDS in youth with and without chronic or complex medical conditions, with mixed results. For instance, compared to youth without these conditions, studies found significantly higher parent-reported CDS symptoms in adolescents classified as extremely low birth weight [44], children with prenatal alcohol exposure [39], and survivors of childhood cancer [32-34]. Conversely, five of these studies reported no significant mean differences in CDS symptoms, including the Willard et al. [33] study comparing CDS in child survivors of ALL to community controls. In a sample of 30 school-aged children with enuresis compared to 30 youth without enuresis, no differences were found in parent-reported CDS symptoms [41]. In a sample of 21 children with fetal alcohol syndrome compared to 87 without, no significant differences were found for parent-reported CDS, though significant elevations were observed for teacher-reported CDS [40]. One study of children with and without idiopathic epilepsy found lower parent-reported CDS symptoms in children with epilepsy compared to those without [38]. Although surprising, it is notable that similar effects were found for parent ratings of anxiety, depression, and social competence [38]. Peterson and colleagues [30] examined differences in parent-reported CDS symptoms in youth with ADHD-IN only, ALL, and brain tumors (BT), finding greater CDS symptoms in youth with ADHD-IN compared to survivors of childhood cancer. Finally, a study of youth with spina bifida reported elevated CDS symptoms compared to samples of typically developing youth but not clinically based samples (e.g., ADHD) [35].

In two large, community-based samples of approximately 2,000 school-aged children and adolescents in the United States [46] and Spain [45] examined associations between CDS and parent-reported history of child medical diagnoses. Among youth in a high parent-reported CDS group, 2.3% had reported a prior diagnosis of blindness/visual impairment, with rates higher than the comparison group but not an ADHD-only group [46]. Finally, in analyses controlling for child sex and an ADHD-IN factor, CDS was uniquely associated with greater mother- and father-reported history of child medical problems [45]. Conversely, the ADHD-IN factor was significantly associated with the absence of reported history of child medical problems.

Associations of CDS with Academic, Neuropsychological, Psychosocial, and Treatment Outcomes

Academic and Neuropsychological Outcomes

Seven studies (24%) examined CDS symptoms in relation to academic (e.g., reading and math abilities) and neuropsychological functioning, including measures of intellectual (IQ) functioning and neurocognitive processes (e.g., working memory, processing speed). In pediatric survivors of ALL one-year post diagnosis, parent ratings of CDS symptoms were correlated with lower IQ scores, basic reading, numerical operations, and math reasoning. Moreover, when controlling for group status (e.g., ALL compared to sibling controls), parent-reported CDS remained uniquely associated with lower reading scores [32]. Willard and colleagues [33] found parent-reported CDS symptoms to be significantly correlated with greater CBCL attention problems, lower working memory, and lower IQ scores, but not processing speed, in a sample of childhood survivors of ALL and BT. Of note, the CBCL attention problems scale includes some DSM ADHD symptoms as well as CDS symptoms and unique effects were not examined. Additionally, in children and adolescents with SCD, parent-reported CDS symptoms were significantly correlated with greater parent-reported learning problems and metacognition after controlling for age, socio-demographic factors, and ADHD symptoms [43]. In contrast to the aforementioned studies, parent ratings of CDS were unrelated to scores on measures of verbal comprehension, perceptual reasoning, working memory (albeit marginally significant), or processing speed six-months following TBI [42]. Finally, in youth with epilepsy, parent-reported CDS symptoms were associated with higher IQ scores but unrelated to academic performance measured on the Academic Performance Test [38].

Two of the seven studies assessed CDS and neuropsychological functioning with a combination of rating scales and performance-based tests [7, 8]. In children and adolescents with spina bifida, mother, father, and teacher ratings of CDS were significantly correlated with lower sustained attention, working memory, cognitive flexibility, and planning/organization as measured by rating scales and performance-based measures [8]. In pediatric BT survivors, Peterson and colleagues [9] found parent ratings of total CDS symptoms and specific daydreamy and low initiation subscales to be correlated with lower verbal ability. Controlling for verbal ability and ADHD-IN symptoms, parent-reported CDS-daydreamy symptoms, but not CDS-initiation or CDS-sleepy/sluggish symptoms, were uniquely related to lower speeded inhibition performance. This study was the only one to consider specific CDS dimensions in relation to processing speed assessed via performance-based tasks.

Psychosocial Outcomes

Six studies (24%) assessed relations between CDS and measures of psychosocial functioning. In children with a history of heavy prenatal alcohol exposure with and without ADHD, parent-reported CDS was significantly correlated with greater internalizing, externalizing, and attention problems [39]. Further, Mahdavi and colleagues [42] found significant correlations between parent-reported CDS and higher internalizing and externalizing symptoms on the CBCL [42]. In a sample of youth with spina bifida, parent-reported CDS symptoms (assessed with the SCT-S) were significantly correlated with greater self-reported anxiety and depressive symptoms [35]. However, unique effects of CDS on psychosocial outcomes when controlling for ADHD inattentive symptoms were not assessed in these studies.

Darow and colleagues [36] conducted the only prospective study that examined the relation between CDS symptoms and peer functioning in youth with and without spina bifida. Across eight years, teacher-reported CDS symptoms at baseline predicted greater social skills difficulties on father- and teacher-report measures, whereas increased mother-reported CDS predicted worse mother-reported social skills [36]. In this study, similar effects were found in the typically developing group on CDS symptoms and social skills.

Two studies assessed relations between CDS and sleep. First, Becker et al. [52] examined parent-reported CDS in relation to a number of academic, sleep, and mental health symptom domains in a sample of youth referred to a pulmonary-based sleep disorders center. Findings showed parent-reported CDS symptoms were significantly correlated with shorter sleep duration and greater symptoms of parasomnias, sleep-disordered breathing, periodic limb movement disorder, delayed sleep phase, and daytime sleepiness. Additionally, after controlling for child characteristics and psychopathology symptoms, greater CDS symptoms remained uniquely associated with greater sleep problem severity, daytime sleepiness, academic impairment, and family dysfunction. Finally, in a sample of pediatric BT survivors, parent ratings of sleep disturbances were associated with increased CDS symptoms above and beyond disease and treatment burden [51].

Treatment Outcomes

No study retained in the review examined whether CDS was associated with any measure or rating of treatment outcomes.

Correlates or Predictors of CDS

In a sample of children admitted for psychiatric hospitalization, thyroid dysfunction was correlated with and uniquely related to greater parent-reported CDS above and beyond participant characteristics, internalizing symptoms, and ADHD symptoms [50]. Although associations of thyroid stimulating hormone (TSH) and CDS were small in magnitude, no significant associations of TSH with internalizing or ADHD symptoms emerged. In two studies examining a longitudinal cohort of community-based infants, iron deficiency at six months of age was examined in relation to parent-reported CDS symptoms in childhood and adolescence. Findings showed that infants with iron-deficiency anemia had higher parent-reported CDS symptoms at age 10 years, with effects remaining even when controlling for youth characteristics, sociodemographic factors, iron supplementation status, and presence of maternal breastfeeding during infancy [47]. Further, the severity of iron deficiency at six months of age exhibited significant bivariate and unique associations with parent-reported CDS symptoms at ages 5, 10, and 16 years, controlling for youth characteristics, sociodemographic factors, and iron supplementation status [48]. Notably, effect sizes were small and similar to those reported for parent-reported ADHD inattentive and hyperactive/impulsive symptoms. Finally, in a community-based sample of school-aged youth, retrospective ratings of maternal smoking during pregnancy and second-hand exposure to smoke were concurrently associated with greater CDS symptoms. When controlling for child characteristics, sociodemographic factors, ADHD symptoms, and other psychopathology symptoms, CDS symptoms remained associated with parent-reported exposure to second-hand smoke [49].

Seven studies examined either medical or sociodemographic risk factors in relation to CDS. Two studies of childhood survivors of pediatric BT did not observe any association between disease-related variables (e.g., age at diagnosis and radiation, location of radiation field, tumor location, history of radiation and/or chemotherapy) and parent ratings of CDS [29, 45]. Conversely, in survivors of childhood cancer, age at diagnosis and history of chemotherapy, surgery, and/or radiation were related to profiles of CDS membership [34]. In children and adolescents with SCD, disease-related (e.g., SCD genotype, current treatment) and sociodemographic factors (e.g., education, family income) were unrelated to parent-reported CDS [43]. However, in one study of childhood survivors of pediatric BT, placement of ventriculoperitoneal shunt to control hydrocephalus was significantly related to CDS [33]. Similarly, patient diagnosis of myelomeningocele and shunt placement was weakly correlated with greater parent-reported CDS in a sample of children with spina bifida [35]. In pediatric and adult survivors of ALL, numerous factors were significantly associated with greater parent or self-reported CDS: chronic health conditions, less living space, poorer family functioning, physical inactivity, fatigue, and greater academic stress [31]. These risk factors were also associated with greater attention problems in general, with the exception of living space. Moreover, Peterson and colleagues [9] found significant correlations between parent-reported symptoms of CDS total and the CDS-low initiation/persistence subscale with longer time since diagnosis of pediatric BT. Finally, regarding neuropsychological predictors, one prospective, longitudinal study found lower working memory to predict increased mother-reported CDS, whereas lower cognitive flexibility predicted increased teacher-reported CDS over eight years in youth with spina bifida [8].

Quality Assessment

Study quality is summarized in Figure 1. The majority of studies were judged to have good sample representativeness (80%) and used well-validated measures for assessing correlates/outcomes (88%). Approximately half of the studies include a priori aims or hypotheses related to CDS (48%) and made appropriate/meaningful inferences and conclusions (56%). Most studies used measures that were not specifically developed to assess CDS/SCT (76%), did not justify sample size for CDS-related analyses (92%), did not include a CDS-specific group or control for ADHD-IN symptoms (76%), and were observational in design (100%). There was variability in studies that did not include any comparison sample (56%), used a clinical comparison sample (24%), or used a typically developing control group (20%).

Discussion

The current systematic review identified 25 studies assessing prevalence rates, mean differences, and/or associations of CDS with academic, neuropsychological, or psychosocial outcomes in samples of individuals with chronic or complex medical conditions. Of these studies, nine assessed biological, medical, or treatment-related, neuropsychological, and/or sociodemographic characteristics as correlates of CDS. The limited research allows for a few primary conclusions. First, CDS symptoms appear significantly elevated in youth with a range of medical conditions (e.g., childhood cancer, epilepsy, spina bifida) and associated with greater neuropsychological and psychosocial difficulties. Second, a handful of studies pointed to biological (e.g., thyroid dysfunction, iron deficiency), sociodemographic risk (e.g., socioeconomic status, second-hand exposure), or treatment-related factors (e.g., shunt history, time since diagnosis) as concurrent and longitudinal predictors of CDS, though many await replication of effects. Finally, there was wide variety in type of medical condition studied, CDS measurement, and statistical design. Below, we discuss key findings, identify limitations, and outline an agenda for future research.

Elevations in Rates of CDS and Associations with Functional Outcomes

According to parent ratings on the CBCL, nearly all the retained studies reported significant elevations in CDS symptoms in youth with chronic and complex medical conditions. Moreover, one study of pediatric and adult survivors of ALL [31] and another of pediatric BT survivors [51] found the highest elevations in CDS symptoms compared to other mental health domains. However, most studies reported increased psychopathology in general [31, 37, 40] and, when comparing CDS symptoms to typically developing youth without medical conditions, findings were more mixed. Additionally, Peterson and colleagues [34] found significantly greater parent-reported CDS symptoms in youth with ADHD-IN compared to youth with ALL and BT. Although findings suggest that CDS symptoms may explain some of the attention and neurocognitive deficits in youth with medical conditions [7], the inconsistencies reported and observational (i.e., non-causal) and largely cross-sectional design of these studies limit our understanding of CDS in youth with chronic or complex medical conditions.

Despite initial calls to study CDS as a marker of neurocognitive and psychosocial effects in pediatric samples of youth [27], surprisingly far fewer studies (11) identified in the review examined CDS in relation to academic, neuropsychological, and/or psychosocial functioning. Collectively, the few studies, primarily in childhood cancer survivors, provided mixed evidence for associations of CDS symptoms with academic and neuropsychological functioning. These limited effects correspond to the literature on CDS in relation to academic and neuropsychological functioning in ADHD-specific samples) [53]. Although preliminary, one prospective study found lower working memory and cognitive flexibility to predict increased ratings of CDS symptoms in youth with spina bifida, controlling for ADHD-IN symptoms, over an eight-year period [8]. In contrast to results from academic/neuropsychological outcomes, several studies consistently linked parent-reported CDS to greater internalizing symptoms in youth with histories of prenatal alcohol exposure [39], TBI [42], spina bifida [35], and sleep-disorders [52]. Importantly, only one of these studies examined the unique effect of CDS controlling for ADHD symptoms, and found CDS associated with greater depressive symptoms [52]. Findings from these few studies converge with the larger literature in community and ADHD-specific samples of youth documenting robust associations of CDS and internalizing symptoms [1], particularly depression [54]. Finally, one prospective study of youth with spina bifida found CDS to be a predictor of fewer social skills [36], coinciding with the growing body of literature in typically developing youth linking CDS to social deficits marked by greater “passive” peer rejection [55].

Additionally, two studies found significant associations between sleeping problems (e.g., sleep disturbance, sleep duration, etc.) and CDS [51, 52]. There is growing evidence in youth with and without ADHD that shortened sleep and daytime sleepiness are uniquely related to CDS [1].

Biological or Medical Correlates of CDS

Preliminary findings suggest that thyroid dysfunction and iron deficiency are associated with CDS symptoms [47, 48, 50]. However, caution is warranted as small effect sizes of thyroid dysfunction on CDS were noted and iron deficiency was also linked to symptoms of inattention, hyperactivity, and impulsivity. Replication is needed for both thyroid function and iron deficiency. Nevertheless, as low iron levels and underactive thyroid functioning contribute to low energy and slow-moving behaviors [47, 50], studies are warranted to further our understanding of possible biological contributions to the CDS phenotype. Further, a few studies pointed to exposure to second-hand smoke and reduced living space as correlates of CDS symptoms [31, 49]. These findings converge with greater CDS symptoms in youth exposed to prenatal alcohol use [39, 40], in additional to studies in non-pediatric samples linking CDS to indices of psychosocial adversity, including lower socioeconomic status [46, 56] and family dysfunction [57]. Although underlying mechanisms have not been explored, future research should consider the role of social and economic factors in the presentation and correlates of CDS in pediatric samples.

Limitations of the Existing Literature

Despite a growing number of studies assessing CDS prevalence rates and associations with functional outcomes in samples of individuals with medical conditions, a number of key limitations emerged across studies. Primarily, the number of studies (N=25) examining these populations in relation to CDS is markedly behind the literature base of CDS in clinical samples of youth with and without ADHD [3]. Although unsurprising given the development of the CDS construct in ADHD-specific samples, this lag prevents a clear understanding of whether, and how, CDS impacts the functioning of youth with chronic or complex medical conditions. Our review also only included peer-reviewed publications, which may be more likely to report significant findings, and we were unable to directly examine the possibility of publication bias or heterogeneity in effect sizes. Additionally, as shown in Table 1, a handful of studies did not report race and ethnicity of participants, clearly limiting interpretation of these findings. Of studies reporting this information, the majority of participants identified as White, which limits our understanding of how the intersection of race and ethnicity impact psychosocial and treatment outcomes in youth with CDS and medical conditions as well as the generalizability of these studies to more diverse samples. Additionally, with few exceptions, the CBCL was the most frequently used rating scale for assessing CDS (72% of studies). The CBCL includes a limited item set to measure CDS and researchers have called for integrating specific measures that capture the multidimensional nature of CDS symptoms that have strong discriminative validity from ADHD-IN [3]. Further, none of the included studies used self-reported ratings of CDS. Given the internal nature of CDS (e.g., “lost in thoughts” and “feel sleepy/tired”) and strong associations with internalizing psychopathologies [1], use of self-report ratings of CDS has been encouraged. Thus, researchers investigating the prevalence of CDS or associations with functional correlates are encouraged to integrate CDS-specific measures with consideration of self-reported ratings.

Next, studies varied in time since medical diagnosis or injury, with retrospective studies ranging from six months [42], one to five years [9, 32, 33], or several years [31, 34] post-condition onset/diagnosis. In survivors of pediatric BT, time since diagnosis was inconsistently associated with CDS symptoms [9, 33]. Additionally, type of treatment (e.g., chemotherapy) was unrelated to CDS [9, 31, 34], with the exception of shunt placement linked to increased CDS symptoms in pediatric BT survivors [33] and youth with spina bifida [35]. Finally, other samples of youth with congenital conditions (e.g., low birth weight) have yet to be replicated. Further, a myriad of other pediatric samples of youth in which attention problems are prevalent and impact functioning, such as asthma [58] and diabetes [59], await investigation.

Call for Future Research

Taken together, we encourage pediatric researchers to leverage clinical databases with measures of CDS to examine prevalence and associations with key outcomes in a variety of pediatric conditions. Although CDS-specific measures with strong psychometric properties would certainly be preferred [3], broadband rating scales like the CBCL have the advantage of being frequently used and, as such, have already been collected in numerous samples. The CBCL SCT (CDS) scale also has been shown to correlate with longer CDS-specific measures [60, 61]. Additional longitudinal and/or test-retest studies across disease groups are needed to determine the stability of the CDS construct and association with disease progression and treatment outcomes. In addition to these limitations and future directions, we now turn to highlighting two critical areas of future research to further our understanding of CDS in medical conditions.

Future Research in CDS and Neuropsychological/Psychosocial Functioning

The extant studies examining CDS in relation to neuropsychological functioning or neurocognitive late effects are both small and limited by sample size, domains of neuropsychological functioning examined, and/or cross-sectional designs. This is an area where far more research is needed, particularly before drawing any firm conclusions. In addition, although CDS symptoms are consistently linked to increased depressive symptoms and peer withdrawal in youth with and without ADHD [1, 55], three studies in this review examined CDS and psychosocial functioning with only one controlling for ADHD-IN symptoms [52]. Due to the psychosocial risk of youth with chronic and complex medical conditions [62, 63], future research is clearly needed to determine whether CDS symptoms independently contribute to or exacerbate risk for poor psychosocial functioning. Moreover, numerous studies document social support and peer functioning as key predictors of treatment responses and health-related quality of life [63] in pediatric patients. As youth displaying CDS, with and without diagnoses or symptoms of ADHD-IN, experience greater peer impairments such as social withdrawal and peer victimization [1], these attentional problems and behaviors might portend long-term risk to psychosocial adjustment during treatment or recovery.

Additionally, fatigue and excessive daytime sleepiness are associated with poorer psychosocial functioning and quality of life in pediatric patients [64]. Additionally, a growing body of evidence suggests that CDS is strongly associated with shortened sleep duration and excessive daytime sleepiness [65], which is unsurprising considering the hypoactive nature of CDS (e.g., “slow-moving,” “sleepy,” “tired”). In the sample of youth presenting to a sleep disorder clinic, parent ratings of CDS were strongly associated with excessive daytime sleepiness [52] and there is experimental evidence pointing to a direct relation between shortened sleep duration and CDS symptoms in adolescence [66]. Thus, if additional studies replicate effects of CDS on sleep functioning in pediatric samples, this might suggest that pediatric psychologists may benefit from assessing signs of CDS as a cause, correlate, or consequence of inadequate sleep and excessive daytime sleepiness.

Future Research in CDS and Self-Management of Disease

Despite the well-documented impact of attention and executive function deficits on self-management of disease and treatment adherence [12, 13, 67], no study specifically tested relations with CDS. Although conclusions cannot be drawn, studies in community and ADHD-specific samples indicate that youth with CDS exhibit difficulties with sustained attention, excessive mind-wandering, slower processing speed, and working memory deficits [17]. Although youth with ADHD-IN may struggle with self-management due to forgetfulness and distractibility, youth with CDS may have difficulty managing the rapid influx of medical information due to internal distractibility and lethargy. Future research is encouraged to integrate assessment of CDS when testing the impact of attention problems on self-management and treatment adherence.

Clinical Implications

Findings from the systematic review indicate elevated CDS symptoms in pediatric samples, and potential associations with academic, neuropsychological, and psychosocial functioning. When integrating findings from the larger evidence base for CDS, we recommend assessing CDS to help identify potential risk of reduced psychosocial functioning. Additionally, given evidence that youth with CDS may not respond to traditional medication and behavioral management of ADHD [19, 20, 22], assessing CDS along with ADHD-IN would be useful for understanding potential treatment response. Despite lack of current-evidence based interventions, recommendations for treatment include cognitive-behavioral therapy, mindfulness, sleep-based interventions, social skills training, and/or homework completion and organizational skills training depending on the nature and associated impairments of youth with CDS [1]. Although treatment studies are scarce, an open trial of cognitive-behavioral therapy for sleep and circadian disturbances found strong effects on reductions in CDS across informants in adolescents with ADHD [68]. Regarding assessment, due to the practical barriers of administering CDS specific measures to patients, we recommend reviewing items endorsed on the CBCL (if administered) as a sign of CDS presence and considering CDS-specific measures in the case of elevation. Further, clinicians can consider administering other CDS-specific ratings scales (e.g., CABI CDS subscale) or available clinician interviews (e.g., Barkley Sluggish Cognitive Tempo scale) [3]. With these efforts, the assessment and intervention of CDS may be enhanced to improve the quality of life among youth with CDS and their families.

Appendix A.