Autistic traits in ADHD index clinical problems

Miriam Cooper Clinical Lecturer, WCAT fellow Section of Child and Adolescent Psychiatry Institute of Psychological Medicine and Clinical Neurosciences Royal College of Psychiatrists Child & Adolescent Faculty meeting Cardiff, September 2014

BACKGROUND

Overlap between ADHD & ASD1 ASD ADHD

Social difficulties

Language Inattention difficulties & Hyperactivity communication deficits Impulsivity Restrictive & repetitive behaviours

High levels of ADHD traits in autism, high levels of autistic traits in ADHD Broader shared phenotype – social cognition, pragmatic language difficulties, executive functions Genetic overlap

1) Rommelse et al (2010) Eur Child Adolesc Psychiatry 19(3): 281-95

Autistic traits in ADHD

Presence well documented, significance not fully understood

Autistic traits in ADHD  What is the significance of ASD traits in ADHD in terms of clinical profile?  So far in ADHD + high ASD traits: ↑ likelihood of combined ADHD subtype1,2 ↑ comorbid ODD & CD3 No clear difference in psychiatric comorbidity1,4 1) Gradzinski et al (2011) J Autism Dev Disord 41(9): 1178-91 2) Reiersen et al (2007) J Child Psychol Psychiatry 48(5): 464-72 3) Mulligan et al (2009) J Autism Dev Disord 39(2): 197-209 4) Kroger et al (2011) Eur Child Adolesc Psychiatry 20(11-12): 561-70

Autistic traits in ADHD BUT - few studies, conflicting results, methodological limitations (sample size, covariates, multiple testing), tentative conclusions Aim: Do ASD traits in ADHD index a more severe clinical profile? •

Methodological considerations from previous studies

METHODS

Sample Using SAGE ADHD sample (clinical)  DSM-IV diagnosis of ADHD, no known clinical ASD diagnoses

SCQ1,2 - autistic traits CAPA interview3 - ADHD symptoms and comorbidities

1) Rutter et al (2003) Western Psycological services 2) Berument et al (1999) Br J Psychiatry 175: 444-51 3) Angold & Costello (1995) J Am Acad Child Adolesc Psychiatry 39(1): 39-48

Total SCQ score (predictor)

Associations measured using regression models

Increasing number of covariates used Unadjusted associations Covarying for IQ, age, gender and family socioeconomic status ADHD severity as further covariate

Clinical characteristics (outcomes) Secondary analyses: SCQ subdomain scores > outcomes

Sample characteristics  ADHD any subtype  Ages 5-18 (mean 10.3, SD 2.9)  M&F (84%M)  IQ 43-124 (mean 84.3, SD 14.0)

 Exclusions:  ASD, Tourette’s, epilepsy, bipolar  >10% missing data on total or any sub-domain score on ASD trait measure (n=107)

 N=711  Mean SCQ score 13.0, SD 6.6, range 0-35

RESULTS

Outcome variable

Unadjusted (max n=711) B (Std. error)

DSM-IV combined ADHD diagnosis

Adjusted a (max n=599)

Adjusted b (max n=599)

OR (95% CI)

p-value

p-value

p-value

1.06 (1.03-1.09)

1.6E-05

2.0E-03

n/a

ADHD symptoms: inattentive *

0.01 (0.00)

1.0E-03

7.1E-03

n/a

ADHD symptoms: hyperactive-impulsive *

0.01 (0.00)

1.1E-08

6.8E-06

n/a

ADHD symptoms: total

0.08 (0.01)

1.4E-08

5.9E-06

n/a

DSM-IV ODD diagnosis

1.03 (1.01-1.06)

4.2E-03

2.1E-03

0.013

DSM-IV CD diagnosis

1.06 (1.03-1.09)

1.1E-04

0.016

0.036

ODD symptoms

0.10 (0.01)

1.5E-14

2.6E-08

6.4E-06

CD symptoms *

0.02 (0.00)

2.3E-09

2.1E-05

3.9E-04

3.9E-03

1.1E-03

2.5E-03

0.02 (0.00)

7.9E-07

5.3E-06

1.2E-04

0.01 (0.00)

1.4E-03

1.5E-03

0.022

DSM-IV anxiety any diagnosis Anxiety symptoms * Depression symptoms

*

1.06 (1.02-1.11)

statistical threshold to account for multiple testing: p-value < 0.003 * transformed a Adjusted for IQ, age, gender and family socio-economic status b Adjusted for IQ, age, gender, family socio-economic status and ADHD severity

Results   autistic trait scores associated with  ADHD-C diagnosis   HI, I and total ADHD symptoms   ODD & CD symptoms, anxiety symptoms

 Autistic sub-domains did not show unique associations with most outcomes  Social deficits > oppositional symptoms  Repetitive behaviours > H-I symptoms

 Robust to inclusion of covariates, correction for multiple testing

Limitations  Potential for subsequent clinical diagnosis of ASD in those with high SCQ scores  Wide age range across the sample – lack of stability of ADHD subtypes across time1  Not yet examined in detail as to whether SCQ has good discriminant validity of autistic symptoms in an ADHD population

1) Willcutt et al 2012

Conclusions  In children with ADHD, increasing levels of ASD symptomatology indexes a more complex clinical phenotype  Severity of ADHD & presence of comorbid psychopathology  In line with but extending previous research  Traits may complicate efficacy of interventions

Thank you for listening Dr Miriam Cooper1,2,3, Mrs Joanna Martin1,2, Dr Kate Langley1,2,4, Dr Marian Hamshere1,2,4, Professor Anita Thapar1,2,4 Child & Adolescent Psychiatry Section, Institute of Psychological Medicine and Clinical Neurosciences, Cardiff University School of Medicine 1

2 MRC

Centre for Neuropsychiatric Genetics and Genomics, Cardiff University School of Medicine

3 Wales

Clinical Academic Track lectureship scheme, Cardiff University School of Medicine

4 Neuroscience

and Mental Health Research Institute, Cardiff University

Published as: Cooper M*, Martin J*, Langley K, Hamshere M, Thapar A (2014) Autistic traits in ADHD index clinical and cognitive problems. European Child & Adolescent Psychiatry, 23(1), 23-34 * Joint first authors