It is widely recognised that anxiety is commonly associated with ADHD; however, it is not known whether anxiety develops secondary to symptoms of ADHD and/or whether anxiety plays a role in the development or exacerbation of ADHD. Better understanding of the developmental relationship between anxiety and ADHD could have significant clinical benefits. This study evaluated these developmental relations in a normative, community-ascertained sample of adolescents to determine whether symptoms of ADHD lead to anxiety and vice versa.
Participants were adolescents from the Zurich Study on Social Development from Childhood to Adulthood.* This study focuses on data collected when participants were aged 13, 15 and 17 years, where both self-reported ADHD and anxiety data were available and measured using an adapted version of the Social Behaviour Questionnaire (SBQ) Anxiety† and ADHD scales.‡ The developmental relations between ADHD and anxiety were examined using an autoregressive latent trajectory model with structured residuals (ALT-SR) fit to the ADHD and anxiety latent factors (i.e. ADHD and anxiety at ages 13, 15 and 17 years).
Across the three follow-up ages, 1483 adolescents (52% male) were included in the study (at age 13 years, n = 1363; at age 15 years, n = 1446; and at age 17 years, n = 1305). The sample was considered diverse in terms of cultural and ethnic background; among the primary caregivers of the participants, 43% were born in Switzerland. A measurement model was used to generate the factor scores for anxiety and ADHD which fit well: root mean square error of approximation (RMSEA), 0.037 (95% confidence interval [CI]: 0.034–0.041); comparative fit index (CFI), 0.95; Tucker-Lewis index (TLI), 0.93; standardised root mean square residual (SRMR), 0.045. The ALT-SR model fit well according to TLI (0.98), CFI (0.93) and SRMR (0.04), but less well based on RMSEA (0.11, 95% CI 0.09–0.12). Standardised autoregressive§ and cross-lagged parameters¥ were based on the ALT-SR.
The autoregressive parameters indicated that symptoms of anxiety and ADHD were only partially stable at the within-person level across adolescence. This meant that the autoregressive effect for anxiety symptoms when adolescents were aged 13‒15 years and the autoregressive effect for ADHD symptoms between the ages of 15‒17 years were not significant. At the within-person level, the cross-lagged parameters suggested that higher levels of ADHD symptoms between the ages of 13 and 15 years predicted relatively higher levels of anxiety symptoms when adolescents were aged 15‒17 years. Higher levels of anxiety symptoms when adolescents were 15 years old also predicted relatively higher levels of ADHD symptoms at the age of 17 years. Within-person symptom levels of ADHD and anxiety were also significantly correlated at each follow-up age (13 years, r = 0.37; 15 years, r = 0.61; 17 years, r = 0.66; all p < 0.001). Gender also had a significant effect on ADHD and anxiety symptom intercepts as females showed higher levels of anxiety (β = 0.50; p < 0.001) and ADHD symptoms (β = 0.07; p = 0.02).
This study had some limitations. Firstly, both ADHD and anxiety can have onset in children; therefore, future studies could trace the developmental relationship between ADHD and anxiety before adolescence and into adulthood. Secondly, as measurements of ADHD and anxiety were administered as part of a broad longitudinal cohort study, these measures were brief and relatively general. Future studies could replicate these findings using more comprehensive measures of ADHD and anxiety. Finally, although ALT-SR can be effective in controlling for between-person, time-stable confounds, this model can be vulnerable to the effects of confounds that co-vary within individuals across time with the focal phenotypes.
To conclude, the authors demonstrated that there is a reciprocal developmental relationship between ADHD and anxiety in adolescence. These data suggest that adolescents presenting with anxiety should be assessed carefully for symptoms of ADHD and sequelae. Moreover, adolescents with ADHD may benefit from continual monitoring of anxiety symptoms, and should have access to preventative interventions. The mechanisms by which ADHD symptoms enhance anxiety symptoms, and vice versa, could be a focus of future studies.
*The study was initiated in 2004 when participants were beginning school at age 7 years, with follow-up waves when participants were aged 8‒13, 15, 17 and 20 years; to date, the study is ongoing
†Anxiety was measured using four items, which were distributed throughout a section labelled “How you feel”; questions on self-harm, depression (also from SBQ) and in later follow-ups, suicidal ideation, were also included in this section. Participants were asked to rate on a 5-point Likert-type scale from ‘never’ to ‘very often’ the extent to which they experienced the following symptoms: “I cried”; “I was scared, fearful or anxious”; “I couldn’t fall asleep”; and “I was worried”. The total reliability values for the anxiety items were 0.63, 0.70 and 0.71 at ages 13, 15 and 17 years, respectively
‡ADHD was measured using four items and were part of a section labelled “Things that you do”, which also included questions on prosociality, conduct problems, oppositionality and aggression, all of which were derived from the SBQ. Questions on ADHD were distributed throughout this section, and participants were asked to rate on a 5-point Likert-type scale from ‘never’ to ‘very often’ the extent to which they experienced the following symptoms: “You were restless and struggled to sit still”; “You struggled to concentrate on a task”; “You were inattentive”; and “You fidget”. The total reliability values for the ADHD items were 0.78, 0.80 and 0.74 at ages 13, 15 and 17 years, respectively
§The autoregressive parameters capture the within-person stability of ADHD and anxiety, i.e. the extent to which individuals’ deviations in symptoms from their base level of symptoms are stable over successive follow-ups
¥The cross-lagged parameters capture the within-person relations between ADHD and anxiety over time i.e. the extent to which within-person deviations from their base level of symptoms in one domain predicts the within-person deviations in the other domain
Murray AL, Caye A, McKenzie K, et al. Reciprocal developmental relations between ADHD and anxiety in adolescence: a within-person longitudinal analysis of commonly co-occurring symptoms. J Atten Disord 2020; Epub ahead of print.