Research has shown that individuals with ADHD demonstrate an increased risk for developing addiction, including alcohol, nicotine and gaming dependence (Fuemmeler et al. 2007; Kessler et al. 2006; Kuss et al. 2012). The mechanism underlying the comorbidity of ADHD with addiction is not clear. One hypothesis suggests that the increased risk could be explained by ADHD and addiction sharing another comorbidity, such as conduct disorder (Biederman et al. 1997). Despite this, studies performed to date do not provide sufficient evidence to support the role of conduct disorder in the relationship between ADHD and addiction. Using a causal discovery algorithm, this exploratory analysis aimed to build a causal model of the relationships between ADHD, comorbid conduct problems, substance use and gaming habits.
This study included 362 participants (81% male; aged 16±2.4 years) with ADHD, who had been recruited by the Belgian, Dutch and German sites of the 2003–2006 International Multicentre ADHD Genetics (IM-AGE) study. ADHD symptoms and conduct problems were rated at baseline using the Parental Account of Childhood Symptoms (PACS), the Strengths and Difficulties Questionnaire (SDQ), and the Long Versions of the Conners’ Parent (CPRS-R:L) and Teacher Rating Scale Revised (CTRS-R:L). The severity of alcohol, nicotine and other drug habits was assessed by patient questionnaires and a number of rating tools.* The exploratory analysis was performed using the Bayesian Constraint-based Causal Discovery (BCCD) algorithm.†
Results demonstrated that the mean symptom count for ADHD-hyperactive/impulsive subtype (ADHD-HI) and ADHD-inattentive subtype (ADHD-In) was 7.8±1.6 and 8.0±1.1, respectively, and the mean conduct problem score was 83±36. ADHD-HI and ADHD-In symptom counts showed a causal dependence with a joint reliability estimate of 52%. Both ADHD-HI and ADHD-In symptom counts were independently linked to conduct problems (joint reliability estimate: ADHD-HI and conduct problems, >99%; ADHD-In and conduct problems, 62%). Importantly, ADHD-HI and ADHD-In were not directly connected to the substance-abuse variables; instead, conduct problems mediated the connection between both ADHD-HI and ADHD-In and nicotine use (joint reliability estimate, 85%), but not alcohol or drug use. Furthermore, ADHD-In demonstrated a directional link to gaming habit (joint reliability estimate, 86%), suggesting that more severe ADHD-In symptoms lead to more severe gaming habits.
Findings from this study should be interpreted in the context of some limitations. Despite the fact that the design of the study tried to account for the effects of stimulant treatment, it remained difficult to examine its effects in the causal model. Furthermore, the ADHD ratings employed in this study could be based on previous medication-free periods; in these instances, ratings relied on parental reporting of symptoms, which could have resulted in a source of bias. Flooring effects due to the low variance in substance abuse may have also biased this analysis. Furthermore, the BCCD does not provide estimates of effect size of the causal influence and, therefore, only reliability estimates are provided in the model. Finally, as the study aimed to identify pathway components that may benefit from further investigation, no conclusions can be made about any underlying biological mechanisms.
The authors concluded that conduct problem severity may mediate the connection between ADHD severity and nicotine use, but not severe alcohol or substance use. Furthermore, ADHD-In severity was a risk factor for gaming, suggesting that this variable has a different causal pathway to substance dependence and should be treated differently. The authors suggested that this work could facilitate further research into the overlap between ADHD and addiction, and help researchers and clinicians to develop more effective treatments.
Find out more about the overlap between ADHD, conduct disorder and addiction here
*Participants >12 years old completed questionnaires to assess the severity of alcohol, nicotine and other drug habits, as well as the severity of gaming habit; continuous scores were used for all measures to increase discriminatory power. Alcohol dependence was assessed using the Alcohol Use Disorders Identification test (scores ranging from 0–40). Nicotine dependence was assessed using the Fagerström Test for Nicotine Dependence (scores ranging from 0–10). Other drug dependence was assessed using the Drug Abuse Screening Test-20 (scores ranging from 0–20). There was no universally accepted standard to assess pathological gaming; therefore, a 24-item gaming questionnaire was constructed from an existing questionnaire and supplemented with additional questions (scores ranging from 0–92)
†The analysis steps of the BCCD were as follows:
1. Data: accepted both discrete and continuous data
2. Preprocessing: input was mapped through a Gaussian transform into a correlation matrix
3. Search: an efficient search was performed to obtain Bayesian reliability scores based on the BGe metric, resulting in a list of weighted independence constraints
4. Logical causal inference engine: local independence constraints were used together with background knowledge to create model
5. Model: coherent output causal model was generated
Biederman J, Wilens T, Mick E, et al. Is ADHD a risk factor for psychoactive substance use disorders? Findings from a four-year prospective follow-up study. J Am Acad Child Adolesc Psychiatry 1997; 36: 21-29.
Fuemmeler BF, Kollins SH, McClernon FJ. Attention deficit hyperactivity disorder symptoms predict nicotine dependence and progression to regular smoking from adolescence to young adulthood. J Pediatr Psychol 2007; 32: 1203-1213.
Kessler RC, Adler L, Barkley R, et al. The prevalence and correlates of adult ADHD in the United States: results from the National Comorbidity Survey Replication. Am J Psychiatry 2006; 163: 716-723.
Kuss DJ, Lopez-Fernandez O. Internet addiction and problematic Internet use: a systematic review of clinical research. World J Psychiatry 2016; 6: 143-176.
Schoenmacker GH, Groenman AP, Sokolova E, et al. Role of conduct problems in the relation between attention-deficit hyperactivity disorder, substance use, and gaming. Eur Neuropsychopharmacol 2018; Epub ahead of print.