Long-term methylphenidate use for children and adolescents with attention deficit hyperactivity disorder and risk for depression, conduct disorder, and psychotic disorder: a nationwide longitudinal cohort study in South Korea

Attention deficit hyperactivity disorder (ADHD) is a common neurodevelopmental disorder that occurs in childhood and adolescence. ADHD is a chronic debilitating disorder that can affect several aspects of an individual’s life including academic performance, peer relationships, and parent–child relationships [1]. Several studies have suggested that 40–60% of affected children continue to show symptoms of the disorder until adulthood [2, 3]. Children and adolescents with ADHD are at an increased risk of comorbidities, including conduct problems, mood and anxiety disorders, and substance abuse [4]. Moreover, patients with ADHD and comorbidities are at an increased risk of adverse outcomes. Youths with ADHD and depression had a greater risk of suicide [5]. Adolescents with ADHD and conduct disorders are more likely to experience substance abuse [6]. Therefore, drug treatment is required to mitigate symptoms and related impairments of ADHD. The American Academy of Pediatrics and American Academy of Child and Adolescent Psychiatry guidelines recommend medication as the first-line treatment [7, 8].

Methylphenidate (MPH) is the most frequently used medication for ADHD treatment. As the prescription of MPH for both children and adults has increased in several countries [9‐11], concerns have been raised regarding the safety of its long-term use [12]. In fact, approximately one-third of patients with ADHD (both children and adults) continue to take MPH for 2 years after treatment initiation [13]. The Committee for Medicinal Products for Human Use has suggested that more studies are needed on the long-term effects of MPH, especially on adverse psychiatric events [14]. A recent review has investigated the association between long-term MPH treatment and adverse neuropsychiatric effects [15]. However, it was shown that the evidence was unclear, and more data are needed on the relationship between long-term MPH use and adverse neuropsychiatric effects. Although the Comparison of Methylphenidate and Psychotherapy in the Adult ADHD Study (COMPAS) was conducted focusing on safety profiles including neuropsychiatric events, it only included adult patients with ADHD [16]. Therefore, research is required to establish real-world evidence for the impact of long-term MPH treatment on adverse neuropsychiatric events in children and adolescents with ADHD.

In this study, we targeted depressive disorders and conduct disorders, which are the most common comorbidities of ADHD, and psychotic disorders, where the risk of MPH use was reported [17, 18]. We aimed to evaluate the risk of depressive disorders, conduct disorders, and psychotic disorders associated with long-term MPH treatment compared with short-term MPH treatment in children and adolescent patients with ADHD.

In this nationwide cohort study, we found that long-term MPH treatment for children and adolescents with attention deficit hyperactivity disorder was associated with a significantly lower risk of depressive and conduct disorders than short-term MPH treatment. No significant differences were noted in the risks of psychotic disorders between the long- and short-term MPH treatment groups. These results were consistent across the analysis using different treatment periods and in the analysis excluding non-stimulant medication. In addition, compared to the non-MPH users, consistent findings regarding long-term MPH users were observed. However, the risk of depressive disorder, conduct disorder and ODD, and psychotic disorder was not significantly different between short-term MPH users and non-MPH users.

Although it is a necessary to investigate the long-term treatment effects of MPH, the current evidence is limited [25]. Depression is known to be the most common comorbidity of ADHD. Depression in children and adolescents with ADHD usually occurs after its onset. Longitudinal studies have shown that MPH treatment may reduce the risk of subsequent depression in adolescents with ADHD [26, 27]. Population-based pharmacoepidemiological studies have found that ADHD medication is associated with the reduced probability of developing depression [28, 29]. Specifically, the patterns observed in our study align with those of Chang et al. showing that the risk of depression was lower for longer duration of ADHD medication [29]. Levels of depression improved during the first year of treatment compared to the levels after a brief MPH treatment. Improvements in academic and social functional domains by long-term stimulant treatment and the resulting alternative development trajectory may be the reason for the reduction in depression [27, 30]. Additionally, long-term MPH treatment was associated with a lower risk of conduct disorder and ODD than short-term MPH treatment. Our results are consistent with previous findings that stimulant treatment reduces the risk of developing conduct disorder in both boys and girls with ADHD [31]. A recent qualitative review of studies examining the effects of ADHD medications has shown a reduced relative risk of injuries, motor vehicle accidents, and substance abuse [32, 33]. It have been reported to have a therapeutic effect of psychostimulants on the management of oppositional behaviour, conduct problems, and aggression in ADHD patients without ODD or CD [34]. It is possible that ADHD medication helps patients to organise their lives better and it contributes to continuing changes at the neuronal level [35]. One of the most common comorbidities associated with ADHD is conduct disorder and/or ODD. Conduct disorder and/or ODD are present in approximately 40–70% of children with ADHD [36]. The symptoms of both disorders usually respond to psychostimulants; however, ADHD with conduct disorder usually has a worse clinical outcome than either of the two conditions alone [37]. Given the poor prognosis associated with conduct disorder and ODD, the effects of stimulants are likely to have beneficial effects in people with ADHD. Dopaminergic excess induced by MPH treatment may trigger psychotic symptoms [38]. Although a previous study showed that MPH treatment led to an increased risk of psychotic disorders, this study did not distinguish between this risk and long- and short-term MPH treatments [39]. Also, a study using Hong Kong population-based electronic medical records reported that risk of psychotic disorders was not found during MPH exposure compared to non-exposure [18]. A possible explanation for the increased risk of psychotic disorders is that ADHD itself is a risk factor for psychosis [39]. In our study, no difference was noted in the risk of psychotic disorder between the long- and short-term MPH treatment groups.

Compared to non-MPH users, long-term MPH users were associated with a significantly lower risk of depressive and conduct disorders. The risk of psychotic disorder was not significantly different between two groups. However, no significant differences were noted in the risks of all outcomes between the short-term MPH users and non-MPH users. Although previous studies suggested beneficial effects of MPH on neuropsychiatric outcomes [25], this study found that the effects of MPH use on depression and conduct disorders were different depending on the duration of MPH use. It is possible that maintaining MPH for more than a certain period may have beneficial effects on depression and conduct disorders. Considering the risk of depression and conduct disorders was significantly lower in the long-term MPH use than in the short-term MPH use in the primary analysis, these results were consistent with the findings of the primary analysis.

This was a population-based cohort study. Although randomised controlled trials (RCTs) are the gold standard for the evaluation of health care outcomes, the cohort’s strengths, such as longer follow-up time and larger sample size, could be an alternative to RCTs [25]. To overcome potential confounding factors in this population-based study, we applied the propensity score matching to the study population. The propensity score was used to reduce the effects of confounding factors [40]. Specifically, a 1:2 nearest neighbour matching for patients with the same propensity scores was used, considering the ratio between the two groups. And we conducted sensitivity analyses excluding patients who were treated with non-stimulant ADHD medications. Meanwhile, we compared the risks of depressive, conduct, and psychotic disorders according to the duration of MPH use. Several previous studies have compared the effects of the treatment duration of MPH. Huang et al. and Liang et al. divided patients with ADHD into long and short-term users according to the cumulative defined daily dose [41, 42]. Schrantee et al. classified patients with ADHD according to the time of drug initiation [43]. Also, several studies classified short-term and long-term based on one year [22, 23, 40]. Likewise, we compared ADHD patients by dividing them into short-term and long-term groups based on one year.

The strengths of this study include the use of national health insurance data that contained data from all patients with ADHD in South Korea. Furthermore, we evaluated the risk of common comorbidities of ADHD according to the length of MPH use. To increase comparability, we matched the doses that may have caused side effects. Considering the sex differences in ADHD symptoms, we also matched the sex ratio between the two groups.

This study has several limitations. First, we were unable to distinguish between the types of ADHD in patients. This was because it was difficult to identify the detailed symptoms owing to the nature of the claims data. Second, we could not include familial factors related to ADHD because of the limitations of the claim database. Given strong genetic background of ADHD, further studies that include familial factors are needed. Third, one year may not be accurate time point to distinguish between short-term and long-term MPH use. In fact, the relationships between 11 and 13 months of use may be closer to that of between 11 and 4 months. However, consistent findings were demonstrated across different treatment periods. Fourth, there was no information regarding the patients’ management except for the number of prescription days. It may not be possible to determine the level of treatment compliance or the effect on the parents of the child.

Our findings suggest a decreased risk of depression and conduct disorders in patients undergoing long-term MPH treatment. No difference in the risk of psychotic disorders was shown between the short- and long-term MPH treatments. These findings support the notion that long-term MPH treatment may not be contraindicated for depression, conduct disorders, and psychotic disorders in children and adolescents with ADHD. There is a possibility of unmeasured confounders; hence, further research is required to clarify the safety of long-term MPH use.