Tolerance and sensitisation can affect the efficacy of pharmacological treatments. This analysis aimed to investigate if the length of treatment can affect the efficiency of pharmacological treatments and impact on the severity of ADHD symptoms.
The study consisted of a systematic review and meta-analysis of randomised controlled trials (RCTs) investigating the efficacy of pharmacological interventions recommended for individuals with ADHD.* The RCTs were identified on 15 April 2019 from the Minerva Database, which comprised 244 RCTs investigating the efficacy and safety of pharmacological interventions for ADHD at that time.† All studies were assessed using the Cochrane Collaboration tool for bias and labelled in terms of risk of bias.‡
The dependent variable was the efficacy of the pharmacological intervention in improving the symptom severity in individuals with ADHD, compared with placebo. The independent variable was treatment duration, defined as time of randomisation to completion of the double-blind phase of the RCT; treatment duration was a minimum of 3 weeks. Covariates that could influence the relationship between treatment duration and efficacy of the treatment were also collected.§ Weighted linear regression, moments-based meta-regression and multivariate meta-regression were used to examine the effect of treatment duration on efficacy. Covariates that were likely to affect the relationship between treatment duration and efficacy of the pharmacological treatment were also analysed.
A total of 87 RCTs, enrolling 18,924 individuals with ADHD (mean age 21.9 years; 67.6% male), were included in these analyses. Comorbid oppositional defiant disorder was noted in 28.4% of individuals. The mean baseline ADHD score ranged from mild–moderate to severe (38.3 [range 26.3‒46.9]); 41.6% of individuals had received prior psychostimulant treatment. Many studies (38.9%) investigated psychostimulant treatments with high doses (65.5%) and a flexible dosing regimen (62.4%), and 12.3% of individuals received concomitant psychotherapy for ADHD. The average duration of treatment was 9.0 weeks (range 3‒28 weeks). The majority of studies had a parallel study design (94.3%) and had a commercial sponsor (87.5%). For most studies, efficacy was rated by a clinician (92.0%); however, comorbidity inclusion criteria were only included in 20.5% of studies and only 9.4% had a placebo lead-in phase. As determined by the Cochrane Collaboration tool, 40.9% of RCTs were noted as having a high risk of bias in ≥1 domain.
In this study, the authors found that pharmacological treatment reduced ADHD symptoms by ‒7.35 points compared with placebo (95% confidence interval ‒8.06, ‒6.64; p < 0.0001). Treatment duration was positively associated with treatment efficacy (p = 0.055). This relationship remained statistically significant when one and two of the potential confounders (baseline ADHD severity [p = 0.007], comorbidity [p = 0.009] and type of drug [p < 0.001]) were included in a bivariate analysis, but not when all three confounding covariates were included in the analyses, as only baseline ADHD severity was negatively associated with treatment efficacy (slope = ‒0.250; p = 0.013). There were no statistically significant effects of treatment duration on the efficacy of the psychostimulant drugs α2-agonists (p = 0.530), amfetamine derivatives (p = 0.105), atomoxetine (p = 0.186) and methylphenidate (p = 0.130).
Publication bias is a limitation of any meta-analysis. In this case, because of the inclusion criteria of RCTs lasting at least 3 weeks and using the Diagnostic and Statistical Manual of Mental Disorders – 4th Edition or subsequent versions, more studies investigating psychostimulant interventions may have been excluded as these therapies are older and their acute effects have been studied more compared with non-stimulants. In this study, dose was treated as a binary variable which can be inaccurate, as it should be treated as a continuous variable; however, an equivalent dose of all drugs for ADHD would have had to be available. Total ADHD rating scale (ADHD-RS) score was analysed, instead of individual inattention, hyperactivity or impulsive symptoms, and the possibility that tolerance or sensitisation occurs separately for these symptoms cannot be ruled out. Finally, this study included RCTs with a duration of 3‒28 weeks, so the possibility of an individual developing tolerance to treatment outside of this timeframe cannot be excluded.
The authors concluded that pharmacological treatment was effective in reducing the symptoms of ADHD and there was no change in efficacy over time. Baseline ADHD severity seems to be an important moderator of treatment, which the authors recommend for further research.
*Amfetamine derivatives, atomoxetine, bupropion, clonidine, desipramine, guanfacine, methylphenidate and modafinil
†Published in almost 500 scientific articles, clinical trial registers, regulatory agencies’ reports and industry web pages
‡Risk of biases is evaluated using this tool based on the description and suitability of the following domains: allocation concealment; blinding; incomplete data; selective outcome reporting; sequence generation; and other biases. The risk of bias for each domain was judged as either ‘low’, ‘high’ or ‘unclear’ risk
§Covariates included: age; baseline ADHD symptom severity; comorbidity as inclusion criterion; comorbid oppositional defiant disorder; concomitant psychotherapy; dose; gender; placebo lead-in phase; prior psychostimulant treatment; rater (clinician vs individual with ADHD); sponsor (commercial vs independent); study design (cross-over vs parallel); treatment regimen (fixed vs flexible dose regimen); and type of drug (psychostimulant vs non-stimulant)
Castells X, Ramon M, Cunill R, et al. Relationship between treatment duration and efficacy of pharmacological treatment for ADHD: a meta-analysis and meta-regression of 87 randomized controlled clinical trials. J Atten Disord 2020; Epub ahead of print.