|How efficacious and safe is short-acting methylphenidate for the treatment of attention-deficit disorder in children and adolescents: a meta-analysis
|Schachter H M, Pham B, King J, Langford S, Moher D
To review the efficacy and safety of short-acting methylphenidate (Ritalin), compared with placebo, for attention-deficit disorder (ADD) in individuals aged 18 years and less.
The following sources were searched for a range of dates, the earliest being 1981: MEDLINE, EMBASE, PsycINFO, ERIC, CINAHL, HealthSTAR, Biological Abstracts, Current Contents, and Dissertation Abstracts. There were no restrictions on either the language or publication status of the reports, and searches included a standardised filter to capture randomised controlled trials. The Cochrane Controlled Trials Register in the Cochrane Library and Current Controlled Trials (http://www.controlled-trials.com) were also consulted. In addition, manual searches of the reference lists from randomised controlled trials included in the meta-analysis and from pertinent reviews, and the files of content experts, were conducted. Details of the search strategy including search terms were given in the appendix of the article.
Study designs of evaluations included in the review
Studies needed to be placebo-controlled randomised trials. The grounds for exclusion included reports published earlier than 1981, i.e. before the publication of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM-III) criteria.
Specific interventions included in the review
Methylphenidate. A trial was eligible for inclusion if it compared short-acting methylphenidate with placebo. Across the included trials, thirty-eight different types of methylphenidate intervention (i.e. combinations of dose potency contrast and schedule) were employed. A cointervention was provided in 25.8% of the studies. In 11.3% of the studies, the participants continued to receive a pre-trial intervention: 57.1% special education services and 42.9% nonstimulant drug therapies. The mean duration of the interventions was 3.3 weeks (range: 2 days to 28 weeks).
Participants included in the review
Children with ADD. Trials were eligible for inclusion if they involved children aged 18 years or less with a primary diagnosis of ADD (with or without hyperactivity), which was made in a systematic and reproducible way. Trials with participants who required highly specialised school and/or home environments were excluded, as were studies with participants who were receiving stimulants other than methylphenidate. The median age of all trial participants was 8.7 years (range: 2.4 to 18); these data were taken from 52 of the trials included. The median percentage of males included in the meta-analysis was 88.1% (range: 0 to 100); these data were taken from 59 trials of the 62 trials included. A co-morbid condition was identified in 59.7% of the trials.
Outcomes assessed in the review
No inclusion criteria were specified with respect to patient outcomes. The primary outcomes were the teacher and parent versions of the hyperactivity index (HI-T and HI-P, respectively). Parent and teacher rating data were also abstracted for other global, core and related features of ADD. Self-reported indices of side-effects were also included.
How were decisions on the relevance of primary studies made?
Two assessors independently screened the title, abstract and keywords for each citation to determine whether to retain it. Potentially relevant citations were retrieved and then subjected to a relevance assessment using the inclusion and exclusion criteria. Potentially relevant full-text articles were independently assessed, and the reasons for excluding reports were noted. Any disagreements were settled by forced consensus.
Assessment of study quality
Two measures were used to assess quality. One was the 3-item Jadad scale for randomisation, double-blinding, and the description of withdrawals and drop-outs (see Other Publications of Related Interest no.1). The other was an index of the concealment of treatment allocation (i.e. adequate, inadequate or unclear). One reviewer assessed the quality of all reports that referred to a trial that was included in the meta-analysis.
Two reviewers abstracted the data using a structured form to capture the following: report details, e.g. language; trial details, e.g. design; population characteristics; intervention characteristics; behavioural efficacy; and adverse events. The reports were not masked. The trials included in the meta-analysis were given in an appendix, which is available on the Canadian Medical Journal website (accessed 07/08/2002). See Web Address at end of abstract.
The standard mean differences for all efficacy outcomes in the trials were derived. For trials with multiple doses or stratification, an average effect size was computed across the dose levels or strata. All of the analyses were performed according to the intention to treat principle. The relative risks of side-effects were derived and the associated numbers-needed-to-harm (NNH) were calculated.
Methods of synthesis
How were the studies combined?
The studies were combined in a series of meta-analyses. The treatment effect and adverse event data were combined in meta-analyses using random-effects models. For adverse event data, the authors employed an approach to multi-treatment trials similar to that of Hasselblad (see Other Publications of Related Interest no.3), whereby the percentage of patients experiencing a side-effect or serious side-effect was treated as a dependent variable in a random-effects analysis of variance.
Publication bias was investigated using funnel plots. The degree of asymmetry in the funnel plots was evaluated using the graphical test of Egger et al. (see Other Publications of Related Interest no.4). The 'trim and fill' method of Duval et al. (see Other Publications of Related Interest no.5) was used to estimate the number of observed trials, and to derive the treatment effect estimates adjusted for publication bias.
How were differences between studies investigated?
Some sensitivity analyses (based on trial quality and design)and subgroup analyses (based on intervention length, cointerventions and dose) were performed.
Results of the review
Sixty-two trials with 2,897 participants were included. The majority of the trials (83.9%) were of a crossover design.
Each primary outcome (HI-T and HI-P) demonstrated a significant effect of methylphenidate. The effect size reported by the teacher was 0.78 (95% confidence interval, CI: 0.64, 0.91, whilst that reported by the parent was 0.54 (95% CI: 0.40, 0.67). In sensitivity analyses based on trial quality, the effect sizes reported by the teacher and parent decreased by 9.0 and 5.6%, respectively, for high-quality trials; however, they increased by 26.9 and 24.1% for low-quality studies. Reports with adequate allocation concealment increased the methylphenidate effect by 39.7 and 13% for teacher and parent data, respectively. All effects remained statistically significant. The analyses of study design revealed that only the two methylphenidate effects for crossover trials were statistically significant.
In subgroup analyses, the original HI-T estimates increased for both age categories (less than or equal to 12, and over 12 years of age), whereas the HI-P estimates increased only for the older participants. For both informants, trials exclusively including males exhibited a stronger methylphenidate effect than those in which both genders were represented. Each of the HI-T and HI-P methylphenidate effect sizes was positively related to intervention length for interventions lasting no more than 4 weeks.
For all categories of primary diagnosis, and for both informants, statistically-significant methylphenidate effects were observed. Both the HI-T and HI-P estimates for attention-deficit hyperactivity disorder diagnosis studies decreased their initial methylphenidate effect estimates. The strongest teacher and parent effects were seen for ADD with hyperactivity trials, and mixed diagnosis trials, respectively. For teacher data, the magnitude of the methylphenidate effect and the dose potency appeared to be positively related, suggesting a possible trend.
The NNH for adverse events showed wide variation between participants taking methylphenidate. Prominent NNH results were derived exclusively from parent or self-ratings of five specific events. For a decreased appetite and insomnia-related event to be identified, only 4 and 7 study participants receiving methylphenidate, respectively, were required. Three other important NNH values were observed: all stomach ache events (NNH=9), all drowsiness events (NNH=10) and all dizziness events (NNH=11).
Short-acting methylphenidate had a statistically-significant clinical effect in the short-term treatment of individuals with a diagnosis of ADD, who were aged 18 years and less. This finding may not be robust or completely valid. The extension of the effect beyond 4 weeks of treatment has not been demonstrated. The treatment exhibited a short-term safety profile that requires further investigation.
The authors addressed a clear research question with defined inclusion and exclusion criteria. Their search strategy appeared to be comprehensive with no language restrictions. Attempts were made to find unpublished material and publication bias was formally assessed. All the included studies were placebo-controlled randomised trials, which provide more reliable evidence than observational studies. The trials were quality assessed using published validated scales, but only one reviewer was involved in this process. Details of the individual studies and the participants were not presented in the paper, which makes it difficult to determine the generalisability of the report's conclusions. The studies were analysed appropriately through meta-analyses. The authors investigated heterogeneity, and also performed a series of predefined sensitivity and subgroup analyses to test the robustness of the treatment effects.
Although the research confirmed the findings of other meta-analyses showing a short-term 'methylphenidate effect', the authors acknowledged the limitations of their results. Their results must be treated with caution given the likely presence of publication bias, the influence of trial quality on efficacy estimates, and the dependence of the 'methylphenidate effect' on the outcome measures used. Additionally, the authors used age as a surrogate for cognitive-developmental level and to derive a dose potency typology. They acknowledged that this is controversial. The hyperactivity index measure, which was the primary outcome measure, is neither necessary nor sufficient to diagnose ADD; this suggests that the findings are not readily generalisable to forms of ADD, which exclude disruptive-externalising co-features, and probably not at all to females.
Despite these limitations, this was a well-conducted review which drew appropriate conclusions based on the available evidence. It highlights the need for further long-term trials in this complex area.
Implications of the review for practice and research
Practice: The authors state that methylphenidate has an adverse event profile that requires consideration. For example, clinicians only need to treat four children to identify an episode of decreased appetite. Practitioners should recognise that the benefits and risks associated with methylphenidate need to be carefully reviewed prior to the start of treatment, and monitored vigilantly during treatment. The evidence largely refers to males with a restricted definition of ADD, so there is a need to avoid generalisation of the findings to other populations.
Research: The authors state that there is a need to undertake a large, long-term trial lasting longer than 14 months, to redress the notable methodological and clinical shortcomings of previous efforts. This should determine to what extent, in which terms, and for whom, specific courses of short-acting methylphenidate work efficaciously and safely in the short- and long-term.
University of British Columbia, Therapeutics Initiative; the British Columbia Ministry for Children and Families; Cochrane Collaboration, Cochrane Child Health Field.
Schachter H M, Pham B, King J, Langford S, Moher D. How efficacious and safe is short-acting methylphenidate for the treatment of attention-deficit disorder in children and adolescents: a meta-analysis. CMAJ: Canadian Medical Association Journal 2001; 165(11): 1475-1488
Other publications of related interest
1. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17:1-12. 2. Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias: dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 1995;273:408-12. 3. Hasselblad V. Meta-analysis of multitreatment studies. Med Decis Making 1998;18:37-43. 4. Egger M, Davey SG, Schneider M, Minder C. Bias in meta-analysis detected by a simple graphical test. BMJ 1997;315:629-34. 5. Duval S, Tweedie R. A nonparametric 'trim and fill' method of accounting for publication bias in meta-analysis. JASA 2000;95:89-98.
Subject indexing assigned by NLM
Adolescent; Adult; Age Factors; Anorexia /chemically induced; Attention Deficit Disorder with Hyperactivity /classification /drug therapy /psychology; Bias (Epidemiology); Central Nervous System Stimulants /pharmacology /therapeutic use; Child; Effect Modifier, Epidemiologic; Evidence-Based Medicine; Female; Humans; Male; Methylphenidate /pharmacology /therapeutic use; Randomized Controlled Trials as Topic; Registries; Research Design /standards; Safety; Sensitivity and Specificity; Severity of Illness Index; Time Factors; Treatment Outcome
Date bibliographic record published
Date abstract record published
This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.