In a thorough and evidence-based commentary on Storebø OJ. et al.’s 2015 Cochrane System Review, Dr. Dr. Tobias Banaschewski et al. pull the Cochrane System Review apart, inch by inch, and finally concludes, that Storebø’s work is based on false results and inadequate conclusions, under the title: Trust, but verify!
That’s a new one …
I read scientific publications for hours on end, every day, and in my database of references, I count them around 4,500, and in all of these, I’ve never read anything that comes close to the critique that Storebø et al. endures from Banaschewski et al. in the referenced commentary.
I am, unfortunately, not allowed to print the full text here (due to the abhorrent copyright laws on scientific publications) so I will provide bits and pieces from the full text, to give you an idea of the scope of this massive criticism of Storebø et al.
Quotes from the article
Despite their great therapeutic importance in the treatment of attention deficit hyperactivity disorder (ADHD), psychostimulants such as amphetamine and methylphenidate (MPH) have repeatedly been subject to public debate. A recent review published by a group of researchers led by O. J. Storebo (2015), which illuminates the efficacy and safety of MPH for the treatment of children and adolescents with ADHD, has now reignited this discussion.Banaschewski, T. et al. (2016)
Almost daily, we can read in the media about how awful our ADHD-medication is, how dangerous it is, and how ‘bad parents’ we are, for medicating our children with the ‘insulin for their diabetes’. When Storebø et al. release statements to the media which re-ignite this prejudicial narrative (which is in fact started by, and maintained by, Scientology), we who work to provide better insight into what ADHD is, does and actually feels like, are yet again back to start.
The conclusion of the Cochrane review by Storebo and colleagues is nevertheless surprising, and in terms of the evaluation of the efficacy of MPH, it contradicts previous evidence.Banaschewski, T. et al. (2016)
The available evidence on this is extensive: Between 1962 and 1993 alone, more than 250 reviews and 3000 further single works on the effect of psychostimulants were published. Between 1975 and 2007, the efficacy of these substances for the therapy of children and adolescents with ADHD was examined in over 180 randomized, double-blind, placebo-controlled studies, comprising over 12,000 children and adolescents, the results of which were published in peer-reviewed journals.Banaschewski, T. et al. (2016)
Not only do Storebø et al. create havoc in the public debate, they likewise discredit the entirety of the scientific evidence on ADHD, which today accounts for more than 34,000 individual publications, all working to solve the puzzle, by cooperating and by ‘Standing On The Shoulders Of Giants’ that have gone before them.
They also claim that possible adverse drug reactions (ADRs) were not, or were only insufficiently, assessed. Moreover, the authors conclude that the fundamental benefit of medical treatment with MPH is not sufficiently substantiated.Banaschewski, T. et al. (2016)
The Cochrane review by Storebo et al. (2015a) comprises several meta-analyses, which included a total of 185 randomized, double-blind, placebo-controlled studies with 12,245 patients overall. However, this analysis included only 19 of the 185 studies, with a total of 1,698 patients. Thus, it must be acknowledged that the selection of studies was in part manifestly flawed, and other relevant studies were excluded without sufficient reasoning.Banaschewski, T. et al. (2016)
On the ADHD rating scale, the severity of symptoms is classified from 0 to 54 points −the authors of the Cochrane review erroneously state a range between 0 and 72 points here.Banaschewski, T. et al. (2016)
This is a classic example of the morale and ethics, that is behind Storebø OJ and Simonsen E, and their work. They pick and choose the facts that support their agenda, while withholding relevant, and sometimes even crucial information, from the public, all in lieu of their own personal agendas.
Storebo’s et al. (2015) evaluation of clinical relevance is dubious and must be rejected for several reasons. First, to determine the minimal clinically relevant difference, it is not methodologically sufficient to generalize from a single study to another class of substances – and to then draw far-reaching conclusions from this, with clinical recommendations. This surely does not correspond to the requirements of a Cochrane analysis.Banaschewski, T. et al. (2016)
Moreover, the comparison of group means with the minimal clinically relevant difference, which only allows an estimation of the relevance of individual changes, is unsuitable and misleading for determining the relevance of group means.Banaschewski, T. et al. (2016)
The interpretation of the clinical relevance of the (erroneously too small) calculated effect sizes in the Cochrane review is inadmissible. Derived from the published data, the benefit of treatment with MPH was called into doubt by the authors. This interpretation contradicts all international guidelines and is not based on sufficient evidence.Banaschewski, T. et al. (2016)
These examples are yet another proof of the tricks, they’ve got up their sleeves, using some random fact to build up elaborate assumptions, which they then go on to present as evidence. This works very well, if the reader is not paying attention or lacks insight, but it clearly illustrate their frame of mind to those of us who can actually see right through their bluffs.
Storebo and colleagues (2015a) added an eighth dimension, “vested interest,”which is mentioned in the Cochrane handbook as optional.All studies funded by the pharmaceutical industry or whose study leaders declared potential conflicts of interest are therefore judged as being subject to a high risk of bias, irrespective of their scientific quality in all other aspects –even if no objections were found in all other assessment criteria.Moreover, the authors classified each study for which the assessability was deemed to be unclear in one of the eight domains as subject to a high risk of bias overall, and thus of insufficient quality.Banaschewski, T. et al. (2016)
Consequently, the studies for which no high risk of bias was found in any of the domains, but which, in the authors’view, failed to mention necessary (detailed) information were also rated as being of poor quality in order to definitively rule out a systematic error in one or more areas.
For seven studies (not six, as was wrongly stated), the risk of bias was rated as low in all areas. However, according to the authors, these seven studies are also possibly subject to a systematic error: They claim that the blinding of raters is not assured, as ADRs occur more frequently in patients treated with MPH than those under placebo, which is easily identifiable and can lead to an overestimation of the efficacy and an underestimation of the risks of MPH (“It is likely that the trials initially judged to be at low risk of bias may, in fact, be trials at high risk of bias because methylphenidate gives rise to various prevalent and easily recognisable adverse events, which can lead to loss of blinding and hence can bias the ratings of symptoms, resulting in an overestimation of benefits and an underestimation of harms”; “it was possible for people in the trials to know which treatment the children were taking”).Banaschewski, T. et al. (2016)
Contrary to these assumptions, ADRs such as reduced appetite, sleep disturbances, and difficulty falling asleep are not specifically associated with MPH, but are also seen under placebo conditions. They are not primarily observable in the school context. Should teachers assume in individual cases that children are being treated with MPH due to these phenomena, this also does not necessarily result in an overestimation of efficacy.Banaschewski, T. et al. (2016)
What they are basically saying here, is that because many children with ADHD who are treated with Methylphenidate will experience some degree of reduced appetite, problems falling asleep and sleep disturbances, the teachers (who rate the children in the studies, and who do not know who is on medication and who is on placebo) should somehow deduct that children with these symptoms (I wonder how a teacher can measure my child’s appetite and sleep quality by seeing them a few hours a day), must be medicated, and that that subsequently influences their rating of efficacy. It is so far-fetched that it would be laughable, if it was not such a serious matter, we are examining here.
In summary, the Cochrane review produced a sweeping devaluation of the quality of decades of international research efforts and was undertaken with a degree of stringency inappropriate even for the high requirements of Cochrane articles.Banaschewski, T. et al. (2016)
On the other hand, the authors overlooked the presence of relevant information. Even using the strictest criteria, the reasoning behind the assessment of bias and quality is not sound. The fact that studies that otherwise met all criteria were ultimately also deemed to be of poor quality due to the potential unblinding danger arising from ADRs of MPH proves that the criteria applied by Storebo and colleagues (2015a) are inappropriate and cannot be satisfied.Banaschewski, T. et al. (2016)
Finally, in their own analyses of the hypothetically assumed bias, the authors found no evidence for the presence of a relevant bias, but did not revise their evaluation accordingly.Banaschewski, T. et al. (2016)
Storebo and colleagues (2015a) erroneously report that 72 of the included studies were funded by the pharmaceutical industry. In actual fact, only 58 studies were sponsored by the pharmaceutical industry, some of which were approval studies and studies that were funded by companies that have no commercial interest in MPH as they produce alternative medications.Banaschewski, T. et al. (2016)
In response to this criticism, Storebo falsely claims that none of the studies funded by the pharmaceutical industry showed a low risk of bias in all other areas: “There were no trials with only the ‘vested interest bias’ domain assessed as ‘unclear risk of bias’or ‘high risk of bias’.” Actually, however, at least 11 of the studies were classified as uncertain risk only in the domain of “vested interest.”Banaschewski, T. et al. (2016)
Moreover, the authors could have statistically tested their hypothesis that studies funded by the pharmaceutical industry show larger effect sizes than studies without such funding, but they neglected to do so, or at least did not report this.Banaschewski, T. et al. (2016)
In at least 11 studies, the classification “unclear”or “high”solely in the domain “vested interest bias” led them to be categorized as being of poor quality. The authors did not provide evidence for their assumption that industry-sponsored studies might generally be associated with more positive results in the studies.Banaschewski, T. et al. (2016)
According to Storebo and colleagues (2015a), the results of their meta-analyses suggest that MPH might improve teacher-rated ADHD symptoms, general behavior, and parent-rated quality of life in children and adolescents with ADHD. They claim, however, that the available evidence is of such poor quality that it is impossible to make any firm statements about the size of the effects. Moreover, they believe that it is generally unclear whether treatment with MPH is beneficial (p. 2): “[A]t the moment, the quality of the available evidence means that we cannot say for sure whether taking MPH will improve the lives of children and adolescents with ADHD.”Banaschewski, T. et al. (2016)
The conclusions of Storebo and colleagues contradict all previous meta-analyses and reviews, including those by NICE, which provide evidence of a substantial efficacy of MPH in the treatment of ADHD.Banaschewski, T. et al. (2016)
Storebø et al. are deliberately trying to discredit the consensus, in order for them to play their ‘divide and conquer’ strategy out. Their goal is to pit the parents against the clinicians, so that they can swoop in with their psychosocial treatments, and convince the ill-informed, that ADHD can be cured by talking about it, not by providing the neurochemicals that science have proven to be between 80 and 90% effective for most with ADHD.
The Cochrane review of the efficacy and tolerability of MPH treatment in children and adolescents with ADHD is marked by numerous inaccuracies, errors, and inconsistencies.Banaschewski, T. et al. (2016)
We have demonstrated that the study selection was flawed and undertaken without sufficient scientific justification.Banaschewski, T. et al. (2016)
Storebø et al. are obviously betting on, that people will not actually read their 475 page long publication, and that they by manipulating the results and their interpretations of them, in the media, can gain traction for their agenda, without any resistance. But they were wrong … on that fact also!
Finally, the conclusion that it is methodologically warranted to conduct so-called Nocebo studies, in which a substance which provokes solely undesired effects, claimed to be comparable with those of MPH, is given to the control participants instead of placebo, is extremely dubious from an ethical perspective.Banaschewski, T. et al. (2016)
I’ve written an article on ‘Nocebo, ADHD and Cochrane’ wherein I go into details on what the nonsense about nocebo is all about, so let me just qoute Dr. Russell A. Barkley, Ph.D. from that article, here: […] Third, the requirement that studies must use a “nocebo” shows that the authors are not aware that such studies would also not be permitted by IRBs not only for the reasons given above but also because you cannot inflict suffering and harm on people, via inducing side effects, when there is no offsetting benefit to doing so. I am not aware that any such nocebo exists that can specifically create MPH side effects in the equivalent proportion of people and to the equivalent degree as MPH. And even if such a currently fantasy drug existed no IRB here would permit its use. So the authors are asking for the impossible while judging the existing database against a fantasy. That is not a very fair assessment of the prevailing evidence even if they are technically correct about the short term duration of most trials, done mostly of necessity. […]
Overall, it can be concluded that even Cochrane works should be critically read and verified.Banaschewski, T. et al. (2016)
When I wrote my original article on this study from Storebø et al. back in 2015 when it was first published, I tried to establish a dialogue with the authors to gain insight into their thoughts, but sadly they have ignored me … since …
Therefore I have now written a letter to The Cochrane Institute International asking them for their help in getting to the bottom of this issue, since the authors themselves do not seem to wish to debate this, with me.
Cochrane International have confirmed that they have received my complaint and that they will look into it, so I expect there will be some sort of statement from the authors, on the way.
Banaschewski, T., Gerlach, M., Becker, K., Holtmann, M., Döpfner, M., & Romanos, M. (2016). Trust, but verify. The errors and misinterpretations in the Cochrane analysis by O.J. Storebo and colleagues on the efficacy and safety of methylphenidate for the treatment of children and adolescents with ADHD. Zeitschrift Fur Kinder- Und Jugendpsychiatrie Und Psychotherapie, 44(4), 307–314. https://doi.org/10.1024/1422-4917/a000433