Response to Damkier

In response to the letter by Damkier 1, we welcome the further data identified from cohort study records as described by Dideriksen et al. 2. We note that data from four cohorts are included. We must declare that two of these 3,4 were among the rejected references in our own systematic review 5. While in both cases secondary descriptions of the data were readily available in the cited textbooks, the source data for methylphenidate were unpublished, and in one case could not be obtained despite us writing to the author of the cited ‘personal communication’ as is standard practice in systematic reviews. Indeed, for this data set the perils of relying on secondary sources without having access to the raw data are well illustrated in that there is a disagreement between textbooks 3 and narrative reviews 6 as to the incidence of malformations in children exposed to methylphenidate. In the other series 4, the available data suggested that there were seven malformations in a group of 96 children exposed to any of 17 sympathomimetic drugs, including 11 exposed to methylphenidate. As Dideriksen et al. acknowledge 2, no data were provided to link the seven reports of malformations to any specific drug, and we judged that in the absence of source data specific to the risks associated with methylphenidate, it was not appropriate to include information from this series in our review. We would also question why the 11 children exposed to methylphenidate are listed in main data table of Dideriksen et al. as having no malformations 2 when the data were described as not being specific to individual drugs.

However, we congratulate Dideriksen et al. 2 for identifying two further useful data sources that were beyond the reach of our systematic review, one being current data from an ongoing Swedish birth registry accessed through a public healthcare website and the other a recent congress abstract 7.

While valuable cohort data languish in sources relatively inaccessible to the clinician, the case reports and case series we identified, which generally constitute a lower level of evidence, were accessible in mainstream journals, underlining the point that a publication bias appears to exist in favour of case reports of serious adverse events. As suggested both in our original review and in that of Diederiksen et al. 2, results based on the population described in the papers we identified do not represent the majority of women taking methylphenidate, because all were abusing methylphenidate intravenously, abusing other drugs or taking other known teratogens. We assume that the cohort participants are much more representative of the general population of pregnant women and their children exposed to methylphenidate. However, any estimation of the incidence of malformations in this group relative to the unexposed must still acknowledge the possibility of residual confounding. When studies are unpublished or the source data are inaccessible, it is particularly difficult to be certain that the most important confounders have been identified.

Given that many clinicians will not access data that are unpublished or unavailable in journal articles searchable in online medical databases, we applaud the authors in bringing their recent publication 2 and the valuable cohort study data it contains to the attention of BJCP readers. In the case of baby K, had these data been as readily available to the treating clinicians as the papers we cited, the balance between the risks of continuing methylphenidate in pregnancy and those of stopping it may have been viewed differently. Further prospective cohort data will be required to elucidate the risks of methylphenidate in pregnancy more precisely, with identification of any residual confounding that could still be causing an overestimation of the impact of methylphenidate use. Nevertheless, Dideriksen et al.'s paper 2 strengthens our view that rather than methylphenidate being automatically stopped in pregnancy, its risks and benefits should be discussed, appraised and evaluated with care on a case-by-case basis.

Competing Interests

All authors have completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organization for the submitted work; SJCD had no financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years; BB-A has received speaking fees on one occasion from Janssen Pharmaceuticals; no other relationships or activities that could appear to have influenced the submitted work.