A feature appearing recently in a local newspaper referred to an article which appeared in the last issue of the Malta Medical Journal that raised concern that general practitioners were prescribing an atypical antipsychotic to children. This was concluded from a survey (in technical terms – a retrospective study) of the referral tickets which family doctors make when referring children to the Child Guidance Clinic. It is no surprise that the conclusions of such an article appear in the media and indeed one wonders how it did not have more impact. Nevertheless, since the article ‘hit’ the media, probably in an unintentional way, it merits some comment from an ethical point of view as well.
The objective of the study was “to assess the accuracy of the diagnosis and appropriateness of antipsychotic medication prescribed to children and adolescents referred to child guidance clinic, and how many of these remained on the said medication after consultant psychiatrist and multidisciplinary team review”. The authors claimed good interrater reliability, but admit to bias and not communicating with the GPs who issued the tickets – the main notion which undermines the study, as we shall see, while raising an ethical concern.
The results of this study showed that 154 (63.1 per cent) out of a total of 244 children referred, carried an admission diagnosis which differed from the working diagnosis after six months of presumed evaluation by the team and review by a child psychiatrist. The study naturally raises a number of concerns, namely:
• Are general practitioners misdiagnosing children to their detriment?
• Are general practitioners prescribing antipsychotics without consultation with psychiatrists?
• Is there indeed a lack of continuing medical education in this area?
What is said here is therefore not intended in any way to undermine any of the results, but intends to review critically some of the interpretations in the article, to evaluate such conclusions.
It should be pointed out that the authors do not make any reference to Maltese general practitioners, but it may be assumed due to the article being published in the Maltese Medical Journal and the authors all being Maltese. The reason for this observation is that the preamble to the study makes reference to studies outside Malta and with systems that are completely different from the local one. It is dubious therefore how relevant they are to the present study. If anything it points in the direction of a general global problem and not a local one.
The authors admit that one of the limitations of the study was that data was collected retrospectively and that “it was solely collected from medical files without actually interviewing the children”, and that moreover, “No contact with the referring GP was made to acquire any missing data”. There is an ethical concern here that even though the results were anonymised, the data used was still identifiable. People often confuse these points as FP-funded EU projects have been trying to hammer in. Yet I have seen it ignored on research ethics committees so often. The importance of this lies in the context of the conclusion. Clearly if GPs were contacted, they may have given a clear explanation of the drug which the child was given. In particular the authors do not concede that within the Maltese system, a practitioner making a referral must indicate the drugs that the patient is on. In many cases this does not mean that the referring practitioner has actually started the drug.
In my experience, when referring a child to the Child Guidance Unit, the child was already on treatment which was not changed as it was started by a psychiatrist seen privately. It is quite unlikely that general practitioners in Malta actually start children on antipsychotics without referring early on in the course of management to a psychiatrist and only refer when this treatment has not worked. Indeed this sole particular puts into question the whole study by these authors as clearly an interview with the referring GPs would not only have obtained the necessary consent, which evidently was not obtained, but would have clarified whether the GPs themselves had started their patients on this treatment. Consent was still necessary as the researchers do not state that an effort was made by themselves to make the referral ticket information anonymised, and indeed non-identifiable. They do not indicate that GPs were not contacted because of the nature of the study being anonymised. This means that the information at hand was of an identifiable nature. Without going into the merits that therefore of whether informed consent should have been obtained, clearly many GPs would have objected to this ‘non-objective’ used of their referral notes. The point is not so much the ethical issue here, but the validity of the whole study itself.
Indeed GPs usually do consult with psychiatrists, while the authors claim the opposite. This could clearly indicate lack of information on the referral note, which, combined with the admitted observer bias and the non-blind nature of the study, would have been important. It is indeed inexcusable to state that one takes referral notes at face value when one knows of such limits and when one makes such important conclusions. The authors indeed go on to question the “number of unrecorded children on sulpiride within the community”, and directly imply this is the fault and negligence of general practitioners. They conclude that “The misuse of sulpiride by the primary level care doctors is of concern considering that this is an antipsychotic with a number of long term and short term side effects.”
The authors point out that the efficacy of the atypical neuroleptics in children and adolescents is in the treatment of schizophrenia and pervasive developmental disorders. It is also used as an antidepressant in children. However it is clear that the conflict did not seem to be about whether the children referred should have been on an antipsychotic; rather it was the choice of drug. “Atypical antipsychotic medication was prescribed as treatment for children prior to admission to CGC. In all cases but one, the antipsychotic started by the consultant psychiatrist was the newer atypical antipsychotic risperidone”, pointing out however that sulpiride is not indicated in children under 14 years of age.
One may make several conclusions from this analysis.
• The authors make the assumption (not stated) that the referring physician was the same who started the patient on sulpiride. In the Maltese system this is not necessarily the case due to patients not being registered with doctors. This is an admitted weakness of the article by the author but indeed puts into question the conclusions and recommendations drawn.
• Many statements are taken from studies done in a diverse range of countries while the prose of the article seems to intend the local scenario. This could be the ‘bias’ the authors are referring to.
• That the prescription of an antipsychotic seems not to be a problem related solely to Malta but, from the references shown it seems to be an across the board international issue. In particular the problem was prevalent in Australian GPs who have very stringent post-graduate qualifications and re-certification programmes.
• The problem is not restricted to general practitioners and neither has it been conclusively identified as being a primary care issue, unless one includes self-referral to specialists as a primary contact.
• The antipsychotic was usually changed to another modern antipsychotic by the child psychiatrist at the CGC.
• There is no evidence that CME is the problem; it would seem more appropriate to accept the part of the conclusion attesting to better intraprofessional communication.
• The study, having been conducted only upon referral notes, is at most an audit; accepting audits at face value, one would have to accept that many patients are ‘misdiagnosed’, in order to benefit from free medicine.
So where does the answer lie after all this? As has been stated so often in this column, Family Medicine is now a speciality for a reason – it needs specialisation and vocational training. It is the authorities’ responsibility to see that the interpretation of the EU law is done correctly. The government is not only responsible for the training of doctors working within the state-sponsored health centres, but is responsible as well for any doctor it warrants to work in the community, private or otherwise. This has been a specific issue, which, if anything shows the value and moral importance of vocational training. Both positive and negative experience in talking with vocational trainees and indeed from personal experience indicates that when it comes to psychiatry it is more of an interprofessional issue. Don Quixote was a kind and chivalrous person, but he still confused the windmill for the dragon, which, I fear, is what this article did.
Pierre Mallia is Associate Professor in Family Medicine, Patients’ Rights and Bioethics at the University of Malta; he is also Ethics Advisor to the Medical Council of Malta. He is also former president of the Malta College of Family Doctors.
