Experiencing scientific dishonesty

Scientific dishonesty is becoming the focus of increasing attention from both regulators and educators. An important part of this debate is how to ensure that new recruits to science are educated and socialised in a manner that promotes scientific integrity. Many previously believed that scientific integrity would be absorbed and inculcated automatically, as by osmosis from older peers, but recent research fraud scandals has shown that this process hasn't worked. One reason for this might be that the older peers themselves lack integrity. We have previously in the JME published case reports about the pressures older academics sometimes put on younger academics and we are happy in this issue to publish a Swedish questionnaire study on the knowledge and experiences of PhD students in the health sciences in relation to scientific misconduct (See page 315). Many results in this study are interesting but let me highlight two. Out of the 134 respondents 11 had during the last 12 months been exposed to unethical pressure in relation to authorship issues, and seven in relation to falsification of data. If this is representative of the environment in which people are inducted into research, it is not strange that some end up with a slightly skewed view of what scientific integrity demands!

What is in a word—‘euthanasia’

Another Swedish study published in this issue investigates the way in which large representative samples of members of the Swedish public and Swedish physicians classify withdrawals of treatment and what arguments they use in deciding whether they are morally acceptable (see page 284). The results show that not all acts of withdrawal are viewed in the same way. The responses to the three cases used in the survey were slightly different, especially in the public. This is not altogether surprising since the acts and omission debate is still rumbling on in the background. What is more surprising, and thus worth noting is that among those who classified one or more of the acts as ‘euthanasia’ (which is illegal in Sweden) many still prioritised the arguments in favour of stopping the treatment as the most important arguments, and as the authors state: ‘This suggests that even a value-charged label such as ‘euthanasia’ does not necessarily mean the act being found ethically unacceptable.’

Do physicians discount adequately for conflict of interest when reading the literature?

Readers of the scientific literature know that they should discount for potential conflict of interests, for instance if a study has been sponsored by a firm that has an interest in the outcome (researcher conflict of interest). Similarly they should also discount for the interests of whoever has brought the paper to their attention, for instance if they have been given it by a representative of a firm that sells the product the study is about (presenter conflict of interest). But do readers actually discount for these conflicts of interest? An ingenious study by Silverman et al tries to answer this question (see page 265). It is well worth reading, just for the beauty of its methodology. The study finds that a majority of physicians report that they will discount for both researcher conflict of interest and presenter conflict of interest when they use scientific evidence in their prescribing decisions. However, the study also shows that when presented with cases with or without conflict of interest, willingness to prescribe does not vary according to whether there is a conflict of interest in relation to the evidence or not. The reasons for this important finding are probably complex and are discussed in detail in the paper. One is worth noting here, that even if you know that you should apply some kind of discount to the value of evidence generated or presented with some kind of conflict of interest there is no algorithm or even heuristic that can tell you how much you should discount.

General practitioners and the paramount interests of the child

It is a fundamental principle of English child protection law that the interests of the child should always be paramount, that is that they cannot be outweighed by the interests of any other party or parties. This creates problems for general practitioners who often care for the whole family and not just the child. Wainwright and Gallagher provides a nuanced analysis of these problems and argues that the paramountcy principle is both philosophically unjustified and practically misleading (see page 302). They conclude that:‘Balancing the interests of individuals in the work of safeguarding children is challenging and ethically demanding. Adopting the paramountcy principle uncritically is likely to be detrimental to the interests of all concerned, implying as it does that no other interests are worthy of consideration and that there are, in effect, no conflicts of interest in child protection work. We would argue that, on the contrary, conflicts of interest are a real and inevitable part of child protection practice as they are in all human relations. Paramountcy is an oversimplification and we need to consider interests as scalar - a child's trivial or replaceable interest should not necessarily outweigh another individual's important or crucial interests.’

Informing patients about expensive unsubsidised drugs—the importance of context

Doctors in many healthcare systems will sometimes be in a situation where the treatment they think is best for the patient is not subsidised by the system. One of the specialities where this occurs quite often is oncology, because there are many new and expensive drugs coming to market. But should doctors tell the patient that the drug is available if the patient pays for it privately?

A previous study of Australian doctors showed that few of them would tell their patients, but a New Zealand study in this issue of the JME shows that many New Zealand doctors would (see page 260). In trying to explain this difference the authors make the interesting observation that whereas the general healthcare systems in Australia and New Zealand are quite similar, the way drug subsidy decisions are made are quite different. This lead to the more general point that results from studies such as these cannot be generalised, but must be read against the background of a specific funding environment.