Most doctors are dedicated clinicians who have worked extremely hard to earn the privilege of practising the art of medicine and caring for their fellow human beings. But there are, unfortunately, always some doctors who don’t live up to the oath they took when they finished medical school. Yet I always feel a mix of uneasiness and sadness when I see a news story in The BMJ about a doctor facing legal or disciplinary action.
It takes a lot of time, hard work, and commitment to build strong patient-doctor relationships, as well as a solid professional reputation as a doctor, but it is far too easy to put everything that has been hard earned on the line or even see it all go down the drain.
Recently, we reported on a headache specialist from the UK who had two charges of dishonesty proved against him by the Medical Practitioners Tribunal Service panel in a research misconduct case. The whole story is very complex, and is apparently making waves because the case is considered “the most important to feature allegations of research misconduct by the General Medical Council since that of Andrew Wakefield, whose work resulted in an international scare over the measles, mumps, and rubella vaccine.”
In our constant efforts to provide educational support to researchers, we have just published a research methods and reporting paper on the ins and outs of the stepped wedge cluster randomised trial. I am aware that the expression “stepped wedge cluster randomised trial” may sound offputting, but it is much simpler than you might think.
Unlike standard parallel cluster trials, in this kind of trial a cluster or group of clusters is randomised at regular intervals to cross from the control to the intervention until all clusters are exposed to the intervention at the end of the study. Why does this matter? Because this study design could be the solution to reaching a compromise when there is a need for rigorous scientific evaluation of an intervention, but when there is also logistical, ethical, or political constraints. I hope some policymakers are reading this, because this study design might just be what they need to evaluate, for instance, the effectiveness of policy changes.
But let’s return for a minute to the good old randomised controlled trial. In his latest endgames statistical question, Philip Sedgwick “dissects” a RCT—which evaluated the effectiveness of a self-management programme for arthritis on the overall function of patients with osteoarthritis—to illustrate the differences between doing an intention to treat or a per protocol analysis of the data.
In the former, the original treatment group composition achieved after randomisation was maintained, regardless of whether participants, for example, withdrew from the trial or were lost to follow-up. An intention to treat analysis reflects real life clinical practice, where patients often do not start, finish, or continue with a given treatment, and it is recommended as standard practice in reporting guidelines.
Finally, I’d like to remind our general practitioner readers to consider sending in cases in the form of case reports or picture quizzes for our upcoming endgames series for GPs. I have explained what it’s all about in a previous blog.
Tiago Villanueva is assistant editor, The BMJ.