Looking backwards, a whole bunch of studies which were undertaken may be seen as either astoundingly obvious, or startlingly pointless. Why on earth, for example, would you want to trial the effects of dexamethasone in suspected preterm labour, or see if doxorubicin was a useful to drug in treating sarcoma, in the 21st century? The answer can be either a) it was a waste of research effort or b) it’s all about the context.

It’s true that sometimes clinical studies are done for the researcher and not for the research. We need, as Archimedes has called for with boring regularity, to explore what we already know before jumping into Doing More Research. With increasingly better use of evidence synthesis before new studies, and the involvement of patients and clinicians with academics in designing trials, we are improving.

Mostly though, it’s the context that matters. Over a fifteen years after a conclusive Cochrane review and the widespread adoption of antenatal steroids in threatened preterm labour, a clinical trial was launched comparing steroids to no treatment. Why? Well, the adoption of this practice in resource limited settings had, in post-implementation observational studies, seen an increase in early neonatal death and maternal infection. (These potentially make sense if you consider the immunosuppresive effect of glucocorticoids and reduced ability to detect and respond to infection.)  The findings were uncertain though – observational data often is – and the subsequent trial included over 2800 participants. It showed benefits, outweighing the harms, of antenatal steroids. As we all ‘knew’ it would. Except, if we appreciate the context, we really didn’t know it would find that. When it comes to doxorubicin and sarcoma, we knew it made the tumours shrink, and shrink quickly, and produce a range of fairly negative effects too. We didn’t know if it actually made major things – like survival and relapse – change. We found it didn’t make much difference to those outcomes, and had to swallow the fact we’d been treating unnecessarily for a decade or more.

When you’re getting your ‘push’ clinical studies updates, and you hit a ‘DUH! OF COURSE’, stop a moment and consider the context.

• Archi