Experiment avoidance syndrome

This morning's Taking A Stand on BBC Radio 4 featured twins who were adopted by separate families at birth. The twins had then been enrolled in a psychiatric study -- indeed, they state that the "experiment" was the reason for the separation. For much of the programme it was implied that the purpose of the study was to characterise the various contributions of biology and upbringing in development. That the study was unethical went unquestioned.

Well, no. When the programme actually got around to talking about the study itself, to which it devoted a minute or so (fair enough, they had lots to fit in the half-hour programme), "nature versus nurture" did not seem to be its purpose at all. The study was in fact trying to determine whether it is better for twins to be adopted together or separately, in terms of their development, education and psychiatric health. This was still considered an unethical thing to do -- the guests spoke of their meeting with the principal investigator and his refusal to apologise -- and again no supporting arguments were made. Perhaps the study was unethical -- there are a series of examples of improper doctors and drug companies through history. But I can not judge this one because the details were not described and the programme gave us no pointers for further reading. But I don't think it's adequate just to give us a brief summary and tell us that it was unethical, when actually, it's not at all obvious that it was.

It reminded me of a story that Skrabanek and McCormick tell in Follies and Fallacies in Medicine (they in turn attribute it to the great champion of evidence based medicine, Archie Cochrane ). The story tells of an early randomised controlled trial which was comparing hospital versus home treatment of heart attacks. It was generally assumed that hospital treatment was the effective way to treat heart attacks, and there were only a few maverick specialists arguing otherwise. At a meeting half-way through the trial the results so far were discussed: eight had died at home and four had died in the hospital. "Well, we told you so, and now it would clearly be unethical to continue the trial!" Oops! On closer inspection, it turned out that the opposite was true: four had died at home and eight had died in the hospital! Some tumbleweed blew past, a cough was stifled, and a distant church bell clanged. It was decided that the numbers so far were too small to draw any conclusions from, and the trial went on.^[1]

I was also reminded of last year's "Durham Fish Trials". Supplement pill manufacturers Equazen conducted a "trial" with GCSE students (15 year-olds) in County Durham which looked at the effect of taking fish oil pills on exam performance. The study did not use control groups, and as far as I know, it was never peer reviewed, though it did generate plenty of (mostly favourable) publicity for the company. Parents were arguing that it would not be ethical to have a control group: students in the control group would loose out in their exams. Clearly they had already decided on the efficacy of the pills in advance, and apparently hadn't considered the possibility that fish oil was harmful in terms of exam results.^[2]

Do we have/need a name for this sort of thing? Perhaps the "mind-made-up fallacy", "objection to evidence fallacy", "unethical control fallacy", or the "gut conclusion fallacy".

 

References

1. ^  Skrabanek and McCormick, 1990. Follies and Fallacies in Medicine. Prometheus.
2. ^  Ben Goldacre, The fishy reckoning. The Guardian, Saturday September 22 2007