Understanding the data.
My research mentor, Professor Michael Hobsley, FRCS, MChir, PhD,DSc
(Med)(1), taught me to play with the data by plotting it and considered
this the real fun in research. That he believed, and I agree, was the best
way of understanding the data.
The editorial board of the NEJM appears to have a different view: the
means of analysis must be decided prospectively for any retrospective
analysis runs the risk of introducing an unacceptable bias. What is more
they have taken a strong stand on the issue retracting at least one
landmark study for this reason and rejecting another which was
subsequently accepted by the Lancet.
Olshansky and Dossey have considered the issue in a different context
(2). "Leibovici", they wrote, "published an intriguing study questioning
conventional notions of time, space, prayer, consciousness, and causality.
The randomised, controlled, double blind, parallel group study (prayer
versus no prayer) included 3393 septic patients and considered the
hypothesis that "retroactive" prayer, offered 4-10 years later, affects
outcomes. Of the preselected outcomes, mortality was similar in both
groups, yet length of stay in hospital and duration of fever were shorter
with prayer (P = 0.01 and P = 0.04). Leibovici, with humour befitting his
style, concluded that remote, retroactive intercessory prayer should be
considered for clinical practice".
Might the target of Leibovici's paper on the effect of retroactive
prayer have been the editorial board of the NEJM? Might the whole issue
even have been manufactured for political or even legal reasons? If so by
whom and for what reason.
2. Brian Olshansky and Larry Dossey
Retroactive prayer: a preposterous hypothesis?
BMJ 2003; 327: 1465-1468
Two or papers rejected by the NEJM
Competing interests: No competing interests