Intended for healthcare professionals

Letters

Ozone and childhood wheezy episodes

BMJ 1996; 312 doi: https://doi.org/10.1136/bmj.312.7046.1607 (Published 22 June 1996) Cite this as: BMJ 1996;312:1607

Needs testing as an a priori hypothesis

  1. Meirion R Evans,
  2. Gillian Richardson,
  3. Layla N Al-Jader,
  4. Michael L Burr
  1. Consultant in public health medicine Registrar in public health medicine Senior registrar in public health medicine Directorate of Public Health, Bro Taf Health Authority, Cardiff CF4 3QX
  2. Senior lecturer in public health medicine Centre for Applied Public Health Medicine, University of Wales College of Medicine, Cardiff CF1 3NW

    EDITOR,—Roger Buchdahl and colleagues conclude from their analysis of associations between air pollution and the incidence of acute childhood wheezy episodes that a critical ozone concentration exists, above or below which children are more likely to develop symptoms.1 Their finding of a relation with very low ozone levels has not been previously described. We have some concerns about the way in which this finding is presented.

    The emergence of the finding from a series of complex statistical tests bears all the hallmarks of a classical a posteriori hypothesis. In effect, the hypothesis was generated by the study rather than the study being set up specifically to test it. Nor do the authors put forward a plausible explanation for the observation. Some caution is therefore necessary in interpreting the finding, and although it is entirely apt to consider its implications in the paper's discussion section, we believe it is inappropriate to give it such prominence in the paper's abstract and key messages. The validity of the finding will remain uncertain until confirmatory evidence is available from further work which tests it as an a priori hypothesis. In the meantime, the finding should be treated with greater circumspection.

    References

    1. 1.