Authors' reply

BMJ 1996; 313 doi: https://doi.org/10.1136/bmj.313.7068.1333a (Published 23 November 1996) Cite this as: BMJ 1996;313:1333
  1. P J Fleming,
  2. P S Blair,
  3. I Smith,
  4. P J Berry,
  5. J Golding,
  6. D Bensley
  1. Professor of infant health and developmental physiology Medical statistician Foundation for the Study of Infant Deaths Unit for Research into Infant Health and Development, Institute of Child Health, Royal Hospital for Sick Children, Bristol BS2 8BJ
  2. Senior lecturer Nuffield Institute, Leeds LS2 9PL
  3. Professor of paediatric pathology Department of Paediatric Pathology, St Michael's Hospital, Bristol BS2 8EG
  4. Professor of paediatric and perinatal epidemiology Unit of Paediatric and Perinatal Epidemiology, Institute of Child Health, University of Bristol, Bristol BS8 1TQ
  5. Statistician Operation Research Division, NHS Executive, Leeds LS2 9UA

    EDITOR,—We used several factors to measure socioeconomic status, three of which—occupational classification, weekly family income, and receipt of income support (wrongly identified as family income supplement in our paper)—are shown in table 1. If we stratify for maternal smoking during pregnancy we find, as expected, that the incidence of smoking increases in the more deprived families. We also find, however, that the risk associated with smoking is maintained in each stratum for all three socioeconomic variables; the Mantel-Haenszel test for homogeneity is strongly significant (P<0.001). If we take into account the matching using conditional …

    View Full Text

    Log in

    Log in through your institution


    * For online subscription