An analysis published in this week's British Medical Journal is set to re-open the debate over sudden unexpected infant death (cot death).
The Lancet recently published a study on repeat infant deaths in 46 families, which suggested that almost 90% of second deaths in the same family are natural.
These findings contrasted with earlier studies, which found a much higher proportion of repeat cot deaths were probably homicide. Yet the Lancet study has proved very influential, being accepted by bodies such as the American Academy of Pediatrics.
In December 2006, the BMJ questioned the way these deaths were classified in the study.
Today the BMJ publishes a re-analysis of the data by two senior (now retired) paediatricians, Christopher Bacon and Edmund Hey. They reviewed the 46 second deaths in the original study to see how many might reasonably be regarded as undetermined.
They suggest that, in the three families in the original study in which both deaths were attributed to specific natural causes, one death should be regarded as undetermined.
In 18 families in the original study, both deaths were attributed to sudden infant death syndrome. The authors stated that all these families were at high risk of cot death and that the second deaths exhibited many untoward features, such as violent family relationships, pathology findings suggestive of asphyxia, and parental mental health problems.
Although Bacon and Hey do not suggest that violence in the family, for example, necessarily implies that a baby's death was unnatural, they believe that when a family has two unexplained deaths this possibility at least has to be considered and may sometimes be true. They therefore estimate that a third of these deaths might be classed as undetermined.
For 13 families in the study, information on the second death was incomplete. The original authors classified all these cases as natural, though they acknowledged that the information was insufficient to enable them to distinguish between sudden infant death syndrome and a specific natural cause.
But Bacon and Hey argue that it seems illogical to exclude covert homicide when there is not enough information to exclude a specific natural cause. They categorise all 13 cases as undetermined.
There were six further cases that could not be reviewed because the Lancet paper gave no details about them.