A recent article by Meadow1 documented some clinical features of 8l cases of infanticide in an attempt to identify particular features that might help. Paediatricians differentiate between natural and unnatural infant deaths.1 In table 1we have documented the prevalence of some of these features, highlighted by Meadow's study, in a contemporary SIDS database (165 cases of SIDS that occurred in the Republic of Ireland between 1994 and 1997) to give paediatricians, pathologists, and coroners, information to assist them in making a judgment on the likely cause of death.

Several features highlighted by Meadow's study as possibly raising suspicions of infanticide are either very common in SIDS (death in the first 7 months, 91%) or, while less common in SIDS than infanticide, are sufficiently common to negate their use as markers of infanticide. These include no previous live healthy child (26%), time of death 1100–2200 (22%), an interval of <2 hours from last been seen and found dead (30%). Blood was found in the mouth, nose, or face in 32/122 cases of SIDS. However, the question did not distinguish between blood stained froth or frank blood and needs to be more rigorously framed in future. Death of a previous child and previous apnoeic episodes or apparent life threatening events were uncommon events in the SIDS group. These situations will undoubtably raise concerns in the future.

Although we are unable to address all the issues raised in Meadow's paper we agree that the way forward must be a thorough investigation of all sudden unexpected infant deaths by professionals experienced in this area, including paediatricians and paediatric pathologists. Anything less will allow the present situation of double jeopardy to continue, with cases of infanticide remaining undiagnosed and, increasingly, many newly bereaved SIDS parents wrongly suspected of fatal child abuse.