Findings from a US study of severe adverse events after sedation in children have been reported in two papers (Charles J Coté and colleagues.Pediatrics2000;105:805–14 and Ibid 2000:106:633–44).  They used the US Freedom of Information Act to obtain data from the Food and Drug Administration about reports of adverse events received between 1969 and 1996 and also obtained reports of adverse events reported to the US Pharmacopoeia and from a survey of paediatric anaesthetists, intensivists, and emergency medicine specialists. Altogether, they reviewed 118 reports, 95 of which proved suitable for analysis. Fifty one of the 95 children died and nine had permanent neurological sequelae. Fourteen were normal on follow up. Death and neurological damage occurred more frequently with sedation outside hospital (93% in hospital v 37% outside) and were attributed to inadequate monitoring and resuscitation. The importance of pulse oximetry monitoring was stressed.  The second paper concentrates on drug use and adverse events. No single drug or class of drug or method of administration was particularly associated with poor outcome. Use of excessive dose was an important factor. Of 39 instances in which the maximum recommended dose was exceeded by between 25% and 500%, 28 resulted in death or permanent neurological damage. The use of multiple sedating drugs was also associated with adverse outcome. Twenty nine of the deaths followed sedation for dental procedures and these were particularly associated with the use of multiple drugs and of nitrous oxide. Fifteen patients died after being sedated with chloral hydrate, as the sole drug in seven (in overdose or unknown dose in six of these). Twelve of the 95 children suffered the adverse event at home (8) or in a car (4); 11 of these died or were permanently neurologically damaged. Drugs with long half lives (chloral hydrate, pentobaritone, promazine, promethazine, and chlorpromazine) were more likely to be associated with adverse events after leaving medical supervision.  These adverse events after sedation were related not to specific drugs or modes of administration but to overdose, multiple drug use, and failures of clinical practice.