OBJECTIVE: To examine the occurrence of clinical lead poisoning in England based on routine sources of data. METHODS: Three routine data sources were examined, over different periods according to availability of data: (a) mortality for England, 1981-96; (b) hospital episode statistics data for England, for the 3 years 1 April 1992-31 March 1995; (c) statutory returns to the Health and Safety Executive under the reporting of injuries, diseases, and dangerous occurrences regulations (RIDDOR), also for the period 1 April 1992-31 March 1995. Also, analyses of blood lead concentrations carried out by the Medical Toxicology Unit, Guy's and St Thomas' Hospital Trust in London during the period 1 January 1991-31 December 1997 were examined. The analyses were performed both for industrial screening purposes and in response to clinicians' requests where lead poisoning was suspected. This is one of several laboratories carrying out such analyses in the United Kingdom. RESULTS: One death, of a 2 year old girl, was coded to lead poisoning in England during 1981-96. Analysis of hospital episode statistics data identified 83 hospital cases (124 admissions) over 3 years with any mention of lead poisoning, excluding two with admissions dating from 1965 and 1969. For these 83 cases the median hospital stay per admission was 3 days (range 0-115 days). Five were coded as having received intravenous treatment. Further clinical details of these cases beyond what is routinely recorded on the hospital episode statistics database were not available, except for blood lead concentrations in cases also identified on the Medical Toxicology Unit database. Eighteen cases (22%) were below 5 years of age of whom 10 (56%) came from the most deprived quintile of electoral wards. There was evidence to suggest spatial clustering of cases (p = 0.02). Six occupational cases were reported under RIDDOR in England during the period of study, two of whom were identified on the hospital episode statistics database. One further occupational case was identified on hospital episode statistics. Blood lead analyses for 4424 people carried out by the Medical Toxicology Unit (estimated at about 5% of such analyses in England over 7 years) found that among 547 children aged 0-4, 45 (8.2%) had a blood lead concentration in excess of 25 micrograms/dl, the action level in the United Kingdom for investigation, or removal of environmental sources of lead. At all ages, there were 419 (9.5%) such people, including 106 adults with no mention of industrial exposure. CONCLUSIONS: Both mortality and hospital admission ascribed to lead poisoning in England are rare, but cases continue to occur and some, at least, seem to be associated with considerable morbidity. Lead poisoning was confirmed as a probable cause of clinical signs and symptoms in only a small proportion of those in whom a blood lead concentration was requested. Where indicated, appropriate remedial action for the safe removal of environmental sources of lead should be taken.