Abstract

The purpose of this study was to describe demographic and clinical characteristics along with medical management and environmental investigation of Montreal children whose laboratory records indicated high blood lead levels. We conducted a community-based case series investigation on Montreal Island, based on laboratories that measured blood lead levels for children sampled at hospital clinics as well as medical records departments of Montreal hospitals that treated children. The study population was made up of children aged 15 and under who had a blood lead level ô25 mg/dL during the years 1981-1989. We measured the children's age, sex and blood lead levels and identified reasons for testing, possible predisposing factors, exposure source, treatment and medical and environmental follow-up. Nine of the twelve children identified were under 3 years old (2 months to 10.75 years). Initial blood lead level had a range of 27-94 mg/dL. Pica and iron deficiency were the most frequently cited predisposing factors. Paint was the principal source of exposure identified. Lead poisoning was the direct cause of hospitalization for three children; two received chelation therapy. Clinical follow-up often appeared incomplete. Only two cases were reported to public health authorities. Between 1981 and 1989, few Montreal children were identified as having a high blood lead level. Management of this problem could be improved by closer collaboration between laboratories, attending physicians and public health authorities.

Key words: Child; environmental exposure; lead poisoning; medical audit; Quebec

Introduction

In 1991, American public health authorities launched a population-wide prevention campaign aimed at eliminating pediatric lead poisoning by the year 2000. 1,2 The extent of this program has elicited various reactions in Canada, where authorities have not considered that lead poisoning constitutes a public health priority for Canadian children.3,4Commentators Godolphin, Schmidt and Anderson recommend that Canada adopt a more modest approach, focusing preventive interventions on the identification and follow-up of children at high risk for overexposure.3

Other than rare case reports,5-7 little is known about how often Canadian children are identified as having high blood lead levels, about the causes of their poisoning and about the quality of their follow-up. In order to describe the level and nature of physician-identified lead poisoning among Montreal children, we instituted a Montreal Island-wide case study covering a nine-year period. Cases were identified through the logbooks of laboratories analyzing blood lead for Montreal hospitals. Information about these cases was extracted from hospital records, from parents and from public health authorities involved in case management. Based on an overview of information gathered about individual cases, we offer a portrait of Montreal children considered to have lead poisoning during the 1980s, and we describe the management of such cases by medical and public health practitioners.

Finally, we discuss the utilization of laboratory registries and of hospital patient records as sources of information to portray management, from a community perspective, of cases of exposure to environmental toxins.

Methods

The target population included children aged 15 and under, living on Montreal Island and having a blood lead level considered as elevated based on public health recommendations current during the study period of 1981-1989. The US Centers for Disease Control's pediatric intervention level for lead in blood was lowered from 30 to 25 mg/dL in 1985 and remained at that level until 1991, when it was further lowered to ô10 mg/dL.2,8,9 We set our case identification threshold at 25 mg/dL to be consistent with public health standards in the study period. Case identification was carried out in two phases, the first in 1986 and the second in 1990.

Telephone calls to all acute-care hospitals on Montreal Island revealed that blood samples for children were taken at the city's two pediatric hospitals as well as at eleven general hospitals. One of the two pediatric hospitals (hospital 1) performed analyses for blood lead both for patients from that institution and for patients from other Montreal-area hospitals. The other pediatric hospital (hospital 2) sent blood samples for analysis to a third Montreal institution (hospital 3), which also analyzed blood samples from other local general hospitals. Finally, the Quebec provincial toxicology laboratory, situated in Quebec (City), occasionally received pediatric blood lead samples from Montreal-area hospitals. Distinct logbook review procedures were implemented in 1986 (phase 1) and 1990 (phase 2). Laboratories in hospitals 1 and 3 reviewed their logbooks themselves for the years 1981-1985, providing us with a list of children (their ages established through medical records) who had a blood lead result ô25 mg/dL (phase 1). We reviewed the hospitals' logbooks for the years 1986-1989. The Quebec provincial toxicology laboratory provided us with computerized lists of all persons having a blood lead level ô25 mg/dL as analyzed between 1987 and 1989, results for previous years not being available (phase 2). Only the earliest entry was retained for each individual identified.

Laboratory logbooks contained no information as to the age of the person sampled. Hospital record librarians at hospitals from which these laboratory analyses originated helped to identify records that came from children 15 years of age and under. The children's medical records were then reviewed.

The American Academy of Pediatrics 10 and the US Centers for Disease Control 9 have issued practice guidelines for the medical and environmental evaluation of children found to have elevated blood lead levels. These guidelines include the following elements: medical history, physical (neurologic) examination, medical investigation and follow-up, environmental sampling, additional case finding and notification of public health authorities. Based on the guidelines, we developed audit criteria (Table 1) to describe patient management.

We contacted Montreal-area community health departments (hospital departments having territorial responsibility for the protection of public health) to determine if cases we identified had been reported to them, such as is stipulated by Quebec law.11 We also determined whether or not public health authorities had been involved in the case investigation.

Finally, when medical chart review and contact with the community health department did not provide all the information necessary, a letter was sent by registered mail to the parents at the child's last address as recorded in the hospital chart. In the letter we asked the parents' assistance in completing and validating information extracted from the hospital records. No attempt was made to contact the child's physician.

Results

Description of Data Sources (Laboratory Logbooks)
The only information available in the three laboratories' logbooks was the date, name of the patient, hospital where the blood sample was taken and that hospital's sample identification number and analytic result. The hospital chart number was recorded only if the sample was taken at the hospital where the testing laboratory was located; hospital chart numbers were not recorded for analyses from other hospitals.

The total number of blood samples received by laboratories for blood lead analysis varied from year to year. For example, hospital 1 received 56 samples in 1988, whereas it received only 9 in 1989. Hospital 3 received 8 samples from pediatric hospital 2 in 1987, 5 in 1988 and 99 in 1989. This rise could be explained by the introduction at pediatric hospital 2 of a program to systematically sample blood lead from children evaluated at the hospital's speech clinic.

Logbook review revealed 125 patients showing a blood lead level ô25 mg/dL. We were able to locate a medical record corresponding to 100 (80%) of the 125 logbook entries. Of these, 24 records pertained to pediatric patients; 12 met our definition of a case and were subject to audit. These 12 cases came from three hospitals, pediatric hospitals 1 (8 cases) and 2 (3 cases) and one acute-care general hospital. The 12 other pediatric reports pertained to children living outside Montreal Island.

We reviewed medical records for the 12 cases. Additional information was obtained from two community health departments. One case was described in the Canada Diseases Weekly Report.7 We asked for additional information from the parents of nine children (including two sisters), and we were able to contact the parents of three children (two families). One parent refused to participate, and five other letters were returned undelivered.

Age, Sex and Blood Lead Level
The study group included seven girls and five boys, of whom nine were 36 months of age or less at the time of their first blood lead analysis (Figure 1). The first blood lead level recorded in the chart varied from 27 to 94 mg/dL. Given that none of these cases had been transferred or referred by another hospital, this can be considered to be the earliest pertinent sample result. The highest result (94 mg/dL) was obtained for an 8-year-old boy under investigation for convulsions.

Reason for Sampling
The reason for testing the children's blood lead levels was ascertained from hospital records and from parents. In six of the twelve cases, blood lead was sampled in response to a history of ingestion of non-food substances ("pica"), particularly paint chips (162). In three of the six cases, anemia was also noted at the time that blood lead was tested. Two cases were family members screened as part of the follow-up of cases previously identified. Blood lead analysis was performed under a variety of scenarios for four cases.

Personal Factors Predisposing to Lead Poisoning
We evaluated factors that might have promoted lead poisoning among the 12 children. It is unlikely that these factors led the treating physician to order a blood lead analysis as they were generally noted only after the blood lead result had been obtained. Pica (10 of the 12 children) and iron deficiency (6) were the factors most frequently noted.

Clinical Manifestations
Clinical signs and symptoms present at the time of first blood lead analysis and that might have been associated with lead poisoning include anemia (10 children), neurological or behavioural problems (3), abdominal pain (3) and vomiting (3). The child with the highest blood lead level was admitted to hospital for convulsions not attributed to lead poisoning by the attending physicians.

Medical and Environmental Follow-up
Six children were admitted to hospital. Lead poisoning related directly to the hospitalization of three of these children; two of them, sampled in 1988 and 1989, were hospitalized on several occasions for chelation. In three cases, blood lead analysis was performed during the course of a hospitalization related to some health problem other than lead poisoning.

Follow-up blood lead samples were taken for seven (58%) of the twelve children; as many as 16 such follow-up analyses were taken per child. No follow-up analysis could be identified for the child who had a blood lead level of 94 mg/dL, despite his having been seen on several occasions as an outpatient during the two years following his first sample (in fact, no note was made in his chart of the blood lead result).

We tried to determine if the source of lead exposure had been looked for, identified and corrected and, where indicated, whether other persons potentially exposed to the same source had been evaluated. According to information available, the source of lead exposure was looked for in eight cases: lead-based paint was the principal source incriminated in seven of these (flaking paint was noted, or mention made of paint dust released during home renovation). Environmental sampling, performed in three cases, confirmed the presence of lead in paint fragments. We are aware of one exposure source having been corrected (in 1989), following the intervention of a community health department. The eighth case was poisoned in utero, following the ingestion by his mother of an Asian folk remedy containing 85 ppm of lead.7

Information retrieved from the hospital charts, from the parents or from community health departments confirmed that blood lead samples were taken from the family members and/or playmates of three children. This sampling allowed elevated blood lead levels to be identified in the sister of one case (this case was also identified by our review of laboratory logbooks). The medical chart indicates that sampling of the family of a fifth child had been planned, but there was no record in the chart of its having been completed. Family members of the child exposed in utero were also tested. For six other children, we were unable to determine if additional case finding was indicated and/or was conducted.

Only two cases, identified in 1988 and 1989, were reported to the local community health departments. In both instances, the community health departments assessed the child's environment and sampled persons potentially in contact with the identified exposure source. Medical charts suggest that two other cases, sampled in 1982 and 1983, were also reported to local community health departments. However, no records of their having been reported could be found at the community health departments named.

Discussion

This study attempted to trace all Montreal Island children for whom laboratory analysis revealed a blood lead level ô25 mg/dL between 1981 and 1989. We offer a demographic and clinical portrait of children so identified and assess the management of pediatric lead poisoning by our local health care system and by public health authorities.

Use of Laboratory Records to Assess the Occurrence of Lead Poisoning in a Community
It is possible that cases of children investigated for exposure to lead escaped our attention. Blood lead analyses may have been performed in public or private laboratories not identified by us or by laboratories not on Montreal Island. Additionally, laboratory results prior to 1987 were not available from the provincial toxicology laboratory. During the first phase of data collection (1986), retrieval of case records from the laboratory of hospital 3 was limited to children whose sample derived from pediatric hospital 2. During the second phase of data collection (1990), the medical records of a non-negligible number of patients sampled at general hospitals could not be found by those hospitals' record librarians. As the laboratory logbooks provided no information on patient age, it is not possible to estimate how many of those patients were children. Despite these gaps, having obtained records of virtually all samples that derived from Montreal's two pediatric hospitals, we believe that ascertainment of cases was as complete as possible given the circumstances.

No epidemiologic studies of children's blood lead levels were conducted in Montreal during the 1980s. Table 2 summarizes results of three studies of blood lead levels in children in Toronto and Windsor (Ontario), Vancouver and Quebec (City) during the study period.12-14 These studies were based on representative samples drawn from entire cities or from neighbourhoods considered at high risk for excessive exposure to lead. Results illustrate the distribution of blood lead levels in the sampled communities. Like our retrospective Montreal Island case series, these epidemiologic studies suggest that few urban Canadian children had blood lead levels ô25 mg/dL during the 1980s.

Still, the 12 cases identified here likely represent only a small fraction of all Montreal children who were exposed to lead between 1981 and 1989 at levels considered excessive by the standards of those years. Because the symptoms and signs related to blood lead at levels around 25 mg/dL are subtle and non-specific (Table 3),15 clinical suspicion alone would likely have led to a limited use of blood lead testing by clinicians.

Audit of Patient Records to Assess Predisposing Factors and to Evaluate Quality of Case Management
Medical audit is widely used to evaluate the management of specific diseases or conditions in institutions.16-18 The identification of cases from several hospitals permits the evaluation of case management at the level of a community rather than at that of an individual institution.

Despite our effort to integrate three sources of information (medical records, parents and community health departments), the medical chart was often the only source of information available. Thus, it is difficult to evaluate the validity and the completeness of information noted in the medical record, particularly related to sampling of family members, investigation of putative exposure source(s) and environmental sampling. It is possible that certain data that we looked for had not been recorded in the medical chart and/or that some children had been followed up in physicians' offices or in other hospitals. Direct contact with the attending physician, which we did not pursue, would allow for more certain and complete assessment of follow-up.

Paint was the exposure source most frequently identified, especially among children aged 3 and under and among those described as having the behaviour called "pica." Although this portrait of exposure corresponds to what has been described in the literature,19,20 it may be biased by the inclination of physicians to evaluate blood lead when presented with a history of pica. As other exposure sources may not initiate the same diagnostic reflex, our portrait may underestimate the importance of sources of lead other than paint. Despite its limitations, this case review provides evidence of deficiencies in individual and community follow-up of children with lead poisoning, as well as offering potential solutions.

The discovery of elevated blood lead should represent a red flag to the health-care system and should trigger a series of interventions both at the individual and community levels. From the point of view of the individual patient, serial blood lead levels should be followed, whether chelation therapy is instituted or not. Further, the source of exposure to lead should be identified and, if possible, eliminated from the environment of the child. There was no evidence in five of the twelve medical records reviewed that blood lead levels were followed up, including that of the child who had a blood lead level of 94 mg/dL. In four cases, we found no information concerning evaluation of an exposure source; in those cases for which an exposure source was identified, we have no record of its having been eliminated.

The management of cases of childhood lead poisoning should also include assessment of risk to siblings and playmates. If they too have been exposed, their blood lead level should also be tested.

From the community's point of view, "tip-of-the-iceberg" cases such as those identified in this study can serve to trigger identification and correction of a larger threat to public health. We found that few cases were reported to public health authorities, but where they were reported, investigation revealed important sources of exposure that went beyond the identified case. An active search for cases at the now recommended blood lead standard of ô10 mg/dL , combined with a high rate of case reporting to public health departments, can be expected to enable the identification of additional exposure sources.

We consider it important both to encourage the recognition of cases by clinicians and to assure that cases of overexposure to lead so identified are reported to public health authorities. According to the regulations falling under the Quebec Law for the Protection of Public Health,11 responsibility for reporting toxic levels of blood lead falls to hospital laboratories. The application of this law is complicated by three factors: the level of blood lead considered toxic is not specified, this level is much lower in children than in adults and patients' age is rarely noted on the laboratory request. In order to increase reporting, demographic information should be included on laboratory requests and treating physicians, as well as hospital laboratories, should be obliged to report.

The collaboration of clinicians is crucial for reporting and for individual case follow-up. Exchange between all groups concerned (laboratories, clinicians and public health authorities) would improve the medical and environmental management of identified cases.

Acknowledgements

We thank Brigitte Côté (Direction de la santé publique de la Régie régionale de la santé et des services sociaux de Laval) for permission to use information gathered during phase 1, and Maureen C-LaPerrière for editorial assistance.

References

1. US Department of Health and Human Services. Strategic plan for the elimination of childhood lead poisoning. Atlanta (GA): Centers for Disease Control, 1991 Feb.

2. Centers for Disease Control. Preventing lead poisoning in young children: a statement by the Centers for Disease Control. Atlanta (GA): Centers for Disease Control, 1991 Oct.

3. Godolphin W, Schmitt N, Anderson TW. Blood lead in Canadian children: a current perspective [see comment]. Can Med Assoc J 1993;148(4):517-9. Comment in: Can Med Assoc J 1993;149(2):138-43.

4. Blood lead levels in children [letter]. Can Med Assoc J 1993;149(2):138-43. Comment on: Can Med Assoc J 1993;148(4):517-9.

5. Tenenbein M. Does lead poisoning occur in Canadian children? Can Med Assoc J 1990;142(1):40-1.

6. Lockitch G, Berry B, Roland E, Wadsworth L, Kaikov Y, Mirhady F. Seizures in a 10 week-old infant: lead poisoning from an unexpected source. Can Med Assoc J 1991;145(11):1465-8.

7. Lecours S, Osterman J, Lacasse Y, Melnychuk D, Gélinas J. Environmental lead poisoning in three Montreal women of Asian Indian origin. Can Dis Wkly Rep 1989;15(35):177-9.

8. Centers for Disease Control. Preventing lead poisoning in young children: a statement by the Centers for Disease Control. J Pediatr 1978;93(4):709-20.

9. Centers for Disease Control. Preventing lead poisoning in young children: a statement by the Centers for Disease Control. Atlanta (GA): Centers for Disease Control, 1985 Jul.

10. American Academy of Pediatrics (Committee on Environmental Hazards and Committee on Accident and Poison Prevention). Statement on childhood lead poisoning. Pediatrics 1987;79(3):457-65.

11. Règlement d'application de la Loi sur la protection de la santé publique, RRQ 1981, c. P-35, r.1, à jour au 6 octobre 1992.

12. O'Heany J, Kusiak R, Duncan CE, et al. Blood lead and associated risk factors in Ontario children. Sci Total Environ 1988;71:477-83.

13. Jin A, Hertzman C, Peck S, Lockitch G. Blood lead levels in children aged 24 to 36 months in Vancouver. Can Med Assoc J 1995;152:1077-86.

14. Levallois P, Weber J-P, Gingras S, et al. Lead exposure of children living in the Quebec city area. In: Trace substances in environmental health-XXIV. Env Geochem Health 1991;13(Suppl):308-14.

15. Toxicological profile for lead. Atlanta (GA): US Department of Health and Human Services, Agency for Toxic Substances and Disease Registry; 1990; ATSDR publ TP-88/17.

16. Abramson JH. Survey methods in community medicine: an introduction to epidemiological and evaluative studies. Edinburg: Churchill Livingstone, 1984.

17. Wood M, Mayo F, Marsland D. Practice-based recording as an epidemiological tool. Ann Rev Public Health 1986;7:357-89.

18. Shannon MW, Graef JW. Lead intoxication in infancy. Pediatrics 1992;89(1):87-90.

19. Garnier R, Chataignier D, Efthymiou ML. Intoxication par le plomb chez les enfants d'âge pré-scolaire: un risque sous-estimé. Toxicologique [Information bulletin published by le Centre de Toxicologie du Québec and le Centre anti-poison du Québec] 1989;5(4).

20. Needleman HL. Childhood lead poisoning: a disease for the history texts. Am J Public Health 1991;81(6):685-7.

Author References

Louise Valiquette, Direction de Santé publique, Régie régionale de la santé et des services sociaux de Montréal-Centre, 1616, boulevard René-Lévesque ouest, 3 e étage, Montréal (Québec) H3H 1P8
Tom Kosatsky, Direction de Santé publique, Montréal-Centre and Department of Occupational Health, McGill University, Montreal

[Table of Contents] [Next]