SciELO - Scientific Electronic Library Online

vol.86 issue5Longitudinal study of sphincter control in a cohort of Brazilian childrenFactors predicting early discontinuation of exclusive breastfeeding in the first month of life author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Jornal de Pediatria

Print version ISSN 0021-7557

J. Pediatr. (Rio J.) vol.86 no.5 Porto Alegre Oct. 2010 



Risk factors contributing to childhood poisoning



Carla Luiza Job Ramos;I Helena Maria Tannhauser Barros;II Airton Tetelbom Stein;III Juvenal Soares Dias da CostaIV

IMSc. Chefe, Núcleo de Prevenção e Educação, Centro de Informação Toxicológica do Rio Grande do Sul (CITRS), Fundação Estadual de Produção e Pesquisa em Saúde (FEPPS), Secretaria da Saúde (SES), Porto Alegre, RS, Brazil
IIMD, PhD. Professora titular, Departamento de Ciências Fisiológicas, Universidade Federal de Ciências da Saúde de Porto Alegre (UFCSPA), Porto Alegre, RS, Brazil. Coordenadora, Programa de Pós-Graduação em Ciências Médicas, UFCSPA, Porto Alegre, RS, Brazil
IIIMD, PhD. Professor titular, Departamento de Saúde Coletiva, UFCSPA, Porto Alegre, RS, Brazil
IVMD, PhD. Professor, Programa de Pós-Graduação de Saúde Coletiva, Unisinos, São Leopoldo, RS, Brazil




OBJECTIVE: To investigate whether the lack of knowledge of toxic agents in households is a risk factor for individual unintentional childhood poisoning.
METHODS: The case group (n = 25) was composed of caregivers of children under 60 months of age who underwent accidental oral poisoning and were treated at two reference hospitals in Porto Alegre, southern Brazil, and recorded in the Toxicology Information Center database. The control group (n = 25) was composed of caregivers of children matched for sex, age, and presence in their homes of the same toxic agents found in the case group, who sought emergency medical care at the same hospitals, but for other reasons. A structured questionnaire was administered to verify the following questions: sociodemographic data, clinical history, behavioral antecedents of caregivers, storage of toxic agents, history of previous poisoning accidents.
RESULTS: The children's mean age was 31.8 months (±0.97) and mean height was 93 cm (±11). Families, in both groups, were aware of the toxic action of agents available in their homes; however, caregivers in the control group were twice as likely to have such knowledge compared to the case group. Distraction was 15 times more likely to occur among caregivers of children who underwent poisoning compared to the control group. Storage of toxic agents below 150 cm was approximately 17 times more likely to occur in the group of children who underwent poisoning compared to children in the control group.
CONCLUSIONS: Lack of knowledge of the toxic action of agents stored in households is not a risk factor for childhood poisoning. The attributable risks described in this study indicated that the elimination of other factors, such as distraction and storage below 150 cm, would lead to the prevention of 13 and 19% of poisonings in childhood, respectively.

Keywords: Risk factors, poisoning, accident prevention, home accidents, preschool child, behavior.








Approximately 5 million people died as a result of an accident in 2000, with a mortality rate of 80.9 per 100,000 inhabitants among children under 5 years of age. Of the total, 7% were due to poisoning, this being a leading cause of emergency care.1-3 In 2008, the American Association of Poison Control Centers' Toxic Exposure Surveillance System (AAPCC-TESS) reported 2,491,049 human poison exposure cases; 49.7% of these events occurred among children under 4 years of age.4 However, the actual number is estimated to be around four million cases, since many poisoning cases may go unreported.4,5 That same year, the Brazilian National Poison Data System (Sistema Nacional de Informações Tóxico-Farmacológicas, Sinitox) of Oswaldo Cruz Foundation (Fundação Oswaldo Cruz, Fiocruz) identified that human poisoning is most often due to medicines (29% of the 81,828 recorded cases), of which 34% were unintentional. Of the 23,700 records of poisoning due to medicines, 32% (7,584) were among children aged 0 to 4 years, most of the cases occurring at home.6 Of the 21,615 cases recorded in the Toxicology Information Center of the State of Rio Grande do Sul (Centro de Informação Toxicológica do Rio Grande do Sul, CITRS), Brazil, in 2008, 19,199 (88.8%) referred to human poison exposures, and of these, 25% involved injuries among children under 5 years of age.7

Children are curious, thus opening, eating and drinking what adults recognize as dangerous. Therefore, drugs and supplies are stored in inappropriate places by neglect or lack of information of parents and caregivers. The habit of self-medication leads to the consumption of drugs, increasing the availability of these agents in the home and the risk of children having access to them, thus undergoing poisoning.8-10 Ramos et al.11 identified characteristics of childhood poisoning, such as time of poisoning (6:01pm-10:00pm) and places where toxic agents were found (bedroom floor and/or living room). Sixty-four percent of these cases were evaluated as nontoxic or probably nontoxic, according to data obtained from CITRS Reporting Form and Medical Record.11

Childhood poisoning involves a complex interplay of factors related to the child, toxic substance, environment, family behavior, and/or access to health services. Thus, prevention requires greater knowledge about the specific conditions involved in childhood poisoning.

In order to obtain more information on the behavior of caregivers and children, we conducted an investigation aiming to verify whether the lack of knowledge by caregivers of toxic agents available in the home is a risk factor for individual unintentional childhood poisoning.



This is a case-control study to identify lack of knowledge of the toxic action of agents available in households as a determining factor for poisonings in childhood, through a questionnaire administered to caregivers. Individual accidental childhood toxic exposure in the home (dependent variable) is any event in which individuals are exposed to an agent that may or may not result in poisoning with presence or absence of symptoms.

The research project was analyzed and approved by the Research Ethics Committees of Universidade Federal de Ciências da Saúde de Porto Alegre (UFCSPA), Santa Casa de Misericórdia de Porto Alegre, and Hospital de Pronto Socorro Municipal de Porto Alegre (HPS), Brazil.

The questionnaires were coded, reviewed and double entered into a database, with subsequent comparison to eliminate the likelihood of typing errors and quality control. The data were then analyzed using the Statistical Package for the Social Sciences (SPSS) version 13. According to the pilot study, in the case group 72.7% of children were exposed to toxic agents stored below 150 cm compared to 27.3% in the control group; for that same exposure pattern, the sample size was calculated to be 22 children in each group for a 95% confidence interval (95%CI) and 80% statistical power (odds ratio 0.14). By adding a 15% margin of error, we obtained a consecutive sample of 25 cases and 25 controls, between September 2005 and December 2006.

To compose the case group, children under 60 months of age attending the emergency departments of HPS and Hospital da Criança Santo Antônio (HCSA), in Porto Alegre, Brazil, were selected from CITRS database when recorded as undergoing individual accidental oral poisoning by agents available in their homes. The control group was composed of age-matched children, seeking care in the same emergency departments, with diagnoses other than poisoning, such as respiratory infection, among others. The emergency departments, HPS and HCSA, were selected for their representativeness of pediatric emergency care, 35.7 and 14.3%, respectively, in the city of Porto Alegre, Brazil. CITRS, in operation for over 30 years, is considered a reference center that provides toxicologic guidance, using a 24-hour telephone advice line, for management of poison exposures. This is a state public center linked to the State Foundation for Research in Health (Fundação Estadual de Produção e Pesquisa em Saúde, FEPPS), which also acts in the prevention and education of health professionals, as well as the general population. The dependent variable was defined as individual accidental poison exposure and/or poisoning.

The evaluation of risk factors for a positive poisoning event was conducted according to Ramos et al.11 The analysis model maintained the hierarchical configuration, privileging socioeconomic determinism over other variables. The univariate analysis compared groups using the chi-square test (95%CI). Models used in the multivariate logistic regression analysis contained variables with statistical significance in the univariate analysis (p < 0.1). Independent risk factors were assessed after adjustment for confounding variables (socioeconomic and educational factors) in the same model, using both crude and adjusted odds ratio (OR). The presence of confounding bias was detected through changes in OR before and after adjustment.

Based on the frequency obtained in the multivariate analysis, in relation to the data found by Brayden for the variable "distraction,” study population attributable fractions were calculated for the variables: storage of agents below 150 cm and caregiver distraction while taking care of children in their homes.12 The attributable fraction was estimated by the formula Pe (OR - 1) / 1 + Pe (OR - 1), where Pe is the proportion of population exposure to distraction (13%), OR is the odds of disease in exposed individuals/odds of disease in non-exposed individuals.12-14

The parents or legal guardians signed a written consent form in the hospital, and a structured questionnaire, with 81 questions and mean duration of 20 minutes, was then administered.11,12 The independent variables analyzed were divided into the following sections: sociodemographic data (questions regarding children's sex, age and height); children's legal guardian (educational level, socioeconomic status, degree of kinship); family composition (presence or not of other children and adults older than 50 years). Socioeconomic status was stratified into seven levels, based on the classification by the Brazilian Association of Survey Companies (Associação Brasileira de Institutos de Pesquisa de Mercado, ABIPEME).15 These were stratified into two categories for further analysis using descriptive statistics.

Clinical history, storage site of toxic agents, and history of previous poisoning were considered as independent variables. Behavioral antecedents of caregivers were composed of six questions regarding the caregiver behavior (distraction, closely watching the children, going through a stressful period, what type of stress, among others). These data were based on a questionnaire administered by Brayden, subjected to the processes of translation and back-translation.12

Regarding knowledge of the toxic action of agents available in the home, the following questions were considered: storage site of the agent (box, cabinet, and box in a cabinet); height at which the agent is or was stored (above 150 cm); believe that the agent is dangerous to the child; be aware of the existence of CITRS; and analyze a list of 14 products considered dangerous to children, in which all or at least 11 possibly toxic agents should be identified. These questions were classified, according to the analysis of data obtained after collecting information from caregivers, as follows: 0 - all questions were incorrect; 1 - one incorrect question; 2 - two incorrect questions; 3 - three incorrect questions; 4 - four incorrect questions; 5 - all questions were correct.



Of all children who sought emergency medical care, the mean age was 31.8±0.97 months (case: 32.6±11.9; control: 31±11.9) and mean height was 93±11 cm (case: 93.4±10.5; control: 93.6±11.7). Children's distribution according to sex was similar: 13 boys and 12 girls in both groups. Most children did not attend day care (76%).

Table 1 summarizes the results of the questionnaire, which revealed that most caregivers were the child's parents (84%), 76% of these were mothers, in both groups. Among the families interviewed, 64% in the case group and 60% in the control group belonged to lower socioeconomic classes. Seventy-two percent of mothers of children seeking medical care after poison exposure had completed high school or higher education, whereas exposure to the same factor in the control group was 68%.
 Although without significant differences, families in which poisoning accidents occurred were twice as likely to have more than one child in the home. In 44% of households where childhood poisoning occurred, families were composed of at least one adult older than 50 years.

Seeking medical care after poison exposure occurred after an average of 12 hours (±3 hours and 34 minutes) in the case group. In the control group, seeking medical care occurred after 16 hours and 20 minutes (±5 hours and 25 minutes) of onset of symptoms. Poisonings were most often due to medicines (72%), more often due to centrally acting drugs. Toxic exposures were evaluated by members of the CITRS team on duty as nontoxic (12%), probably nontoxic (36%), poisoning not excluded (8%), and poisoning (44%). Of the 11 poisoning cases, 72% had mild and 18% had moderate poisoning. All children, in the case and control groups, were cured. In poisoning due to centrally acting drugs, 44% of cases required hospitalization and 67% were rated as mild or moderate poisoning. 

Most caregivers believed that the agents available in the home were dangerous (82%), did not overestimate the child's ability to handle potentially toxic products (90%) and reported not to be experiencing stress (60%). The analysis of mother distraction revealed that mothers concomitantly performed household chores, such as cleaning the house (40%), cooking (15%), and taking care of another child (9%). As for the stress that mothers were experiencing, we identified problems related to difficulties at work or unemployment (40%), diseases (20%), marital problems (10%), death of relative or friend (5%), birth (5%), and others (20%).

Regarding storage of agents, products were most often found in the bedroom (56%) (Table 1).

Storage of toxic agents below 150 cm was 13 times more likely to occur in the group of children who underwent poisoning compared to children in the control group, when analyzing agents separately.

The data showed that families, in both groups, were aware of the toxic action of agents available in their homes; however, caregivers of children seeking care for reasons other than poison exposure were twice as likely to have greater knowledge compared to caregivers of children who underwent poisoning (Table 1). There is evidence of an association of behavioral antecedents of parents (distraction), storage of agents (height), and knowledge of the toxic action of agents available in the home with childhood poisoning, when controlled for the variables that make up the risk factor "lack of knowledge.” We also found an association for the variable "mother's education,” reclassified as having studied up to incomplete high school (1) and high school up to higher education (2), as a risk factor.

Distraction was 15 times more likely to occur among caregivers of children who underwent poisoning compared to the control group (OR = 15.39; 95%CI 2.31-102.30). Storage of toxic agents below 150 cm was approximately 17 times more likely to occur in the group of children who underwent poisoning compared to children in the control group (OR = 16.59; 95%CI 2.86-96.20).

The attributable fractions were 19% for storage of agents below 150 cm and 13% for caregiver distraction while taking care of children in their homes.



Childhood poisonings occur most often among both boys and girls under 5 years of age, with a peak between 1 and 2 years of age, due to medicines, mainly painkillers. However, in our study, there was increased frequency of poisoning due to centrally acting drugs, especially antidepressants. These events occur regardless of socioeconomic class, educational level of mothers, seasonality, age, and sex.10,11,16,17 Studies have reported that children who are poisoned are more likely to belong to families with few social resources.8,10

The lack of significant difference between groups with respect to socioeconomic status of families suggests that the magnitude of this variable has no statistical power. Other factors may be attributed, such as the presence of more than one child or an adult older than 50 years in the home, caregiver distraction, and improper storage of agents. These findings run contrary to variations in the relationship between socioeconomic status and childhood injuries and to data from the literature.18,19

Although most mothers in the group of children who underwent poisoning had higher educational level, there was no significant difference in comparison to mothers in the control group. Similar data were found in a study on ingestion of medicines among children, conducted by Béria et al.,20 in which 62% of mothers of children who ingested medicines had 4-10 years of schooling. It is worth mentioning that children have the habit of imitating what adults do. Toxic exposures occur as a result of various factors and understanding these factors and their interrelationships allows a better interpretation of data and more accurate preventive measures.

Accessibility (improper storage site), i.e., storage of agents below 150 cm, was 16 times more likely to occur in the case group than in the control group. This was also identified by Petridou et al.,21 who, through a questionnaire administered to parents of hospitalized children, concluded, in their study, that absence of at least one parent appeared to be associated with poison exposures, as well as product accessibility, increasing the risk of poisoning among children.

Although most caregivers showed higher frequency for the variable "be aware of the toxic action of agents available in the home,” parents of children who did not undergo poisoning were twice as likely to have greater knowledge, as well as the fact that the variable "caregiver distraction” proved to be a significant risk factor for these events. In this sense, Brayden et al.12 found that 13% of parents were distracted at the time of the poisoning accident. The authors also noted that poisonings occur due to several factors, such as improper storage and curiosity of children. However, lack of knowledge by caregivers of the toxic action of agents available in the home and of preventive measures does not appear to be an important condition associated with poisoning between 3 and 7 years of age. Likewise, Baltimore & Meyer,22 in a case-control study of 104 poisoned and non-poisoned children, found no significant differences in storage pattern or knowledge of toxicity of ingested substances. De Souza & Barroso,23 as well as Nedel,24 report that carelessness of those responsible for protection and safety of children often becomes a cause of hospital records.

The use of frequency rate ratio of risk factors in multivariate analysis allows us to calculate attributable fractions among categories, which may represent an important contribution to the projection of public health programs as they can estimate the percentage of poisoning events occurring among children, which could be reduced if exposure was reduced as well.25,26 Thus, the results show that intersectoral collaboration is important and necessary. A possible contribution concerns behavioral changes in families of children younger than 5 years through correct storage (above 150 cm) of potentially toxic agents available in households, as well as the importance of providing caregivers guidance on distraction while taking care of children at their homes.

Medicines are usually the main cause of poisoning among children, which led to the law requirement that any medicine package and package insert must contain the warning label: "Keep out of reach of children.” Poisoning usually occurs because these products are left within the reach of children, but also due to other factors. In the Cochrane systematic review, Kendrick et al.,27 in relation to countermeasures and safety equipment against injuries and poisoning, reported as a variable of protection against drug poisoning the storage of medicines in enclosed spaces, such as cabinets and drawers, at a height above the waist of an adult.27,28 However, even if these interventions prove to be effective, many children still get involved in poisoning accidents.29-31 The fact is that, thus far, in countries such as Brazil, laws have not been sanctioned requiring the mandatory packaging with a protective cover for children: specific laws to protect children should be immediately implemented. This study presented evidence that reinforces and provides a basis for public campaigns targeting the prevention of childhood poisoning.



The authors indicate that the elimination of factors such as distraction of caregivers of children under 5 years of age and storage of toxic agents below 150 cm represents the prevention of 13 and 19%, respectively, of poisonings among children aged 0 to 4 years.



The authors thank Dr. Robert Brayden, MD, for having encouraged this study and provided the data from his research.



1. Penden M, Mcgee K, Sharma G. The injury chart book: a graphical overview of the global burden of injuries. Geneva, World Health Organization; 2002.         [ Links ]

2. United Nations Children's Fund. A league table of child deaths by injury in rich nations, Innocenti Report Card n° 2. Florence, Italy: UNICEF Innocenti Research Centre; 2001.         [ Links ]

3. Jorge MH, Koizumi MS. Gastos governamentais do SUS com internações hospitalares por causas externas: análise no Estado de São Paulo, 2000. Rev Bras Epidemiol. 2004;7:228-38.         [ Links ]

4. Bronstein AC, Spyker DA, Cantilena LR Jr, Green JL, Rumack BH, Giffin SL. 2008 Annual Report of the American Association of Poison Control Centers' National Poison Data System (NPDS): 26th Annual Report. Clin Toxicol (Phila). 2009;47:911-1084.         [ Links ]

5. Faraoni F, Protano C, Bini V, Lizzi R. The prevalence of accidental poisoning in a hospital pediatric unit of Latium. Ann Ig. 2006;18:207-13.         [ Links ]

6. Sistema Nacional de Informações Tóxico-Farmacológicas. Casos registrados de intoxicação humana e envenenamento: Brasil, 2008. Acesso: 23/04/2010.         [ Links ]

7. Rio Grande do Sul. Secretaria Estadual da Saúde. Centro de Informação Toxicológica. Toxicovigilância - toxicologia clínica: dados e indicadores selecionados Rio Grande do Sul 2008-2009. Porto Alegre, Brazil: CITRS; 2009.         [ Links ]

8. Bucaretchi F, Baracat EC. Exposições tóxicas agudas em crianças: um panorama. J Pediatr (Rio J). 2005;81:S212-22.         [ Links ]

9. Bochner R. Papel da Vigilância Sanitária na prevenção de intoxicações na infância. Rev Bras Vigilancia Sanit. 2005;1:50-7.         [ Links ]

10. Mintegi S, Fernández A, Alustiza J, Canduela V, Mongil I, Caubet I, et al. Emergency visits for childhood poisoning: a 2-year prospective multicenter survey in Spain. Pediatr Emerg Care. 2006;22:334-8.         [ Links ]

11. Ramos CL, Targa MB, Stein AT. Perfil das intoxicações na infância atendidas pelo Centro de Informação Toxicológica do Rio Grande do Sul (CITRS), Brasil. Cad Saude Publica. 2005;21:1134-41.         [ Links ]

12. Brayden RM, MacLean WE Jr, Bonfiglio JF, Altemeier W. Behavioral antecedents of pediatric poisonings. Clin Pediatr (Phila). 1993;32:30-5.         [ Links ]

13. Last JM. A dictionary of epidemiology. New York: Oxford University Press; 2001.         [ Links ]

14. Rumel D. "Odds ratio": algumas considerações. Rev Saude Publ. 1986;20:253-8.         [ Links ]

15. Associação Brasileira dos Institutos de Pesquisa de Mercado. Critério de Classificação Econômica. São Paulo: ABIPEME; 1996. Acesso: 25/06/2005.         [ Links ]

16. Uziel Y, Adler A, Aharonowitz G, Franco S, Fainmesser P, Wolach B. Unintentional childhood poisoning in the Sharon area in Israel: a prospective 5-year study. Pediatr Emerg Care. 2005;21:248-51.         [ Links ]

17. Schmertmann M, Williamson A, Black D. Stable age pattern supports role of development in unintentional childhood poisoning. Inj Prev. 2008;14:30-3.         [ Links ]

18. Gilbride SJ, Wild C, Wilson DR, Svenson LW, Spady DW. Socio-economic status and types of childhood injury in Alberta: a population based study. BMC Pediatr. 2006;6:30.         [ Links ]

19. Williamson DL, Fast JE. Poverty status, health behaviours, and health: Implications for social assistance and health care policy. Canadian Public Policy. 1998;24:1-25.         [ Links ]

20. Béria JU, Victora CG, Barros FC, Teixeira AB, Lombardi C. Epidemiologia do consumo de medicamentos em crianças de centro urbano da região sul do Brasil. Rev Saude Publica. 1993;27:95-104.         [ Links ]

21. Petridou E, Kouri N, Polychronopoulou A, Siafas K, Stoikidou M, Trichopoulos D. Risk factors for childhood poisoning: a case-control study in Greece. Inj Prev. 1996;2:208-11.         [ Links ]

22. Baltimore CL, Meyer J. A study of storage, child behavioral traits, and mother's knowledge of toxicology in 52 poisoned families and 52 comparison families. Pediatrics. 1968;42:312-7.         [ Links ]

23. de Souza LJ, Barroso MG. Revisão bibliográfica sobre acidentes com crianças. Rev Esc Enferm USP. 1999;33:107-12.         [ Links ]

24. Nedel MN. Acidentes por domissanitários em crianças de zero a quatro anos no município de Balneário Arroio do Silva, Santa Catarina. [Dissertação]. Canoas: ULBRA; 2002. 106 p. Mestrado em Saúde Pública.         [ Links ]

25. Hippisley-Cox J, Groom L, Kendrick D, Coupland C, Webber E, Savelyich B. Cross sectional survey of socioeconomic variations in severity and mechanism of childhood injuries in Trent 1992-7. BMJ. 2002 May 11;324:1132.         [ Links ]

26. Rothman KJ, Greenland S. Modern Epidemiology. Philadelphia: Lippincott Williams & Wilkins; 1998.         [ Links ]

27. Kendrick D, Coupland C, Mulvaney C, Simpson J, Smith SJ, Sutton A, et al. Home safety education and provision of safety equipment for injury prevention. Cochrane Database Syst Rev. 2007:CD005014.         [ Links ]

28. Clamp M, Kendrick D. A randomised controlled trial of general practitioner safety advice for families with children under 5 years. BMJ. 1998;316:1576-9.         [ Links ]

29. Paes CE, Gaspar VL. As injúrias não intencionais no ambiente domiciliar: a casa segura. J Pediatr (Rio J). 2005;81:S146-54.         [ Links ]

30. Kelly B, Sein C, McCarthy PL. Safety education in a pediatric primary care setting. Pediatrics. 1987;79:818-24.         [ Links ]

31. Watson M, Kendrick D, Coupland C, Woods A, Futers D, Robinson J. Providing child safety equipment to prevent injuries: randomised controlled trial. BMJ. 2005;330:178-81.         [ Links ]



Carla Luiza Job Ramos
Centro de Informação Toxicológica do Rio Grande do Sul
Rua Domingos Crescêncio, 132, 8º andar - Santana
CEP 90650-090 - Porto Alegre, RS - Brazil

Manuscript submitted Mar 08 2010, accepted for publication August 03 2010.



Financial support: The authors Helena Maria Tannhauser Barros and Airton Tetelbom Stein received Research Productivity Scholarships from CNPq.
No conflicts of interest declared concerning the publication of this article.
Suggested citation: Ramos CL, Barros HM, Stein AT, da Costa JS. Risk factors contributing to childhood poisoning. J Pediatr (Rio J). 2010;86(5):435-440.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License