Acute poisoning in the pediatric age group is an important cause of preventable mortality and morbidity. The circumstances of poisoning and related risk factors vary widely across different geographic regions globally due to variable accessibility and availability and varied environmental factors. Mortality due to acute unintentional poisoning among children under 4 years of age varies from 0.3 to 7 per 100,000 people in different countries of the world [
Acute poisoning is an important clinical problem in Sri Lanka and it has a significant economic impact on the health service of the country. The data on financial costs of managing children with unintentional poisoning in rural Sri Lanka, however, are currently unavailable. The significance of this problem lies mainly in the factors predisposing to acute unintentional poisoning. These factors are diverse and include both situational factors (geographic location, social and economic barriers, and culture) and person related factors (personality, lifestyle, parenting style, and education level of parents).
An Asian study [
The literature on risk factors for acute poisoning among children is sparse in the South Asian region. Ahmed et al. (2011) [
This hospital based case-control study was conducted over a period of two years (from February 2012 to January 2014) at Anuradhapura Teaching Hospital. Anuradhapura Teaching Hospital is the largest hospital in the north-central province in which the majority of people belong to a rural community.
This study involved all inpatient children who presented with acute unintentional poisoning and who were between 1 and 5 years of age. Children were recruited as “cases” after their poisoning events were confirmed by caregivers following the initial evaluation at the hospital’s emergency department and subsequently at general pediatric wards. Children with doubtful poisoning and with no clear etiology were excluded from the study. Children who had intentional poisoning were also excluded. Children with past history of poisoning were excluded from both groups. Children with food poisoning, snake envenomation, allergic reactions, and adverse drug reactions which can be considered in the purview of toxicology were also excluded from the study.
Three hundred children were recruited as “cases” over the two-year study period. The “control” group was selected from the same pediatric wards over the same study period. Children, who presented with acute medical illnesses and without any history of chronic medical illnesses or accidental or deliberate poisoning, were interviewed as controls to compare the prevalence of risk factors among the two groups. The acute medical illnesses considered included viral fever, acute upper respiratory tract infection, and urticaria. All other acute conditions including nonspecific symptoms without a definitive diagnosis were excluded. All “cases” were matched for age and gender on individual patient basis. Both groups comprised three hundred children adding to a total of six hundred children. Minimum sample size required for this matched case-control study was 248 pairs (
Data were collected from the caregivers of children recruited to the “case” and “control” groups. Mothers were interviewed in most encounters and fathers or other caregivers (grandparents/other related caregivers) were interviewed only when mothers were not available to participate in the study. Data collection from all six hundred caregivers was done by the principal investigator himself to minimize interviewer bias. Interviews with the caregivers were conducted on the same day of admission to minimize possible recall bias. Data were collected using a pretested structured questionnaire which comprised questions to identify demographic data, type and circumstances of poisoning, and risk factors for acute poisoning (Appendix) and qualitative evaluation via focused group discussions. The study instruments were pretested by administration of the questionnaire to fifty caregivers (twenty-five in each group) in the same study setting over a two-month period prior to commencement of the study. Risk factors were categorized under four domains, environmental, psychosocial, and family related factors and personal characteristics, and twenty-three risk factors were proposed. Extensive literature survey was done to identify previously reported risk factors in other geographic regions. The investigators proposed and designed a risk factor questionnaire themselves and the questionnaire was administered following careful pretesting and expert review.
The outcome of interest was medically attended acute poisoning from medicines, household chemical agents, garden plants, and pesticides present at the child’s home or home garden resulting in hospital admission. Suspicious and doubtful poisoning events were excluded. Twenty-three proposed risk factors were considered in terms of “exposures.” The proposed 23 risk factors were broadly categorized to environmental, psychosocial, and family related factors and personal characteristics. Table
Exposures of interest: the twenty-three proposed risk factors for acute poisoning among children aged 1–5 years.
Risk factor category | Proposed risk factor |
---|---|
(1) Environmental risk factors | (1) Unsafe storage of medicines |
(2) Unsafe storage of household chemicals | |
(3) Unsafe use/storage of agrochemicals | |
(4) Inadequate space in the house | |
(5) Inadequate supervision of the child | |
(6) Poisonous plants in the | |
neighborhood/home garden | |
|
|
(2) Psychosocial risk factors | (1) Psychological problems in parents |
(2) Lack of social support | |
(3) Lack of schooling/education to the child | |
|
|
(3) Family related factors | (1) Father using alcohol or illicit drugs |
(2) Problems with the siblings | |
(3) Incorrect parenting styles | |
(4) Mother employed outside the country | |
(5) Economic problems in the family | |
(6) Mother working during the daytime | |
(7) Poor education in the mother (<grade 8) | |
(8) Young mother (<21 years old) | |
(9) Marital problems among parents | |
(10) Lack of family support | |
(11) Single parent status | |
|
|
(4) Personal characteristics | (1) Developmental problems in the child |
(2) Personality abnormalities in the child | |
(3) Behavioral abnormalities in the child |
Each risk factor was defined prior to inclusion of those risk factors in the questionnaire.
The presence of childhood personality abnormalities was identified by parents’ subjective judgment of the child’s personality as being one or more of the following: shy, timid, aggressive, avoidant, antisocial, overdependent, or any psychiatric illness related personality disorder. Child behavioral abnormalities were defined for the study as one of abnormal behaviors including nightmares, night terrors, nail biting, stammering, abnormal eating or sleeping habits, hyperactivity, impulsivity, and attention seeking behavior. The presence of unsafe storage (medicines, household chemicals, and pesticides) was identified when those compounds were not stored in a lockable container or an inaccessible location to the child. Caregivers’ judgment was considered in determining the presence of economic and marital problems, sibling related problems, inadequate house space, and lack of family and social support. Harmful alcohol use was considered as an adverse exposure. A young mother was defined as a person who was nineteen years old or less at the time of assessment. The presence of a psychological illness in a parent was defined for the study as being diagnosed with a psychiatric illness or having sufficient clinical criteria for a diagnosis of a psychiatric illness based on DSM-V (Diagnostic and Statistical Manual of Mental Disorders). Inadequate supervision was defined for the study as the lack of consistent presence of a principal caregiver (either mother or father) during the child’s stay at the home premises. It was considered as an environmental risk factor based on cultural circumstances given that the home environment and family structure were determining the status of supervision. Incorrect parenting style was defined for the study as one of the nonauthoritative parenting styles including neglectful, permissive, and authoritarian parenting styles.
Developmental delay was defined as a delay of more than six months in achieving milestones in one of four domains of child development up to five years as identified in the Child Health Development Record (CHDR) published by the World Health Organization (WHO). The four domains included gross motor, fine motor and vision, speech, language and hearing, and social and behavioral development. Investigators understood that CHDR provides only crude assessments; however, it was more convenient to use and familiar to the parents of children recruited in the study. Sophisticated developmental assessment tools were not feasible with the study due to limitation of resources and time factor and lack of cooperation from participants.
In order to perform an in-depth analysis of the proposed predisposing risk factors of acute accidental poisoning and to ensure that all the proposed risk factors meet the study definitions, a qualitative study was conducted by the principal investigator himself, recruiting parents of all participant children concurrent with the administration of the pretested, multistructured questionnaire. Each risk factor was evaluated in qualitative terms and in relation to study definitions before the response was recorded in the data collection questionnaire. Data in the qualitative study was recorded as field notes and had emphasis on study definitions. Data collection was done prospectively over two years via focused group discussions.
All data were analyzed using SPSS version 19.0. All twenty-three proposed risk factors were used to create a binary logistic regression model adjusted for age and gender. In this model, each factor was initially evaluated using univariate analysis for significance levels. Controls were kept as the dependent variables and all proposed risk factors were submitted as categorical covariates. Stepwise backward conditional method was applied in the model. Probability for stepwise method was set as entry 0.5 and removal 0.20. Odds ratios were calculated for each risk factor along with 95% confidence intervals (CI). Independent risk factors were identified in the same model by multivariate analysis. Odds ratios were calculated for each of the risk factors along with 95% confidence intervals for each ratio similar to univariate analysis.
Six hundred participants comprising 300 children in each of the “case” and “control” groups were available for analysis. All poisoning events occurred by ingestion of a poison. Household poisoning substances were the commonest type of poison (
All children were between 1 and 5 years of age and had similar sex distributions. The age and sex distributions of the two groups are illustrated in Tables
Age distributions of two groups: “children with acute unintentional poisoning” and “control group.”
Age | Children with unintentional poisoning | Control group |
---|---|---|
1-2 years old | 107 (35.7%) | 107 (35.7%) |
2–4 years old | 153 (51.0%) | 153 (51.0%) |
4-5 years old | 40 (13.3%) | 40 (13.3%) |
|
||
Total | 300 (100%) | 300 (100%) |
Sex distribution of the two groups with “acute poisoning” and “controls.”
Sex | Children with unintentional poisoning | Control group |
---|---|---|
Male | 169 (56.3%) | 169 (56.3%) |
Female | 131 (43.7%) | 131 (43.7%) |
|
||
Total | 300 (100%) | 300 (100%) |
Nine out of twenty-three risk factors showed a significant effect at
Univariate unadjusted analysis of risk factors in the binary logistic regression model.
Proposed risk factor | Cases | Controls | Odds ratio | 95% CI (OR) |
|
|
---|---|---|---|---|---|---|
Low | High | |||||
Environmental risk factors | ||||||
(1) Unsafe storage of medicines | 142 (47.3%) | 72 (24%) | 2.85 | 2.04 | 4.00 | <0.001 |
(2) Unsafe storage of household chemicals | 168 (56%) | 54 (18%) | 5.80 | 4.00 | 8.40 | <0.001 |
(3) Unsafe use/storage of agrochemicals | 58 (19.3%) | 38 (12.7%) | 1.65 | 1.06 | 2.57 | 0.027 |
(4) Inadequate space in the house | 102 (34%) | 67 (22.4%) | 1.79 | 1.24 | 2.57 | 0.002 |
(5) Inadequate supervision of the child | 249 (83%) | 39 (13%) | 32.26 | 20.83 | 52.60 | <0.001 |
(6) Poisonous plants in the home garden | 95 (31.7%) | 23 (7.7%) | 5.58 | 3.42 | 9.09 | <0.001 |
Psychosocial risk factors | ||||||
(1) Psychological illness in parents | 11 (3.7%) | 0 | — | — | — | <0.001 |
(2) Lack of social support | 43 (14.3%) | 42 (14%) | 1.02 | 0.64 | 1.62 | 0.907 |
(3) Lack of schooling/education to the child | 5 (1.7%) | 7 (2.3%) | 0.70 | 0.22 | 2.26 | 0.562 |
Family related risk factors | ||||||
(1) Father using alcohol or illicit drugs | 61 (20.3%) | 87 (29%) | 0.62 | 0.43 | 0.91 | 0.014 |
(2) Sibling disharmony | 12 (4%) | 18 (6%) | 0.65 | 0.31 | 1.38 | 0.260 |
(3) Nonauthoritative parenting styles | 35 (11.7%) | 2 (0.7%) | 19.6 | 4.69 | 83.3 | <0.001 |
(4) Mother employed outside the country | 2 (0.7%) | 5 (1.7%) | 0.40 | 0.08 | 2.06 | 0.270 |
(5) Economic problems in the family | 115 (38.3%) | 155 (51.7%) | 0.58 | 0.42 | 0.81 | 0.001 |
(6) Mother working during the daytime | 68 (22.7%) | 14 (4.7%) | 5.98 | 3.28 | 10.87 | <0.001 |
(7) Primary level education in the mother | 51 (17%) | 15 (5%) | 3.89 | 2.13 | 7.09 | <0.001 |
(8) Young mother (<19 years old) | 37 (12.3%) | 19 (6.3%) | 2.08 | 1.16 | 3.70 | 0.013 |
(9) Marital problems among parents | 23 (7.7%) | 13 (4.3%) | 1.83 | 0.91 | 3.69 | 0.090 |
(10) Lack of family support | 122 (40.7%) | 42 (14%) | 4.20 | 2.82 | 6.28 | <0.001 |
(11) Single parent status | 2 (0.7%) | 3 (1%) | 0.67 | 0.11 | 4.00 | 0.656 |
Personal characteristics | ||||||
(1) Developmental delay in the child | 18 (6%) | 3 (1%) | 6.31 | 1.84 | 21.68 | 0.003 |
(2) Personality abnormalities in the child | 3 (1%) | 2 (0.7%) | 1.51 | 0.24 | 9.09 | 0.656 |
(3) Behavioral abnormalities in the child | 6 (2%) | 2 (0.7%) | 3.03 | 0.61 | 15.10 | 0.175 |
Table
Multivariate analysis of independent risk factors for acute unintentional pediatric poisoning in the binary logistic regression model.
Proposed risk factor | Odds ratio | 95% CI (OR) |
|
|
---|---|---|---|---|
Low | High | |||
Environmental risk factors | ||||
(1) Unsafe storage of household chemicals | 2.70 | 1.49 | 4.90 | 0.001 |
(2) Inadequate supervision of the child | 28.50 | 15.38 | 52.60 | <0.001 |
(3) Poisonous plants in the home garden | 3.67 | 1.70 | 7.93 | 0.001 |
Family related risk factors | ||||
(1) Mother working during the daytime | 7.14 | 2.89 | 17.54 | <0.001 |
(2) Nonauthoritative parenting styles | 12.34 | 2.14 | 71.40 | 0.005 |
(3) Primary level education in the mother | 2.73 | 1.05 | 7.14 | 0.039 |
(4) Lack of family support | 17.54 | 3.65 | 83.3 | <0.001 |
Personal characteristics | ||||
(1) Developmental delay in the child | 7.93 | 1.57 | 40.00 | 0.012 |
In step 8 of the backward conditional approach, eight risk factors were left in the model showing significance of at least
The current study evaluated a broad range of potential risk factors for unintentional poisoning in children aged 1–5 years. The risk factors were broadly categorized under environmental, psychosocial, and family related factors and personal characteristics.
Unsafe storage of medicines and household chemicals [
A European study [
The current study observed no direct or significant association between the economic problems and subsequent poisoning in compliance with other studies [
Unsafe storage of household chemicals and medicines was three and two times more reported among cases than among controls and it was consistent with findings of studies from Pakistan [
In the current study, inadequate supervision was observed six times more commonly among children with unintentional poisoning compared to the control group, and this is consistent with other Asian studies which reported a fivefold higher risk for unintentional poisoning [
The study did not identify a significant association between childhood behavioral characteristics and acute unintentional childhood poisoning. This was consistent with other studies published from Europe [
Other significant findings of the current risk factor study are the implications of the presence of poisoning plants in the home garden, nonauthoritative parenting styles, and developmental delay in the child as major and independent risk factors for acute unintentional poisoning. Evidence from properly controlled studies is scant in the currently available literature regarding these variables and needs further studies. As the majority of unintentional poisoning occurred within home premises in the current study, a holistic approach which targets the household environment would help in managing the burden of poisoning as suggested by other studies [
This study has several limitations in its methodology. The study was hospital based rather than community based. It is likely that the study has not addressed the poisoning events which were not brought to medical attention during the period of the study. Further, the study was conducted only at Anuradhapura Teaching Hospital which has a wider drainage area within the north-central province of Sri Lanka. Though most children with acute poisoning are transferred from local hospitals to the teaching hospital for further management, a fraction of acute poisoning cases are likely to have been not transferred, thus being not taken into account in the current study. The investigators studied twenty-three risk factors in the case-control study design. The questionnaire underwent expert review, piloting, and evaluation by a psychometrician. However, principal component analysis was not performed and internal consistency was not calculated. Thus, the study instrument may have deficiencies in its validity. The questions were carefully selected following expert review and each risk factor was defined for the study. Data collection from all six hundred participants in the current study was carried out by one investigator and interviews were administered as soon as patients were admitted to the pediatric wards. The authors believe that this likely minimized interviewer bias and recall bias in the questionnaire administration. The sample size was determined for the study based on regional data and most risk factors showed acceptable distributions of 95% confidence intervals of odds ratios; however, given the wide distribution of respective parameters in the variable “inadequate supervision of the child,” the authors suggest reanalysis of the same parameter, recruiting a larger sample to increase reliability.
Children become victims of acute unintentional poisoning mostly secondary to inadequate supervision by caregivers, unsafe storage of potentially poisonous substances, and unsafe environment. As these risk factors are significantly associated with unintentional poisoning, the effect of community education to enhance vigilance, safe storage, and assurance of safe environment should be evaluated.
Date of data collection: …/…/… Hospital: … Date of admission: …/…/… Ward: …
Name: … BHT Number: … Age: … Gender: male/female Residential address: … Medical officer of health (MOH) division: … Public health midwifery (PHM) division: … Parent’s education level: Father: … Mother: … Parents’ occupation: Father: … Mother: … Ethnicity: … Religion: …
…
(2.2.1) General/trade name of the poison: …
(2.2.2) Chemical/scientific name of the poison (if available): …
Quantity of poison: … Location of the poisoning event: … Route of poisoning: ingestion/inhalation/direct skin contact/other Transferred hospital:…/not applicable
Proposed risk factors
Present Absent Inadequate space in the house
Present Absent Nonauthoritative parenting styles
Present Absent Unsafe use/storage of agrochemicals
Present Absent Lack of schooling/education to the child
Present Absent Inadequate supervision of the child
Present Absent Unsafe storage of household chemicals
Present Absent Young mother (<19 years old)
Present Absent Marital problems among parents
Present Absent Sibling disharmony
Present Absent Mother employed outside the country
Present Absent Father using alcohol or illicit drugs
Present Absent Unsafe storage of medicines
Present Absent Mother working during the daytime
Present Absent Economic problems in the family
Present Absent Single parent status
Present Absent Personality abnormalities in the child
Present Absent Behavioral abnormalities in the child
Present Absent Primary level education in the mother
Present Absent Lack of family support
Present Absent Psychological illness in parents
Present Absent Lack of social support
Present Absent Developmental delay in the child
Present Absent Poisonous plants in the home garden
Present Absent
The study was granted ethical approval by ethical review committees of the Faculty of Medicine, University of Kelaniya, and Rajarata University of Sri Lanka.
Parents of all participant children provided written consent for participation of their children in the study and publication of results.
The authors declare that they have no conflicts of interest.
M. B. Kavinda Chandimal Dayasiri designed the study, carried out data collection following the appropriate methodology, analyzed data, and wrote the manuscript. Shaluka F. Jayamanne and Chamilka Y. Jayasinghe designed the study, analyzed data, and supervised the manuscript writing process.
The authors of this study would like to thank Dr. Suneth Agampodi, Head, Department of Community Medicine, and Dr. Lalith Senarathna, Senior Lecturer, Faculty of Applied Sciences, Rajarata University of Sri Lanka, for providing technical advice in data analysis. The authors also thank Dr. Thilini Hemachandra and Dr. Chamila Dissanayaka of Anuradhapura Teaching Hospital, Sri Lanka, for providing their support in entering of data into statistical databases.