Original Research: A profile of acute poisoning at selected hospitals in South Africa

A profile of acute poisoning at selected hospitals in South Africa

N Malangu, GA Ogunbanjo

N Malangu, National School of Public Health, University of Limpopo.

GA Ogunbanjo, Department of Family Medicine and Primary Health Care, University of Limpopo.
Correspondence to: N Malangu, National School of Public Health, University of Limpopo (Medunsa Campus), Box 215, MEDUNSA 0204, Pretoria, South Africa.
E-mail: gustavmalangu@gmail.com

The aim of this study was to characterise acute poisoning cases admitted to a number of selected hospitals in South Africa. All cases
admitted to eight hospitals, from January 2005 to June 2005, were evaluated retrospectively. Data obtained from the hospital medical
records included the following: demographic characteristics, toxic agents, length of hospital stay, circumstances of poisoning, morbidity
and mortality information on the poisoned patients. From a total of 424 patients admitted for treatment, whose median age was 17.6 years,
57.8 % were females, and 89.6% black Africans. Fifty-nine percent of the poisonings were accidental, and the involved toxic agents were,
in descending order: household chemicals (45.7%), modern medicines (17.5%), animal/insect bites (15.8%), agrochemical chemicals
(9.7%), food poisoning (5.4%), drugs of abuse (3.3%), traditional medicines (2.4%), and plants (0.2%). Poisoning by drugs of abuse was
commoner in males than females, but the percentage of females poisoned by all other toxic agents was higher than in males. Most patients
spent less than two days in hospital, but more females (70.1%) than males (29.9%) stayed for more than two days. The overall case fatality
rate was 2.4%. Of those who died, 80% were black Africans, aged 13 to 19 years and it was deliberate poisoning through drugs of abuse,
carbon monoxide and agricultural chemicals. Acute poisoning reviewed in some selected hospitals in South Africa revealed that more black
African females were involved, who spent more than two days hospitalised. The case fatality rate was 2.4%, mainly due to drugs of abuse,
carbon monoxide and agricultural chemicals. These findings suggest that further studies are needed to understand the motivation(s) for this
emerging problem and that these should focus primarily on the female black African.

South Afr J Epidemiol Infect 2009;24(2):14-16

Introduction
Acute poisoning is a cause of both morbidity and mortality in many
parts of the world. The toxic agents associated with the morbidity
and mortality vary from place to place, and over time, due to the
availability and use of various chemicals and other poisoning
agents. In South Africa, recent reports suggest that acute poisoning
is responsible for up 17% of total ward admissions in children.1,2
It is generally known that children under 10 years represent up to
80% of all victims of poisoning, and the majority of these poisoning
incidents are unintentional.3,4 It is also known that more male than
female children are victims of poisoning.5-7 Although mortality from
acute poisoning is reportedly low, it is usually high in patients who
are victims of suicide. 8-10 The aim of this study was to characterise
acute poisoning cases admitted to a number of selected hospitals in
South Africa during the first half of 2005. The objectives of this study
were to determine the common toxic agents involved in poisoning,
length of hospital stay, to compare outcomes with regard to gender
and the case fatality rate of the acute poisoning cases.
Methods
The study reviewed 423 cases of acute poisoning admitted to eight
conveniently selected hospitals as part of the experiential attachment
of a group of Master of Public Health students from the University of
Limpopo (Medunsa Campus). Four hospitals were in KwaZulu-Natal,
while the rest were in Gauteng province. Data were collected using
a pre-tested data collection form from records of patients admitted
between January to June 2005. Data obtained included age, gender,
toxic agents, length of hospital stay, circumstances of poisoning,
and whether these poisoned patients had survived or died as a
result of the poisoning. All patients with a diagnosis of poisoning
were included. Toxic agents involved were classified based on their
characteristics. The information regarding the circumstances of the
poisoning incident was obtained from the patients’ files. One data
collector per facility collated the data. Two data capturers entered
the data using MS-Excel 2003 and the data were analysed using
SPSS version 13. Descriptive statistics were used to describe the
sample and p-values less than or equal to 0.05 were considered to
be statistically significant. The statistical test for comparison used
was the chi-square test.11 This test was performed to compare
actual frequency counts against the null hypothesis for bivariate
tabular analyses. Ethics approval for the study was obtained from
the Research Ethics and Publications Committee of the Faculty of
Medicine, University of Limpopo (Medunsa Campus).
Results
The majority of the acute poisoning cases were female, black
Africans, whose median age was 17.6 years (range 1-74 years). The
age group mostly affected was below 12 years of age (36.9%), and

South Afr J Epidemiol Infect 14 2009;24(2)

 Original Research: A profile of acute poisoning at selected hospitals in South Africa

the least affected were those over 30 years (Figure 1). Household aged 13 to 19 years, due to poisoning from carbon monoxide (40%),
chemicals were the most commonly implicated poisons (45.7%), drugs of abuse (40%), and organophosphates (20%), respectively
followed by medicines (17.5%), animal/insect bites (15.8%), (Table 4).
agricultural chemicals (9.7%), food poisoning (5.4%), traditional
Figure 1: Acute poisoning by age group and gender (n=423)
medicines (2.4%) and plants (0.2%) (Table 1). Among the household
chemical agents, paraffin was the most commonly ingested (26.9%),
28.0
0 to 12 years
while among the agricultural chemicals, organophosphates were the 48.6
most frequently implicated agents. Regarding the length of hospital
stay, 70% of the victims of acute poisoning stayed for less than two 13 to 19 years
26.3
13.4
days with the median duration of stay of 1.9 days (Table 2). Based on
the variables reflected in Table 3, the majority of the patients were 28.0
20 to 30 years
black Africans, with more females hospitalised longer than two days, 22.3
and the majority of poisoning incidents due to drugs of abuse (64.3%)
17.7
were in males. The majority of poisoning occurred accidentally, but Over 30 years
15.6
deliberate poisoning was more prevalent in females (48.4%) than in
males (31.3%). The case fatality rate was 2.4%. Of those who died, 00.0 10.0 20.0 30.0 40.0 50.0 60.0
60% were females (Table 3). Eighty percent were black Africans, Percent
Male Female

Table 1: Frequencies of study variables Table 2: Age and length of hospital stay (LOS) parameters
Study variables Percentage Study variables Percentage
Parameters Age in years (n= 423) LOS in days (n=290)
Age category (n=423) Medicines 17.5
Mean 18.1 2.4
0 to 12 years 36.9 Unspecified medicine
10.0
overdose Median 17.6 1.9
13 to 19 years 20.8
Paracetamol 3.5 Minimum 1.0 1
20 to 30 years 25.5
Carbamazepine 1.7 Maximum 74 26
Over 30 years 16.8
Aspirin 1.4
Gender (n=423)
Erythromycin 0.2
Female 57.8 Table 3: Frequencies of study variables by gender
Cotrimoxazole 0.2
Male 42.2 Variables Male Female
Chloroquine 0.2
Race (n=423)
Ascabiol 0.2 Race
Black 89.6
Animal/insect bites 15.8 Black (n=379) 41.2 58.8
Caucasian 7.8
Snake bites 15.1 Caucasian (n=33) 51.5 48.5
Coloured 1.7
Bee stings 0.2 Coloured (n=7) 42.9 57.1
Asian 0.9
Scorpion bites 0.2 Asian (n=4) 75.0 25.0
Length of hospital stay (n=290)
Spider bites 0.2
Less than or equal to 2 days 70.0 Length of hospital stay
Agricultural chemicals 9.7
More than 2 days 30.0 Less than or equal to 2 days (n=203) 46.8 53.2
Organophosphates 5.9
Circumstances of poisoning (n=423) More than 2 days (n=87) 29.9 70.1*
Carbamates 2.6
Accidental 59.0 Circumstances of poisoning
Pyrethroids 0.9
Deliberate 41.0
Insecticides (Baygon®) 0.2 Accidental (n=249) 68.7* 51.6
Outcomes (n=423)
Food poisoning 5.4 Deliberate (n=174) 31.3 48.4
Survived 97.6
Drugs of abuse 3.3 Outcomes
Died 2.4
Cocaine 2.8 Survived (n= 413) 42.3 57.6
Toxic agents involved (n=423)
Marijuana 0.5 Died (n=10) 40.0 60.0
Household chemicals 45.7
Traditional medicines 2.4
Paraffin 26.9 Toxic agents involved
Plants 0.2
Unspecified 9.6 Household chemicals (n=185) 42.7 57.3
Sodium hypochlorite 2.4 Modern medicines (n=74) 44.6 55.4
Alcohol 2.4 Animal/insect bites (n=67) 40.3 59.7
Carbon monoxide 2.1
Agricultural chemicals (n=41) 36.6 63.4
Thinners 1.3
Food poisoning (n=23) 39.1 60.9
Potassium
0.5 Drugs of abuse (n=14) 64.3 35.7
permanganate
Power soap (Omo®) 0.2 Traditional medicines (n=10) 50.0 50.0
Glue 0.2 Carbon monoxide (n=9) 22.2 77.8
Chlorhexidine 0.2 *p<0.05

South Afr J Epidemiol Infect 15 2009;24(2)

 Original Research: A profile of acute poisoning at selected hospitals in South Africa

Table 4: Frequencies of study variables by fatal outcome (n=10) and public education on the topic of acute poisoning is warranted and
Variables Percentage
the involvement of primary healthcare, namely general practitioners,
nurses and pharmacists, is needed. Specific messages directed at
Age category
creating more awareness on the danger of this unnecessary scourge
0 to 12 years 0.0
must be designed and widely circulated.15 Secondly, legal approaches
13 to 19 years 80.0*
should include the enforcement of the existing legislation relating to
20 to 30 years 10.0
the control of medicines. One practical regulatory approach should
Over 30 years 10.0 include measures aimed at strengthening the ethical responsibilities
Race of pharmacists, dispensing doctors and nurses regarding advising
Black 80.0 patients on the safe use and storage of medicines bought over-the-
Caucasian 20.0 counter or dispensed by them. In this study, 17.5% of poisoning
Coloured 0.0 incidents were due to medicines, namely aspirin, carbamazepine,
Asian 0.0 and paracetamol. This finding is consistent with previous reports
Circumstances of poisoning from other institutions in South Africa.1-3, 16
Deliberate 70.0 Finally, this study has some limitations. As a cross-sectional study
Accidental 30.0 design, cause and effect relationships could not be determined. In
Toxic agents involved addition, the use of convenience sampling technique to select facilities
Drugs of abuse (cocaine) 40.0 did not make our findings generalisable as they are not representative
Carbon monoxide 40.0 of all hospitals in South Africa. But the findings provide vital information
Agricultural chemicals 20.0 on the spectrum of acute poisoning, and common causes of mortality
Traditional medicines 0.0
due to the phenomenon as seen in some of our hospitals. Hence
further studies are needed, preferably cohort or interventional studies
Modern medicines 0.0
involving victims of poisoning to reduce the problem.
Household chemicals 0.0
Food poisoning 0.0
Conclusion
Animal/insect bites 0.0
*p<0.05 Acute poisoning affected more black African women, who spent
more than two days hospitalised, and resulted in a case fatality rate
Discussion of 2.4%, mainly due to carbon monoxide, drug of abuse (cocaine) and
organophosphates. These findings suggest that further studies are
The majority of those poisoned were female, black Africans, whose
needed to understand the motivation(s) for this emerging problem
median age was 17.6 years. This demographic distribution of
and that these should focus primarily on the female black African.
poisoning cases is consistent with previously reported findings in
which children younger than 12 years, and black females were the
Acknowledgements
most affected by acute poisoning attributed to their developmental
and contextual vulnerability.12,13 As reported by other investigators, We thank the 2005 Master of Public Health students who were
young boys are twice as likely to be victims of poisoning than girls, involved in the field work.
but during the teenage period, the prevalence of acute poisoning
in females increase almost two-fold compared to their male References
counterparts.1-5 1. Malangu N. Poisoning in children from a rural community in South Africa. South Afr J Epidemiol
Infect 2005; 20(3): 97-102
In this study, the majority of poisoning cases were accidental but 2. Reed RP, Conradie FM. The epidemiology and clinical features of paraffin poisoning in rural African
children. Ann Trop Paediatr 1997; 17(1): 49-55
70% of deaths occurred among those who committed suicide. This
3. Joubert PH. Poisoning admissions of black South Africans. J Toxicol Clin Toxicol 1990; 28: 85-94
finding is consistent with those reported in the literature.10,14 As a 4. Eaton-Jones SE, Fields KB, Rasco T, Akintemi OB. Preventing childhood poisoning: an intervention
in a family medicine residency program. NCMJ 2000; 61(6): 332-334
result of these poisoning incidents, most victims in this study stayed
5. Van der Merwe PJ, Botha JJ. An epidemiological study of poisoning in children under 18 years old
in hospital for less than two days, but the majority of females as in the Bloemfontein area. S Afr Med J 1991; 79(5): 253-255
compared to males stayed for more than two days. This suggests 6. Ho L, Heng JT, Lou J. Accidental ingestions in childhood. Singapore Med J 1998; 39(1): 5-8
7. De Wet B, Van Schalkwyk D, Van der Spuy J, Du Plessis J, Du Toit N, Burns D. Paraffin poisoning in
that women, by virtue of being more likely victims of deliberate childhood - is prevention affordable in South Africa? S Afr Med J 1994; 84(11): 735-738
poisoning than males, experienced somehow more severe incidents 8. Du Plooy WJ, Jobson MR, Osuch E, Mathibe L, Tsipa P. Mortality from traditional medicine
poisoning: a new perspective from analysing admissions and deaths at Ga-Rankuwa Hospital.
that required more observation and/or treatment. This could be S Afr J Science 2001; 97: 70
9. Ellis JB, Krug A, Robertson J, Hay IT, MacIntyre U. Paraffin ingestion: The problem. S Afr Med J
explained by the types of toxic agents involved. The three chemicals 1994; 84(11): 727-730
responsible for deaths, namely carbon monoxide, drug of abuse 10. Abebe M. Organophosphate pesticides poisoning in 50 Ethiopian patients. Ethiop Med J 1991;
29(3): 109-118
(cocaine), and organophosphates are all highly toxic substances. 11. Armitage P, Berry G. Statistical methods in medical research. 3rd ed. Oxford: Blackwell Scientific
Publications; 1994
The findings of this study have a number of important implications. 12. Mathee A, Barnes B. Children, environment and health in South Africa (Editorial). Urban Health and
Firstly, it is crucial to understand the motivation(s) of female black Development Bulletin 2001; 4(2): 1
13. Eddleston M. Patterns and problems of deliberate self-poisoning in the developing world.
Africans involved in acute poisonings before interventions can be QJM 2000; 93(11): 715-731
designed to address the problem as women are both victims and 14. Stewart MJ, Moar JJ, Mwesigwa J, Kokot M. Forensic toxicology in urban South Africa.
J Toxicol Clin Toxicol 2000; 38(4): 415-419
caretakers of children and teenagers who need to be protected from 15. Malangu N, Ogunbanjo GA. Predictors of topical steroid misuse among patrons of pharmacies in
Pretoria. SA Fam Pract 2006; 48(1): 14
acute poisoning. These interventions should be multi-dimensional
16. Young RJ, Critchley JA, Young KK, Freebairn RC, Reynolds AP, Lolin YI. Fatal acute hepatorenal
involving educational, regulatory, and managerial approaches. Media failure following potassium permanganate ingestion. Human Exp Toxicol 1996; 15(3): 259-261

South Afr J Epidemiol Infect 16 2009;24(2)