Abstract
Background
Acute poisoning is a significant cause of emergency department visits and mortality, particularly in low- and middle-income countries. However, evidence from urban tertiary care settings in Ethiopia remains limited. Therefore, this study assessed treatment outcomes and determinants of in-hospital mortality among patients admitted to selected governmental hospitals in Addis Ababa, Ethiopia.
Design and Methods
A multicenter retrospective chart review was conducted using medical records of all patients diagnosed with acute poisoning who presented to emergency departments between January 2021 and January 2023 and met the inclusion criteria. Among 372 records reviewed, 371 (99.7%) were eligible for analysis.
Result
A total of 371 patients were included. The median age was 22 years (IQR: 18–28), and 59.3% were female. Intentional poisoning accounted for 82.5% of cases. Pharmaceutical agents (56.1%) were the most common exposures, while organophosphates (11.9%) were associated with the highest proportion of deaths. Overall, in-hospital mortality was 4.3%. Mechanical ventilation was required in 5.9% of patients, and it is a marker of severe poisoning (AOR = 24.01, 95% CI: 5.49–41.96).
Conclusion
Acute poisoning in this urban tertiary care setting predominantly affects young adults and is largely intentional. Organophosphate poisoning and severe clinical presentations, particularly those requiring mechanical ventilation, were associated with increased in-hospital mortality. Mechanical ventilation should be interpreted as a marker of severity rather than a direct determinant. Strengthening early recognition, improving emergency care capacity, and expanding mental health services are essential to improve outcomes.
Background
Poisoning is defined as exposure to a chemical or biological substance capable of disrupting normal human physiology through cellular injury or death. 1 Acute poisoning occurs when toxic agents enter the body through ingestion, inhalation, dermal absorption, or parenteral routes, resulting in damage to tissues and organs. 2 Globally, acute poisoning remains a major cause of emergency department (ED) visits and hospitalizations, contributing substantially to morbidity and mortality. 3
Despite improvements in emergency care, poisoning continues to be the third most common pediatric emergency, leading to significant childhood morbidity and mortality worldwide.4,5 Acute poisoning is typically defined as exposure lasting less than 24 hours and constitutes a major reason for ED admission.4,5 The clinical severity varies depending on the type, quantity, and toxicity of the substances involved, ranging from mild symptoms to life-threatening complications. 6
Low and middle-income countries (LMICs) are disproportionately affected due to limited access to medical care, antidotes, and poison control systems. 7 According to a recent World Health Organization (WHO) report, more than 3 million acute poisoning cases occur annually, with an estimated 200,000 deaths, nearly 90% of which occur in developing countries. 8 Occupational poisoning, particularly from pesticides, remains common in LMICs due to inadequate protective equipment, lack of regulatory enforcement, and limited awareness about toxicity. 9
Poisoning can be intentional (deliberate self-harm) or accidental, with intentional poisoning more common in developing countries.10,11 In Africa, self-poisoning with pesticides accounts for approximately 7,800 deaths annually.10,11 Ethiopian hospital-based studies similarly show frequent intentional poisoning, with case fatality rates ranging from 1.5% to 18.6%. 12 Organophosphates, herbicides, and rodenticides are consistently reported as leading causes of poisoning-related deaths. 13 In Ethiopia, rising poisoning cases are linked to greater availability of toxic substances, lifestyle transitions, psychosocial distress, and limited public awareness. 14 Household detergents and pesticides are among the most commonly implicated agents in multiple studies.15,16
Effective management of acute poisoning requires a multidisciplinary approach involving emergency physicians, toxicologists, nurses, and mental health professionals. Treatment typically includes gastric decontamination, activated charcoal, antidotes when available, and supportive care. 17 Treatment success depends on factors such as the type of toxin, route of exposure, time to presentation, age, comorbid conditions, and timely administration of appropriate interventions. Treatment outcomes range from full recovery to severe complications or death. 18
Most studies conducted in Addis Ababa have been single-center investigations, and no multicenter study has been carried out in the city. Considering the rising incidence of poisoning and the variation in outcomes across different Ethiopian settings, this study aimed to evaluate treatment outcomes and identify determinants of in-hospital mortality among patients admitted to the emergency departments of major government hospitals in Addis Ababa.
Methods and materials
Study setting and participants
This study was conducted in four major government hospitals in Addis Ababa: Tikur Anbessa Specialized Hospital (TASH), St. Paul’s Hospital Millennium Medical College (SPHMMC), Addis Ababa Burn, Emergency and Trauma Hospital (AaBET), and St. Peter Specialized Hospital (SPSH). These facilities were deliberately selected because they are the largest referral and toxicology centers in Ethiopia, receiving patients from Addis Ababa and other regions, thereby providing broad representativeness for urban settings. This purposive inclusion enhances generalizability to similar tertiary-care environments but may not fully represent rural poisoning patterns.
A multicenter retrospective design was conducted and all medical records of patients diagnosed with acute poisoning and treated in the emergency departments between January 2021 and January 2023 were considered. Eligible participants included patients who presented within 24 hours of exposure with complete documentation of essential variables such as demographic characteristics, type of poisoning, time of exposure, treatment provided, and clinical outcomes. Patients with chronic poisoning (>24 hours), incomplete charts, or those brought in dead or dead body on arrival were excluded due to the inability to assess their clinical course or treatment outcomes, which may otherwise introduce classification errors in mortality analysis. Therefore, the primary outcome was in-hospital mortality among admitted patients. The reporting of this multicenter study adheres to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology statement. The completed STROBE checklist is included as Supplementary File 1.
Sample size determination
The sample size was determined using a single population proportion formula, assuming a 32.5% 16 prevalence of poor treatment outcomes based on a previous Western Ethiopian study with a 95% confidence level, 5% margin of error, and 10% allowance for incomplete charts, the final sample size was 372.
Sampling technique
A total of 715 acute poisoning cases were recorded across the four hospitals during the study period (TASH=140, SPHMMC=135, AaBET=100, SPSH=340). Proportional allocation was used to distribute the sample size across facilities. Random selection of charts was ensured through computer-generated random numbers, minimizing selection bias and ensuring representativeness within each hospital’s caseload.
Data collection tool and techniques
A pretested structured checklist format adopted from other similar studies was used to record the necessary information from patients’ medical records.17,18 The checklist was designed in English, and it includes variables such as sociodemographic data: sex, age, marital status, residence, educational level and occupation of the patients; whereas factors affecting the treatment outcome includes route of poisoning, type of poisoning, status of patients when coming to the hospital, reason for taking the poison, time of arrival to the emergency hospital, source of poisoning agents, treatment given for acute poisoning; and treatment outcome. The checklist’s reliability was confirmed with a Cronbach’s alpha of 0.81, indicating strong internal consistency. Content validity was reviewed by emergency medicine and toxicology experts. Two trained data collectors and one supervisor reviewed charts. To ensure inter-rater reliability, they received standardized training, and ambiguous entries were resolved through consensus or supervisor review. Charts with unresolved inconsistencies were excluded.
Data quality control and management
A pretest involving 5% of the sample was conducted at Yekatit 12 Hospital, outside the study area. Necessary modifications were made to improve clarity and completeness. Daily supervision ensured adherence to data extraction procedures.
Data entry, process and analysis
Data were collected using Kobo Toolbox and exported to SPSS version 26 for analysis. Descriptive statistics, including frequencies, percentages, medians, and interquartile ranges (IQRs), were computed to summarize patient characteristics and poisoning patterns. Due to the limited number of outcome events (n=16 deaths), inclusion of multiple variables in the multivariable logistic regression model could lead to overfitting. Therefore, a reduced model was constructed by including only variables with strong clinical relevance and statistical significance in bivariate analysis, in order to maintain model stability and reliability.
Operational definitions
Treatment outcome is defined as either discharge alive or in-hospital death; Poor treatment outcome is defined as in-hospital death due to acute poisoning; Organ damage is defined as renal and/or respiratory injury documented in the patient’s medical charts during hospitalization.
Ethical approval
This study involves human participants and was approved by Institutional Review Board (IRB) of Addis Ababa University, College of Health Sciences with a reference number 08/chs/esn/2024. The need for consent to participate was waived by the IRB since the study was retrospectively chart review and difficult to get a consent of each participant.
Results
Sociodemographic characteristics of acute poisoning patients in Addis Ababa emergency departments (2021–2023) (N= 371)
Sociodemographic characteristics of acute poisoning patients in Addis Ababa emergency departments (2021–2023) (N= 371).
Clinical characteristics of acute poisoning patients in Addis Ababa emergency departments (2021–2023) (N= 371)
Among the 371 patients, 173 (46.6%) were referred from other health facilities for further management, while 198 (53.4%) were not referred. Among the referred patients, fluid resuscitation was the most common intervention, provided to 102 (59.0%), followed by gastrointestinal decontamination in 47 (27.2%) and antidote administration in 42 (24.3%); 40 (23.1%) received other medical interventions. Most patients, 343 (92.5%), arrived at the emergency after two hours of the exposure. Clinically, gastrointestinal symptoms were the most frequent presentation, reported in 267 (72.0%) patients, followed by altered mental status in 76 (20.5%).
Of the poisoning cases analyzed, pharmaceutical drugs were the most common agents, accounting for 208 (56.1%) cases, followed by home detergents 56 (15.1%), rodenticides 45 (12.1%), organophosphates 44 (11.9%), and herbal remedies 18 (4.9%). The majority of poisonings were intentional 305 (82.5%), while 64 (17.3%) were accidental. Oral ingestion was the predominant route of exposure 310 (83.6%), followed by inhalation reported in 58 (15.6%) cases. A documented history of comorbid illness was present in only 21 (5.7%) patients. Hypertension was the most common comorbidity, reported in 10 (47.6%) patients, followed by cardiac disease and epilepsy each in 8 (38.1%). Diabetes mellitus and chronic kidney disease were present in 2 (9.5%), and 1 (4.8%) patient, respectively, while other comorbidities seen in 3 (14.3%) patients.
Clinical characteristics of acute poisoning patients in Addis Ababa emergency departments (2021–2023) (N= 371).
Others *: Psychosocial intervention Others **: Stroke, peptic ulcer disease.
Treatment outcomes of acute poisoning patients in Addis Ababa emergency departments (2021–2023) (N= 371)
Overall, 355 patients (95.7%) were discharged alive, while 16 patients (4.3%) died. Among those who died, 10 (62.5%) developed organ damage during treatment; specifically, 5 (31.25%) experienced renal complications and 5 (31.25%) developed respiratory organ damage.
Treatment outcomes by type of poisoning
Treatment outcomes by type of poisoning, clinical complaints, and mechanical ventilation requirement among acute poisoning patients in Addis Ababa emergency departments (2021–2023) (N= 371).
Treatment outcomes by main presenting complaint
Mortality differed considerably according to the main presenting clinical features. Patients presenting with respiratory symptoms experienced the highest mortality, with 11 deaths among 26 patients (42.3%). Those presenting with altered mental status had a lower but still notable mortality rate of 5.3%. In contrast, gastrointestinal symptoms, the most frequent presentation were associated with a very low mortality rate (0.4%). All asymptomatic patients survived. This gradient suggests that respiratory compromise at presentation is a strong marker of severe poisoning and poor prognosis (Table 3).
Treatment outcomes by mechanical ventilation requirement
Mechanical ventilation requirement was strongly associated with mortality. Nearly half of intubated patients died (45.5%), compared with only 1.7% mortality among those who did not require intubation. Although intubated patients represented a small proportion of the cohort, they accounted for the majority of deaths. This finding underscores the role of respiratory failure and critical illness severity in predicting poor treatment outcomes among acutely poisoned patients (Table 3).
Bivariate and multivariable analysis of factors associated with treatment outcomes of acute poisoning patients in Addis Ababa emergency departments (2021–2023) (N= 371)
Bivariate and multivariate analysis of factors associated with treatment outcomes among acute poisoning patients in Addis Ababa emergency departments (2021–2023) (N= 371).
Discussion
This is the largest and most recent multicenter study in Addis Ababa and provides updated urban poisoning patterns that differ from prior rural and regional reports. In this multicenter study of acute poisoning at four tertiary hospitals in Addis Ababa, the majority of patients were young adults (median age 22 years). This age distribution is consistent with recent Ethiopian and regional reports and likely reflects greater psychosocial vulnerability in this group, including academic pressure, relationship conflict, economic stress, and limited access to preventive mental-health services. These factors increase susceptibility to intentional self-poisoning and are consistent with findings from other urban centers.6,16–19
We observed a female predominance, which aligns with several Ethiopian studies but contrasts with some reports where males predominate.2,7,16,17,20,21 The sex difference likely reflects a mix of sociocultural influences (e.g., differential exposure to stressors, restricted social outlets for women, and differing help-seeking behavior), as well as potential reporting and care-seeking biases. We have therefore reframed this finding as context-sensitive: sex differences exist, but their interpretation requires consideration of local gender roles and healthcare access.
Regarding intentional poisoning, the current study found a remarkably high proportion (82.5%), exceeding the rates reported from Gondar (75%), St. Peter Hospital (78.4%), and Jimma (50.5%).2,22–26 This suggests a potentially growing burden of self-harm and psychosocial distress in urban populations, especially among young adults, and highlights the urgency of strengthening mental-health services in Addis Ababa. It reinforces the urgent need to integrate mental-health screening and brief psychosocial interventions into emergency care pathways, and to strengthen community-level suicide prevention programs, especially targeting adolescents and young adults. Early identification of at-risk individuals at primary care and pharmacy contacts (where many pharmaceuticals are obtained) could reduce access to means and lower self-harm admissions.
Gastrointestinal symptoms were the most frequent clinical presentation (72%), which corresponds with the dominant route of oral ingestion (83.6%) in our sample and is similar to patterns reported in other ingestion-predominant settings.17,25 This contrasts with a study in Tanzania, where neurological manifestations (62.3%) were more prevalent 27 Regional differences exist: studies from areas with inhalation or occupational exposures report different symptom profiles (for example, neurological dominance), demonstrating how toxin type and route shape clinical presentation. A poisoning agents and outcomes across major Ethiopian studies (Gondar, Jimma, Adama, Felege Hiwot, St. Peter) would help readers visualize these regional variations.15,18,23–25,28
Only 20.3% of patients underwent gastric lavage, a rate likely influenced by delayed presentation, as 92.5% arrived more than two hours after ingestion. This stands in contrast to findings from Gondar, Jimma, and India where up to 90% of patients presented within two hours.1,23,25 This lower rate likely reflects delayed presentation reducing the clinical indication for lavage and activated charcoal. The finding emphasizes the need for community education on early facility presentation, improved pre-hospital referral networks, and clinician training on time-sensitive decontamination guidelines.
Despite pharmaceuticals being the most frequent agents in our setting which is supported by studies from St. Peter Hospital, Iran and Egypt,24,29,30 organophosphates remained the leading cause of mortality, accounting for 62.5% of deaths. This trend aligns with reports from Gondar, Jimma, St. Peter Hospital, Tikur Anbesa and Zambia23–26,31 where organophosphates are consistently associated with the highest fatality rates due to their rapid progression to cholinergic and respiratory failure and need for intensive care. The pattern underscores the disproportionate lethality of pesticide exposure, even where overall prevalence is lower. The urban predominance of pharmaceuticals and the rural predominance of pesticides reported elsewhere underline the importance of contextualized prevention: urban strategies should prioritize safe dispensing and storage of medications, while rural policy must focus on pesticide regulation and safe agricultural practices.
Mortality patterns also demonstrate both consistency and contrast. The overall mortality rate of 4.3% in our study falls within the range reported across Ethiopia (1.5%–10%).26,32 However, this finding is different from the study done in Western Ethiopia which indicates 32.5% poor outcome. 16 The difference might be that the prior study used mortality and complications whereas this finding is only about mortality. The discrepancy between expected and observed outcome proportions is likely due to differences in outcome definitions and exclusion of pre-hospital deaths in our study.
Although time to presentation was recorded and is clinically important in poisoning management, it did not meet the statistical criteria for inclusion in the regression analysis. Mortality among mechanically ventilated patients was notably high, with nearly half of intubated patients dying. In multivariable analysis, respiratory failure requiring mechanical ventilation was strongly associated with increased mortality (P-value <0.001). This is consistent with international and regional literature showing that respiratory compromise whether from organophosphate toxicity, aspiration, or sedative overdose is the proximate cause of many poisoning deaths. In this case, mechanical ventilation was serving as an indicator of severe poisoning accompanied by respiratory failure rather than a predictor of death.
The study period overlaps with the COVID-19 pandemic and periods of healthcare system strain in Ethiopia. These conditions may have influenced emergency department capacity, referral patterns, and availability of intensive care resources such as mechanical ventilation. Therefore, some observed associations, particularly the relationship between mechanical ventilation and mortality may partially reflect health system constraints rather than purely biological severity.
Significance for public health
This study is of great public health importance because acute poisoning is a preventable yet serious cause of emergency admissions, disability, and death, particularly in low- and middle-income countries like Ethiopia.
Limitations of the study
This study has several limitations. First, its retrospective design limits causal inference. Second, exclusion of patients brought dead on arrival may underestimate mortality and introduces survivorship bias. Third, reliance on chart documentation may result in information bias, particularly for poisoning agents and treatment details. Fourth, the small number of death events limited the complexity of multivariable analysis and may result in model instability. Fifth, lack of toxicological confirmation and incomplete data on antidote administration restricted detailed clinical interpretation. Finally, as a hospital-based study, findings may not be generalizable to the broader population.
Conclusion
Acute poisoning in this urban tertiary care setting predominantly affects young adults and is largely intentional. Organophosphate poisoning and severe clinical presentations, particularly those requiring mechanical ventilation, were associated with increased in-hospital mortality. Mechanical ventilation should be interpreted as a marker of severity rather than a direct determinant. Strengthening early recognition, improving emergency care capacity, and expanding mental health services are essential to improve outcomes.
Supplemental material
Supplemental material - Treatment outcomes and determinants of in-hospital mortality among acute poisoning patients in Addis Ababa, Ethiopia: A multicenter retrospective study
Supplemental material for Treatment outcomes and determinants of in-hospital mortality among acute poisoning patients in Addis Ababa, Ethiopia: A multicenter retrospective study by Fasil Biyadigilign Ayalew, Wagari Tuli Nora in Journal of Public Health Research
Footnotes
Acknowledgment
We would like to express our gratitude to the Addis Ababa University College of Health Science Department of Emergency Medicine, the management of each hospital and the data collectors.
Ethical considerations
This study involves human participants and was approved by Institutional Review Board (IRB) of Addis Ababa University, College of Health Sciences with a reference number 08/chs/esn/2024.
Consent to participate
The need for consent to participate was waived by the IRB since the study was retrospectively chart review and difficult to get a consent of each participant.
Author contributions
Study conception and design: FB and WT. Both authors took part in data collection and questionnaire development. Analysis and interpretation of data was done by FB and WT. Writing the introduction and methods and material: WT and FB. Drafting the article: WT and FB. Final approval of article: WT and FB. Guarantor: FB. Critical review was done by: WT. The corresponding author attests that both listed authors meet ICMJE authorship criteria.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The datasets created and analyzed in this study are not publicly accessible because of the confidentiality of the participants, but can be obtained from the corresponding author upon reasonable request.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
