The risk factors including age less than 2 years, malnutrition, delayed referral, and hypokalemia are significantly associated with increased morbidity and mortality among paediatric patients presenting with severe diarrhea at Muhimbili National Hospital’s Emergency Medicine Department.

This was a prospective cohort study of paediatric patients presenting with severe diarrhea at EMD MNH and followed for outcomes at EMD and after 24-hours.

The study area was at the EMD of MNH in Dar-es-Salaam, Tanzania and all the pediatric wards at MNH and Pediatric Intensive Care Unit (PICU) where patients’ data was traced for 24-hours follow-up. MNH is located in Ilala district, along Malik Road, in Upanga. MNH has a bed capacity of 1500 and serves as the top referral hospital in Tanzania. The EMD was established in 2010 via a partnership between the Ministry of Health and Social Welfare and the Abbott Fund Tanzania. The EMD is the first full capacity public Hospital in Tanzania and currently the only training site for emergency medicine residency program in the country.

The department is staffed 24 hours, seven days a week by locally trained specialist emergency physicians, who oversee the care of patients and training of interns, registrars and emergency medicine residents. The pediatric patients are seen in pediatric wing D, where stabilization is done before disposition to the ward. The department receives about 90 pediatric patients with acute diarrhea and related complications per month. The age limits of paediatric patients seen at EMD are those who are 12 years and below. Those who are 12 years above are attended as adult patients, hence the age limit for this study was adopted based on the hospital age limit criteria for easy tracking and follow-ups. There is one machine in the EMD for electrolyte testing and acid base disorders, as well as diagnostic imaging, including portable x-ray and point of care ultrasound, fluids, antibiotics, two ventilator machines with airway devices, one CT scan machine accessible for all patients and laboratory services. The EMD has 19 emergency medicine specialists who attend patients yearly.

All paediatric patients presenting with severe diarrhea at EMD MNH, Tanzania

The accessible population were the paediatric patients with severe diarrhoea attending EMD MNH.

All paediatric patients twelve years of age and below presenting with severe diarrhoea at EMD MNH during the study period.

The estimation of sample size was based on a study conducted in Nigeria by Okoronkwo et al on “Pattern of morbidity and mortality of childhood illnesses at children′s emergency room of Abia State University Teaching Hospital: a descriptive prospective study” that reported 10.5% of deaths at the emergency room among paediatric patients were due to acute diarrhoea (14). This proportion and other assumptions (confidence interval (Cl=95%), Z=1.96 and margin error (Ɛ) =5% were used to calculate the sample size in this study. Therefore, in consideration of the study done, the sample size was calculated by the formula below;

Formula

N=Z²P (1-P)

Ɛ²

Where

N=sample size

Z=standard normal deviate at 95% confidence interval

P= mortality of 10.5% among pediatric patients with acute diarrhea death at the emergency room.

Ɛ=tolerable error 0.05

Substituting for the formula above

N=1.96² *10.5(100-10.5)/5²

N=144

The estimated sample size of pediatric patients with acute diarrhea presented at MNH EMD for this study was 144

A simple random sampling technique was used

This included the age and gender of the study participant, the location of the home of the study participant (region, district, ward, and street), the age of the guardian/care giver, the source of water for cooking and drinking, the clinical presentation on arrival, (such as level of consciousness measured as AVPU, heart rate, work of breathing, and capillary refills abnormalities associated with severe diarrhea such as hypokalemia, and acidosis, and the management done at EMD MNH.

EMD morbidity and mortality, dehydration and 24 hours′ outcomes (disposition to PICU, pediatric ward or discharged home for outpatient care) and their status at 24 hours were traced for continuity of care. Note, dehydration was determined using a WHO flowchart for assessing hydration status; ASK: What about for how long has the child had diarrhea? Ask: Is there blood in the stool? Look at the child's general condition. Is the child lethargic or unconscious? Restless and irritable? Look for sunken eyes, offer the child fluid. Is the child not able to drink or is drinking poorly? Drinking with zeal and thirst? Pinch the skin of the abdomen. Does it go back? Very slowly (longer than 2 seconds)? Slowly? Classify the child's dehydration as severe dehydration if it has two or more of the following: Lethargic or unconscious, sunken eyes, not able to drink or drinking poorly, skin pinches go back very slowly (longer than 2 seconds). Some dehydration if you have two or more of the following: restless, irritable, sunken eyes, drinking eagerly, thirsty, skin pinching goes back slowly (less than 2 seconds but longer than normal) and there is no dehydration if two or more of the signs of some dehydration are not present 15.

The risks of acquiring severe diarrhoea, complications-such as electrolyte imbalances, acid base abnormality and dehydration status, management strategies of the complications at EMD, and predictors of outcomes.

Data obtained was de-identified, coded, cleaned, and entered into the Statistical Package for Social Science (IBM SPSS version 25, IBM, LTD, Carolina, USA). Data was then summarized in the form of proportions and frequency tables for categorical variables. Mean, median with inter quartile range and proportions were used to summarize continuous variables. A Modified Poisson log linear model was used to test for the significance of associations between the predictors and outcome variables for continuous variables. Multivariate logistic analysis was used to determine the categorical predictor variables that are associated with the outcome. Significance was defined as a p-value of less than 0.05. In addition, subgroup analyses were done for age categories < 2 years, 2-5 years and > 5 years but less or equal to 12 years, age of care takers < 25years, 26 – 35 years and > 35 years, duration of passing stool per day; < 5 days and ≥ 5 days and frequency of passing stool per day; < 4 frequency per day and ≥ 4 frequency per day

Objective number 1. To determine the risk factors for acquiring severe diarrhea that were present among pediatric patients presenting to the EMD MNH from December 2021 to April 2022. The number/proportion of patients with each risk factor was analyzed with frequency.

Objective number 2. To describe the complications of severe diarrhea among pediatric patients presenting at EMD MNH from December 2021 to April 2022.

The complications were analyzed using the frequency of any complication from severe diarrhea.

Objective number 3. To describe the management strategies of severe diarrhea among pediatric patients presenting at EMD MNH from December 2021 to April 2022.

The management strategies of severe diarrhea among pediatric patients presenting at EMD MNH were analyzed using frequency

Objective number 4. To determine the predictors of morbidity & mortality from severe diarrhea among pediatric patients attending EMD MNH from December 2021 to April 2022.

The predictors of morbidity and mortality from severe diarrhea among pediatric patients attended at EMD MNH were analyzed using a modified poison log linear regression model and logistic regression.

Ethical approval for this study was obtained from the Muhimbili University of Health and Allied Sciences Institutional Review Board, Approval No. MUHAS-REC-02-2022-951.

A total of 6,570 paediatric patients were attended at EMD MNH during the study period. There were 164(2.50%) paediatric patients with severe diarrhoea. Of all patients, twenty (12.20%) among those who met the criteria had missing information and 144(87.80%) were enrolled and all were able to be followed up for 24 hours (Figure 1).

Among 144 paediatric patients with severe diarrhoea who were attended at EMD MNH, the median age was 1 (IQR 0.7, 2.0) years, 86 (59.7%) were male and most of the study participants were under 2 years of age 103 (70.8%). Of all study subjects’ majority 137 (95.1%) had been exclusively breastfed for 6 months or longer, 125 (86.80%) were from Dar es Salaam, and 92 (63.90%) were referred cases. The primary caretaker aged between 26 and 35, married in most cases, and 87.5% had completed secondary school (Table 1).

The children and environmental characteristics related to severe diarrhoea; most of the children were using tap as drinking water, only 13 (9.0%) reported using antibiotics 7 -14 days prior to severe diarrhoea, 102 (70.8%) had passage of loose stool more than 5 days and with 95 (66.0%) with 4 or more stools per day (Table 2).

Most of the paediatric patients with dehydration were under 2 years of age 91 (89.2%) among the paediatric participants which it was statistically significant with p-value of 0.036, and 114 (83.2%) had completed breast feeding for 6 months, though it was not statistically significant, 103 (81.7%) completed secondary school, majority of care givers were between 26-35 years of age 77 (80.3%), most of paediatric patients did not reports use of antibiotics 110 (84.0%) prior to onset of severe diarrhoea, 82 (80.4%) had passage of loose stool for less than 5 days and 82 (86.3%) had more or equal to 4 motion of loose stool per day, all these were not statistically significant (Table 3).

Most of the participants, 120 (83.3%) had dehydration among the paediatrics with acute diarrhoea, electrolyte imbalances-hypokalaemia was 76 (53.1%), acute renal failure contributed to 11 (7.6%) among 144 paediatric patients, leucocytosis had 16 (11.2%) paediatric patients and 8 (5.6%) had hypoglycaemia, most patients had point of care venous blood gases were done as an investigation to rule out electrolytes imbalances and monitor treatment outcomes, point of care random blood glucose, creatinine and urea and full blood count which were sent to central laboratory (Table 4).

Management strategies that were used by the study participants. Among 144 paediatric patients, 131 (90.97%) were given IV crystalloids, IV potassium chloride for correction of hypokalaemia 52 (36.1%), IV dextrose 10% was given to 6(4.2%) for corrections of hypoglycaemia., and IV antibiotics was given to 16(11.2%) of the study participants (Table 5).

Dehydration, hypokalemia, Acidosis, Hypoglycemia, Acute renal failure, Anemia, Leukocytosis, abnormal nutrition status, Duration of diarrhea, and referral status were all found be predictors of admission to the ward or Pediatric ICU in the univariate analysis, but in the multivariate analysis, only hypokalemia and referral status were found to be independent predictors of severe diarrhea among paediatric participants in in this study (Table 6).

Among all the pediatric patients who were enrolled,144 were alive, 82.64% among them were admitted to the ward, 9.03% were admitted to PICU, and 8.33% were managed and discharged home, there were no deaths at EMD and after 24hours of following (Figure 2)

The main strength was a prospective cohort study with consecutive recruitment of participants. Hence participants had a good chance to follow up on data and record findings. A structured questionnaire was used to capture all the necessary information. Missing data was minimized as a large number of variables were captured.

Some of the limitations expected while doing this study include accuracy in making a definition of acute diarrhea, but this was mitigated by using the Bristol stool chart to assess the consistency of the stool.

In this study, fewer patients were seen either as referral cases or self-referral due to the seasonal fluctuations of pediatric patients who presented with acute watery diarrhea during the study period

The author declares no conflict of interest

The ethical approval was obtained from MUHAS IRB, with an ethical approval No. MUHAS-REC-02-2022-951, and permission from MNH research and consultancy bureau to use its patients and file notes. Consent was obtained from the patients before the start of the study and confidentiality was maintained during and after the study by safely keeping the study materials. De-identification at data entry was done to minimize direct association of patients’ details with the data. Only personnel directly involved with the research had access to the data. Information gathered should be used only for purposes of research and resultant publication.

From this study, there was no benefit to the patient, but the information obtained from the study will help in knowing where more efforts should be directed so as to improve the care that the paediatric patients presenting with severe diarrhoea receive in the EMD MNH, Tanzania.

We anticipated that no harm would come to those participants of the study as all the patients had received all the care as per the protocol of the EMD MNH where the study was conducted.

Not applicable

The data set supporting the conclusion of this article is available from the authors on request

This was a non-funded project; the principal investigators used their own funds to support the data collection and logistics

MJY was involved in the study design conceptualization, data collection, analysis and interpretation, drafted the manuscript, and made all necessary changes to the manuscript.

HRS, SSK and EW were involved in the study design conceptualization, review of the data analysis and interpretation, and critical review of the manuscript. CS, MAG and WM were involved in interpretation and revision of the manuscript. JM was involved in study design, data analysis and interpretation and review of manuscript.