Study area, design, and period
Institutional based cross-sectional study was conducted from January 1, 2015, to December 31, 2019, at public Hospitals found in south Gondar Zone, Amhara Regional State, Ethiopia. South Gondar zone is one of the zones’ in the Amhara region, which is located in the Northern part of Ethiopia. Currently, there are 8 governmental hospitals and 94 functional health centers. These hospitals and health centers serve a population of approximately 2.3 million people, from which children account for about 980,490. Among 8 hospitals, the study was conducted in four randomly selected hospitals. The average numbers of pediatric surgical bed in each hospital were fifteen. From the four selected hospital, Debre Tabor General Hospital has well organized malnutrition treatment Centers, while others have not well organized malnutrition treatment Centers.
Study participants
All pediatrics with burn injury and admitted to south Gondar Zone public Hospitals between.
January 1, 2015, and December 31, 2019, were the target population. However, charts of burn victim pediatrics with none readable handwriting were excluded from the study.
Sample size, and sampling technique
The sample size was determined by using the single population proportion formula with the assumptions of prevalence(P), 27% which is the proportion of mortality among pediatrics burn victims in Central Malawi [9], level of confidence (CL), 95%, the margin of error (d) =5%, \( \mathrm{n}=\frac{{\left(\mathrm{Za}/2\right)}^2\mathrm{p}\left(1-\mathrm{p}\right)}{{\left(\mathrm{d}\right)}^2}=\frac{(1.96)^2\ 0.27\left(1-0.27\right)}{(0.05)^2}\kern0.75em =296. \)
W Where, n = the required sample size, Zα/2 = Standard normal variation for type 1 error, p = prevalence (0.5) & d = Margin of sampling error tolerated (0.05).
The calculated sample size was 296. After considering a 15% incomplete charts, the final sample size of our study was 348.
This study was conducted in four randomly selected public health institutes in South Gondar Zone. From the beginning, a sampling frame was prepared using the patient’s medical registration number from each hospital’s registration logbook. Then, the total sample sizes were allocated proportionally for each hospital. Finally, charts of study participants were taken from each of the four selected hospitals using a computer-generated simple random sampling technique.
Data collection tool and procedure
The data were collected and registered by using a structured checklist. The checklist was prepared by reviewing different literatures done on the same problems [8, 12, 18,19,20,21]. The checklist was focused on socio-demographic characteristics of injured children, clinical related factors, and treatment-related factors. Data were collected by four trained BSc nurses and were supervised by two physicians. Training about the objectives of the study, the contents of the tool, and data collection procedures was given for data collectors and supervisors for one day. To check the incompleteness of charts, a pre-test was conducted in 5% of the total sample size in the hospital where the area is not selected for actual data collection. The assigned supervisors and principal investigator closely monitored and supervised the whole data collection process.
Dependent variable
Mortality status
Independent variables
Socio-demographic related factors, clinically related factors, and Management-related factors.
Data management and statistical analysis
The consistency and completeness of the collected data were examined during data management and analysis. Data were entered into Epi Data Version 3.1 and analysis was done using Statistical Package for Social Science (SPSS) Version 25. Frequencies and cross-tabulations were used to check for missed values of variables and to describe the study population concerning relevant variables. Furthermore, percentages, proportions, and summary statistics were used to summarize the study population characteristics. Binary logistic regression analysis was used to determine the association between the dependent and independent variables. All variables with P-value ≤0.25 in the bivariable analysis were included in the final model of multivariable analysis to control all possible confounders. The goodness of fit was tested by Hosmer-Lemeshow statistic and Omnibus tests. The direction and strength of association were measured by the odds ratio with 95% CI. The adjusted odds ratio along with 95% CI were estimated to identify factors associated with mortality by using multivariable analysis and finally P-value < 0.05 was considered to declare as statistically significant.