ICU-Managed Patients' Epidemiology, Characteristics, and Outcomes: A Retrospective Single-Center Study

Background Resources are limited, and it is exceedingly difficult to provide intensive care in developing nations. In Somalia, intensive care unit (ICU) care was introduced only a few years ago. Purpose In this study, we aimed to determine the epidemiology, characteristics, and outcome of ICU-managed patients in a tertiary hospital in Mogadishu. Methods We retrospectively evaluated the files of 1082 patients admitted to our ICU during the year 2021. Results The majority (39.7%) of the patients were adults (aged between 20 and 39 years), and 67.8% were male patients. The median ICU length of stay was three days (IQR = 5 days), and nonsurvivors had shorter stays, one day. The mortality rate was 45.1%. The demand for critical care services in low-income countries is high. Conclusion The country has a very low ICU bed capacity. Critical care remains a neglected area of health service delivery in this setting, with large numbers of patients with potentially treatable conditions not having access to such services.


Introduction
Despite continuing to be one of the world's lowest-income countries, Somalia has improved its healthcare over the last two decades. Its citizens' life expectancy has increased from 51 years in 2000 to 58 years in 2020 [1]. However, until recently, there had been little investment in secondary and tertiary care services.
Given the disproportionate burden of diseases like malaria, tuberculosis, HIV/AIDS, and trauma, the prevalence of critical illness in developing countries is disproportionately high. 25% of the world's disease burden is carried by sub-Saharan Africa [2]. Critically ill patient management necessitates substantial human, material, and fnancial resources. Low-income nations like Somalia often have fewer of these resources. Large hospitals in urban or metropolitan settings are the most common places to fnd major intensive care units (ICUs) [3].
Te intensive care unit (ICU) facility is a signifcant and expensive part of modern healthcare. With extensive ICU infrastructures in all developed countries, the number of ICU beds available varies greatly between and among nations [4].
Evaluating the characteristics and outcomes of critically ill patients admitted to ICUs in low-income nations may contribute to determining the priorities and resources needed to enhance the care of critically ill patients in resource-constrained areas of the world. So, in this study, we aimed to determine the characteristics, admission diagnoses, and outcomes of patients admitted to the Mogadishu Somali Turkish Training and Research Hospital ICU from January 2021 until December 2021. Te information collected will be used by other ICUs in the country to improve services and assist institutions that are establishing new ICUs, and the fndings are believed to add something valuable to the growing amount of research showing the diferences between critical care in high-and low-income nations.

Methods
Tis retrospective study was approved by the Institutional Review Board of Mogadishu Somali Turkey Training and Research Hospital. Te health information system was reviewed, and anonymity was preserved for each case record. We evaluated all adults patients admitted to the intensive care units from January 2021 to December 2021.
Data analysed during this study came from the demographics, characteristics, and outcomes of these patients admitted to Mogadishu Somali Turkey Training and Research Hospital's ICU.

ICU Admissions.
Te criteria for admission to an intensive care unit were used to identify hospitalizations with ICU admissions by the medical or surgical teams of the hospital.

2.2.
Outcomes. Te primary outcome measure was shortterm mortality of ICU-admitted hospitalizations, defned as the ICU admission till discharge from the ICU to inpatient.

Study
Covariates. (a) Te frst covariate is demographics like age and gender, (b) the second is comorbid conditions, (c) third variable is the patients' need for hospitalization for medical or surgical procedures (based on the primary diagnosis-related grouping), (d) the fourth is organ failures, (e) the ffth is hospital length of stay (f ), and the sixth is the interventions made during the ICU stay, such as intubation, hemodialysis, and so on.

Data Analysis.
Continuous variables were reported as the mean (standard deviation (SD)) or the median, while data on categorical variables were summarized as numbers and percentages (interquartile (IQR)). Continuous variables were compared using the t-test, Mann-Whitney test, and Kruskal-Wallis test, as appropriate. We used the Kaplan-Meier estimator test to show the survival outcome.

Results
Te majority (39.7%) of the patients were adults (aged between 20 and 39 years); 734 (67.8%) were males. 595 (54.9%) were transferred to the inpatient, while 488 (45.1%) died in the intensive care unit. Te median ICU length of stay was 3 days (IQR � 5 days), and nonsurvivors had a shorter stay of one day. Te median no. of organ failure is one (IQR � 1). 73.4% of patients were in the ICU for less than 6 days, whereas only 4.2% of patients were in the ICU for more than 20 days. Table 1 describes sociodemographic and clinical characteristics among patients admitted to the ICU. When it comes to diagnosis on admission, complications related to acute and chronic kidney injuries top the list of medical patients, where perforation due to organ injury top the list of surgical patient, as Somalia is a war-torn country, where gunshot injuries are prevalent. Table 2 shows sociodemographic factors and their association with outcome using the Chi-square test. Since the p values for age group, comorbidity, type of organ failure, clinical intervention, type of admission, duration, and number of organ failures were less than 0.05, we may infer that all these variables signifcantly afect the outcome.
Te survivor function S(t) is the Kaplan-Meier estimator. From Figure 1, we can see that most patients died in the frst 20 days, as indicated by the steep slope of the estimated survival function in the frst 20 days ( Figure 2). S(10) � 0.3 means that the probability of a patient surviving longer than ten days is 30%. Kaplan-Meier estimator showed that the probability of a patient surviving longer than two days in the ICU is 76%; however, the probability of surviving longer than 20 days is dropped to 23%. Survival curves do not go head-to-head, indicating that there is a diference between groups being studied (Figures 3  and 4).
Te factors associated with outcome in univariate logistic regression are displayed in Table 3. It can be seen that the p value of hypertension, renal failure, respiratory failure, clinical intervention, and duration is below <0.005. Terefore, these variables are statistically signifcant. On univariate analysis, adult patients had a considerably higher likelihood of experiencing the outcome (OR � 1.707, 95% CI: 1.271-2.285). Surprisingly, compared to patients with three or more organ failures, patients without organ failure were more likely to have a outcome (OR � 4.586, 95% CI: 2.174-9.674). Te outcome was signifcantly lower in patients with hypertension as compared to patients without hypertension (OR � 0.629 95% CI: 0.463-0.855). Te patients who did not have respiratory failure were less likely to have outcome compared to the patients who had respiratory failure (OR � 1.472 95% CI: 1.111-1.950). Patients without renal failure were more likely to have outcome compared to patients with renal failure. Patients admitted through medical had lower odds of outcome than those who admitted through surgical (OR � 0.6140, 95% CI: 0.497-0.786). Patients who stay less than 7 days in ICU were less likely to have outcome in comparison with patients who stay more than 7 days (OR � 2.114, 95% CI: 1.559-2.868).
Te Cox-proportional hazard model estimates are shown in Table 4. Te variables respiratory, clinical intervention, intubation, and duration have highly statistically signifcant coefcients. Te p value for clinical intervention is <0.005 and HR is 1.292, indicating a strong relationship between the clinical intervention and increased risk of death. HR is 1.292 for respiratory; this means that a patient's hazard ratio increases by a factor of 1.292 (versus the baseline).

Discussion
We conducted a single center based analysis of ICUadmitted patients in Mogadishu. Te center has 21 tertiary level ICU beds, the only one in Mogadishu, with a population of around 2.5 million [5]. Tis clearly indicates the scarcity of tertiary ICU levels in Somalia.
In this retrospective study, we aimed to determine admission patterns in our ICU during the year 2021. Te overall mortality rate was 45.1%, somewhat higher than the reports from the neighbouring country, Ethiopia [6]. Tis could be caused by patient care-seeking delays, a lack of treatment protocol, a pharmaceutical shortage, or other factors. Te outcome of intensive care depends on many issues given by the surgeons and doctors who make the frst judgments that lead to their patients needing intensive care, as well as the amenities ofered in the unit and the competence and timing with which they are administered [7]. Te length of stay was comparable to some US hospitals (3 days) [8], but lower than studies from Austria and Switzerland (7.6 in survivors and 11.7 in nonsurvivors) [9] and higher than Scandinavian countries (1.9 in nonsurvivors) [10]. Males were admitted to our ICU in greater numbers (67.8%) than females. Te majority of males accessing health facilities, which is also observed in all hospital admissions as recorded by other studies in Ethiopia, could be one of the causes [11,12]. However, this is in contrast to the general demographic        [13]. Identifying patients who should be targeted for interventions early in their hospital or ICU course is one of the challenges to improving the quality of end-of-life care in the ICU. Tere are several potential methods for identifying the most appropriate patients. Te SUPPORT study, for example, is aimed at seriously ill patients with one or more of the following illnesses: acute respiratory failure, multiple organ system failure with sepsis, multiple organ system failure with malignancy, coma, chronic obstructive pulmonary disease with respiratory failure, decompensated congestive heart failure, severe cirrhosis, metastatic colon cancer, and non-small-cell lung cancer [14]. As our ICU was a level 3 ICU, most of the interventions needed to support vital organs were available such as mechanical ventilation and hemodialysis.
When it comes to the type of organ failure, renal failure topped among all, with respiratory failure being second in the rank. Tis is in consistent with a previous report by Sari and Bashir [15], about the patients admitted to the medical ward, which showed that 45% of the admissions were due to renal failure. However, it is in contrary to the fndings of Sakr et al. from Belgium that the cardiovascular failure was the most common type of organ failure [16].

Conclusion
In this frst ICU-related study from Somalia, we demonstrated the demographic characteristics of and the outcome of patients admitted to a level 3 ICU in Mogadishu. It also demonstrated the scarcity of ICU beds in Somalia's capital city of Mogadishu. Although there are facilities for tracheal intubation, mechanical ventilation, hemodialysis, and patient monitoring, the survival rate of patients in our ICU is uncomfortably low. Tere is a lack of data in our region on workload, outcomes, costs, and the heterogeneity of ICUs; any recommendation about future provision will be highly speculative.

Limitations
Tis study has limitation. Due to the retrospective nature of the study, we could not get all data relating to APACHI Score and SAPS.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon request.

Consent
As our study is retrospective, consent was waived by the Ethical Committee of Mogadishu Somali Turkey Training and Research Hospital.