1. Introduction
In many African countries, Acinetobacter baumannii has emerged as a concerning and significant cause of healthcare-associated infections, particularly in intensive care units (ICUs) and other healthcare settings. Its ability to thrive in healthcare facilities and persist on various surfaces contributes to its reputation as a troublesome nosocomial pathogen. However, the lack of standardized surveillance systems and inconsistent data reporting practices across different African countries may result in underestimation or variability in reported prevalence rates. As a result, obtaining accurate and comprehensive epidemiological data on Acinetobacter baumannii prevalence in Africa becomes an ongoing challenge, hindering the development of targeted interventions and infection control measures. Despite these challenges, available data suggests a notable rise in Acinetobacter baumannii infections in African healthcare settings. For example, a study done in a major tertiary hospital in West Africa reported that Acinetobacter baumannii was the common ubiquitous pathogen in majority of the specimens from clinical setups, accounting for approximately 40% of all gram-negative bacteria [1] . In another study from East Africa, conducted in a referral hospital’s ICU, the prevalence of Acinetobacter baumannii infection was found to be 24.5%, with a considerable proportion of the isolates showing resistance to multiple antibiotics, including carbapenems [2] .
In Kenya, as in many other regions, the major risk factor associated with the emergence of drug-resistant Acinetobacter baumannii (Ab) variants is prior exposure to antibiotics. Patients with a history of carbapenem exposure have a higher likelihood of acquiring strains of Ab that are resistant to multiple drugs. Moreover, the constant exposure of Ab to antimicrobial drugs within the hospital environment creates a breeding ground for the development of mutant forms, leading to infections that pose serious challenges for patient management [3] . Despite the importance of understanding and addressing this issue, there remains a significant knowledge gap in Kenya regarding the risk factors contributing to the carriage of CRAB in patients admitted to ICU. Gaining comprehensive insights into these aspects is crucial for implementing targeted interventions to mitigate the spread of drug-resistant A. baumannii and improve patient outcomes in Kenyan healthcare settings. This study therefore determined risk factors associated with MDR and CR Acinetobacter baumannii carriage among ICU patients hospitalized at MOI Teaching and referral hospital, Kenya.
2. Materials and Methods
2.1. Study Area
This study was carried out in the ICU section of MOI teaching and Referral hospital (MTRH). The facility is the second largest public hospital in Kenya. It is located in Eldoret, Uasin Gishu county (311 kilometers Northwest of Nairobi), Kenya. It serves residents from 24 counties in Kenya, Southern Sudan, parts of Eastern Uganda and Democratic Republic of Congo. At the time of this study, MTRH had a bed capacity of 1020 serving as a level six referral hospital. It offers inpatient, outpatient and specialized care services such as the ICU [4] . The intensive care unit had a bed capacity of 20 beds at the time of study. These hospital demographics made the facility more ideal for the current study.
2.2. Study Design
This study adopted a cross-sectional descriptive study design among patients admitted to the ICU at MTRH between January and December, 2020.
2.3. Study Population
Patients admitted in the ICU at MTRH between July 2019 and July 2020 constituted the study population for this study. This cohort serves as high-risk individuals for A. baumannii co-infection in the hospital setup and ICU environs within a hospital settings [5] . One hundred and thirty two (132) were recruited in this study.
2.4. Sample Size Determination
The study sample size was calculated according to the Chow et al. 2007 [6] formula. The prevalence rate used was 9.5% [7] with a sample size of 132 patients participating in this study.
2.5. Sampling Technique
Adult patients (≥18 years) admitted to the ICU for at least 48 hours and consenting to participated in this study through close relatives or legal representatives were enrolled. Purposive sampling technique according to Palinkas et al. 2013 [8] was used to select subjects who satisfied the inclusion criteria. Sampling interval Kth was calculated as; where, MTRH has intensive care units (ICUs) services with an annual capacity of 1500 patients and the sample size is n (132), giving a sampling interval (K) of 12. Every day for the year of study, intervals of 12 was used to sample participants. The recruitment process was done continuously until 132 study participants were recruited.
2.6. Data Collection
Demographic data, such as gender, age and hospital stay duration, comorbidities, and treatment outcomes, were collected using a standardized questionnaire. The questionnaire was designed and standardized to capture essential variables and clinical information of the study subjects and the factors associated with MDR and CR Acinetobacter baumannii among ICU patients. Pretesting of the questionnaire was done at Nakuru County referral hospital (level 6) with 14 (10%) of the study sample patients. The results of pretesting were used in improving data collection tool (Appendix).
2.7. Ethical Considerations and Permit
The study was conducted with approval from the School of Medicine, MOI University and the superintendent at MOI Teaching and Referral Hospital (ELD/MTRH/R&P/10/2/V.2/2010), while ethical clearance was granted by the Institutional Research and Ethics Committee (IREC) at MOI University under reference number (IREC/2019/126 - 0003392). The National Commission of Science Technology and Innovation (NACOSTI) also issued a permit for data collection under licensure number (NACOSTI/P/19/75649/29865). Prior to sample collection, informed consent was obtained from all participants or their representatives and anonymized codes were used to ensure confidentiality of patients’ data.
2.8. Data Analysis
Mean values and standard deviations were used to present data in tables. Statistical analysis was carried out using SPSS Statistics version 25.0 [9] . Percentages and proportions were utilized to characterize the social and clinical demographics of the subjects. Additionally, odds ratios (aORs) were computed to assess the presence or absence of Acinetobacter baumannii carriage. Furthermore, multivariate logistic regression analysis was employed to further explore and identify significant associations between these variables. Statistical significance was set at p ≤ 0.05.
3. Results
3.1. Socio-Demographic and Clinical Characteristics of the Study Population
A total of 132 patients were enrolled in the study. Majority of the participants (51.5%, 68/132) were male. The median age was 52 (IQR = 36 - 58) years with (38.6%, 51/132) of the participants aged between 45 and 59 years. Most of the participants (74.2%, 98/132) were married. In investigating education level, (39.4%, 52/132) had primary level of education with equal percentage having secondary level education. Almost half of the participants, (44.7%, 59/132) were self-employed. All of the participants had at least one comorbidity with renal related conditions (50.8%, 67/132) and respiratory related conditions (42.4%, 56/132), Table 1.
3.2. Factors Associated with A. baumannii Carriage among Patients Admitted to the ICU at MOI Teaching and Referral Hospital
Bivariable analysis showed that employed participants were 3.4 times more likely to have A. baumannii compared to the unemployed (cOR = 3.38, 95%, CI: 1.09 - 10.43, p = 0.035), and patients referred to the study site was 83% less likely to have infection compared to non-referral (cOR = 0.17, 95%, CI: 0.06 - 0.51, p = 0.002). Multivariable analysis showed an independent association between A. baumannii carriage with occupation and referral status, where the employed patients were 4.4 times more likely to harbor A. baumannii (aOR = 4.41, 95%, CI: 1.32 - 14.79, p = 0.016), while those referred from other facilities were 86% less likely compared to non-referred patients (aOR = 0.14, 95%, CI: 0.05 - 0.45, p = 0.001). Patients who were having high BMI were likely to be infected by A. baumannii compared to those who had normal/low BMI (aOR = 11.2, 95%, CI: 3.57 - 21.11, p = 0.004). However, gender, age, marital status and education factors were not significantly associated with A. baumannii carriage among patients admitted to the ICU (Table 2).
3.3. Factors Associated with Carbapenem-Resistant Acinetobacter baumannii (CRAB) among ICU Patients Hospitalized at MOI Teaching and Referral Hospital
Bivariable analysis established that those who were aged ≥ 50 years were 21 times more likely to be carbapenem-resistant Acinetobacter baumannii, COR = 21.0, 95% CI: 1.83 - 240.52, p = 0.011. Those who stayed in ICU for more than 30 days were 16 times more likely to be carbapenem-resistant Acinetobacter baumannii compared to those who had been admitted (COR = 16.0, 95% CI:1.45 - 176.45, p = 0.019). Those who had underlying comorbidity were 46 times more likely to be carbapenem-resistant Acinetobacter baumannii compared to those without underlying comorbidity, COR = 46.0, 95% CI:3.33 - 634.88, p = 0.003. Variables were subjected to multivariable model where none of the variables were statistically significant (Table 3).
4. Discussion
This study found that employed participants exhibited a 4.4-fold higher likelihood of harboring A. baumannii, suggesting occupational exposure’s potential role in colonization. Consistent findings emerged in previous studies across various healthcare settings, employing retrospective cohort designs and cross-sectional
![]()
Table 1. Socio-demographic and clinical characteristics of the study population.
![]()
Table 2. Factors associated with A. baumannii carriage among patients admitted to the ICU at MTRH.
BMI: Body Mass Index, LOHS: Length of Hospital Stay, Ref: Reference, CI: Confidence Interval, cOR: crude Odd Ratio, aOR: adjusted Odd Ratio.
![]()
Table 3. Factors associated with Carbapenem-Resistant Acinetobacter baumannii (CRAB) among ICU patients hospitalized at MOI Teaching and referral hospital.
CRAB: Carbapenem-Resistant Acinetobacter baumannii, ICU: Intensive Care Unit, cOR: Crude Odds Ratio, aOR: adjusted Odds Ratio, CI: Confidence Interval.
surveys. These investigations consistently reported increased A. baumannii colonization rates among healthcare workers, supporting the notion that occupation may influence colonization risk in hospital environments [10] [11] .
Participants referred to MTRH were at a lower risk of being colonized with A. baumannii as compared to those that were initially being treated at MRTH. This is because, in most cases, patients referred to the study’s site had already been managed with a course of antibiotics without positive outcomes, necessitating their referral to MTRH. The prior management with a course of antibiotics lessens the probability of referral patients to be colonized with bacteria such as A. baumannii as at the time of screening. Moreover, this may be attributed to differences in patient populations, infection control practices, or colonization patterns in different healthcare settings [12] .
Patients with a high BMI were more likely to be colonized with A. baumannii as compared to those who had low or normal BMI. This is consistent with the knowledge that a high BMI predisposes an individual to having a weakened immune system and common health complications like hypertension and diabetes among many others. Gender, age, marital status and a participant’s level of education were found to have no significant association with carriage of A. baumannii among patients admitted to the ICU, explainable by the fact that A. baumannii is an opportunistic pathogen that will almost randomly colonize any suitable host, without regard to either Gender, age, marital status and a participant’s level of education [13] .
On factors associated with carbapenem resistance among ICU patients, this study established that participants with a longer ICU stay such as that of more than 30 days were 16 times more likely to be colonized with carbapenem-resistant A. baumannii compared to those who had just been recently admitted to the ICU. This is because patients with long stays in the hospital generally are more predisposed to contracting hospital acquired infections that are adamant to treatment by commonly prescribed drugs, as the causative pathogens circulating within the hospital environment are better adapted for survival even against agents designed to eliminate them [14] .
It was also interesting to note that participants that had underlying comorbidities were 46 times more likely to harbor isolates of A. baumannii that were resistant to carbapenems as compared to those without underlying comorbidities. This is because underlying comorbidities complicate treatment of the target disease, as the clinician has to most often than not consider the drug interactions of the primary disease with the underlying conditions, which most often leads to use of superior drugs of higher critical activity. This, when often done even in undeserving situations goes a long way to cause the increase in numbers of bacteria strains that exhibit insensitivity to the highly active antibiotics like Colistin and Carbapenems [15] .
5. Limitations of the Study
The study was conducted only to patients who gave their informed consent. The study was also purposively limited to patients admitted to MTRH-ICU only and not in other hospitals in the county where patients could have been admitted and maybe different data could have been obtained.
6. Conclusion
Increased length of hospital stay, obesity and marital status were the factors found to be significantly associated with A. baumannii infections among ICU admitted patients. On the other hand, gender, age, level of education, occupation, referral status and presence of infection were found to have no significant association with A. baumannii infections among ICU admitted patients. Therefore, all patients admitted to the intensive care units should be screened for colonization with A. baumannii, owing to the poor treatment outcomes associated with carriage of this multidrug resistant pathogen. Proper infection control in the ICU settings should be upheld to mitigate the spread of A. baumannii in the intensive care units. To improve treatment outcomes, special attention should be given to patients noted to have risk factors established to have significant association with A. baumannii colonization while in the ICU.
Acknowledgements
We would like to extend our sincere gratitude to Mr. Ben Kipturgo, Laboratory In-Charge Technologist at MTRH, for his immense support. Our appreciation also goes to Mr. Hadan Baiwo for his diligent work in the laboratory and to Dr. Josephat Tonui for providing essential protocol guidelines. Their contributions have been instrumental in the successful execution of this study.
Funding
The study did not receive any funding.
Authors’ Contributions
The authors made significant contributions to this study. FK was responsible for data collection, cleaning, and analysis, as well as writing the manuscript. AM assisted with data analysis and manuscript review. NM conducted data analysis and reviewed the manuscript.
Appendix: Questionnaire
PART I: Demographic Information
1) Patient No…………………………..
Age: ………………………… Gender: ………………………
Nationality: ……………………………………
Date of Admission: ……………………………
Occupation:
a) Employed……………. b) Self-employed………… c) Unemployed………..
Highest Level of education:
a) Primary………. b) Secondary……… c) Tertiary……….. d) None………..
Marital status:
a) Married………………………… b) Single………………………
2) Referral status:
a) Referral…………………………. b) Non Referral…………………………
3) Which ward are you currently admitted to………………………………
4) Length of stay at the hospital: ……………………………………………
PART II: Clinical Information
1) Prior admission what other disease were you suffering from or diagnosed with? …………………………
2) Any comorbidity apart from the one diagnosed you may be suffering from?
Yes………….. No………………….
If yes: examples:
Diabetes……………Renal related conditions…………
CNS related conditions……. Autoimmune related conditions……..
Respiratory related conditions……. Metabolic disorder…….….
Injuries………. Asthmatic………. Burns…… Blood stream conditions……..
Specify (others conditions)……………………….
3) Are you an immunocompromised patient?
Yes……………… No………………………
If yes! PITC (HIV)…………………………
Cancer………………………………………
Specify (Others)…………………
4) What is your current diagnosis? ………………………….
5) Tests done before/during or after Admission? Yes/No. When was it done?………………
If yes! State tests done……………………………………
What were the diagnosis? ………………………………
What organisms were isolated? …………………………
Antibiotics prescribed? State……………………………….
6) Antibiotics used before admission preferably 7 days prior to admission?
Yes………………………… No……………………….
If yes! Which antibiotics have you used? ………………
7) Were you hospitalized in another health facility prior to your admission?
Yes…………………………. No………………………….
If Yes! Which facility and for how long………………….
Diagnosis…………………………………………
Antibiotics prescribed……………………………
Reason for Discharge? …………………………….
8) Referral from another hospital?
Yes……………….. No…………….
If Yes! Reason/Need for referral…………………………………………..
9) What are your current prescribed antibiotics?…………………………..
Are the antibiotics effective?
Yes……………….. No…………….
If Yes! State them……………………………………………………………..