Health systems in sub-Saharan Africa (SSA) face numerous challenges, including inadequate human resources, insufficient funding, poor infrastructure, and a high burden of disease (1–5). On the other hand, antimicrobial resistance has emerged as a global health concern that requires concerted effort to control or mitigate (6–8). Nevertheless, SSA continues to face weak healthcare systems, poverty and inadequate infrastructure, which have compounded the fight against antimicrobial resistance (9). However, for this fight to realize success, efforts need to be directed toward strengthening surveillance systems, promoting responsible antimicrobial use, improving infection prevention and control, and fostering collaboration across various sectors. It is with this vision that the Kakamega–Cambridge health partnership was established.

The Kakamega County Teaching and Referral Hospital (KCTRH), like many healthcare provision facilities in SSA, is faced with limited resources (both financial and human). KCTRH is a level-5 hospital with a 500 in-patient bed capacity and around 1,000 out-patients seen per day. As a referral center, most of the surgical procedures in Kakamega County are done at KCTRH. However, KCTRH does not have an established AMR surveillance program. Therefore, and although it has an established antimicrobial stewardship team, their efforts toward effective delivery at the healthcare facility remain limited. The absence of these data continues to undermine the fight against AMR and exacerbates the spread of hospital-acquired infections and antibiotic resistance. This, in turn, has a negative clinical and cost implication for patients, communities and the healthcare system (10, 11). To circumvent this challenge, staff at KCTRH teamed up with staff at the Cambridge University Hospitals NHS Foundation Trust (Cambridge, United Kingdom) to engage in an effort to improve and strengthen antimicrobial stewardship (AMS) at the facility. An AMS program aims at: (i) optimizing antibiotic use, (ii) promoting appropriate prescribing practices, (iii) improving patient outcomes, (iv) reducing healthcare costs, and (v) minimizing the development and spread of antimicrobial resistance (AMR) (12).

The team carried out surveys to assess the knowledge and awareness of healthcare staff at the facility regarding AMS and antibiotic use in the surgical wards. This study focused on the surgical wards, because in our settings, the surgical department is one of the heavy consumers of antibiotics (12). To qualify antibiotic use, the global Point Prevalence Survey (PPS) was used. The PPS is a study that measures the proportion of individuals in a population who have a specific characteristic such as a disease or condition at a single, specific point in time. It can be used to monitor healthcare-associated infections and antimicrobial use in hospital. Additionally, it can also show how health facilities comply with the AWaRe classification system, which categorizes antibiotics under three groups of Access, Watch and Reserve as part of AMS (13). The Access group contains antibiotics used in the first- and second-line treatment of infections. The Watch group contains broad –spectrum antibiotics with a higher potential of developing resistance. The Reserve group contains last-resort antibiotics used for multi-drug resistant infections.

This study reports on knowledge and awareness among healthcare workers of these five key strategic objectives of the Global action plan on AMR prevention, and on efforts made by the partnership to establish a program for monitoring and quantifying antimicrobial use and compliance on the surgical wards. The five strategic objective on AMR include: Objective 1: Improve awareness and understanding of AMR through effective communication, education and training. Objective 2: Strengthen the knowledge and evidence base through surveillance and research. Objective 3: Reduce the incidence of infection through effective sanitation, hygiene and infection prevention measures. Objective 4: Optimize the use of antimicrobial medicines in human and animal health. Objective 5: Develop the economic case for sustainable investment that takes account of the needs of all countries, and increase investment in new medicines, diagnostic tools, vaccines and other interventions (14).

This work contributes to the current global health agenda on antimicrobial resistance by demonstrating how international partnerships can bolster local health systems, especially those in lower and middle-income countries (LMICs) where resource constraints often frustrate the implementation of good AMS.

The main goal of this partnership was to quantify the antimicrobial use at KCTRH surgical wards, a challenge that many health facilities in LMICs have yet to accomplish. Our findings suggest slightly higher overall antibiotic use at KCTRH (48.7%), compared with Kenyatta National Hospital (KNH; 43% antibiotic usage) and Moi Teaching and Referral Hospital (MTRH; 47% antibiotic usage) but lower antibiotic usage than Coast Provincial General Hospital (CPGH; 52%) (16). These differences may be attributable to the levels of care at each facility and patterns of referral. Variation in antibiotic overuse across different types of healthcare settings is a global phenomenon, observed not only in the specific settings of the study but also in developed countries (17, 18). These persistent variations in antibiotic overuse can be rooted in several underlying systemic and social issues (17)^, including: structural risks and perverse incentives, social norms, missing infrastructure and patient poverty. Additionally, variation in antibiotic usage can be attributed to several factors, largely revolving around the prescriber and the patient namely; physician experience and individual practice patterns, time constraints, physician perceptions and attitudes, colleagues’ influence, and patient-related factors (perception and attitudes toward antibiotics) (18). In this study, findings point to an increased use of antibiotics as the healthcare facility level (based on Kenyan classification) is descended, with tertiary facilities being at the peak of the health facility classification pyramid in Kenya. Tertiary hospitals (Level 6) typically operate with larger budgets due to their national scope and highly specialized services. Their funding is primarily sourced from national government transfers, whereas County hospitals (level 5) receive a specific allocation from their respective county budgets (19). Significantly, Level 6 facilities enjoy greater autonomy in deploying their allocated resources than Level 5 hospitals, potentially enabling them to dedicate more funding to antimicrobial stewardship programs.

Importantly, this study shows the value of transnational collaboration in addressing the challenge of AMR. Crucially, in the Kakamega-Cambridge partnership, shared methodologies have been used to change practice and build capacity in the area of antimicrobial stewardship. Interestingly, our survey suggested above average knowledge about appropriate use of antimicrobials among healthcare workers at KCTRH, but lower levels of awareness about training in antimicrobial resistance and stewardship, or about the importance of surveillance and monitoring. These may be possible factors driving high antimicrobial usage at the facility. These challenges are further compounded by the limited information available to clinicians to guide the treatment and management of patients. In Kenya, public health facilities below the tertiary hospitals are poorly resourced, which again, could influence the healthcare services provided at these levels. For instance, clinical microbiology laboratories are poorly supported, particularly in terms of human resources, as well as diagnostic equipment (20). Irrational antibiotic use presents a complex problem globally, affecting not only LMICs but also developed nations, necessitating a multifaceted approach (21, 22). A promising approach for optimizing antibiotic use is improving education for healthcare professionals (23, 24). This improvement could include enhancing medical training curricula in both medicine and pharmacy. Specifically, it is becoming important to incorporate education on pharmaceutical promotion, including: strategies used by companies; the impact of these strategies on prescribing and dispensing; and practicing evidence-based medicine (25)^. Unfortunately, these emerging educational issues have not been sufficiently emphasized in Kenya. Addressing them is becoming increasingly vital for stakeholders in the health sector, particularly within the context of partnerships like the Kakemega–Cambridge collaboration. Additionally, training of healthcare personnel in specific courses, workshops and events designed explicitly for healthcare workers (HCWs) involved in managing patients (such as surgeons, general practitioners) is becoming more critical (26, 27). The Kakemega–Cambridge collaboration will therefore aim to develop such programs to strengthen AMS compliance at KCTRH.

Multifactoral approaches have nevertheless been suggested to maximize implementation of interventional efforts targeting HCW. Developing intrahospital protocols and guidelines have been shown to promote strong adherence to programs such as influenza vaccination among splenectomy patients of over the years (28). This partnership therefore will aim to develop guidelines and intrahospital protocols to strengthen antimicrobial stewardship programs including prudent antimicrobial use. Currently the absence of hospital antibiograms and infections prevention and control (IPC) policy continues to limit clinicians choices on the most effective optimal empirical treatment regimens (29), and prevention of healthcare-associated infections (30). Implementing such programs in healthcare settings in Kakamega County can reduce the rise of antimicrobial resistance, minimize healthcare costs and ensure a safe and high-quality care environment.

In our study, only 37% of patients had a recorded diagnosis, and most of these diagnoses were based on hematology. Some of the hematological parameters monitored include changes in white blood cell count (leukocytosis or leukopenia), specific immune cell ratios like the neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR), and changes in hemoglobin and platelet levels. Nevertheless, these tests are dependent on availability of laboratory reagents, and therefore not regularly reported for all patients. This limits diagnostic capacity, which subsequently impacts on the ability to provide targeted therapy. Consequently, even though knowledge on antimicrobial use scored highly, it remains the case that healthcare workers have limited information on which to make informed decisions regarding antibiotic use, but relied on the antibiotic prescription algorithms provided by the hospital to guide the empirical antibiotic prescription (31)^.

In this survey, all (100%) of the antibiotic treatments targeting hospital-acquired infections (HAIs) and community-acquired infection (CAIs) were based on empirical decisions. These high rates of empirical antibiotic usage at the healthcare facility serves to bring to the fore weaknesses in implementation of institutional or national antibiotic usage guidelines at secondary healthcare facilities in Kenya. Empirical prescription for both CAIs and HAIs underscores the lack of microbiology-based services, which limits the implementation of antibiotic usage guidelines (32, 33). This challenge may have been made worse in the wake of COVID -19, with mean rate of antibiotic use in COVID-19 management is 74.0% and only 17.6% of patients had secondary infections (34). Overall, proper management strategies need to be established by healthcare system institutions and policymakers, to sustain requirements for healthcare delivery and mitigate persistence misuse of antibiotics in LMIC healthcare institutions. This is urgent as witnessed during the COVID -19 pandemic, healthcare demand will probably overload the insufficient capacity of the public health service, previously weakened by resource limitations and pandemic bursts (35). Therefore, implementing measures to track the use of antibiotics and comply with the WHO’s guideline to promote antibiotic stewardship are critical strategies for combatting antimicrobial resistance and bolstering pandemic preparedness.

The use of third generation cephalosporins (3GC) specifically ceftriaxone recorded in this study is commensurate with other studies conducted in referral hospitals in Kenya and the region (16, 17, 33, 36) and is a major concern in healthcare systems in SSA. This is because ceftriaxone is not indicated for surgical prophylaxis (37). However, in SSA, poor infection prevention and control practices have been attributed to result in extensive inappropriate use of 3GC to increase clinicians’ confidence while conducting surgical operations (33, 38–40). This practice may contribute to the increased resistance to 3GCs among microbes in hospitals and the environment in the region (41–44). Resistance to 3GCs undermines treatment and exacerbates poor outcomes in SSAs with limited alternative therapies, and calling for immediate interventions including enhanced surveillance, improve IPC practices and strengthen AMS programs (45, 46).

Our findings show that the majority of antibiotic prescriptions for surgical prophylaxis were for more than 1 day of treatment. This appears to be common practice in Kenya, as also shown in previous studies (32, 47, 48). This is considered to be high, as per WHO recommendations, which recommends a single dose to be administered 1 to 2 h before surgery (37). Prolonged use of antibiotics for surgical prophylaxis may lead to unfavorable drug reactions and further enhance selection of antibiotic resistance. KCTRH has an established Medicines and Therapeutics Committee (MTC) responsible for guiding and approving formularies in the hospital. The choice of surgical prophylactic prescriptions documented in this study presents a picture of the challenges faced by such committees in poorly- resourced settings. The high levels of ceftriaxone usage may also be linked to behavioral factors, since clinicians may prefer to use antibiotics with a broader spectrum of action to ensure patients are covered for both Gram-positive and Gram-negative bacteria.

This study revealed a high prevalence of antibiotic usage at KCTRH, and highlights several areas where improvement is needed. Key emerging issues are that the majority of prescriptions remain empiric and do not follow the WHO recommended targets. Additionally, we note the prolonged duration of antibiotic use in surgical prophylaxis, as well as over prescription of broad spectrum agents such as ceftriaxone. These issues appear linked with the resource limitation of the facility, which has limited access to surveillance and monitoring information. Our study also emphasizes the need for ongoing awareness and educational support regarding AMS.

Our data are commensurate with the notion that inadequate resourcing in the LMIC healthcare setting is associated with greater levels of empirical antibiotic prescription, low microbiological diagnostics and poor enforcement of standard treatment guidelines. Our study also provides actionable data for policy makers, and shows that there is a need for greater public awareness and education to promote improved antibiotic stewardship, especially among healthcare workers. In addition to its international collaborative nature, another key strength of this collaboration lies in the deployment of validated tools such as the PPS. However, the study also has limitations, which we hope to address in future work. These limitations include its cross-sectional design and the lack of accompanying microbiological diagnostic data that can be used in the confirmation of treatment indications. Additionally, our findings primarily apply to surgical wards and should not be extrapolated to the entire hospital without caution.