Artificial intelligence for antimicrobial resistance surveillance in public hospitals: evidence of equity gaps in Nigeria
1
Program, Society for Family Health Nigeria, Kano, Nigeria
2
Research & Program, Kano Youths Champion Development Initiative, Sabuwa gandu Kano State, Nigeria
Popul. Med. 2026;8(Supplement Supplement 1):A169
ABSTRACT
BACKGROUND:
Antimicrobial resistance is a major public health concern in low- and middle-income countries, where surveillance systems are often weak. Many public hospitals face limited laboratory capacity, incomplete antimicrobial susceptibility testing, and poor data systems. Artificial intelligence has been proposed to improve antimicrobial resistance surveillance, but most existing evidence comes from high-income countries with strong health information systems. There is limited evidence on how these tools perform in public hospitals in resource-limited settings, or whether they may worsen existing gaps between better- and lower-resourced facilities.
METHODS:
A mixed-methods, retrospective study was conducted using routine hospital data from April 2024 to March 2025 in Kano State, Nigeria. Data were collected from four public hospitals, including two tertiary and two secondary facilities. Records of patients with bacterial infections were reviewed, yielding 2,480 selected through consecutive sampling. Model performance was assessed using accuracy, sensitivity, and area under the receiver operating characteristic curve, with comparisons by facility level. 16 key informant interviews with clinicians, laboratorist, and pharmacists explored data quality, and equity concerns. Quantitative data were analysed using Python-based tools, while qualitative data analysed thematically.
RESULTS:
Model performance was higher in tertiary hospitals than in secondary hospitals. Prediction accuracy was 82.4% in tertiary facilities and 68.7% in secondary facilities. Sensitivity for detecting resistance was 79.1% in tertiary hospitals compared to 61.3% in secondary hospitals. Facilities with more than 30% missing susceptibility data showed poor model performance. Interview participants highlighted limited laboratory coverage, incomplete records, and weak digital systems especially in secondary hospitals as key barriers to effective use.
CONCLUSIONS:
AI-based antimicrobial resistance surveillance is feasible in Nigerian public hospitals, but its effectiveness depends on existing system capacity. Without improvements in laboratory services and data quality, AI tools risk reinforcing existing inequities. Equity-focused implementation is essential for effective antimicrobial resistance surveillance in low-resource health systems
Antimicrobial resistance is a major public health concern in low- and middle-income countries, where surveillance systems are often weak. Many public hospitals face limited laboratory capacity, incomplete antimicrobial susceptibility testing, and poor data systems. Artificial intelligence has been proposed to improve antimicrobial resistance surveillance, but most existing evidence comes from high-income countries with strong health information systems. There is limited evidence on how these tools perform in public hospitals in resource-limited settings, or whether they may worsen existing gaps between better- and lower-resourced facilities.
METHODS:
A mixed-methods, retrospective study was conducted using routine hospital data from April 2024 to March 2025 in Kano State, Nigeria. Data were collected from four public hospitals, including two tertiary and two secondary facilities. Records of patients with bacterial infections were reviewed, yielding 2,480 selected through consecutive sampling. Model performance was assessed using accuracy, sensitivity, and area under the receiver operating characteristic curve, with comparisons by facility level. 16 key informant interviews with clinicians, laboratorist, and pharmacists explored data quality, and equity concerns. Quantitative data were analysed using Python-based tools, while qualitative data analysed thematically.
RESULTS:
Model performance was higher in tertiary hospitals than in secondary hospitals. Prediction accuracy was 82.4% in tertiary facilities and 68.7% in secondary facilities. Sensitivity for detecting resistance was 79.1% in tertiary hospitals compared to 61.3% in secondary hospitals. Facilities with more than 30% missing susceptibility data showed poor model performance. Interview participants highlighted limited laboratory coverage, incomplete records, and weak digital systems especially in secondary hospitals as key barriers to effective use.
CONCLUSIONS:
AI-based antimicrobial resistance surveillance is feasible in Nigerian public hospitals, but its effectiveness depends on existing system capacity. Without improvements in laboratory services and data quality, AI tools risk reinforcing existing inequities. Equity-focused implementation is essential for effective antimicrobial resistance surveillance in low-resource health systems
| eISSN: | 2654-1459 |
We process personal data collected when visiting the website. The function of obtaining information about users and their behavior is carried out by voluntarily entered information in forms and saving cookies in end devices. Data, including cookies, are used to provide services, improve the user experience and to analyze the traffic in accordance with the Privacy policy. Data are also collected and processed by Google Analytics tool (more).
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
