Integration of large language models in Nigeria's healthcare: doctors early perspectives, operational barriers, and privacy concerns

Abstract The advent of large language models (LLMs) like ChatGPT has the potential to revolutionize healthcare, offering opportunities for improved patient engagement, clinical decision support, and administrative efficiency. This pioneering study aimed to capture the early perspectives of Nigerian medical professionals on LLMs, particularly ChatGPT, immediately following its 2022 introduction by OpenAI. It explores initial responses, anticipated barriers, and data privacy concerns during this transformative period. A nationwide cross-sectional survey was conducted between February and March 2023, involving 406 medical doctors across Nigeria's six geopolitical zones. Among those aware of ChatGPT, 66.9% were willing to use it. Most (41.4%) were "not very confident" using LLMs for patient care. The majority (54.0%) rated LLMs "somewhat important" for healthcare integration, and 76.6% believed they could improve outcomes. No explored factors significantly influenced willingness to use LLMs. Primary perceived benefits included improved patient information access (65.2%), clinical decision-making efficiency (57.7%), and patient-provider communication (46.4%). Major concerns were misinterpretation (80.1%), technology dependence (66.3%), liability (65.5%), lack of human interaction (60.7%), and privacy/security (51.3%). These findings highlight both optimism and apprehension towards LLMs, recognizing potential benefits and concerns about misinterpretation, overreliance, legal implications, and privacy, thus underscoring the need for educational programs, ethical frameworks, and robust data protection.