Awareness, adoption, and predictors of artificial intelligence use among public health researchers in Bo District, Sierra Leone: a cross-sectional study

Abstract

Background Artificial intelligence (AI) is increasingly applied in global health research, yet empirical evidence on its uptake in low-resource settings is scarce. This study assessed awareness, adoption, and predictors of AI use among public health researchers in Bo District, Sierra Leone. Methods A cross-sectional survey was conducted with 200 public health researchers and research-related personnel across university, hospital, government, and non-governmental organisation (NGO) settings, selected by proportional stratified random sampling. Data were analysed with descriptive statistics, chi-square tests of independence, and binary logistic regression at a 5% significance level. Results Awareness of AI was high (149/200; 74.5%), but use of AI tools was low (71/200; 35.5%), and only 26.5% (53/200) had received any formal AI training. AI use was descriptively higher among trained respondents (43.4% versus 32.7%), but the association between training and adoption was not statistically significant (χ² = 1.964, df = 1, p = 0.161). The association between AI use and perceived improvement in research quality was likewise non-significant (χ² = 0.821, df = 2, p = 0.663). In binary logistic regression, none of the examined predictors AI tool use, AI training, institution type, or years of experience significantly predicted perceived improvement (all p > 0.05). Conclusions Despite high awareness, AI adoption among public health researchers in Bo District is at an early stage and is not explained by training or institution type alone. The awareness–adoption gap points to structural and infrastructural determinants that individual-level factors do not capture, underscoring the need for structured capacity-building embedded within supportive institutional environments.