Using machine learning algorithms to predict COVID-19 vaccine uptake: A year after the introduction of COVID-19 vaccines in Ghana.

The impact of vaccine hesitancy on global health is one that carries dire consequences. This was evident during the outbreak of the COVID-19 pandemic, where numerous theories and rumours emerged. To facilitate targeted actions aimed at increasing vaccine acceptance, it is essential to identify and understand the barriers that hinder vaccine uptake, particularly regarding the COVID-19 vaccine in Ghana, one year after its introduction in the country. We conducted a cross-sectional study utilizing self-administered questionnaires to determine factors, including barriers, that predict COVID-19 vaccine uptake among clients visiting a tertiary and quaternary hospital using some machine learning algorithms. Among the findings, machine learning models were developed and compared, with the best model employed to predict and guide interventions tailored to specific populations and contexts. A random forest model was utilized for prediction, revealing that the type of facility respondents visited and the presence of underlying medical conditions were significant factors in determining an individual's likelihood of receiving the COVID-19 vaccine. The results showed that machine learning algorithms can be of great use in determining COVID-19 vaccine uptake.

A cross-sectional study on antibiotic prescription in a teaching hospital in Ghana.

INTRODUCTION: Antibiotic misuse is the paramount factor for antibiotic resistance. Tamale Teaching Hospital (TTH), located in Ghana's Northern Region, is the biggest tertiary hospital in the Northern half of the country and consequently one of the biggest prescribers of antibiotics. Understanding the use of antibiotics in the TTH and providing information that could be inferred to develop strategies for antibiotic prescription is of extreme importance in this era of multiple and pan-resistant strains of pathogenic microorganisms. METHODS: A cross-sectional study on the use of antibiotics at TTH in the Northern region of Ghana was performed. Data were collected by reviewing 10% of patients' files from January to June 2015 and then assessed for its appropriateness against the criteria based on the British National Formulary (BNF) 2015 and BNF children 2013-2014. Results were expressed in frequencies and percentages. RESULTS: A total of 617 patients' records were included in this study. Up to 385 cases of different antibiotic misuse were found, comprising of 335 errors in prescriptions and 50 non-completed treatments. The most common prescription error was made on treatment duration (29.6%). The potential interactions were 16.7%. CONCLUSION: The study revealed a high burden of antibiotics misuse in TTH. This suggests a need for the development of an antibiotic stewardship programme for the hospital.