Understanding the healthcare provider costs of antimicrobial resistance (AMR) in lower-middle-income countries would motivate healthcare facilities to prioritize reducing the AMR burden. This study evaluates the extra length of stay and the associated healthcare provider costs due to AMR to estimate the potential economic benefits of AMR prevention strategies. We combined data from a parallel cohort study with administrative data from the participating hospitals. The parallel cohort study prospectively matched a cohort of patients with bloodstream infections caused by third-generation cephalosporin-resistant enterobacteria and methicillin-resistant Staphylococcus aureus (AMR cohort) with two control arms: patients infected with similar susceptible bacteria and a cohort of uninfected controls. Data collection took place from June to December 2021. We calculated the cost using aggregated micro-costing and step-down costing approaches and converted costs into purchasing power parity in international US dollars, adjusting for surviving patients, bacterial species and cost centres. We found that the AMR cohort spent a mean of 4.2 extra days (95% CI: 3.7-4.7) at Hospital 1 and 5.5 extra days (95% CI: 5.1-5.9) at Hospital 2 compared with the susceptible cohort. This corresponds to an estimated mean extra cost of $823 (95% CI: 812-863) and $946 (95% CI: US$929-US$964) per admission, respectively. For both hospitals, the estimated mean annual extra cost attributable to AMR was approximately US$650 000. The cost varies by organism and type of resistance expressed. The result calls for prioritization of interventions to mitigate the spread of AMR in Ghana.