Background: Antimicrobial resistance (AMR) is a global health threat responsible for increased healthcare costs and mortality. The World Health Organization (WHO) global action plan on AMR recommends antimicrobial sensitivity testing (AST) and surveillance of antibiotic use to address this threat. We conducted a cross-sectional study to determine the utility of AST in three major referral hospitals in Uganda. Methods: We used abstraction checklists to collect data on AST requests, AST performed, AST turnaround time, and dispensed antibiotics, from in-patient files and laboratory and pharmacy records. Antibiotic data were summarized using proportions. The relationship between dispensed antibiotics and AST antibiotic-disks was analyzed using spearman’s rank correlation and simple linear regression. Results: Of the 607 in-patient files reviewed, AST was requested in 24 (4.0%), and done in 13 (2.1%). All three hospitals used the Kirby-Bauer disk diffusion method for AST, with a median turnaround time of 5 days (IQR 4-8). While the frequently used AST antibiotic-disks were cotrimoxazole (13.0%), ampicillin (11.7%), and ceftriaxone (10.5%), the most dispensed antibiotics were metronidazole (30.3%), amoxicillin (19.6%) and ceftriaxone (14.8%). There was a weak correlation (r=0.313, p=0.120) between dispensed antibiotics and AST antibiotic-disks, and AST performance was not associated with antibiotic consumption. Conclusion: We report an underutilization of AST, inconsistent with the hospital antibiotic consumption, and may be related to the use of the disk diffusion method. We recommend alternative faster and better coordination in the procurement of AST diagnostics in Ugandan hospitals.