Infections with multidrug resistant gram-negative bacteria is a growing problem especially in health care settings. Colistin is one of the last resort antibiotics for such infections in which treatment options are limited. Increasing resistance to colistin is a global problem. Clinical and Laboratory Standards Institute (CLSI) and the European Committee on Antimicrobial Susceptibility Testing (EUCAST) study groups have recommended the ISO-standard broth microdilution method (20776-1) as the reference method for the determination of colistin susceptibility. Since the broth microdilution method is not a practical method, it is rarely used in routine clinical microbiology laboratories, yet simple and accurate phenotypic detection methods for the determination of colistin resistance in routine microbiology laboratories are not precisely defined. The aim of this study was to evaluate BD Phoenix100 (Becton Dickinson, USA) system and colistin broth disk elution method for the detection of in vitro activity of colistin against gram-negative bacteria. A total of 419 gram-negative bacteria, including 199 Klebsiella pneumoniae, 163 Acinetobacter baumannii, 34 Escherichia coli, 20 Enterobacter spp., and three Citrobacter spp. isolates which were isolated from various clinical samples in our hospital between 2016-2018 were tested. The broth microdilution method was used as the reference method applying ISO-standard broth microdilution methods (20776-1) and CLSI/EUCAST recommendations. For colistin broth disk elution method, final concentrations of 0 (growth control), 1, 2 and 4 μg/ml were obtained by adding 10 μg colistin disks to four tubes containing 10 ml cation-adjusted Mueller Hinton broth per isolate. After incubation at room temperature for 30 minutes, 50 μl of standardized inoculum suspensions were added to the tubes. Colistin minimum inhibitor concentration (MIC) values were read visually after 16-20 hours of incubation at 35°C in ambient air. Manufacturer's recommendations were followed for BD Phoenix100 system. The categorical agreement between the reference broth microdilution method and the colistin broth disk elution method was 99.3%, very major error and major error rates were 0.2% and 0.5%, respectively. For BD Phoenix100 system, the categorical agreement was 95%, with a very major error rate of 5%. Our results showed that colistin broth disc elution method worked well compared to the reference broth microdilution method. The BD Phoenix100 system, with a high very major error rate, does not reliably distinguish colistin-resistant and colistin-susceptible strains.