Objectives: The objective of this study was to conduct a systematic review of evidence for the accuracy of the Kimberley Indigenous Cognitive Assessment (KICA) tool in supporting the diagnosis of dementia in Indigenous Australian populations. Methods: Cross-sectional diagnostic accuracy studies of the KICA with an appropriate reference standard published to November 2015 were included. Comparison to an alternative cognitive assessment tool was required in nonremote populations. Case control analyses were excluded. Results: Four studies were included: one of the KICA-Cog and KICA-Carer, one of the KICA Screen, and two of the modified-KICA. All tools developed for remote populations had a sensitivity of ≥76% and a specificity of ≥71% for the diagnosis of dementia. The KICA-Cog and KICA-Carer conducted in series had the highest sensitivity and specificity (91% and 94% respectively). In an urban and regional population, the mKICA had similar accuracy to the Mini-Mental State Examination (MMSE) (AUC 0.93, 95% CI 0.88-0.99 vs 0.94, 95% CI 0.89-0.99). Key risk of bias limitations related to lack of pre-determined cut-points and population selection methods. Conclusion: The use of the KICA in remote Indigenous Australians may assist in timely diagnosis of dementia in this population. Using the KICA-Cog and KICA-Carer in series may maximise specificity, decreasing false positive results without compromising sensitivity.