Oral cancer is a major health problem, with over 500000 patients diagnosed each year. Although progress has been made in regards to both diagnosis and treatment, the overall 5-year survival rate has not changed much in the last 30 years and still relatively static at around 50 %. This high mortality rate is related to late diagnosis of oral malignancies. At stages III and IV the 5-year survival rate is as low as 30 %, however if diagnosed at stage I the survival rate increases substantially to approximately 80 %. Research in to viable methods for early detection of oral malignancies and potential malignancies has the potential to save lives and reduce suffering for many people word wide. The purpose of this systematic review is to find an effective and practical diagnostic test for early detection of malignancies and potential malignancies in the oral cavity. Electronic database searches were conducted in English on the 4th of February 2017, using PubMed medical database, publication date 2005 or later. Inclusion criteria: Diagnostic Test Accuracy studies for oral malignancies and potential malignancies in human patients with gold standard reference test. Search yielded 166 records, titles and abstracts was screened and evaluated, 19 records was included. Included studies were assessed in detail regarding methodological quality and diagnostic accuracy. 19 studies with a total of 11575 participants were included. The studies were subdivided in to groups based on the specific index test assessed in each study. Diagnostic accuracy results: 4 records assessed Tissue autofluorescence; sensitivity ranging from 65.5 % to 100 % and specificity ranging from 41.7 % to 97.4 %. 10 records assessed Brush biopsy sampling; sensitivity ranging from 45 % to 100 % and specificity ranging from 90 % to 100 %. 1 record assessed Blood sampling; sensitivity 64 % and specificity 80 %. 2 records assessed saliva sampling Saliva sampling; sensitivity 100 % and specificity 96.7 %. 1 record assessed Metachromatic dye staining; detection rate of experimental group was 5 % higher than control group. 1 record assessed Narrow-band imaging; sensitivity 95 % and specificity 97 %. The Brush biopsy sampling diagnostic test methods has a body of evidence that far exceeds any other category presented in this review. As for methodological quality, diagnostic accuracy and risk of bias, we deemed the category as a whole to be at an acceptable level.