Background & Aims: We assessed the ability of 3 simple biochemical tests to stage liver fibrosis in patients co-infected with human immunodeficiency virus (HIV) and hepatitis C virus (HCV). Methods: We analyzed liver biopsy samples from 324 consecutive HIV/HCV-positive patients (72% men; mean age, 38 y; mean CD4+ T-cell counts, 548 cells/mm3). Scheuer fibrosis scores were as follows: 30% had F0, 22% had F1, 19% had F2, 23% had F3, and 6% had F4. Logistic regression analyses were used to predict the probability of significant (≥F2) or advanced (≥F3) fibrosis, based on numeric scores from the APRI, FORNS, or FIB-4 tests (alone and in combination). Area under the receiver operating characteristic curves were analyzed to assess diagnostic performance. Results: Area under the receiver operating characteristic curves analyses indicated that the 3 tests had similar abilities to identify F2 and F3; the ability of APRI, FORNS, and FIB-4 were as follows: F2 or greater: 0.72, 0.67, and 0.72, respectively; F3 or greater: 0.75, 0.73, and 0.78, respectively. The accuracy of each test in predicting which samples were F3 or greater was significantly higher than for F2 or greater (APRI, FORNS, and FIB-4: ≥F3: 75%, 76%, and 76%, respectively; ≥F2: 66%, 62%, and 68%, respectively). By using the lowest cut-off values for all 3 tests, F3 or greater was ruled out with sensitivity and negative predictive values of 79% to 94% and 87% to 91%, respectively, and 47% to 70% accuracy. Advanced liver fibrosis (≥F3) was identified using the highest cut-off value, with specificity and positive predictive values of 90% to 96% and 63% to 73%, respectively, and 75% to 77% accuracy. Conclusions: Simple biochemical tests accurately predicted liver fibrosis in more than half the HIV/HCV co-infected patients. The absence and presence of liver fibrosis are predicted fairly using the lowest and highest cut-off levels, respectively. © 2009 AGA Institute.