OBJECTIVES: To compare cost-effectiveness model input influence on incremental net monetary benefit (INMB) across three methods of uncertainty analysis: 1) one-way sensitivity analysis; 2) probabilistic analysis of covariance (ANCOVA); and 3) expected value of partial perfect information (EVPPI). METHODS: We replicated and expanded a published HIV/AIDS cost-effectiveness Markov model (monotherapy vs. combination therapy) using TreeAge®.  Case 1 assumed a willingness-to-pay of £20,000/QALY (relatively low decision uncertainty in this application).  Case 2 assumed a willingness-to-pay of £8,000/QALY (relatively high decision uncertainty).  For Cases 1 and 2, one-way sensitivity analysis identified the ten most influential inputs.  From these ten inputs, we estimated ANCOVA results (10,000 Monte Carlo draws) and EVPPI for each input (1,000 inner and 1,000 outer draws).  For each case and method, we ranked inputs based on their influence on variation of INMB and compared input ranks within case using Spearman’s rank correlation.    RESULTS: Mean INMB was £9,740 (Case 1) and £179 (Case 2) in favor of combination therapy.  Case 1: The two most influential inputs were the same across all uncertainty methods, contributed 78% of variation in outcome (ANCOVA), and were the only inputs with non-zero EVPPI values.  Case 2: All inputs had non-zero EVPPI values, with the two most influential inputs accounting for 49% of variation in outcome (ANCOVA). For Cases 1 and 2, the influential input rank order correlations across uncertainty methods ranged from 0.70 to 0.99 (all p-values <0.05 for pairwise uncertainty method correlations for both cases). CONCLUSIONS: For both cases, the influential input ranks were positively correlated between one-way and more advanced uncertainty analyses, indicating influential input rank agreement.  Although each method provides unique information, the additional resources needed to generate and communicate advanced analyses should be weighed, especially when the outcome decision uncertainty and therefore value of information is low. (i.e. Case 1).