OBJECTIVES: Starting point (SP) bias has been reported in some willingness-to-pay (WTP) analyses using a bidding game (BG) approach. The objective of the present study was to systematically review WTP analyses which used more than one SP to ascertain if any bias was observed.  METHODS: Using OVID MEDLINE® for 1996-2011 and the keywords [Bidding AND (Willingness to pay OR Willingness-to-pay OR WTP)], two reviewers independently identified and examined WTP analyses which applied more than one SP to a BG approach and which discussed any associated bias.  Additional publications were also extracted from the references of relevant articles. Any discrepancy between the two reviewers was resolved through consensus. RESULTS: From 51 articles yielded by the literature search, 12 met the inclusion criteria and a further 5 were classified as secondary evidence (publications reporting on SP bias without presenting numerical data). Relevant publications were based in Asia (k=4), North America (k=3), Africa (k=3) and Europe (k=2). The target population of these publications were the general public (k=7), patients (k=4) or both (k=1).  A slight majority of the 12 primary evidence publications (k=7) and a larger majority of the 5 secondary evidence publications (k=4) reported a SP bias on the results of the BG. Various parameters such as male gender, higher education and higher income levels were, in some instances, associated with higher WTP amounts. Other factors analysed are the population surveyed (patients vs. general population), and the location of the study.  Association between these factors and the occurrence of starting point bias is examined and will be reported. CONCLUSIONS: There is evidence in the literature of a SP bias on the results of BGs, without however a full consensus on the matter. Further research is warranted in order to evaluate the conditions under which such bias appears.