A large part of the scientific community has accepted the paradigm that a simulated body solution (SBF) can be used to test the bioactivity of a material. This is exemplified by the rapidly increasing number of publications using this test. The aim of this document is to demonstrate that (i) there is presently not enough scientific data to support this assumption, and(ii) even though the assumption was valid, the way the test is generally conducted leaves room for improvement. Theoretical arguments and facts supporting these statements are provided, together with possible improvements of the proposed bioactivity test.