We report the case of a 45-year-old man who was diagnosed in June 1969 9 years old as having mild haemophilia A following a traumatic left shoulder bleed when his factor VIII (FVIII) activity was 11 IU dL(-1) based on a two-stage assay. The bleed resolved following treatment with intravenous cryoprecipitate. There was no family history of haemophilia. Cryoprecipitate infusions were required to treat further traumatic bleeds between 1971 and 1981. During this time, his FVIII activity was confirmed at 14 IU dL(-1). He defaulted many hospital appointments until 1991 when his FVIII activity had risen to 42 IU dL(-1). There was no evidence of infection, inflammatory or liver disease to account for this change. By 2005 he had a normal FVIII activity of 62 IU dL(-1) based on a one-stage assay. FVIII gene analysis confirmed a codon 531 mutation. It appeared that the discrepant FVIII results related to whether a two-stage or one-stage assay was used that has been previously reported for other patients with these mutations. We felt it important to raise awareness that this phenomenon may lead to apparent correction of haemophilia A.