The fatigue failure of hip joint prostheses will be expected to assume more importance in second generation implants aimed at younger, more active patients. Furthermore, new designs and material combinations including coatings (e.g. hydroxyapatite) may introduce fatigue problems that as yet have not been considered. The current research makes an initial attempt to develop accelerated fatigue testing procedures to enhance the methodology of hip implant lifetime prediction. Tests conducted on a 'model' four point bendbar testpiece (mill-annealed Ti-6A1-4V) highlighted that the accelerated test must be conducted in a physiological solution such as Ringer's at 37 degrees C. The introduction of superimposed block overloads (50 cycles) to signify stair ascent/descent or fast walking and single overloads to signify sit/stand movements or stumbling were found to reduce fatigue life by > 50%. The findings of this fatigue study were combined with biomechanics studies to construct a variable amplitude 'in-service' load spectrum for testing hip implants. Using ambulatory trial data, a simple load sequence was designed containing 4 single (sit/stand movements) and 3 block (stair ascent/descent) overloads that repeated ten times gave one days activity; single overloads repeated every 110 base cycles (normal walking) and block overloads 80, 110 (morning/evening) and 250 (daytime) base cycles.