Trials of manipulative treatment have been compromised by, amongst other things, different definitions of the therapeutic procedures involved. This paper describes a spinal manipulation package agreed by the UK professional bodies that represent chiropractors, osteopaths and physiotherapists. It was devised for use in the UK Back pain Exercise And Manipulation (UK BEAM) trial--a national study of physical treatments in primary care funded by the Medical Research Council and the National Health Service Research and Development Programme. Although systematic reviews have reported some beneficial effects of spinal manipulation for low-back pain, due to the limited methodological quality of primary studies and difficulties in defining manipulation, important questions have remained unanswered. The UK BEAM trial was designed to answer some of those questions. Early in the design of the trial, it was acknowledged that the spinal manipulation treatment regimes provided by practitioners from the three professions shared more similarities than differences. Because the trial design specifically precluded comparison of the effect between the professions, it was necessary to devise a homogenous package representative of, and acceptable to, all three. The resulting package is 'pragmatic', in that it represents what happens to most people undergoing manipulation, and 'explanatory' in that it excludes discipline-specific variations and other ancillary treatments.