Manipulative therapy as part of a multidimensional approach may be more effective than standard physical therapy in treating Acute Nonspecific Low Back Pain. 64 participants, 29 women and 35 men, with Acute Nonspecific Low Back Pain and a mean age of 40 yr. (SD=9.6) were randomly assigned to two groups: an experimental group (manipulative therapy plus physical therapy) and a control group (only physical therapy). A multicentre, nonblinded, randomised clinical trial was conducted. Pain relief was the main performance criteria measured together with secondary criteria which included functional status and mobility of the lower back. Fritz, Childs, and Flynn's clinical prediction rule--a duration of symptoms less than 16 days, no pain distal of the knee--was used to analyse the results. In combination with an age >35 years, results showed a statistical significant effect for disability, but no statistically significant benefit of additional manipulative therapy over physical therapy found for pain and mobility within 4 treatments. Controlled for the applied clinical prediction rule, there were statistically significant interaction effects with low effect size for disability and sex, but no significant effects were found for pain of mobility.