This is consistent with the two trials (Kjellman and Oberg

2002, Viljanen et al 2003) that reported medium- (WMD –2, 95% CI –7 to 4) and long-term (WMD –0.1, 95% CI –6 to 6) pain outcomes. Pooled results from the two trials that reported disability outcomes (Kjellman and Oberg 2002, Viljanen et al 2003) from general strength and conditioning exercise showed no significant difference compared with minimal intervention at the conclusion of treatment (WMD 1, 95% CI –3 to 5) or medium- (WMD 1, 95% CI –3 to 5) or long-term (WMD –3, 95% KU-57788 clinical trial CI –7 to 2) follow-up. Manual therapy: In the three included trials of manipulation, there were four sham-controlled comparisons of the immediate analgesic effect of a single high-velocity manipulation. One trial ( Cleland et al 2005) investigated the effect of thoracic spine manipulation on neck pain and two trials ( Martinez-Segura et al 2006, Pikula 1999) investigated cervical spine manipulation. The three-arm trial by Pikula

and colleagues (1999) compared two different manipulation techniques with sham. The two manipulation groups in this trial were combined to create a single pair-wise comparison. Three trials learn more ( Hemmila 2005, Hoving et al 2002, 2006, Skillgate et al 2007) were identified that compared manual therapy with minimal or no intervention. Pooled outcomes from three trials (Cleland et al 2005, Martinez-Segura et al 2006, Pikula 1999) show a significant analgesic benefit from a single manipulation compared with control (WMD –22, 95% CI –32 to –11). Medium- and longterm outcomes were not reported in these trials. Disability was not assessed. Pooled outcomes from two trials (Hoving et al 2002, Skillgate

et al 2007) show that manual therapy provided better pain relief after a course of treatment than minimal treatment (WMD –12, 95% CI –16 to –7). A similar benefit was not reported in the single trial (Hoving et al 2006) that reported medium- (MD –7, 95% CI –16 to 2) and long-term (MD –1, 95% CI –11 to 9) pain outcomes. Pooled outcomes from three trials (Hemmila 2005, Hoving et al 2002, Skillgate et al 2007) show that manual therapy resulted in significantly better disability old outcomes at the conclusion of treatment than control (WMD –6, 95% CI –11 to –2). A similar benefit was not demonstrated in the two trials (Hemmila 2005, Hoving et al 2006) that reported medium- (WMD –8, 95% CI –24 to 7) and long-term (WMD –1, 95% CI –12 to 9) disability outcomes. Multimodal physical therapies: Two trials compared multimodal physical therapies, which included exercises, massage, and various electrotherapies, with minimal treatment. One trial excluded manual inhibitors therapies ( Hoving et al 2002, 2006), and one trial included manual therapies ( Palmgren et al 2006) in the range of treatments provided.