Fourteen studies (n=801) were included in the review, of which three were RCTs (n=453), two were cohort studies (n=101), one was a case study (n=1), five were described as experimental studies (n=124) and three were described as quasi-experimental studies (n=122).
The largest RCT (n=258) found that pregnant women randomised to a water gymnastics group had significantly lower pain intensity than women in the control group in the second half of pregnancy (p=0.031). There was no significant difference between the groups in number of days of sick leave, but significantly fewer women in the APT group were on sick leave due to LBP after weeks 32 to 33 of gestation (p=0.031).
A second RCT (n=95) assessed group hydrotherapy in patients with chronic LBP and back and leg pain. A number of measures were assessed, of which only functional status showed a significant difference between the groups, favouring the intervention (p=0.04).
The third RCT (n=100) compared balneotherapy with balneotherapy plus exercise or exercise alone. Significant improvements on a number of measures of ROM and flexibility were observed in the balneotherapy groups (p<0.001) but not in the exercise group, which showed improvements only on pain score (p<0.05). Statistical comparisons between the groups were not reported.
Other study designs reported variable results for a range of different forms of APT. The most frequent positive outcome was pain reduction; full results were reported in the paper.