BackgroundIndividuals suffering from chronic low back pain (CLBP) experience major physical, social, and occupational disruptions. Strong evidence confirms the effectiveness of Alexander technique (AT) training for CLBP.ObjectivesThe present study applied an integrative model (IM) of behavioral prediction for improvement of AT training.MethodsThis was a quasi-experimental study of female teachers with nonspecific LBP in southern Tehran in 2014. Group A contained 42 subjects and group B had 35 subjects. In group A, AT lessons were designed based on IM constructs, while in group B, AT lessons only were taught. The validity and reliability of the AT questionnaire were confirmed using content validity (CVR 0.91, CVI 0.96) and Cronbach’s α (0.80). The IM constructs of both groups were measured after the completion of training. Statistical analysis used independent and paired samples t-tests and the univariate generalized linear model (GLM).ResultsSignificant differences were recorded before and after intervention (P < 0.001) for the model constructs of intention, perceived risk, direct attitude, behavioral beliefs, and knowledge in both groups. Direct attitude and behavioral beliefs in group A were higher than in group B after the intervention (P < 0.03).ConclusionsThe educational framework provided by IM for AT training improved attitude and behavioral beliefs that can facilitate the adoption of AT behavior and decreased CLBP.