Aims• To determine whether a basic set of Pilates exercises improves the efficiency of load transfer through the pelvis • To compare the effects on chronic, mild low back pain (LBP) symptoms of three slightly different Pilates based regimes MethodsA between subjects equivalent group experimental design was used -Independent variable: type of exercise training (three groups) -Dependent variables: efficient load transfer through the pelvis as measured by the stork test in weight bearing; low back pain symptoms At entry, to establish baseline values, subjects completed an Oswestry DisabilityQuestionnaire and recorded the frequency, intensity and duration of their back pain in an average week. Also, a Stork test was recorded.Thirty-nine volunteers with mild chronic low back pain (CLBP) were taught four Pilates based exercises before being randomly allocated to one of three groups for the addition of other interventions.• Groups A received four basic exercises• Groups B and C received an additional relaxation posture using a specific spinal support • Group C received an additional postural training exercise Exercises were performed three times per week for six weeks and recording was done once per week for eight weeks. For each of the eight weeks, subjects recorded frequency, intensity and duration of back pain ii At the final assessment, subjects completed another Oswestry Disability Questionnaire, returned their recording sheets, had a stork test recorded and their exercises checked. Results Oswestry Disability QuestionnairePre and post comparison of answers showed only one statistically significant improvement among subject groups. This was for question one, 'Do you have back pain at present?' where Group B reported significantly less pain post program compared with pre program (Wilcoxon, z=-2.496, p=0.013). Number of days of painGroup B experienced a statistically significant reduction in the number of days of pain between Week 1 and Weeks 6 to 8 (F 7,84 =6.4, p=0.0001). Post hoc analysis using Scheffé showed significant differences between Week 1 and Weeks 6, 7 and 8 and between Week 2 and Week 8 (p<0.05). There were statistically significant differences by week within Group C (F 7,77 =3.29, p=0.0041), but they only show up with Fisher (p<0.05) and they were between Week 1 and Weeks 6, 7 and 8, Week 2 and Weeks 6, 7 and 8, Week 3 and Weeks 6 and 7 and between Week 4 and Week 7.Some of the improvements were lost once exercising ceased at the end of week 6 Duration of back pain episodesAll groups experienced a reduction in the mean length of the shortest, longest and average pain episodes. At week eight all groups had subjects who were pain free (Group A: 7.7%, Group B: 30.8%, Group C: 25%) and in Group B, no subjects reported pain episodes longer than six hours. However, differences were not statistically significant for the duration of this study.Some of the improvements were lost once exercising ceased at the end of week 6iii Intensity of pain across all lengths of pain episodes While, all groups exper...