Background: There are 880.000 people living with low back pain in Denmark. The majority of the patients are diagnosed with so called nonspecific low back pain. The social and economic consequences are huge both for the individual patient and the society. In the region of Central Jutland the Process Program for low back pain is an attempt to strengthen the cooperation between different parts of the health care system. As a part of the process program the Center for Health Courses provides a Pilates-based exercise therapy program. This program is aimed at patients with chronic nonspecific low back pain living in the municipality of Aarhus. Although there is strong evidence for exercise therapy as treatment for low back pain, there is yet not established a gold standard. There is also a lack of knowledge about predictors for the outcome for patients participating in an exercise therapy program for low back pain. Objective: To investigate the efficacy of and predictors for outcome of a Pilates-based exercise therapy program for chronic nonspecific low back pain in a municipality setting in Denmark. Method: In the years 2012-2015 data from 148 patients with chronic nonspecific low back pain participating in the exercise therapy program at the Center for Health Courses was collected. All patients were asked to fulfill the Oswestry Disability Index (ODI) and to rate Self-rated health at the beginning and at the end of the program. The Pilates-based exercise therapy program consisted of two sessions of 75 minutes per week for nine weeks. Patients were encouraged to combine the two sessions with home based exercises. Results: For the 98 patients that fulfilled the ODI both at the beginning and at the end of the program there was a significant (P=0,001) increase of the score of 3,5 percentage points. The only significant (P<0,001) predictor of the outcome was a baseline ODI score under 11,5%. For the 84 patients that rated Self-rated health both at the beginning and at the end of the program there were a significant (P=0,01) improvement of Self-rated health. There was no significant (P=0,064) correlation between change in Self rated health and change in ODI score. Conclusion: The Pilates-based exercise therapy program did worsened pain and disability, but the mean change was under the level for minimal clinically important difference for the Danish version of the ODI. Patients with low baseline scores increased their scores the most. Self-rated health improved, and there was no correlation between improvement in self-rated health and decrease of ODI score.