A health promoting-care approach includes supporting the individual to increase control over and to improve his or her health to obtain the best possible wellbeing and quality of life. In prediabetes and type 2 diabetes regular physical activity is of pronounced importance for preventing complications and premature death as well as to enhance quality of life. Supporting the individual in self-management of physical activity and establishing new routines in daily life is a major challenge for the health-care system. There is a need for feasible, low-cost intervention programs to support physical activity in type 2 diabetes care.

The overall aim of this thesis was to design and evaluate the implementation of an intervention in primary care intended to support individuals with prediabetes or type 2 diabetes to become physically active on a regular basis. A secondary aim was to explore the theoretical associations of replacing prolonged sedentary time with time being sedentary but taking breaks, engaging in light intensity physical activity or moderate and vigorous physical activity with health parameters.

Study I was a study protocol describing the assessment protocol and the theoretical framework underlying the randomized controlled trial Sophia Step Study.

Study II was a process evaluation study describing the context of the Sophia Step Study and evaluating the implementation during the first 6 months of intervention. Intervention delivery and dose received were obtained through the process of continuous dialogue with the diabetes specialist nurses, attendance records and data on the number days for step registration and step goal. Changes in physical activity behavior were measured objectively for 159 participants by accelerometers and health outcomes by blood samples, anthropometry and blood pressure. Study III was a qualitative interview study reporting the experiences of 18 adhering participants' after attending 2 years of the Sophia Step Study. Study IV employed a cross-sectional design that included 124 individuals with prediabetes or type 2 diabetes. The study investigated the associations of reallocating sedentary time in bouts (>60 min) to sedentary time in non-bouts (<60 min), light intensity physical activity and to moderate and vigorous physical activity with cardiometabolic risk factors.

The participants in the Sophia Step Study were randomized into one of three parallel groups: a multicomponent intervention (A) entailing individual consultations with a diabetes specialist nurse based on motivational interviewing and physical activity on prescription, group meetings and self-monitoring of steps; a single component intervention (B) encompassing self-monitoring of steps; or a control group (C) entailing usual care except for the assessments that were included in the study (study I). Between April 2013 and October 2016 159 persons were recruited. The interventions were feasible to implement in primary care with a low dropout rate (3%) at 6 months and high fidelity to the study protocol, except for the physical activity on prescription component. There was a high grade of delivery and dose among those enrolled in the study. Group A increased mean daily steps with 1097 steps (CI: 232, 1962), group B increased mean daily steps with 1242 steps (CI: 313, 2171) and the control group (C) decreased mean daily steps with 457 step (CI:-1164, 250). Clinical improvements after 6 months were found in the two intervention groups, while the control group showed mixed results (study II). In total 18 participants with high adherence to the interventions and study assessments were interviewed in study III. The health check-ups were described as personalized, giving feedback on health outcomes and positive reinforcement. Overall, the participants felt that they received good care throughout the study. The self-monitoring of steps, the group sessions and the health check-ups were recalled as resources that increased motivation for physical activity, led to the establishment of new daily routines and in empowering them to take control over their own health. Study IV demonstrated that modeling reallocation of 30 min of objectively measured time from sedentary time in 60 min bouts to moderate and vigorous physical activity was beneficially associated with BMI (b = -1.46 95% CI: -2.60, -0.33 kg/m2), waist circumference (b = -4.30 95% CI: -7.23, -1.38 cm), and HDL cholesterol b = 0.11 95% CI: 0.02, 0.21 mmol/l). No associations of reallocating sedentary time from long bouts to shorter bouts or to light intensity physical activity were seen. Accumulating time in moderate and vigorous physical activity was beneficial, independent of the behavior it replaced.

In conclusion, this thesis shows that it is possible to implement self-monitoring of steps both with and without counseling support as a strategy to support individuals with prediabetes or type 2 diabetes in the primary care setting. The two interventions were effective in increasing physical activity after 6 months. Self-monitoring of steps, feedback on health outcomes and a personalized approach were highly valued by the individuals and should be considered in primary care. The thesis also confirmed that modeling replacement of sedentary time with time in moderate and vigorous physical activity showed beneficial associations with HDL cholesterol, waist circumference and BMI levels in individuals with prediabetes or type 2 diabetes.