Physical activity behaviors, associated parameters, and the feasibility of an online physical exercise intervention in individuals with diabetes

Introduction
Diabetes has become one of the fastest growing public health challenges of the 21^st century. The incidence of disability and morbidity caused by diabetes has increased significantly in the last decades. Engagement in regular physical activity is a cornerstone of diabetes management, not only to prevent long-term complications and premature mortality, but also to improve overall well-being and health. While the evidence and recommendations are clear on providing physical activity in prevention and treatment of diabetes, increasing and maintaining physical activity in individuals with diabetes remains challenging. Understanding physical activity behaviors and how to improve it in individuals with diabetes needs to be investigated from different perspectives to understand why regular engagement in physical activity can be a major challenge. 

Aim
The overall aim of this thesis is to provide detailed descriptions of physical activity characteristics, determinants, and patterns among Danish individuals with diabetes sampled from recent populationbased studies and to evaluate the feasibility, fidelity, and acceptability of an online physical exercise intervention combined with group sessions to increase and maintain physical activity in individuals with type 2 diabetes.

Methods
Paper I was performed as a nationwide cross-sectional study using data from the Danish National Health Survey from 2017. The primary outcome measure was weekly moderate to vigorous physical activity (MVPA). Exposures included self-reported number of chronic conditions, body mass index (BMI), perceived stress, and health-related quality of life (HRQoL). Mean differences in MVPA across exposures were estimated by multiple linear regression analyses.
Paper II and III were performed as cross-sectional studies, both using data from the Danish household-based, mixed rural-provincial population study, The Lolland-Falster Health Study from 2016-2020. Participants were categorized into diabetes, prediabetes, and no diabetes based on their HbA1c level and self-reported use of diabetes medication. For Paper II, the outcome was physical activity in terms of intensities and timing of engagement in activity assessed with a lower-back worn accelerometer, and exposure was diabetes status (diabetes, prediabetes, no diabetes). The outcomes in Paper III were total daily sedentary activity, sedentary bouts, and breaks in sedentary activity assessed with thigh-worn and lower back-worn accelerometers, and exposures were wellbeing and stress. Finally, Paper IV was designed as a one-armed feasibility study and the intervention was developed using a co-creation approach. Individuals with type 2 diabetes were recruited for eight weeks of 30-minute online physical exercise intervention followed by 30-minute online group meetings in smaller groups once a week. Outcomes included pre-defined research progression criteria, secondary measurements of health parameters, and participant feedback.

Results
Paper I: Among individuals with diabetes, 40% were insufficiently physically active. Having three or more comorbidities were associated with lower weekly MVPA (-0.48 h/week, 95% CI -0.88; - 0.07) compared with participants with no comorbidities. Furthermore, overweight or obese individuals engaged in less weekly MVPA (obese class III vs. normal weight: -1.98 h/week, 95% CI -2.49; -1.47). Higher perceived stress was associated with lower weekly MVPA (-1.76 h/week, 95% CI -2.18; -1.34) vs. low perceived stress. An association between low HRQoL and lower weekly MVPA was found when compared to those with moderate or high HRQoL (physical HRQoL: -0.93 h/week, 95% CI -1.19; -0.66 and mental HRQoL: -0.39 h/week, 95% CI -0.71; - 0.08).
Paper II and III: Of participants with diabetes, 63% did not adhere to the WHO recommendations of weekly MVPA, while numbers of participants with prediabetes and no diabetes were 60% and 50%, respectively. Around a third of those with diabetes were highly inactive (<5 min/day of MVPA) daily and had >2 consecutive days of inactivity. Mean time spent physically active daily at any intensity was lowest among participants with diabetes. Higher well-being was associated with lower total sedentary activity in participants with diabetes (-1.1 min/day, 95% CI -2.0; -0.2) and participants with prediabetes (-0.6 min/day, 95% CI -1.1; -0.05). No association was found between stress and sedentary activity.
Paper IV: Most research progression criteria reached a level of acceptance, with exception of participant recruitment, burden of objectively measured physical activity, and adverse events related to the intervention, where changes are needed before continuing to an RCT.

Conclusion
Our findings showed that on a nationwide level, 40% of individuals with diabetes do not meet the WHO recommendations of physical activity. The objectively measured physical activity findings showed that 63% of individuals with did not adhere to the WHO recommendations. The associations analyses revealed that the prevalence of comorbidities, higher BMI, higher perceived stress, and lower HRQoL were significantly associated with lower engagement in weekly MVPA in individuals with diabetes. Further, higher well-being contributed to lower daily sedentary activity. This underscores the necessity for tailored interventions that recognize these factors to enhance physical activity engagement. Thus, an intervention combining online physical exercise and group sessions may be a feasible approach to accommodate some of these factors and other barriers to participating in physical activity. Altogether, the findings of this thesis emphasize the importance of acknowledging differences in physical activity and sedentary behaviors among individuals with diabetes and how these differences could be accounted for and integrated into future tailored interventions to enhance physical activity engagement in this population.