Patterns of physical activity and the effect of accelerometer wear on physical activity participation in people with Type 2 diabetes

Data were taken from a trial comparing three physical activity interventions, in 134 people with T2D (age=61.3±10.3yrs; BMI=33.32±6.9kg/m2). The interventions were a one-to-one consultation, a written-delivered pack and a leaflet. Physical activity was measured over seven days, using the GT1M accelerometer, pre-intervention and 6 and 12 months post-intervention. Weekly and daily total accelerometer and step counts were recorded then analysed using analysis of variance. Significance was set at p < 0.05. At baseline men had greater accelerometer counts than women. Accelerometer and step counts were greater in participants <61yrs and in employment. Greatest counts were on day 1 of accelerometer wear, lowest counts on day 7 at baseline and 6 months, and day 5 at 12 months. At baseline an interaction of gender and day of wear for step count and at 12 months for step and accelerometer count was found. Women, those >61yrs and retired individuals are the most inactive subgroups of people with T2D and are priority for intervention. The 'wear effect,' from measurement of physical activity with an accelerometer, should be considered when evaluating the effectiveness of interventions, with possible removal of the first day of data.