Socio-demographic and behavioural correlates of physical activity perception in individuals with recently diagnosed diabetes: results from a cross-sectional study

Abstract

Background: Physical activity (PA) levels in type 2 diabetes mellitus (T2DM) patients are generally low. Poor PA perception may impede healthy behaviour change in this high risk group. We describe (i) objective PA levels, (ii) the difference between objective and self-reported PA ('PA disparity') and the correlates of (iii) PA disparity and (iv) overestimation in recently diagnosed T2DM patients.

Methods: Cross-sectional analysis of 425 recently diagnosed T2DM patients aged 42 to 71, participating in the ADDITION-Plus study in Eastern England, UK. We define 'PA disparity' as the non-negative value of the difference (in mathematical terms the absolute difference) between objective and self-reported physical activity energy expenditure (PAEE in kJ · kg-1 · day-1). 'Overestimators' comprised those whose self-reported- exceeded objective-PAEE by 4.91 kJ · kg-1 · day-1(the equivalent of 30 minutes moderate activity per day). Multivariable linear regression examined the association between PA disparity (continuous) and socio-demographic, clinical, health behaviour, quality of life and psychological characteristics. Logistic regression examined the association between PA overestimation and individual characteristics.

Results: Mean objective and self-reported PAEE levels ± SD were 34.4 ± 17.0 and 22.6 ± 19.4 kJ · kg-1·day-1, respectively (difference in means =11.8; 95% CI=9.7 to 13.9 kJ · kg-1 · day-1). Higher PA disparity was associated with male sex, younger age, lower socio-economic status and lower BMI. PA overestimators comprised 19% (n=80), with those in routine/manual occupations more likely to be overestimators than those in managerial/professional occupations.

Conclusions: T2DM patients with poor physical activity perception are more likely to be male, younger, from a lower socio-economic class and to have a lower BMI. PA overestimators were more likely to be in lower socio-economic categories. Self-monitoring and targeted feedback, particularly to those in lower socio-economic categories, may improve PA perceptions and optimise interventions in T2DM patients. Our findings suggest that strategies for enabling realistic assessment of physical activity levels, through self-monitoring or feedback, warrant further investigation and may help refine and improve physical activity interventions.

Figure 1

Distribution of physical activity difference. The difference between objective minus self-reported physical activity energy expenditure (PA difference) in recently diagnosed (A) female and (B) male T2DM patients is presented. Data were collected in Eastern England, UK between 2003 and 2007 (ADDITION-Plus study).