A web-based physical activity intervention benefits persons with low self-efficacy in COPD: results from a randomized controlled trial

Abstract

Promoting physical activity (PA) is of top priority in chronic obstructive pulmonary disease (COPD). This study examines the influence of an internet-delivered intervention on the relationship between exercise self-efficacy and changes in PA, physical health, and exercise capacity in COPD. 112 U.S. Veterans with COPD were randomized to either a comparison (pedometer alone) or an intervention group (pedometer plus access to an internet-mediated PA intervention). There was a significant interaction between baseline exercise self-efficacy and randomization group on change in PA. In the comparison group, there was a significant relationship between higher baseline exercise self-efficacy and greater change in PA, whereas in the intervention group, improvements in PA were independent of level of baseline self-efficacy. Similar patterns were found with physical health and exercise capacity as outcomes. The use of an internet-mediated intervention significantly benefited persons with COPD who had low baseline self-efficacy to increase PA and physical health.Clinical trial registration The randomized clinical trial was registered on ClinicalTrials.gov (NCT01772082).

Keywords: COPD; Chronic obstructive pulmonary disease; Physical activity; Randomized trial; Self-efficacy; Technology.

Fig. 1

The interaction between baseline self-efficacy and change in steps by randomization group controlling for FEV₁ % predicted, age, education, and season, p = 0.016. Simple slope analyses revealed a significant positive relationship between baseline self-efficacy and change in steps for the comparison group (p = 0.006) and no significant relationship between baseline self-efficacy and change in steps for the intervention group (p = 0.655)

Fig. 2

The interaction between baseline self-efficacy and change in PCS (VR-12 Physical Component Score) by randomization group controlling for FEV₁ % predicted, age, education, and season, p = 0.021. Simple slope analyses revealed a significant positive relationship between baseline self-efficacy and change in PCS for the comparison group (p = 0.011) and no significant relationship between baseline self-efficacy and change in PCS for the intervention group (p = 0.597)