Behaviour change interventions and habitual physical activity in people with intermittent claudication: a systematic review and meta-analysis

Introduction: Systemic atherosclerosis, termed peripheral arterial disease (PAD), leads to a reduction in the blood flow to muscles in the legs and pain when walking (intermittent claudication; IC). People with IC demonstrate a 40-45% reduction in habitual physical activity (PA) compared to peers. Supervised exercise therapy (SET) is the primary treatment, but access is limited and adherence low. In addition, SET programmes are typically short-lived and do not target habitual PA. Integrating behavioural change techniques (BCTs) may enhance exercise interventions effectiveness and support sustained habitual PA.

Aim and Methods: Systematic review to determine, in people with IC: 1) effectiveness of interventions incorporating BCTs compared to SET and other controls in increasing or maintaining habitual PA; 2) association between BCTs and short-term and medium-term increase/maintenance of habitual PA. We searched 11 databases from inception to November 2022. Risk of bias was assessed using the ROB2. BCTs were coded using the BCT taxonomy v1. Random effects meta-analysis was used to pool outcome data.

Results and Discussion: Fourteen RCTs of BCT interventions involving 2420 adults with IC (mean age 67.2 years), including three RCTs with a high risk of bias, were included. Meta-analysis (11 trials; 15 comparisons, 952 patients) indicates that BCT interventions result in increased habitual PA in the short term (< 6 months) compared to non-supervised controls (increase of 0.20 SMD (95% CI: 0.07 to 0.33), ~473 steps/day). Medium term (≥6 months) habitual PA change (6 trials) is unclear (increase of 0.12 SMD (95% CI: -0.04 to 0.29); ~288 steps/day). Comparison with SET was inconclusive, short term (-0.13 SMD, 95% CI: -0.43 to 0.16) and medium term (-0.04 SMD, 95% CI: -0.55 to 0.47). Sensitivity analyses without studies at a high risk of bias did not change the results. It was unclear whether any BCT domains or number of BCTs were independently related to an increase in treatment effect for habitual PA.

Conclusion: BCT-based interventions show promise in increasing short-term habitual PA compared to non-supervised interventions in people with IC, but medium to long-term effectiveness and comparison to SET require further research. It is unclear whether any specific BCTs are more strongly related to treatment effects.