A cost-consequences analysis of the SAFE trial: a comparative, effectiveness trial evaluating low versus high level of exercise supervision over a 12-week intervention period for women with breast cancer

Abstract

Aims: To compare the costs and health benefits of a 12-week individualised, exercise intervention delivered via five (5EP) versus 20 (20EP) face-to-face sessions with an Exercise Physiologist (EP) in women with breast cancer.

Methods: Women with stage II or higher disease, who were insufficiently active and had one or more comorbidities or persistent treatment-related side-effects were eligible to participate in this comparative, effectiveness trial. Outcomes (including exercise self-efficacy, health-related quality of life, anxiety, depression, fatigue, pain, physical and social functioning, sleep, lymphoedema, arm symptoms and weekly physical activity levels) were assessed at baseline, immediately post-intervention and 12-week follow-up. To assess potential benefit, a composite outcome score was created by calculating the mean of z-scores computed from the raw data for all outcomes per participant. Total per participant costs included program development, training, implementation, equipment, and participation. Generalised estimating equations were used to test the effects and interaction of group and time.

Results: Sixty women (mean age 50±9) were randomised to the 5EP (n = 30) or 20EP (n = 30) groups. Both groups showed significant (P < 0.001) improvement over time (post-intervention and 12-week follow up, mean±SD): 5AEP 0.08±0.49 and 0.13±0.63; 20AEP 0.28±0.55 and 0.20±0.70, respectively). Despite higher mean composite scores in the 20EP group, no significant differences were seen between groups (P = 0.56). Per participant costs totalled $1871 and $5142 for those the 5EP and 20EP groups, respectively. Greater cost-efficiencies would come with delivery of the program to higher numbers of women with breast cancer (e.g., if delivered to 100 women, per participant costs for the 5EP and 20EP group would equate to $562 and $1543, respectively).

Conclusions: Clearly future analysis will explore the proportion of participants within both groups who benefited from the intervention versus those who did not, and whether the additional benefits sustained by those in the 20EP group represents value for money.