Monitoring aerobic capacity in cancer survivors using self-reported questionnaires: criterion validity and responsiveness.

Abstract

BACKGROUND

Evaluating the criterion validity and responsiveness of the self-reported FitMáx©-questionnaire, Duke Activity Status Index (DASI) and Veterans Specific Activity Questionnaire (VSAQ) to monitor aerobic capacity in cancer survivors.

CONCLUSION

The self-reported FitMáx©-questionnaire has sufficient validity to estimate aerobic capacity in cancer survivors at group level. The responsiveness of the FitMáx©-questionnaire for absolute change is limited, but the questionnaire is able to detect whether aerobic capacity improved. The FitMáx©-questionnaire showed substantial better values of validity and responsiveness compared to DASI and VSAQ.

RESULTS

Seventy participants were included. Outcomes at T₁ were available for 58 participants (83%). Mean CPET-VO_2peak significantly improved at T₁ (Δ1.6 mL·kg^- 1·min^- 1 or 8%). Agreement between questionnaire-VO_2peak and CPET-VO_2peak at T₀ was moderate for the FitMáx©-questionnaire (ICC = 0.69) and VSAQ (ICC = 0.53), and poor for DASI (ICC = 0.36). Poor agreement was found between ΔCPET-VO_2peak and Δquestionnaire-VO_2peak for all questionnaires (ICC 0.43, 0.19 and 0.18 for the FitMáx©-questionnaire, VSAQ and DASI, respectively). ROC analysis showed that the FitMáx©-questionnaire was able to detect improvements in CPET-VO_2peak (area under the curve, AUC = 0.77), when using a cut-off value of 1.0 mL·kg^- 1·min^- 1, while VSAQ (AUC = 0.66) and DASI (AUC = 0.64) could not.

METHODS

Cancer survivors participating in a 10-week supervised exercise program were included. The FitMáx©-questionnaire, DASI, VSAQ and a cardiopulmonary exercise test (CPET) were completed before (T₀) and after (T₁) the program. Intraclass correlation coefficients (ICC) were calculated between VO_2peak estimated by the questionnaires (questionnaire-VO_2peak) and VO_2peak measured during CPET (CPET-VO_2peak), at T₀ to examine criterion validity, and between changes in questionnaire-VO_2peak and CPET-VO_2peak (ΔT₀-T₁) to determine responsiveness. Receiver operating characteristic (ROC) analyses were performed to examine the ability of the questionnaires to detect true improvements (≥ 6%) in CPET-VO_2peak.