The relationship between patient‐reported and objective measures of physical function among cancer survivors receiving rehabilitation care: A correlation analysis

Abstract Background Physical function is associated with cancer‐related quality of life and survival. The correlation of patient‐reported and objective measures of physical function has not been quantified previously in the cancer rehabilitation medicine setting. Objective To correlate patient‐reported physical function with objective measures of physical function. The secondary aim is to correlate fatigue and social participation levels with objective measures of physical function. Design Retrospective. Setting Tertiary university cancer center. Participants A total of 226 cancer survivors referred to an outpatient cancer rehabilitation medicine clinic. Intervention Not applicable. Main Outcome Measure Correlation of Patient‐Reported Outcomes Measurement Information System (PROMIS) Cancer Function Brief 3D Profile domains with objective physical function measured by the 30‐second sit‐to‐stand and hand grip strength. Results Pearson correlation analysis revealed a moderate correlation between patient‐reported physical function and the 30‐second sit‐to‐stand test ( r = 0.57; p <.001), and a weak correlation of patient‐reported physical function and handgrip strength ( r = 0.27; p <.001). Multivariable regression modeling controlling for age, gender, and receipt of systemic therapy demonstrated a significant association between patient‐reported physical function and the 30‐second sit‐to‐stand test (estimated confidence interval 0.76 0.60, 0.92, p < .01), in addition to patient‐reported function and handgrip strength (estimated 0.22 0.10, 0.34, p < .01). Multiple myeloma but no other cancer type was inversely associated with lower physical function. Conclusions Patient‐reported physical function correlated moderately with the 30‐second sit‐to‐stand test, whereas hand grip strength demonstrated a weak correlation. The 30‐second sit‐to‐stand test is a suitable substitute for patient‐reported physical function. Further work evaluating how measures relate to each other across cancer populations is needed before recommending a standardized set of outcome measures.