Physical function predicts mortality in patients with cancer: a systematic review and meta-analysis of observational studies.

PURPOSE This systematic review and meta-analysis aimed to determine whether physical function can predict mortality in patients with cancer. METHODS Literature searches were conducted in Web of Science, CINAHL, the Cochrane Library, ProQuest, PEDro, and PubMed for articles published before September 2020. Four review authors retrieved studies using predetermined eligibility criteria and conducted quality assessment and data extraction. RESULTS A total of 1356 titles and abstracts were screened; ultimately, 26 studies were determined to be suitable for meta-analysis. Grip strength was significantly associated with mortality risk in patients with cancer (hazard ratio [HR] = 1.15, P = 0.005). Gait speed was also associated with mortality risk (HR = 1.58, P = 0.0004). In grip strength and gait speed, the subgroup comprised of patients with cancer aged ≥ 80 years had a higher effect size than that of patients aged < 80 years. The short physical performance battery measurement was markedly associated with mortality risk, showing the largest effect size (HR = 2.37, P < 0.00001). The 6-min walking test distance was significantly associated with mortality risk (HR = 1.55, P = 0.001). The timed up and go test was significantly associated with mortality risk with a high effect size (HR = 2.66, P < 0.00001). CONCLUSION This systematic review and meta-analysis demonstrated that physical function predicted mortality in patients with cancer. Furthermore, physical function outcomes in patients aged 80 years and above reflected a higher mortality.