Several anthropometric measurements and cancer mortality: predictor screening, threshold determination, and joint analysis in a multicenter cohort of 12138 adults

Background Anthropometric measurements (AMs) are cost-effective surrogates for evaluating body size. This study aimed to identify the optimal prognostic AMs, their thresholds, and their joint associations with cancer mortality. Methods We performed an observational cohort study including 12138 patients with cancer at five institutions in China. Information on demographics, disease, nutritional status, and AMs, including the body mass index, mid-arm muscle circumference, mid-arm circumference, handgrip strength, calf circumference (CC), and triceps-skinfold thickness (TSF), was collected and screened as mortality predictors. The optimal stratification was used to determine the thresholds to categorize those prognostic AMs, and their associations with mortality were estimated independently and jointly by calculating multivariable-adjusted hazard ratios (HRs). Results The study included 5744 females and 6394 males with a mean age of 56.9 years. The CC and TSF were identified as better mortality predictors than other AMs. The optimal thresholds were women 30 cm and men 32.8 cm for the CC, and women 21.8 mm and men 13.6 mm for the TSF. Patients in the low CC or low TSF group had a 13% (HR = 1.13, 95% CI = 1.03–1.23) and 22% (HR = 1.22, 95% CI = 1.12–1.32) greater mortality risk compared with their normal CC/TSF counterparties, respectively. Concurrent low CC and low TSF showed potential joint effect on mortality risk (HR = 1.39, 95% CI = 1.25–1.55). Conclusions These findings support the importance of assessing the CC and TSF simultaneously in hospitalized cancer patients to guide interventions to optimize their long-term outcomes.