随着癌症幸存者数量的增加，维持癌症幸存者的健康相关生活质量是当务之急。这就需要准确可靠的方法来评估癌症幸存者的感受和功能。来自可穿戴传感器的现实世界数字测量提供了监测癌症生存中的健康和身体功能的潜力，但围绕这些测量的临床效用的问题仍有待解答。在这项二次分析中，我们使用了 2 个现有数据集来检查使用可穿戴加速计捕获的现实世界身体行为测量结果如何与已完成癌症治疗的个体样本中的有氧健身、自我报告的健康状况和身体机能相关。总体而言，86 名无病癌症幸存者年龄21-85 岁的人完成了健康和身体功能的自我报告评估，以及用于估计其有氧适能的次最大运动测试，量化为预测的次最大摄氧量 (VO2)。使用大腿佩戴的加速计来监测参与者在现实世界中的身体行为 7 天。加速度数据用于计算以下身体行为指标的平均值：久坐时间、步数、轻度和中度至剧烈身体活动的时间、超过 1 分钟的踏步时间和加权中位数步频，以及 30 秒峰值步频斯皮尔曼相关分析表明，7 种由加速度测量得出的现实世界身体行为测量值中的 6 种 (86%) 与癌症治疗功能评估 - 一般总体幸福感或相关的患者报告结果测量信息系统 - 没有显着相关性 -身体功能评分（Ps≥.08）。相比之下，除一项身体行为测量外，所有其他测量均与次最大摄氧量 (Ps≤.03) 显着相关。使用似然比检验比较这些关联，我们发现步数、超过 1 分钟的步数时间以及中等到剧烈活动的时间与次最大摄氧量的相关性比与自我报告的健康或身体功能的相关性更强（Ps≤ .03）。相比之下，超过 1 分钟的步频和 30 秒峰值步频与次最大摄氧量的相关性并不比与自我报告测量值的相关性更大 (Ps≥.08)。在无病癌症幸存者的样本中，我们发现现实世界身体行为的一些衡量指标与有氧健身的相关性更大，而不是与自我报告的健康和身体机能的相关性。这些结果强调了这样一种可能性：在已完成癌症治疗的个体中，与自我报告和表现测量相比，现实世界身体行为的测量可能提供更多信息。为了促进数字测量在肿瘤学临床研究中的适当使用，有必要进一步研究评估随着时间的推移，在癌症幸存者的大量代表性样本中真实世界身体行为的临床效用。ClinicalTrials.gov NCT03781154； https://clinicaltrials.gov/ct2/show/NCT03781154.©Shelby L Bachman、Emma Gomes、Suvekshya Aryal、David Cella、Ieuan Clay、Kate Lyden、Heather J Leach。最初发表于 JMIR Cancer (https://cancer.jmir.org)，2024 年 7 月 15 日。

As the number of cancer survivors increases, maintaining health-related quality of life in cancer survivorship is a priority. This necessitates accurate and reliable methods to assess how cancer survivors are feeling and functioning. Real-world digital measures derived from wearable sensors offer potential for monitoring well-being and physical function in cancer survivorship, but questions surrounding the clinical utility of these measures remain to be answered.In this secondary analysis, we used 2 existing data sets to examine how measures of real-world physical behavior, captured with a wearable accelerometer, were related to aerobic fitness and self-reported well-being and physical function in a sample of individuals who had completed cancer treatment.Overall, 86 disease-free cancer survivors aged 21-85 years completed self-report assessments of well-being and physical function, as well as a submaximal exercise test that was used to estimate their aerobic fitness, quantified as predicted submaximal oxygen uptake (VO2). A thigh-worn accelerometer was used to monitor participants' real-world physical behavior for 7 days. Accelerometry data were used to calculate average values of the following measures of physical behavior: sedentary time, step counts, time in light and moderate to vigorous physical activity, time and weighted median cadence in stepping bouts over 1 minute, and peak 30-second cadence.Spearman correlation analyses indicated that 6 (86%) of the 7 accelerometry-derived measures of real-world physical behavior were not significantly correlated with Functional Assessment of Cancer Therapy-General total well-being or linked Patient-Reported Outcomes Measurement Information System-Physical Function scores (Ps≥.08). In contrast, all but one of the physical behavior measures were significantly correlated with submaximal VO2 (Ps≤.03). Comparing these associations using likelihood ratio tests, we found that step counts, time in stepping bouts over 1 minute, and time in moderate to vigorous activity were more strongly associated with submaximal VO2 than with self-reported well-being or physical function (Ps≤.03). In contrast, cadence in stepping bouts over 1 minute and peak 30-second cadence were not more associated with submaximal VO2 than with the self-reported measures (Ps≥.08).In a sample of disease-free cancer survivors, we found that several measures of real-world physical behavior were more associated with aerobic fitness than with self-reported well-being and physical function. These results highlight the possibility that in individuals who have completed cancer treatment, measures of real-world physical behavior may provide additional information compared with self-reported and performance measures. To advance the appropriate use of digital measures in oncology clinical research, further research evaluating the clinical utility of real-world physical behavior over time in large, representative samples of cancer survivors is warranted.ClinicalTrials.gov NCT03781154; https://clinicaltrials.gov/ct2/show/NCT03781154.©Shelby L Bachman, Emma Gomes, Suvekshya Aryal, David Cella, Ieuan Clay, Kate Lyden, Heather J Leach. Originally published in JMIR Cancer (https://cancer.jmir.org), 15.07.2024.