De Ridder Roel, De Blaiser Cedric
Department of Rehabilitation Sciences and Physiotherapy, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium.
Department of Rehabilitation Sciences and Physiotherapy, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium.
Gait Posture. 2019 May;70:30-32. doi: 10.1016/j.gaitpost.2019.02.009. Epub 2019 Feb 14.
The use of activity trackers has been proposed in rehabilitation where resuming physical activity is deemed crucial, e.g. after total knee arthroplasty (TKA). As patients initially often walk with crutches, it is of importance that clinicians can rely on the information provided by activity trackers.
To determine concurrent validity of 2 activity trackers for step count, positioned on different locations on the body during gait with crutches.
Thirty healthy participants performed normal gait and gait with one crutch and two crutches over a distance of 400 m while wearing a Garmin Vivofit 3 and Nokia Go on both wrists and both sides at the waist (only Nokia Go). The gold standard was manual step count. Inter-device reliability (within brand) was assessed by calculating Intraclass Correlation Coëfficients (ICC) and concurrent validity was determined by performing paired sample t-tests, ICC and Bland-Altman Plots with % bias and 95% CI Limits of Agreement (LoA).
During normal gait, both the Nokia and Garmin showed good to excellent inter-device reliability (ICC > 0.75). Both devices showed concurrent validity compared to manual step count, with slightly better results for the Garmin compared to the Nokia at the wrist (% bias = respectively 0.0% and -1.4% with 95% CI LoA: respectively -1.7%;1.7% and -8.6%;5.8%; ICC: respectively 0.995 and 0.859). During gait with crutches, however, overall 95% CI of LoA were beyond clinically acceptable differences and ICC values with the gold standard were poor. Therefore, notwithstanding a sometimes reported small average % bias, validity of the activity trackers for step count during gait with crutches was not established, independent of tracker position.
Activity trackers showed no concurrent validity when monitoring step count during gait with crutches. This should be taken into account when implementing this technology in e.g. post-operative goal setting in patients with TKA.
在康复治疗中，活动追踪器已被应用于恢复身体活动至关重要的情况，例如全膝关节置换术（TKA）后。由于患者最初常使用拐杖行走，临床医生能够依赖活动追踪器提供的信息至关重要。
确定两种活动追踪器在使用拐杖行走时位于身体不同部位时计步的同时效度。
30名健康参与者在佩戴佳明Vivofit 3和诺基亚Go的情况下，分别在手腕两侧和腰部两侧（仅诺基亚Go）进行400米的正常步态、单拐步态和双拐步态行走。金标准是人工计步。通过计算组内相关系数（ICC）评估设备间可靠性（同一品牌内），并通过配对样本t检验、ICC以及带有偏差百分比和95%一致性界限（LoA）的布兰德-奥特曼图确定同时效度。
在正常步态期间，诺基亚和佳明均显示出良好至优秀的设备间可靠性（ICC＞0.75）。与人工计步相比，两种设备均显示出同时效度，在手腕处佳明的结果略优于诺基亚（偏差百分比分别为0.0%和 -1.4%，95% CI LoA分别为 -1.7%；1.7%和 -8.6%；5.8%；ICC分别为0.995和0.859）。然而，在使用拐杖行走时，总体95% CI的LoA超出临床可接受差异范围，且与金标准的ICC值较差。因此，尽管有时报告的平均偏差百分比较小，但无论追踪器位置如何，活动追踪器在使用拐杖行走时计步的效度均未确立。
活动追踪器在监测使用拐杖行走时的计步情况时未显示出同时效度。在例如TKA患者术后目标设定等方面应用该技术时应考虑到这一点。
