Personal Health Informatics, College of Medicine and Public Health, Flinders University, Adelaide, Australia.
Personal Health Informatics, College of Medicine and Public Health, Flinders University, Adelaide, Australia; South Australian Health and Medical Research Institute, Adelaide, Australia.
J Affect Disord. 2018 Aug 15;236:31-36. doi: 10.1016/j.jad.2018.04.099. Epub 2018 Apr 22.
Monitoring is integral to adequately recognise and track mental health indicators of symptoms and functioning. Early identification of warning signs from digital footprints could facilitate adaptive and dynamic just in-time monitoring and care for individuals with common mental disorders.
Self-report data on mental health and lifestyle behaviour from 120 male and female Australian young adults experiencing psychological distress were collected online. API software was used to download participant's daily activity duration measurements over eight months from linked commercial activity tracker apps and wearables in real time. An independent samples t-test was conducted to compare the differences in daily durations of recorded physical activity between wearable devises and smartphone apps. Entropy techniques using R interpol package were used to analyse volatility in daily activity duration.
DASS-21 depression, stress and anxiety sub-scale scores indicated the study sample on average, had a moderate level of psychological distress. Daily activity duration was significantly greater from wearable devices when compared with smartphone apps (t-test = 25.4, p < 0.001). Entropy indices were not related with any of the DASS-21 measures. However, significant correlation between DASS-21 anxiety subscale scores and entropy of those with over 45 days measurements (r = 0.58, p = 0.02) was observed.
The observational nature of this study prohibits causal inference. As a convenience sample was used, the results may lack generalisability to the wider population.
Continuous monitoring using commercial apps and wearables as a resource to help clinicians augment clinical care for common mental disorders appears viable.
监测对于充分识别和跟踪心理健康症状和功能指标至关重要。从数字足迹中及早发现预警信号，可以为常见精神障碍患者提供适应性和动态的即时监测和护理。
从 120 名经历心理困扰的澳大利亚年轻成年男女中在线收集了心理健康和生活方式行为的自我报告数据。使用 API 软件实时从链接的商业活动跟踪应用程序和可穿戴设备中下载参与者 8 个月的日常活动持续时间测量值。进行独立样本 t 检验比较可穿戴设备和智能手机应用程序记录的体力活动日常持续时间的差异。使用 R 插值包中的熵技术分析日常活动持续时间的波动性。
DASS-21 抑郁、压力和焦虑分量表得分表明，研究样本平均具有中度心理困扰。与智能手机应用程序相比，可穿戴设备的日常活动持续时间明显更长（t 检验=25.4，p<0.001）。熵指数与任何 DASS-21 测量值均无关。然而，观察到 DASS-21 焦虑分量表得分与超过 45 天测量值的熵之间存在显著相关性（r=0.58，p=0.02）。
本研究的观察性质禁止因果推断。由于使用的是方便样本，因此结果可能缺乏对更广泛人群的普遍性。
使用商业应用程序和可穿戴设备进行连续监测，作为帮助临床医生增强常见精神障碍临床护理的一种资源，似乎是可行的。
