Qu Chengcheng, Sas Corina, Daudén Roquet Claudia, Doherty Gavin
School of Computing and Communications, Lancaster University, Lancaster, United Kingdom.
School of Computer Science and Statistics, Trinity College Dublin, Dublin, Ireland.
JMIR Ment Health. 2020 Jan 24;7(1):e15321. doi: 10.2196/15321.
In the last decade, there has been a proliferation of mobile apps claiming to support the needs of people living with depression. However, it is unclear what functionality is actually provided by apps for depression, or for whom they are intended.
This paper aimed to explore the key features of top-rated apps for depression, including descriptive characteristics, functionality, and ethical concerns, to better inform the design of apps for depression.
We reviewed top-rated iPhone OS (iOS) and Android mobile apps for depression retrieved from app marketplaces in spring 2019. We applied a systematic analysis to review the selected apps, for which data were gathered from the 2 marketplaces and through direct use of the apps. We report an in-depth analysis of app functionality, namely, screening, tracking, and provision of interventions. Of the initially identified 482 apps, 29 apps met the criteria for inclusion in this review. Apps were included if they remained accessible at the moment of evaluation, were offered in mental health-relevant categories, received a review score greater than 4.0 out of 5.0 by more than 100 reviewers, and had depression as a primary target.
The analysis revealed that a majority of apps specify the evidence base for their intervention (18/29, 62%), whereas a smaller proportion describes receiving clinical input into their design (12/29, 41%). All the selected apps are rated as suitable for children and adolescents on the marketplace, but 83% (24/29) do not provide a privacy policy consistent with their rating. The findings also show that most apps provide multiple functions. The most commonly implemented functions include provision of interventions (24/29, 83%) either as a digitalized therapeutic intervention or as support for mood expression; tracking (19/29, 66%) of moods, thoughts, or behaviors for supporting the intervention; and screening (9/29, 31%) to inform the decision to use the app and its intervention. Some apps include overtly negative content.
Currently available top-ranked apps for depression on the major marketplaces provide diverse functionality to benefit users across a range of age groups; however, guidelines and frameworks are still needed to ensure users' privacy and safety while using them. Suggestions include clearly defining the age of the target population and explicit disclosure of the sharing of users' sensitive data with third parties. In addition, we found an opportunity for apps to better leverage digital affordances for mitigating harm, for personalizing interventions, and for tracking multimodal content. The study further demonstrated the need to consider potential risks while using depression apps, including the use of nonvalidated screening tools, tracking negative moods or thinking patterns, and exposing users to negative emotional expression content.
在过去十年中，声称支持抑郁症患者需求的移动应用程序大量涌现。然而，尚不清楚抑郁症应用程序实际提供了哪些功能，以及它们的目标用户是谁。
本文旨在探讨排名靠前的抑郁症应用程序的关键特征，包括描述性特征、功能和伦理问题，以便为抑郁症应用程序的设计提供更充分的信息。
我们回顾了2019年春季从应用市场检索到的排名靠前的适用于抑郁症的iPhone操作系统（iOS）和安卓移动应用程序。我们采用系统分析方法来审查所选应用程序，数据从这两个市场收集，并通过直接使用这些应用程序获取。我们报告了对应用程序功能的深入分析，即筛查、跟踪和提供干预措施。在最初识别的482个应用程序中，有29个应用程序符合本综述的纳入标准。如果应用程序在评估时仍可访问、属于心理健康相关类别、超过100名评论者给出的评分高于4.0（满分5.0）且以抑郁症为主要目标，则将其纳入。
分析表明，大多数应用程序指定了其干预措施的证据基础（18/29，62%），而描述在设计中接受临床投入的比例较小（12/29，41%）。所有选定的应用程序在应用市场上均被评为适合儿童和青少年使用，但83%（24/29）没有提供与其评级相符的隐私政策。研究结果还表明，大多数应用程序提供多种功能。最常实现的功能包括作为数字化治疗干预或作为情绪表达支持提供干预措施（24/29，83%）；跟踪（19/29，66%）情绪、想法或行为以支持干预措施；以及进行筛查（9/29，31%）以辅助决定是否使用该应用程序及其干预措施。一些应用程序包含明显的负面内容。
主要市场上目前排名靠前的抑郁症应用程序提供了多样化的功能，可使不同年龄组的用户受益；然而，仍需要指南和框架来确保用户在使用这些应用程序时的隐私和安全。建议包括明确界定目标人群的年龄，并明确披露用户敏感数据与第三方的共享情况。此外，我们发现应用程序有机会更好地利用数字特性来减轻危害、个性化干预措施以及跟踪多模态内容。该研究进一步表明，在使用抑郁症应用程序时需要考虑潜在风险，包括使用未经验证的筛查工具、跟踪负面情绪或思维模式以及让用户接触负面情绪表达内容。
