The massive use of digital tools and mobile devices such as smartphones and wearables offer the opportunity to revolutionise the diagnosis and treatment of mental health. It is like having a mental health professional in your pocket at all times, who interprets the data extracted from the sensors as if she reads a digital diary of the patient, associating the data with symptoms related to mental health disorders. According to a systematic review carried out by Valeria De Angel and collaborators at King’s College London in the UK, there is evidence linking symptoms of depression with objective data collected passively with sensors on smartphones and wearable devices. The digital diaries of a patient collected in a non-intrusive way by current technology are an ally for research in mental health.

What would happen if that personal digital diary had a blank page every two? This does not happen with the default apps and services of the operating system or manufacturer of a mobile device, but it does with the specific data collection apps developed by third parties. Those of us who are dedicated to the research and development of mobile device applications for passive data collection that complement therapies for mental health cause us the same frustration. In JMIR, Boonstra and his colleagues encountered a loss of more than 50% of data collected from mobile devices. In a study with a cohort of subjects with schizophrenia – a mental disorder that leads to a lack of perception of reality – Torous and colleagues also found a massive loss of more than 50% of scheduled observations on mobile devices. This ratio of one missed observation out of every two attempts is repeated in a recent study by Bähr and colleagues with 625 participants using Android-based mobile devices. Their results show that the losses of half of the scheduled data measurements are explained by various reasons, but mostly by the restrictive measures introduced in the recent versions of the Android operating system itself and other restrictive policies imposed by the manufacturer. In other words, the most recent versions of Android are more restrictive for a passive data collection app to systematically access the sensor values ​​of the mobile device itself. In addition to this, the manufacturers themselves (Xiaomi, BQ, etc.) wrap the Android base operating system with their own customization software layer (of the company) that often includes more access restrictions to collect sensor data reliably.

Android-based mobile manufacturers defend their new restriction policies in favor of increasing user privacy. Ok, fine. It makes sense to protect users from malicious apps that can happily access mobile device sensors. User’s privacy conditions improve, but worsen for passive sensing apps for mental and public health research purposes. For researchers in mobile devices and mental health, these Android restrictions represent a considerable handicap to reliably and systematically capture data on patients.

A group of researchers from GEOTEC at the Universitat Jaume I have developed an alternative way of collecting data using mobile devices, which circumvents the restrictions of the operating system and manufacturers, in order to guarantee reliable and sustained passive detection over time. This research, recently published in the scientific journal Pervasive and Mobile Computing, represents an advance to ensure that the data collected is really useful for mental health therapies, and prevents mental health professionals from finding blank pages in the personal digital diaries associated to patients.

The work presents two novelties. First, it provides a series of recommendations and configurations so that app developers can optimize systematic data collection through mobile devices.

Unfortunately, these recommendations are not trivial and can be quite time-consuming and difficult to implement in an app. Secondly, to make life easier for developers, we have developed a library or software component called “NativeScript Task Dispatcher” that greatly simplifies the application of recommendations and optimizations in passive systematic data collection apps. The results show that the use of the library on various mobile devices avoids many data losses that occur with apps that do not take the library into account.

With the NativeScript Task Dispatcher, the research community and developers of mobile device applications for passive data collection for mental health have at their disposal an effective tool to prevent data loss in the digital diaries of patients and improve consequently the diagnosis and treatment of mental health disorders . Try it! We look forward to your comments and experiences!