HEALTH Platform Offers Open, Location-Aware Smartphone Tool for Geographic Ecological Momentary Assessment Research

A researcher who wants to know whether an afternoon in a park lifts someone's mood faces a practical problem: asking people to remember later is unreliable. A paper published in PLOS Digital Health on December 11, 2025, describes a tool built to solve that problem. It is called the Healthy Environments and Active Living for Translational Health Platform, or HEALTH Platform, and it is designed to ask the question while the person is still standing in the park.

The platform has three parts. The first is a smartphone app for Android and iOS that can collect audio, photos, surveys, and video. The second is a logic server, which evaluates the data coming from the app and responds to it. The third is a monitor server, where researchers or clinicians review summaries of what has been collected. The authors configure the logic and monitor servers through low-code development environments, and they say the app is meant to work alongside any web or smartphone-based application, so that researchers can keep their preferred collection tools.

The method underneath is ecological momentary assessment, or EMA. Participants are prompted, on a random or fixed schedule or after an event such as a craving, to answer a short survey. Before smartphones, studies relied on diaries or on pagers and watch alarms. The authors note that these were prone to recall bias, since people might answer long after the event, and to desirability bias, since some participants filled in surveys all at once near the end of a study to secure their full incentive or to look compliant. Smartphones help by letting researchers limit how long a prompt stays open and measure the delay between prompt and response.

The platform extends EMA into geographic ecological momentary assessment, or GEMA, which adds background location data. That allows prompts to be linked to environmental and contextual information by place and time. The paper describes four ways of triggering prompts: time-contingent, event-contingent, location-contingent using geofences or proximity points, and response-contingent, where earlier answers shape later prompts. One example given is a visit to a greenspace that queues a later question about mental wellness. The authors also caution that high-frequency location data without any motivational context, collected over a short period, may not reflect a person's typical movement. Surveys that capture reasons, run over a longer span, may do better.

The authors say earlier GEMA studies ran into recurring obstacles. Many required bespoke software that needed constant updating to stay compatible with iOS and Android, which sometimes meant excluding participants by operating system or running different protocols for each. Existing tools named in the paper include Ethica Data, Happyhier, Itinerum, MASS, SmartAPPetite, and VERITAS. Others were built by stretching Qualtrics or REDCap to their limits. According to the authors, these tools generally rely on static proximity buffers rather than complex dynamic geofences, are limited to numerical or text responses, and integrate sensor data with phone data only at the analysis stage rather than during collection.

To show what the platform can do, the team applied it in a study of how exposure to parks may improve well-being. The excerpt available describes the demonstration only in outline, and does not report its results or participant numbers. The authors describe the platform overall as effective, flexible, and secure, and suitable for a wide range of digital health and behaviour change research.

Whether it earns that description will depend on use beyond its creators. The paper positions the platform for wider adoption, and independent studies that use it and validate it will show how far it travels.