Abstract
In June 2017, we embarked on a study to test the efficacy of available smartphone applications (apps) for gout, a type of inflammatory arthritis, which is associated with severe joint inflammation causing significant joint pain. Several noteworthy challenges arose as we grappled with issues such as apps no longer being available (or functional), the poor quality and misinformation of apps, and the rapid and unforeseen updating (or complete lack of updating) of existing apps. These are critical issues that researchers and health professionals working in the rapidly developing digital health field must learn how to contend with. In this study, we offer some of our insights regarding issues that need to be considered when working in m-Health, especially when the research is being conducted in the health sector or with patients.
Arecent study 1 reviewed 57 commercially available smartphone apps for managing gout, demonstrating the rapid increase in apps tailored for different patient groups. Gout is a treatable form of inflammatory arthritis, which develops in close to 4% of the American population and is increasing worldwide. 2,3 Interventions delivered through smartphone apps may reduce barriers to care, in comparison with conventional self-management interventions, which are both time and cost-intensive and often do not reach the population who need them most.
Before recommending health apps to patients, commercially available apps need to be properly reviewed and tested. As discussed by Nguyen et al., 1 the review process itself can be challenging. Their review found that 57 apps related to gout were available to download from iTunes (Apple's digital distribution service), out of which only 6 apps met their inclusion criteria of including information inline with current gout management guidelines. There were surprising limitations in the apps reviewed, in particular a lack of reminders to help patients take their medication, a lack of functions allowing patients to monitor both serum urate and urate-lowering therapies, and lastly a lack of tools for recording and managing flares. They concluded that only one app met all the recommendations set out in the patient-focused gout management guidelines. This app, however, was not entirely electronic as it had features that could only be actioned manually.
When our research team initiated a pilot randomized controlled trial in June 2017 to assess the efficacy of existing gout apps, we began by conducting our own review of gout apps. We searched both iTunes and Google Play (Google's digital distribution services) to ensure we did not miss apps developed for Android phones and not to exclude patients with Android smartphones. In total, we reviewed 62 gout-related apps. Of the 62 apps we reviewed, 48 were available on Android, 25 were available on iOS, and only 11 apps were available on both platforms. We also found that 10 of these apps were not in English. We were amazed to discover that half of the apps, which had been reviewed only a few months prior by Nguyen et al. in 2016, were either no longer available to download or could be downloaded, however, would fail to open or freeze upon launching the app, preventing users from accessing any of its features and rendering it entirely nonfunctional. We had not expected such dramatic changes to have occurred within such a short timeframe; however, this is a demonstration of the rapid pace of digital health research. This is one, if not the biggest key issues that should be considered when designing or conducting app research, or when promoting an app to patients, as care needs to be taken to select an app that will be available to the user for the duration of the research project or for an extended period of time.
After reviewing the 11 apps, which were available to download on both platforms, we came across many of the same issues as Nguyen et al. We were surprised to find how few gout self-management apps existed (only three were available from both iTunes and Google Play), how poor most of these apps were in terms of their ability to track health parameters specific to gout, and the concerning number of apps that had incorrect and misleading information. For example, some apps labeled foods such as red meat and alcoholic beverages such as beer as food items that can be consumed without limitation. In addition, some of these apps, which were available, required users to create profiles to subscribe and pay either a one-off or an annual fee to access features and required Wi-Fi to open and use the app, and some were constricted such that they were only available to download on certain mobile platforms (e.g., available on iOS, but not Android).
We also discovered that one of the limitations of using an app already available on the market is that the developers are able to update the app as little or as often as they wish. Nguyen et al. 1 found that many of the apps reviewed in their study lacked version updates and similarly, we found that at least half of the apps we reviewed had not been updated since their release on the app market (50 of the apps had not been updated for 1 year or more). App updates are needed to improve functionality, bug fixes, or to simply improve the content. When designing a study, or when advising patients to use an app, it is important to consider in advance how one would manage, adapt, and respond to updates, especially when the app is not under your control, and how these updates may affect the study or the patients.
The potential to harness mobile technology for health promotion, including targeting chronic health conditions such as gout, is exciting and can potentially improve access to effective interventions, reduce cost, and personalize interventions. However, there are many significant challenges for health professionals wishing to recommend apps to their patients or for researchers attempting to evaluate and test existing apps. In addition to acknowledging these significant challenges, we need ongoing collaboration between academia, industry, and health sectors to ensure we are giving patients the best chance of success.
Footnotes
Acknowledgment
This work was supported by the University of Auckland Faculty Research Development Fund (grant no. 3713121).
Disclosure Statement
No competing financial interests exist.
