Medicate or Meditate; the App Store’s Solutions for Anxiety and Stress

How the App Store Proposes to Solve Mental Health Issues

Team Members

Lead: Anne Helmond, Fernando van der Vlist, Esther Weltevrede (alphabetical)

Participants: Taylor Geiger, Ine van Zeeland, Ana Pop Stefanija, Fernanda Ibanez, Julia Wolny (alphabetical)


1. Introduction

The number of mobile health (mHealth) apps is rising in an unprecedented manner, and as the American Psychiatric Association notes: “Psychiatry and mental health are no exception, and there are thousands of apps targeting mental health conditions that are directly available for patients to download and use today.” [1] However, there is very little review or oversight for these apps, and as a consequence, users of these apps can receive incorrect or ineffective advice, while the mental health effects of using the apps are often overstated by their developers.

Smartphones are turning into an epistemological device, we turn to them for solutions. When you detect an issue, you turn to your smartphone to find out more. Nearly half of the queries in Google Play Store are broad searches by topic [2] (rather than specific searches for a particular app), showing that users generally turn to their smartphone app store for relevant solutions to broad issues.

When it comes to regular Google search, according to Noble (2018: 155): “In practice, the higher a web page is ranked, the more it is trusted. Unlike the vetting of journalists and librarians, who are entrusted to fact check and curate information for the public according to professional codes of ethics, the legitimacy of websites’ ranking and credibility is simply taken for granted.” Similar to website search results ranking, users accord a certain degree of authority to relevance rankings in app stores, meaning that the order and ranking presented by app stores confers some sort of recommendation to the apps based on the app store’s search results presentation.

In an attempt to make the app store affordances work for them, app developers engage in app store optimization (ASO), trying to end up highly in an app store's search results. With millions of apps available in the bigger app stores, like Google’s (>3 million apps) and Apple’s (>2 million apps), the possibility of a particular app being found is dropping. Common ASO tactics that developers deploy to improve discoverability among millions of other apps, are focused on finding popular keywords to include in the app’s name and subtitle, its ID, and its description.

The growing number of mental health apps, many of which undoubtedly engage in ASO, raises a number of questions: How is mental health represented in the Google Play Store and the Apple App Store? Which solutions does a smartphone user find for mental health issues in these app stores? How do technologists look at the issue of mental health? Which tactics are developers deploying to rank higher? What solutions do they promise and can they deliver? This study addresses these questions by exploring the sphere of mental health apps in the two biggest app stores, focusing on store-mediated ‘relatedness’ between apps and recommendations in the app stores. We glean how the app search engine and how it is manipulated influence what users will find. Lastly, we gauge what kinds of solutions users are presented with when they search for mental health issues.

2. Initial Data Sets

We focused on five queries for mental health issues:
  1. stress,
  2. anxiety,
  3. depression,
  4. insomnia and
  5. mental health.
Using a research browser, we collected the first ten apps presented in the Play Store and the App Store for each of these queries, and then ran them through the DMI’s Similar Apps tool to create an overview of apps recommended for mental health in general and the issues of stress, anxiety, insomnia and depression in particular.

Play Store datasets

For the Play Store we found the following numbers of unique similar apps per query:
  • stress: 240
  • anxiety: 161
  • depression: 199
  • insomnia: 330
  • mental health: 205
In total, for all five queries combined, we have found 728 unique apps in the Play Store. The top 5 categories in the Play Store for these apps were:
  1. Health & Fitness
  2. Tools
  3. Medical
  4. Education
  5. Lifestyle

(Apple) App Store datasets

The number of unique similar apps per query is:
  • Stress: 66
  • Anxiety: 55
  • Depression: 51
  • Insomnia: 73
  • Mental health: 78

It has to be outlined that the initial set for finding the similar apps in Apple App Store was country-specific, in our case the Netherlands, because of the affordances of the tool we used itself. Namely, the iTunes store Tool allows for only a particular country search. In that regard, the question of how (if possible at all) to compare the findings for the different app stores emerges, if we keep in mind that the Google Play Store Tool does a universal, country non-specific search. Due to the same reasons outlined above, it is not possible to investigate the categories these similar apps belong to, as a result of the constraints imposed by the app store and the tool itself.

3. Research Questions

Q1) How does the Google Play Store cluster various issues related to mental health?

Q2) How does the Google Play store categorise apps related to mental health?

Q3) Which apps are recommended most often for mental health concerns within the app store?

Q4) What advice is given to people suffering from mental health concerns on the app store?

4. Methodology

1. We began by brainstorming words related to mental health issues. As several team members have previously conducted research on mental health-related subjects, we were able to identify multiple critical keywords quickly. We confirmed the accuracy of these keywords within the mental health app space, by querying Google for “top health apps”, which returned app review pages with the words “anxiety” and “depression”. We additionally queried “top stress apps” and again were returned results with they keywords “mental health”, “anxiety”, and “depression” in the title. We decided to include “insomnia” as a fifth term as it frequently appeared in the app descriptions as we were reviewing the top search results in this category, so we wanted an understanding of how insomnia is related to mental health issues by the app store. Additionally, the keywords “anxiety”, “stress”, and “depression” are used in multiple academic research articles addressing mental health concerns in a variety of populations (Gallego et al., 2014, Gu et al, 2015). We thus focused on five queries for mental health issues:
  • stress,
  • anxiety,
  • depression,
  • insomnia and
  • mental health.

2. For each query, we collected the first 10 results given in the Google Play Store and we ran the DMI’s Similar Google Apps Tool (in English).

3. The full results provided data which allowed for three types of analysis:
  • Top referred apps across all queries, and how they were categorised in Google Play (Tree Map and Alluvial)
  • Cortext analysis of full descriptions with network visualisation of co-occurrent phrases
  • Qualitative coding of the advice and suggestions provided for people suffering from mental health issues by the App Descriptions was completed. The qualitative coding was done in Atlas.ti. First, we copied and pasted the descriptions from the Top 10 Apps from the Google Play store for our queries “anxiety”, “depression”, and “stress”. Additionally, we created a separate document for the Top 10 Overall Apps. We then used a snowball coding technique to inspect the descriptions for first order codes. As we came upon recurring themes, we were able to categorize them into the same first order codes. Upon completing the first order codes, we then analyzed the total sum of 91 first order codes and grouped them into 9 different second order code groups. We used the frequencies of the first order codes and overall groups to analyze the descriptions to get a better understanding of what advice is given to people with mental health issues via the app store descriptions.

5. Findings

Q. 1 How does the Play Store cluster various issues related to mental health? We found a distinct cluster of Play Store apps that are related to all the queries that we set. This means that the top related apps are optimized in order to be discoverable for a wide variety of mental health issues. It also means that there are shown as solutions and source of information for different problems. When we focus on the apps that have the most queries connected we found that hypnosis, meditation, mental health tests, and music to relax are the top solutions offered. We can say that these apps have achieved better optimization tactics and created a deep network that helps them to be regularly at the top of the results. We found sub-clusters around specific mental health issues like anxiety, depression, and stress. Insomnia is a query that showed apps related to horror games that are separate completely from mental health app solutions.

Q1_GP-Network alla cluster.png

Q.2 How does the Play Store categorise apps related to mental health? Categorisation must be distinguished from issue clustering. Developers choose a category when they submit an app to the Play Store, and Play Store optimizes categories to help users find apps. Consequently, picking the ‘right’ category to improve discoverability is an ASO tactic [6].

Q2_Number of Mental Health Apps by Category in <a class="foswikiNewLink" href="/bin/edit/Dmi/PlayStore?topicparent=Dmi.SummerSchool2018AppStoresBiasMedicateMeditate" rel="nofollow" title="Create this topic">PlayStore</a>.png

It should come as no surprise that mental health apps would mostly be categorised as ‘Health & Fitness’ (most common category; 47%) or ‘Medical’ (third most common; 10%). Looking at apps that were found for all search queries (an indicator for centrality in the mental health sphere), 86% are in the ‘Health & Fitness’ category and 11% in the ‘Medical’ category.

Having the ‘Tools’ category in second place (13%) is interesting. Looking at the apps in the Play Store dataset that are categorised as ‘Tools’, we see a predominance of screen dimmers and various other smartphone functionality-related apps. Almost all of them were found through the ‘insomnia’ query (92), except for a few related to the ‘depression’ query (4). Clearly, this is an effect of the research design; the keyword ‘insomnia’ seems to be an outlier for mental health. Coincidentally, Insomnia is the name of a horror game, which also explains for a number of horror games in the dataset.

‘Health & Fitness’ and ‘Medical’ are the most common categorisations (57% together). Number 4, ‘Education, and 5, ‘Lifestyle’, are only used for 4% and 3% respectively of the apps found, with other categories showing a prevalence of less than 3%. This may signify:
  • Mental health app developers overwhelmingly choose to categorise their apps as ‘Health & Fitness’ or ‘Medical’.
  • The Play Store considers ‘Health & Fitness’ as the most relevant category for mental health searches.
The following visualisation shows how the Google Play Store categorises the apps in relation with the queries we used: anxiety, stress, depression, insomnia and mental health. Looking at an issue (query) level allows for additional insights in more detail.

Q2_Categorisation by query-01.png

Although the majority of apps (as discussed above) belong to the categories of Health and Fitness and Medical, it is interesting to see that apart from most categories characterized as “health & fitness”, there are also apps belonging to the categories of casual (like Stress Ball, Anti Stress Bubble Wrap), puzzle (Antistress - relaxation toys, Infinity Loop, Brain Yoga Brain Training Game), action (Kick the Buddy), arcade (Free Fidget Spinner for Relieve Stress), simulation (TRI Fidget Spinner, My Oasis - Tap Sky Island) etc.

Looking at the outliers might also give an interesting perspective. There are categories which have a small number of apps (n=1) such as shopping (like for the app Åsane Storsenter, related to insomnia), house & home (ZenCrate app, anxiety category), auto and vehicles (app G1 Test Genie Ontario 2018, related both to the issue of depression and anxiety), trivia ( the apps Tricky Test 2™: Genius Brain? and Can you escape the 100 room IV, belonging to three different issues: insomnia, depression and mental health) and similar. This once again raises the issue of arbitrariness of categorization.

Q.3 Which apps are recommended most often for mental health concerns within the app store? We found the following 10 apps were recommended most often in the Play Store:


Another way to look at the types of categories offered is to look at the issues they mention the most. That can be accomplished by analysing the description of the apps. Using CorText, and analyzing frequency of co-occurrence of phrases and their interrelatedness, we can see also few clusters of “types of apps”. Meditation and Therapy, Mental Disorders and Treatment, White Noise and Melodies, Anxiety Cures Apps, Brain Wave Therapy - are some of the detected (implied) issues and solutions the apps focus on.

Q3_Final Co-Occurence with text.png

Looking at the phrases co-occurrences also allows for an insight into the different clusters of what apps focus on the most. We can detect the following clusters of apps: apps focusing on the solutions they offer, apps focusing on the medical authority, on the symptoms and how to treat them, on the performance of the app and one distinctive cluster of terms and conditions. However, this should be taken as just an illustration and overview of the apps in the dataset. An in-depth analysis will have to be done, relating the clusters with particular apps and even queries. That would offer an opportunity to get more granular overview and see if maybe particular queries or categories of apps are focusing on specific issues (like - more on detecting the symptoms or on solutions; are there apps, like depression for example, that outline the medical authority more or not etc).

Q. 4 What advice is given to people suffering from mental health concerns on the app store? As previously mentioned, the advice which is presented to users regarding mental health issues in the app stores is not always evidence-based or created by doctors or therapists. Instead, many of the apps are created by for-profit organizations who may have little or limited qualifications to prescribe solutions and advice for mental health issues. Interestingly, the advice which was returned for each of our distinct five queries differed slightly, but due to time limitations we were unable to complete the qualitative coding of all queries. Therefore the most interesting findings in regards to this question come from the qualitative coding of the top 10 most recommended apps on the Google Play Store.

As can be seen below, the advice of “using meditation” occurs first, with 47 occurrences. The next most commonly occuring suggestion is to track or monitor yourself in some way; either tracking your mood, feelings, time, activities, or sleep, with 13 mentions. However, the descriptions of these apps promise more outcomes than offering solutions; with promised outcomes occurring 30 times. Why is it that the app descriptions promise more results than offer potential solutions? Why is the mentioning of beneficial outcomes more interesting than the recommendations for solving mental health issues in the descriptions? Perhaps app developers and marketers understand that people searching for mental health apps are more interested in the solution, rather than the strategy, or advice; and thus wish to offer “quick fixes” to users, an idea which is discussed further in the following Discussion section.

Q4.png Q4b.png

Additionally, although we were primarily interested in the solutions and advice being offered by the apps, it is worth mentioning the surprising and slightly counter-intuitive finding that the outcomes a user could achieve via using apps in these query categories seem to go above and beyond even the parameters of the problems themselves. This can be seen in the Graph below, which provides an overview of the 30 different outcomes promised in the top 10 most-recommended apps in the Google Play store. To provide a compelling example, the description of the app most often recommended for our search queries, Hypnosis for Anxiety, Stress Relief, & Depression [7], “Listen to audio affirmations, programs, courses, and workshops to help you reduce anxiety, get over your fears, reduce stress and stressful situations, overcome your phobias, become more confident in social settings, and relax more.” The long description goes on to promise:

“OVERCOME THE FEAR THAT CRIPPLES YOU: Do you feel like your fears keep you in a box of familiarity? Are you tired of letting your anxiety rule your life? With the help of the expert-facilitated hypnosis courses, meditation tracks, and affirmations, you can step into your role as the courageous creator within the parameters of your life.

CONQUER YOUR PHOBIA OF SOCIAL SITUATIONS: Do social situations make you nervous? Do you wish you were more confident and brave when out in public? Discover your ability to swoon a person of the opposite sex, perform in front an audience, get over your fear of heights, and hold a tarantula through the power of hypnosis and meditation.

BE COURAGEOUS AND FACE YOUR FEARS: Have you wished that you would live your life with more courage? Are you tired of feeling afraid of new things? With the plethora of audio programs, hypnosis workshops, and guided meditations in this app, you can develop the courage to go skydiving, travel alone, and learn to live more fearlessly.”


6. Discussion and conclusion

Our findings map the ongoing conversation regarding mental health issues, both on and off of app stores. Mental health issues are widely gaining media attention, public attention, academic attention, and thus it follows that app developers are identifying opportunities to develop apps which address mental health issues. Research has shown that reducing stress has a large impact on disease prevention, as well as subsequent psychological disorders (Gomez-Lopez et al., 2010). Additionally, high levels of stress have been demonstrated to contribute to “burnout” and absenteeism (Moriana and Herruzo, 2004). Mindfulness and meditation have been offered as a promising solution to resolving mental health issues, especially within the context of university students (Franco et al., 2011), and research continues to be done on these types of interventions on a wide variety of populations. Interestingly, app developers are also aware of the suggestion of mindfulness and meditation as a solution for mental health issues; many of the top apps within these categories suggest Meditation as a way to resolve the issue. For “stress”, the suggestion to use meditation appears 109 times within the app descriptions of the top 10 apps for this query. For “anxiety”, the suggestion to use meditation appears 86 times and when it comes to the Top 10 apps, the suggestion to use meditation to resolve mental health related queries (across our 5 queries) appears 47 times. Overall, when considering the total sum of app descriptions we coded, the suggestion to use meditation occurs an astounding 257 times! Additionally, the second most popular solution for mental health issues is to use some sort of tracker or tracking to measure your progress, time meditated.

These results also shed some light on knowledge structuring by app stores. As Noble (2018: 148) remarks, “Search does not merely present pages, but structures knowledge, and the results retrieved in a commercial search engine create their own particular material reality.” In the reality of the big app stores, mental health problems appear to be something that can primarily be fixed by hypnosis and meditation; though, to be fair, several of the top 10 recommended apps in the Play Store were careful to point out in their descriptions that professional help would need to be sought as well or that the app could not replace therapy. Yet, the message that users of the app store get when searching for mental health-related issues is that meditation and hypnosis are top solutions for what ails them; these are solutions focused on their personal response to stressors, regardless of the stressors themselves - contextual causes that may lead to mental health problems. This may reflect a particular vision of developers that is focused on individual self-reliance and proactive agency, which may be especially impaired in people suffering from mental health issues.

In a sense our results also point to ‘technological solutionism’ or technocentrism. This is most apparent for people who search with the term ‘insomnia’; basically, the app store responds: ‘turn off the light (of your screen).’ Though less apparent for the other queries we have used, the overall focus on meditation and self-help with technological support again reflects a preference for technological solutions, rather than e.g. psychiatric treatment or addressing the causes of mental distress in the real world.

7. References

Franco, C., Mañas, I., Cangas, A. J., & Gallego, J. (2011). Exploring the effects of a mindfulness program for students of secondary school. International Journal of Knowledge Society Research (IJKSR), 2(1), 14-28.

Gallego, J., Aguilar-Parra, J. M., Cangas, A. J., Langer, Á. I., & Mañas, I. (2014). Effect of a mindfulness program on stress, anxiety and depression in university students. The Spanish journal of psychology, 17.

Gómez-López, M., Gallegos, A. G., & Extremera, A. B. (2010). Perceived barriers by university students in the practice of physical activities. Journal of sports science & medicine, 9(3), 374.

Gu, J., Strauss, C., Bond, R., & Cavanagh, K. (2015). How do mindfulness-based cognitive therapy and mindfulness-based stress reduction improve mental health and wellbeing? A systematic review and meta-analysis of mediation studies. Clinical psychology review, 37, 1-12.

Moriana Elvira, J. A., & Herruzo Cabrera, J. (2004). Estrés y burnout en profesores. International journal of clinical and health psychology, 4(3).

Noble, Safiya Umoja (2018). Algorithms of Oppression. How Search Engines Reinforce Racism. NYU Press, 2018.

8. Endnotes








9. Appendix

Top Apps by Indegree Relatedness Appendix.png
I Attachment Action Size Date Who Comment
Appendix.pngpng Appendix.png manage 1 MB 08 Jul 2018 - 13:35 AnneHelmond Appendix
Q1_GP-Network alla cluster.pdfpdf Q1_GP-Network alla cluster.pdf manage 751 K 08 Jul 2018 - 13:12 AnneHelmond Q1
Q1_GP-Network alla cluster.pngpng Q1_GP-Network alla cluster.png manage 2 MB 08 Jul 2018 - 13:13 AnneHelmond Q1
Q2_Categorisation by query-01.pngpng Q2_Categorisation by query-01.png manage 326 K 08 Jul 2018 - 13:20 AnneHelmond Q2
Q2_Number of Mental Health Apps by Category in <a class="foswikiNewLink" href="/bin/edit/Dmi/PlayStore?topicparent=Dmi.SummerSchool2018AppStoresBiasMedicateMeditate" rel="nofollow" title="Create this topic">PlayStore</a>.pngpng Q2_Number of Mental Health Apps by Category in PlayStore.png manage 51 K 08 Jul 2018 - 13:15 AnneHelmond Q2
Q3.pngpng Q3.png manage 180 K 08 Jul 2018 - 13:21 AnneHelmond Q3
Q3_Final Co-Occurence with text.pngpng Q3_Final Co-Occurence with text.png manage 3 MB 08 Jul 2018 - 13:29 AnneHelmond Q3
Q4.pngpng Q4.png manage 173 K 08 Jul 2018 - 13:31 AnneHelmond Q4
Q4b.pngpng Q4b.png manage 91 K 08 Jul 2018 - 13:34 AnneHelmond Q4b
Q5.pngpng Q5.png manage 959 K 08 Jul 2018 - 13:35 AnneHelmond Q5
Topic revision: r4 - 12 Jan 2019, FernandoVanDerVlist
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback