How does an app become a “top app”?

Both the Android Market and Apple’s App Store have a “Top Apps” section – but do we know what inclusion in this section signifies ?

Perhaps the idea is to list the current most popular apps (think Billboard’s Top 100 song chart).

On the other hand, maybe the idea is to feature apps that are the highest in quality as judged by experts or consumers (like the ratings one finds on CNET.com for electronic devices).  It would make sense that consumers are interested in both popularity and quality.

In an attempt to understand which medical apps made it into the Top Apps section of the Android Market, we extracted and analyzed the Top Free and Top Paid apps listed in the medical category.  We looked at who these apps target as potential users, which diseases and medical specialties they address, and gathered information on each app’s popularity and quality rating.

We found that the Top Apps section of the Android Market does a poor job in directing users to the best apps available for health care professionals and their patients. Read below to learn what are the most important problems of the medical section of the Android Market.

The data below is a snap-shot of what the Android market looked like in the middle of July.  We note that the Top Apps and their ratings, downloads, and rankings change frequently. While we did not study the iTunes App Store, we suspect that many of the observations below likely also apply in the iOS realm.

5) Many “top apps” are not popular at all.

Apps with a high number of downloads were generally rated higher than those with fewer downloads.  Greater popularity, however, did not necessarily mean higher ranking.  Several of the top free apps did indeed have >500,000 downloads (for example, Medscape and Epocrates both have >1 million downloads).  Similarly, the number of ratings was generally higher in the top half of both top 100 app lists.  However, several apps with less than 50 downloads were ranked higher than apps with >1000 downloads without an apparent reason for the higher ranking.

4) App rating is not clearly related to ranking.

The only direct measure of app quality available on the Market is the customer rating (from 1 to 5 stars).  Generally, apps in the top 100 apps are rated between 3.5 and 4.5 stars.  Yet, strangely eight apps in the Top 100 had ratings less than 3.0 stars.  Furthermore, the value or usefulness of these ratings is limited by how few users have actually rated each app.  Twelve percent of the free market’s apps and 31% of the paid market’s apps have less than ten total ratings.  That is not a large sample size from which to infer user satisfaction.

3) Consumer apps dilute the market for health care professionals.

Consumer apps – those targeted primarily at patients, not health care providers – make up 41% of the Top Free apps and 18% of the Top Paid apps.  Apps targeted at physicians made up only 32% of both the free and paid medical apps while the remaining are targeted toward nurses, students, EMS/paramedics, and other providers.  With so many consumer apps, particularly in the free market, it becomes very difficult to find good apps for doctors, nurses, students, and other health care professionals

2) Many non-medical apps are in the medical category.

Apps completely unrelated to health care are placed within the medical category.  Examples include an app that emits a sound to repel insects, an app with romantic quotes to send via text message and an app that displays optical illusions.  Apps like these simply do not belong in the medical category – and certainly do not belong in the Top 100, but unfortunately make up 15% of the top free apps and 6% of the top paid apps. 

1) Filtering apps by category, specialty, disease type, and even popularity or rating is impossible.

The Android Market cannot be filtered.  It can only be searched by title or browsed in order of top ranking.  Yet the “Top Apps” lists are almost meaningless, and do not showcase the best apps as far as we at iMedicalApps can determine.  Without the ability to sort apps by their category or rating, browsing the Market is frustrating to the point of futility. 

Filtering is fundamental – just think of PubMed.  If PubMed was just a collection of every medical papers ever published – without the ability to sort or filter by category (Mesh terms), date, journal, and numerous other criteria – it would be almost impossible to use… much like the Android Market in its current incarnation.

Conclusion:

The Medical category of the Android Market fails to be useful, even relevant, to health care providers.  It misrepresents the “top” apps and the current medical apps are diluted by non-medical and look-alike apps. Worse yet, it is unfilterable and unsortable.  Therefore, it does not currently work for the health care community.