Mobile health: the Good, the Bad and the Unknown
Hi! I’m Natalie and I’m a research fellow at mHabitat working on the evaluation of mobile health. In other words, my role is assess whether apps or websites are genuinely improving our health, rather than simply claiming to. I will be writing a few blog posts reflecting on what I’ve learned, and inviting your input. In the blog post below, I will introduce the field of mobile health as I see it. You can give your own opinion either by filling out the short questionnaire at the end or by posting your comments.
Apping it up
Mobile health is a fast growing field, with currently over 259,000 apps available on the commercial app stores which have been downloaded over 3.2 billion times (1). In addition to apps, there is also a wealth of websites and wearable devices (for example step counter, heart rate monitor) that (cl)aim to improve or sustain health or well-being. An American survey found that 66% of people would use a mobile app to manage health-related issues and 79 % would be willing to use a wearable device (2).
So there is plenty on offer and of interest. But how do you choose the right app? How do you know which ones to trust? How do you make sure you’re not wasting your time but are actually getting a real benefit out of using the app?
Out of those 259.000 apps, we estimate that less than 0.1% (3) (that’s roughly the size of your fingernail compared to Big Ben!) have actually published any data that support their claims of being safe and beneficial. Dr Spence (a GP in Glasgow) sums it up quite well: health apps are perhaps mostly harmless and likely useless (4).
Others have been less gentle and have compared mobile health to the Wild West (for example reference 5). Hence the title of this blog post. While there will be no cowboy hats or Clint Eastwood, there is a rush for the gold, the risk of getting injured, and lawlessness that translates into uncertainty.
An app a day…
… does have the potential to keep the doctor away. Or, rather, keep the doctor in your pocket. We don’t think digital will replace the bulk of face to face healthcare anytime soon, but it offers a valuable addition or in some cases even an alternative.
Mobile technology is always with you, unlike that diary you are supposed to record all your food intake in (just snap a picture), or your partner who reminds you it’s time to take your medication (‘ding’ goes your phone). For some people, it solves the problem of having to travel to appointments (for example living in a remote area, experiencing severe anxiety). Often, you get to pick not only the place, but also the time and frequency that works best for you (for example to do guided exercises).
Mobile health also offers an alternative for people who do not get or want access to face to face healthcare. If a disease isn’t considered severe enough to be referred to treatment(yet), instead of being resigned to waiting, the patient has the opportunity to actively try and manage their condition. For people faced with potential stigma who are reluctant to see a healthcare professional, an app or website offers the anonymity and non-judgmental support that may be the first step to seeking help.
Mobile health also offers unparalleled options for prevention of ill health and increasing wellness. According to a recent study (6), we spend about 5 hours per day glued to our mobile devices. While somewhat worrying in terms of health implications (decreased physical activity, straining eyesight etc), there is also the potential for your device to become a health coach or even a game. For example, the popular Pokemon Go seems to increase the amount of physical activity of its gamers (for a short time anyway (7)).
There is certainly no lack of choice. If you can imagine it, there’s probably an app for it, from Apple HealthKit to Zombies Run…
As mentioned before, while the number of developers creating and users downloading mobile health apps are staggering, the number of apps that have any evidence for their safety and benefits are scarily low.
So, what could possibly go wrong with an app? Isn’t the greatest risk that you will run into a wall or under a car while glued to your screen? Below are some real examples of how apps have endangered their users.
‘What is Bipolar Disorder’ and ‘iBipolar’ were two apps for people with a diagnosis of bipolar disorder, a condition characterised by severe mood swings (8). The former app suggested that bipolar disorder could be contagious and the latter advised people in the middle of a manic episode to drink hard liquor to help them to sleep. Both statements are incorrect, and in the latter case could directly cause serious harm.
Most other examples are about indirect risks and unintended misuse. For example, the Swedish government’s app Promillekoll offers real-time estimated blood alcohol concentration calculated based on entered drinks. It was originally aimed at reducing university students’ alcohol intake but one study actually found that, in reality, it increased the frequency of their drinking occasions (8).
Closer to home, researchers pointed out that two-thirds of the apps collected in the original 2013 NHS health apps library sent identifying patient information over the internet without using proper encryption. If Paypal would do the same with your bank details, they would have been bankrupt and possibly you too. In addition, out of the 14 apps for depression or anxiety treatment, only four provided any evidence to support their claims.
This lack of evidence is a common issue, but only when the app makes explicit claims in its advertising, can regulators step in. This was the case for Lumosity, a paid-for suite of brain games that claimed to improve performance in everyday life and reduce or delay cognitive decline. It was widely promoted in the USA (for example on CNN, Fox News, the History Channel, Spotify) before it was sued by the Federal Trade Commission for unfounded claims and deceptive advertising (9).
From absence to app-sense
So, why is there so little evidence for the safety and effectiveness of mobile health?
Many apps are being developed by people with no medical or scientific background, … and little money. Doing research into safety and effectiveness takes a significant investment of time, money and expert knowledge. Due to the current lack of regulations, developers can get away with publishing their app without testing it. In addition to the absence of a stick, there is no carrot in plain sight either. Though there may be a glimmer of hope on the horizon. NHS England has recently launched its accreditation scheme for health apps, which may provide the incentive to convince at least some of the developers (10).
The only regulation currently in place is that for medical devices by the MHRA, similar to FDA regulations in the USA. It isn’t entirely straightforward to decided whether an app may be a medical device, and most apps can probably argue it doesn’t apply to them, for example by not specifying a particular disease or claiming they only offer general rather than personalised advice.
One reason for the lack of comprehensive regulations applicable to all of mobile health could be the lack of consensus in the scientific community as to how such apps or websites should be evaluated. This is a blog post in itself, but the root of the problem is that existing study designs don’t work well for the field of mobile health, and new study designs are currently being suggested and tested.
Meanwhile, efforts are being made to provide guidelines, a first step towards more prescriptive rules. The European Commission is the first international body to provide draft guidelines on mobile health based on a variety of existing (limited impact) guidelines and public consultation (11). Within the UK, the aforementioned NHS England accreditation process aims to offer guidelines for developers (12) and recommendations to the public (8). Programmes to increase the digital confidence and knowledge of medical practitioners are underway (for example reference 13), hopefully paving the way for clinicians to be more involved in advising their patients and maybe ultimately prescribing apps as they do other therapies.
Patience, kemosabe. Or is there something all of us can do to contribute and make the process easier? We think there is.
Mind the app
Given the accelerated growth of mobile health, it seems unfeasible for a limited number of experts to evaluate each and every one within a reasonable time frame or in sufficient depth. And there will always be one thing even the experts can get wrong. Have you ever heard of the Apple Newton, Microsoft Bob or Flooz? Even the best, most effective tool is useless if no one wants to use it for example because it’s not engaging enough or people don’t see the point of it.
Market research is crucial for companies, and – it seems – a lot of consumers as well. Consumer review surveys have reported that 84% of people trust online reviews as much as a personal recommendation (14), and 47% have reviewed at least one product online (15). Moreover, 39% of sharers seem to be altruists, who post simply to help people without needing to be rewarded (15).
Crowdsourcing, enlisting the services of a large number of people, is what powers reviews on commercial websites such as Amazon and TripAdvisor, but it also created the most used not-for-profit encyclopedia in the world, Wikipedia. Could we harness some of this power to help us evaluate mobile health?
We think we can, and you can contribute! By providing a public forum for evaluation of mobile health, including clear explanations and a transparent way of scoring, we want to give power to you. The power to evaluate apps without having to wait for the experts. The power to help experts select the most promising apps to go through their review process. The power to save the government time and money. In theory anyway.
Interested? Watch this space. Excited and eager to contribute? Fill out our 10 minute questionnaire and maybe even sign up to become a tester. Click here to take part.