Suzi Gage, over in her Guardian science blog Sifting the Evidence, suggests that based upon a single pilot study (that didn’t even use the app she’s recommending), you should purchase an app for your phone that purports to treat depression. It’s a glorious 771 word advertisement for a for-profit company’s app.
Now Gage, a PhD student, I’m sure is well-intentioned with her recommendation. Even if she has an undisclosed conflict of interest in writing about this app.
But if you’re going to write a blog called “Sifting the Evidence,” one would hope you’d dig a little deeper into “the evidence” before recommending an unproven treatment for something as serious as depression.
Especially when the research shows, in my opinion, that the app doesn’t work.
There are dozens upon dozens of apps available to help you track your mood and some that purport to help reduce your depressive symptoms. Prior research has shown that the simple process of actually tracking your mood over time can be helpful not only in better understanding yourself and your emotions, but to help in your recovery efforts.
Most of these apps are downloaded, tried a few times (most are opened only once), and then just as quickly discarded. The apps are not engaging enough to use everyday, especially from someone already suffering from depression, which usually comes with a lack of energy, motivation, and focus.
For scientific support of the app, called HappyPlace and developed by a UK startup called Jericoe, Gage cites a pilot study published in BJPsych back in 2012. In the study, 77 participants looked at faces on a computer screen and judged whether the face they saw on the screen was a happy or a sad face. They were then told if they were correct in their judgments in the experimental portion of the study. Depression was measured before and 2 weeks after the intervention with a standard measurement of depression in research called the BDI-II.
The study found no statistical significance between the experimental group and the control group. That means, quite plainly, that the intervention didn’t work to reduce depressive symptoms.
Despite the lack of data demonstrating the intervention’s effectiveness, that didn’t stop the researchers from optimistically concluding,
Our results provide preliminary evidence that modification of emotional perception may lead to some increase in positive affect. This provides some support for the hypothesis that biases in the perception of emotional facial expressions play a causal role in the maintenance of low mood.
Umm, no, it didn’t. The data clearly show not only did you not achieve any kind of clinical significance, you didn’t even get to the lower bar of statistical significance in your experiment. With undergraduate students — not even representative of the general population.
Shouldn’t a blog called “Sifting the Evidence” actually, you know, go through the scientific research and examine the credibility of such claims in context of the larger picture? In this case, I would think it would have been looking at the research more generally into apps for depression, and whether there’s more potential there.
Perhaps the reason Gage wrote such a puff piece on HappyPlace is because she worked on prior research with the app’s developers, Marcus Munafò and Ian Penton-Voak. This conflict of interest is something she fails to disclose directly in this blog piece (which you’d discover only if you clicked on and read a previous blog of hers).
So no, based upon this single study, I wouldn’t recommend anyone purchasing HappyPlace, because in my opinion, the evidence simply isn’t there.
As for the Guardian, I would hope your “science” blogs read more like objective editorial content from people who actually digest the research they’re sharing, and less like an advertisement from a writer that appears to have a direct and undisclosed conflict of interest.
Editor’s note: After this blog entry appeared, Gage updated her article to add “(full disclosure: this is the same group as the above research, and so conducted in the lab where I work)” in the second paragraph of the piece.
If you’d like to use a depression app that actually has positive scientific findings backing it (Watts et al., 2013) , I’d recommend the Get Happy Program. Better yet, it’s free. And if you want to understand how vast the number of “depression apps” there are, take a look at the Martinez-Perez, et al. (2013) article below — they found over 1,500… most of which had no or little scientific value.
References
MartÃnez-Pérez, B., de la Torre-DÃez, I. & López-Coronado, M. (2013). Mobile Health Applications for the Most Prevalent Conditions by the World Health Organization: Review and Analysis. JMIR, 15, e120. doi:10.2196/jmir.2600
Penton-Voak, IS, Bate, H, Lewis, G., & Munafò, MR. (2012). Effects of emotion perception training on mood in undergraduate students: randomised controlled trial. BJPsych.
Watts, S. et al. (2013). CBT for depression: a pilot RCT comparing mobile phone vs. computer. BMC Psychiatry, 13:49 doi:10.1186/1471-244X-13-49.
Read the advertisement article here: Shiny ‘app-y people: Treating depression from your smart phone
2 comments
Thanks for pointing that that my conflict of interest is unclear, I thought it was apparent from the first sentence but I guess not. I’ll add it in explicitly.
And, to put an end to your speculation as to why I wrote the post: When the press release went out about the app, I tweeted a link to it and people were interested. So I asked if anyone would want a blog about it, and a few people said yes. It’s not meant to be an advert in the least. I would argue I’m not recommending the app, more using it to discuss the issues in question. Namely, does cognitive bias modification of this kind work? Would people use an app as part of their attempts to combat depression?
As for the evidence from the pilot study not being strong enough, it was enough to gain funding to run an RCT in to the technique. As I discuss. And you’re right, the app isn’t identical to the task in the pilot, as the app also contains a lot of other mood monitoring features and the like.
Hi Ms. Gage, thanks for the reply.
If you had mentioned other apps, or other research that uses this technique, I believe I would’ve seen it as a more balanced piece. As it reads today, however, it seems like an advert for one set of researchers’ efforts (and their app, which isn’t free).
When the data don’t reach statistical significance, you can spin that any way you want (and sadly for the state of science today, researchers often do). But this is Science 101 — if your data show no significance, your experiment has confirmed the null hypothesis. Sure, you can do (and get funded for) more research, but that doesn’t change the existing data. Today, right now, the data show that the techniques or theory behind this app simply isn’t that effective.
So while it’s a potentially interesting idea, there’s no research backing for it to be used as a treatment for depression today. In the future, who knows? But of course, even a followup study isn’t conclusive; science needs to be replicated by other researchers (who don’t have a direct financial interest in its outcome) in order to be considered a robust or solid finding.
People already use their phones to lift their mood, with no special app required. Since socialization has been shown to lift people’s moods — and most of what people do on their phones nowadays is to socialize — voila! Moods are being lifted.
Cheers.