App-Based Interventions for Moderate-To-Severe Depression.

Good early news and food for thought if you're building a mobile intervention or treatment app.

Helpful Chicken

This came across my Linkedin Profile yesterday, showing published on November 20th, 2023. Full Text is available here

A Brief Summary:

• Paper provides an analysis of results of 13 different studies using 16 different intervention apps for patients with moderate to severe depression.

• Population is 1470 patients.

• Reports a “Significant” (meaning statistically significant) medium size effect for moderate to severe depression.

  
  

This is good news  - high level it supported the idea that mobile intervention is a viable path, either as a standalone program or as an adjunct to an existing treatment plan with a professional and a medication regime.

Some notes:

The scoring was based on self-reported changes in symptom severity. This made it easier to get objective results but also means that there could be bias introduced by the subjects themselves.

So here’s where things start to get fascinating and on occasion counter- intuitive.

1. Applications with in-app or professional support (defined as “providing feedback, chat messages, phone calls, and supportive coaching”) were reported to be less effective by the participants.

2. Interventions of longer than 8 weeks were seen to be less effective.

3. The apps were effective, but at a smaller scale, for people with “Western Ethnicities.

   1. Put another way, people in marginalized communities with less regular access to professionals and a higher stigma had much higher self-reported efficiency.

4. It didn’t matter whether the person had a formal diagnosis of depression from a professional or not. 

5. Programs with incentives and rewards had higher efficiency.

6. People who had outside medication regimes and access to mental health professionals got value out of the programs, but not as much as those for whom it was the only (one might extrapolate the first) part of a program.

7. ACT, Mindfulness and mood tracking apps had smaller effect sizes than behavioral analysis and evidence-based therapies.

If we twist this around a little bit.

People in cultures where they either don’t have regular access to mental health professionals or a cultural stigma against admitting to mental health struggles did not respond as well to apps that included those types of messages. I wouldn’t go so far as to say they want to pretend they’re talking only to a phone, because group forums aren’t mentioned as a type of professional support.  But there’s a balance here to be struck with what our ability to abide risk with participants who have admitted to a depression and just allowing the therapeutic app to do its work without creating a specter of a mental health professional.

Note: The researchers were texting the participants about the study, so there was non- therapeutic engagement happening outside the apps.

There’s evidence that the programs are most efficient when they’re finite, but that to me is tricky because 8 weeks might be sufficient to come out of a depression but short enough so that you don’t have a material relapse factoring into the symptoms. The longer programs might have been capturing a rebound from the initial euphoria of having fewer depression symptoms.

There’s some amount of bringing coals to Newcastle inherent in adding more intervention apps on top of a relatively therapy-conversant western population, although no one is going to argue whatsoever that there’s enough provider access. If this is true, that they didn’t really need a provider to see gains, it does speak to a potential force multiplier in the use of an almost completely hands off app.

Incentives work, though monetary is probably a bridge too far for much scale. Non-monetary seems to work as well.

Finally, one of the things this study managed to do is highlight, in its own way, the difference between UX research, and what passes for a significant sample size in those rooms, and scientific research. When I asked a group of the latter form of researchers what they thought of the study, it was a resounding “way too soon to tell.”

Interesting to keep an eye on.