Goodbye, DiGAs: The DiGA Bubble Is Bursting

DiGAs are reimbursable Health apps in Germany. Here’s a quick primer on how they work.

Let’s say you show up at your doctor’s office, said doctor can say “hey, you’re suffering from depression and I only have 5 minutes to talk to you because public health insurance only pays me a tiny amount for this visit and even if you want to see a licensed therapist, the wait times are multiple months, so you know what, how about I prescribe you a fancy DiGA app which you can download to play around in the meantime and maybe your depression will get better, at least that’s what the manufacturer-sponsored study says”.

You see where this is going. But minor sarcasm aside, here’s what’s going on behind the scenes: The doctor gives you a prescription, you download the app, the DiGA manufacturer gets reimbursed and receives a slightly-higher-than-what-a-normal-human-would-pay-for-an-app sum, usually a three-digit sum in Euros.

Okay. So far, so “good”.

There are a lot of bad things to unpack here (and some good ones, too), and wrote about those in May 2023.

But instead, here, I’d like to talk about the elephant in the room:
Most DiGAs suck because they are no real innovation, and that’s because the DiGA regulation initially limited them to be mostly-low-risk medical devices.

What do I mean by that? Roughly speaking, the apps can be subdivided into two categories: Better Googling and true innovations.

The First Type of Health Apps: Better Googling

The first type of health apps, and the vast majority of DiGA apps are what I call “better Googling”.

Let’s assume again you’re a patient suffering from depression and the wait times for therapist sessions are way too long . You want to get started now. Here’s what you could do (and what similar people have done already): You could go ahead and start Googling about possible self-administered treatments for depression, e.g. cognitive behavioral therapy (CBT) and similar methods, and come up with a rough step-by-step plan with interventions like getting up at the same time every morning.

All of that would be rather cumbersome and you’d be at risk of relying on “non-medical-grade” information which you found via Google (and now, ChatGPT), but if you execute this well, you would be able to make some reasonable progress. Probably less than working with a therapist, but still better than nothing.

And that’s exactly where the DiGA comes in. Instead of you having to manually research and assemble this step-by-step “self-treatment” plan, there’s a handy DiGA app for that which already comes pre-loaded with the right content. This reduces a lot of friction: Less effort for you, and a better treatment as the information is vetted by professionals.

So it’s better than Googling yourself. But.. how much better exactly?
10x? No, probably not. 3x? Not sure. 1.5x? Maybe. Not a huge improvement.

Suffice to say, I think it’s fair to call these apps “Better Googling” DiGAs.

Now there’s also a second type of apps, and they are entirely different. They are true innovations. Let’s look at those next.

The Second Type of Health Apps: True Innovations

If you were a software developer planning to build a health app, instead of building a “better Googling” app, you could go for something way riskier instead: A true innovation. What would that be? The best example I can come up with would be an app which monitors values from continuous glucose monitors (CGMs) and directly controls insulin pumps, thereby creating a “closed-loop” insulin system.

Those sort of systems exist, but they are very expensive. Imagine an app which comes along and allows patients to combine multiple low-cost components into a viable closed-loop system. That would be huuuge!

.. and it would be hugely risky. Because if things go wrong, you’re literally sending people into a hypoglycemic coma.

But then again, this is exactly the point: In almost all cases, big innovations entail big risks. And this is exactly where the legislation failed, in my opinion, by initially only allowing (up to) class IIa medical device software to go for DiGA approval.

You see, class IIa software, in simplified terms, is still mostly low risk. Our closed-loop system above would most likely be class III, so it would currently not qualify for DiGA approval.

The DiGA people later widened this scope to include class IIb devices, too, but my gut feeling is that this is “too little, too late”. Let’s look at two “policy” scenarios:

• Scenario A: Regulators are wary of this new thing called “software” in healthcare. They want to create reimbursable apps but, dear god, not take any risks. Let’s only allow class IIa (and lower) software to be DiGAs.
  Result: After a few years, people realize the apps mostly aren’t very useful and insurances are becoming wary of reimbursing them. A prominent blogger even calls them “better Googling”. The program ends.
• Scenario B: Regulators “go big or go home” and allow all medical device software, i.e. up to class III, to go for DiGA approval.
  Result: Some crazy innovations become available. People build software which directly treats diabetes. Some things go wrong and some patients are hurt, possibly causing a public backlash. But there are still a few apps which provide crazy medical benefits.

As you see, there’s no perfect scenario here.

Germany opted for Scenario A, the low-risk scenario. I’d argue that, in retrospect, this was the wrong decision. Why?

Scenario A has less risks, sure, but it also doesn’t have any upside at all – best case, you get some apps which are “better Googling”. So while you don’t hurt patients, you also predetermine the outcome of the DiGA program as a whole to fail.

Scenario B, on the other hand, is of course more risky, but has significant upside. Here, you would only have needed one blockbuster app (closed-loop system!) for the DiGA concept to be widely regarded as a crazy success.

But that’s not how things went.

Here are some predictions of what’s going to happen next.

DiGAs: What Will Happen Next

Here’s what will happen next:

1. Patients, insurances and manufacturers (in that order) will realize that DiGAs are not super useful.
2. Manufacturers and their venture capital investors will realize that they can’t “scale” their revenue, because patients and insurances have realized that DiGAs are not super useful.
3. In a sort-of-last-ditch effort, the scope of DiGAs is expanded from class IIa to class IIb. But it’s too little, too late. (This has already happened)
4. The first DiGA companies go bankrupt (<– we are here).
5. More DiGA companies go bankrupt.
6. The program ends.

As always, I might be wrong. DiGAs might turn out to be a huge success, but.. right now, I’m not seeing any reasons why that would happen.

I guess the moral of story is that you have to take real risks to make real innovation happen, and by any indication, German politics of the last few years have not been very favorable towards taking any risks at all.

We will see.

But not everything’s bad. Personally, I was really excited about the DiGA program. It moved a lot of startup founders 1) into Germany and 2) into Healthcare. And, even now, I see more founders considering building a company in Healthcare in the future. So, at the very least, there are some really positive second-order-effects here.

But will the DiGA program itself survive? I don’t think so. We will see.

What are your thoughts? Leave a comment below 🙂

(Oh, and if you are a founder who still wants to bring a medical device to market, check out my company OpenRegulatory which offers lots of free document templates, consulting and a software for medical device compliance. I do see the irony that this post is likely terrible marketing, but it’s worth a try, right?)