Free sounds great, right? But as they say in economics, ‘there’s no such thing as a free lunch’. When considering a health data integration solution, there’s always an opportunity cost to using something that’s free.
Last week we announced Shimmer — the first open-source health data integration tool.
Use it. Extend it. Add to it. And host it in your own infrastructure. There are no strings attached to using Shimmer. But despite it being free to use, we wanted to dive deeper into some of the true costs to using Shimmer (e.g. setup and maintenance). Whether you’re a small company wanting to leverage health data or a large hospital system looking to ingest lots of new kinds of data, you’ll see that Shimmer is the most cost-effective (and customizable) data integration solution out there.
Shimmer’s economics
The direct costs to use Shimmer are $0. All you have to do is head over to GitHub and follow the instructions outlined there.
In order to implement Shimmer, however, it’s going to take a little bit of your software engineer or devops engineer’s time.
Some assumptions on our end:
- In the current market, we estimate a cost of about US$100 per hour for an engineer’s time, this includes salary, benefits and taxes.
- Pessimistically, integrating with Shimmer will take approximately 40 hours of the engineer’s time, though it could take far less depending on the developer’s skill set. An important point here is that regardless if you use Shimmer or another proprietary health data aggregator, you’ll spend the same amount of time with your engineering trying to hook up their API to your infrastructure. (This is $4,000 in setup for all levels).
We’re not done yet.
You’re going to have to maintain the integration. Every once and awhile, one of the API vendors is going to update their API, or Shimmer itself will be updated. Although we anticipate that the community will handle the actual development, your engineer will still need to keep an eye on these changes so they don’t break your app or service. API maintenance is estimated at 5 hours per month, which is $500, or $6,000 for a full 12 months.
Though Shimmer is relatively stable to use over time, another 5 hours of time per month are estimated for other infrastructure maintenance. This adds another $500 a month, or $6,000 for the year. If you are managing 10,000 (or more) users, you will likely have to devote an additional 5 hours a month. But this is normal when you begin to scale. Let’s add an additional yearly cost of $6k.
Here’s the full breakdown over time for 500, 5,000, or 10,000 users:
Shimmer – Cost Breakdown
If you have less than 5,000 (active) users in your application, using Shimmer can cost nearly 6x less than some of the proprietary health data aggregators. Imagine what you can do with all that money you’ve saved in the long-run using Shimmer!
What Our Analysis Doesn’t Include
These numbers do come with a few important caveats:
Once you’ve configured Shimmer for your infrastructure, you’re probably going to want to do something with that data. You’ll then need to build batch processing, analytics, visualization and some business logic capabilities. The price to build these features vary considerably, so we don’t include this in our analysis, and that cost is also incurred using proprietary aggregators. Luckily we’ve built some tools to help you do more with this data.
One thing’s for sure, because the data coming out of Shimmer is clean and clinically-valid via the Open mHealth data format, it should be simple to consume Shimmer data.
We also don’t include hosting in the cost analysis because it is highly dependent on your usage and whether or not you need it be HIPAA-compliant (we highly recommend Catalyze and Aptible for HIPAA compliant hosting). The good news is that these costs continue to fall as competition for the cloud storage market has intensified.
More Control Is Better For You
I know you’re probably thinking it’s scary to run your data integration platform in-house, but let’s consider the advantages that total data ownership has, starting with APIs.
Other data aggregators give you one pipe into many different data sources. But you are making an agreement with them and them only. It’s convenient to not bother setting up developer accounts on the 3rd party services you care about, but you end up losing the relationship with the tools themselves. This means you have to wait for the aggregators to update their API based on changes from the tools and rely on them for your service. That’s really scary to us especially considering how fast companies are being acquired and going out of business. I wouldn’t want to rely on a service for access to my user’s mobile health data.
Because Shimmer is open source, you alone get to decide which APIs to implement in the initial setup by registering with each of the various companies. You are also in control of which to add later and which to update when companies change their APIs. There’s no need to spend unnecessary time or resources on APIs that don’t make sense for your company.
You’re also probably wondering, “Is there support? If something changes on the Open mHealth side, will it break my app?”. With Shimmer you have the added advantage of belonging to a vibrant community of developers and clinical experts who are continually working to improve and expand upon the code. You don’t get that level of community involvement in any other data integration service. We provide excellent documentation, attend to our developer forums and GitHub issues, and squash bugs as fast as we can. For those that want extra assurance, we’re exploring offering paid implementation and support options at prices that would fall far below the cost of proprietary data aggregators.
—
Ultimately, we’re in an era where we’re getting increased access to data we’ve never had before. All at our fingertips. Would you rather be in control or let someone else dictate your future? Get Shimmer today to have the lowest-cost, clinically validated, open-source health data integration toolkit. No other platform can compete with that.