The Elf Movie and the Health Information Exchange Tragedy of the Commons

By Buff Colchagoff, CEO, RosettaHealth
Twitter: @rosettahealth
Twitter: @BuffColchagoff

The 2003 Will Ferrell movie “Elf” has become a new Christmas classic. It also represents a cautionary tale when viewed through the lens of a “tragedy of the commons.”

Spoiler alert for those who have actually never seen the movie: Santa can’t fly his sleigh because there’s not enough Christmas spirit and everybody is just thinking of themselves. It’s not until NYC comes together, full with Christmas spirit that Santa can fly and complete his mission.

But what does this have to do with Health information Exchange for Elf’s sake? First, let’s do a quick review of something you probably learned in college sociology class, brought to you by Wikipedia (you’re welcome).

“The tragedy of the commons is a situation in which individual users, who have open access to a resource unhampered by shared social structures or formal rules that govern access and use, act independently according to their own self-interest and, contrary to the common good of all users, cause depletion of the resource through their uncoordinated action in case there are too many users related to the available resources.

In our industry, the commons is represented by the national exchange networks, which are eHealth Exchange, CommonWell and Carequality. The users are the organizations that connect to those networks.

At the birth of the first operational national network ONC’s Nationwide Health Information Network (NHIN), created in 2009 and eventually became the eHealth Exchange. During this time, there was a known problem.

Because the NwHIN network had everyone connect directly to everyone else without a central traffic cop, a user would be required to call every other user to see if there were any health records for a patient across the entire network. This was known to be inefficient and wasteful.

But, there were other much more pressing concerns at the time, especially making a viable network within the available time and funding. The early network forefathers decided to push this problem onto later generations to solve.

This is familiar territory for me because I was the Director of Operations for the NwHIN at ONC at the time of its birth.

Fast forward to today. It has been many years since those initial network connections. There are multiple exchange networks with interconnections between them — and yet, that avoided problem still exists. Granted, the use of hubs has mitigated some of the issues, but there’s still a problem that my clients face, which is the network tragedy of the commons:

• Network participation costs are asymmetrical between consumers and providers.
• Larger consumers project costs onto information providers.
• Small information provider must therefore pay disproportionate costs to their share of the commons.

More specifically: The cost to query for a record is relatively inexpensive compared to the cost to do the demographics search for a patient. Probabilistic matching of patient demographics uses a fair amount of computing resources, and when an information provider receives many requests per second, it can overload server resources.

The records provider then has to add resources to service the large amount of requests — stretching the resources and budget — all while just trying to be a good participant on the network. I have had this issue with more than one of my clients when they join a national network:

• They don’t anticipate the load from a flood of network requests.
• Some of their services start to fail and timeout because of loads put on resources.
• Adding compute resources is just adding costs, often on budgets that are already stretched.

One real world example demonstrates this. A smaller HIE (SmallHIE) started operations on a national network. The SmallHIE was almost immediately inundated with large loads of requests (hundreds of thousands of requests nightly) from a large nationwide organizations (NatlOrg) . This caused a number of resourcing issues. The SmallHIE tried to lower the demanding load by only processing patients with addresses within their state. But the NatlOrg told them that this was not acceptable, what if someone had moved? Ultimately, the SmallHIE had to process all the requests.

The good news is that these 2 actors were eventually able to find a workable solution. However, these sorts of issues are common on national networks.

Consumers have almost no costs, and those providing records pay a disproportionate cost. Still today, the SmallHIE receives two orders of magnitude more requests than they make on the network. Most outside observers would consider this “unfair”… that in the commons, there needs to be a balance for good of all users.

What’s the answer?

Ultimately, we need updated rules of the commons, and this is the responsibility of the network governing bodies. Sure, there are architectural solutions like record locator services that would take some time to inject, and QHINs will inevitably shift the costs equations.

However, in the near term, I think all players could use some Christmas spirit. If network users treated each other with the Christmas spirit that lifted Santa’s sleigh in Elf, then much of the pain being felt would melt like March snow.

Large organizations need to appreciate the limited resources of smaller organizations and make accommodations.

Growing network governance needs to include the needs of the many, not just the most influential or vocal. And everyone should appreciate that we’re working for a shared goal of health information exchange — some have little means, some have great resources… but in the end we really do need “peace on earth, goodwill toward [all]”.