The national health care software fiasco
Second in a five- part series
By UYLESS BLACK
Special to The Press
It is not unreasonable for taxpayers to ask Uncle Sam: How could you spend $36 billion of our money on a software system that a huge number of its users do not want to use? This writer believes the answer to this question is twofold: One, Obama’s administration set in place the necessity to implement this software initiative on a timeline beyond any semblance of reality. Two, regardless of the timeline, the incompetence of its implementation staggers the imagination.
Given this introduction, hold on to your taxpayer ire. Here we go.
The Tail Wags the Dog
Because this enormous infusion of money was tied to the 2008 financial meltdown, it was to be used as an immediate stimulus to the economy. Thus, the adoption of electronic health records (EHR) took place rapidly, as shown in the figure.
Show me the money! It was already being shown. Physicians could “qualify for federal subsidies...[but] only if they [used] a government-certified system.” As mentioned, time was important. Because the EHR initiative was part of the nation’s strategy to stimulate the economy, the funds had to be distributed as quickly as possible.
As the figure shows, the EHR industry was small potatoes in 2008, valued at about $2 billion. A $36 billion infusion into this industry was akin to a sudden gold rush, but only if a company could ramp up its operations to produce and sell its software quickly.
As a consequence of this sudden software godsend, 700 EHR firms sprang into existence. Seven hundred different systems were created that were islands of isolation from communicating with one another. It was as if 700 different languages were created within a few short years.
All well and good, if the systems (using the language metaphor) were not to speak with one another. But they should have interworked. After all, that was one of the major reasons for their creation, to be able exchange health care records between different hospitals and physicians.
In an astounding display of ignorance and ineptitude, the head of the organization responsible for managing the federal EHR program offered these ideas: “[Seema Verma, chief of the Centers for Medicare and Medicaid Services] shudders at the thought of the billions of dollars spent building software that doesn’t share data — an electronic bridge to nowhere.”
She told her interviewers (Fortune magazine and KHN analysts): “...we didn’t think about how all these systems connect with one another. That was the real missing piece.”
Perhaps Ms. Verma was not in charge of national EHR operations in 2008/09. Regardless of the boss, how could the requirement for having different health care records communicate with one another be “missing?” With this mentality, it is no wonder the national EHR program has floundered.
A software programmer’s dream is to be able to automate a manual system; to create the software from scratch. At the other end of the desirability spectrum is making changes to an existing software system. This effort requires the programmer to learn about the code that was generated by another programmer. The result is often the addition of code that introduces errors or leads to klutzy systems and unfriendly user interfaces.
Granted, in 2008 each hospital and physician had their own way of creating and maintaining their records. But the national EHR initiative provided a golden opportunity to start with no existing (or little) software to manipulate.
Take another look at the bottom picture in the figure accompanying this article. An opportunity was lost because the managers of this program did not have the “missing piece.”
The $36 billion windfall had to be exploited, and exploited soon. Get that software out the door! As one EHR company executive said, “This is a one-time opportunity to expand our share, focus everything there, and then we’ll go back and fix it.”
Fixing software is not like fixing an automobile carburetor. Fixing faulty software sometimes ends up making the software even more error-prone.
One of the readers of my work sent these thoughts to me: “So all the government needed to do to make this work was specify at the start that the software had to be compatible to a certain standard. Is that right?” My response: “Yes, and especially to make the software physician-friendly. They did neither.”
My reader continued, “I’m certain none of the competing companies wanted their software to be compatible with others.” My response: “Why would they want compatibility? Such an action would diminish their potential market share.”
The next article in this series examines the EHR industry, with a focus on the major vendors of EHR software.
These articles have used the following sources: khn.ort/ehr, fortune.com/longform/medical records (which cites Stanford Medicine’s 2018 National Physician Poll), Fortune magazine (April, 2019), U.S. Department of Health and Human Services.
• • •
Uyless Black has retired from writing computer code. Living in Coeur d’Alene with his wife Holly and pup Lilli, he now spends much of his time writing books about a variety of subjects ... none include the subject of computers.