Healthcare delivery should be standardized. Medicine is, after all, primarily a science. Providers must diagnose and treat patients. Clinicians must form hypotheses, test hypotheses, and act. As providers obtain new information, they must adjust their thesis and repeat the cycle until patients are treated. Although there is an art to patient interaction, the medical process itself is scientific.
Science is based on repeatable, nullable hypotheses. Diagnostics and treatments are too.
And yet, it’s widely known that healthcare delivery is anything but standardized. Even basic pre-operative checklists vary dramatically across locations. Although some of this variation can be accounted for by physical constraints and capital limits, most of the aberrations can be attributed to management and culture. Checklists and protocols attempt to standardize care, but even the protocols themselves are widely debated within and between organizations.
It’s also widely known that most innovations take the better part of two decades to roll out through the US healthcare system. For an industry that should be at the cutting edge, this is painful to acknowledge.
There’s a famous saying that vendors represent their clients. It should be no surprise that major health IT vendors are slow to innovate and respond. Providers are used to slow changes, and have come to expect that of their vendors. Since providers often cannot absorb change that quickly, vendors become complacent, the pace of innovation slows, and innovations slowly disperse.
In the same light, health IT vendors are equally unstandardized. In fact, health IT vendors are so unstandardized that there’s an entire industry dedicated to trying to standardize data after-the-fact. The lack of standards is pathetic. A few examples:
Claims – Because insurance companies want to reject claims, they have never agreed on a real standard for claims. As such, an entire industry has emerged – clearing houses – to help providers mold claims for each insurance company. In an ideal world, clearing houses would have no reason to exist; all claim submissions, eligibility checks, and EOBs should be driven through standards that everyone adheres to.
HL7 – It’s commonly cited that every HL7 integration is just that: a single HL7 integration. Although HL7 integrations share the same general format, they accommodate such a vast array of variety and choice that every integration must be supported by developers on both sides of the interaction.
As a technologist, the lack of interoperability is insulting. Every computer on this planet – Windows, Mac, iOs, Android, and other flavors of Linux – communicate via the TCP/IP and HTTP protocols. Even Microsoft, Apple, and Google play nicely within enterprises. But because of the horribly skewed incentives within healthcare, none of the vendors want their customers to interact with other vendors, even though cooperation is vital.
Perhaps the most ironic observation is that technology is widely considered to be hyper-competitive. Despite hyper-competition, the tech giants have coalesced around a common set of standards for communication and interoperability. Yet health IT vendors, who operate within a vertical that prides itself on its scientific foundations, fail to communicate at the most basic levels.
Kyle,
I find the debate between the art vs science of medicine really interesting. I think the challenge with medicine being science is that in many cases we lack the information. So, there’s an art to knowing how to make decisions based on limited information. Plus, there’s an art to getting the information that you need. Once we improve the science of collecting better information, medicine will be more of a science.
An excellent descriptions of the situation that many, who are dealing with the frustrations of the current generation of health IT systems, find themselves.
Until we fix the current health IT mess, especially interoperability, nothing else of consequence will be possible. And forget analytics. No data, no data.
We are a community-based approach to improve health IT so we can get on with the more important things, like the triple aim. Visit us at; http://www.TheCUREProject.org.
Kyle, John has already commented on the art and science of delivering healthcare, so I’ll skip that.
From a technology perspective, the most common analogy I can think of is – as a consumer you go to any ATM, anywhere, irrespective of the Bank you use, you can pull out cash.
To use your word ‘hypercompetitive’, healthcare IT vendors are hypercompetitive to keep it Non-Standard. And the constituents; providers and patients don’t and can’t influence that.
In the world of Pharma, I was part of an organization called CDISC – a data standards organization created and influenced by two important data consumers – the Pharma industry and FDA. When a clinical trial is done for new drugs, there are many players involved and many moving parts.
The industry decided it was time for standardization. The people that rallied were Pharma and FDA – the beneficiaries of standardization. It was super successful.
Neither patients, nor Providers are united to influence standardization. The only hope is with CMS, because the industry is not doing to do it on it’s own.
RE: “It’s also widely known that most innovations take the better part of two decades to roll out through the US healthcare system.”…
There’s nothing really innovative about developing a standardized system — merely a willingness to borrow and share existing ideas. Amazon works well and is scalable and easy to use; Healthcare.gov doesn’t and isn’t.
I don’t know what it will take before the EMR vendors come together for their own self-interest, but it most likely won’t be a government referee — perhaps a revolt by those doctors who refuse to use EMRs, a situation that will likely soon become critical as the government starts to impose penalties that cause them to quit practicing medicine altogether.
“Perhaps the most ironic observation is that technology is widely considered to be hyper-competitive. Despite hyper-competition, the tech giants have coalesced around a common set of standards for communication and interoperability. Yet health IT vendors, who operate within a vertical that prides itself on its scientific foundations, fail to communicate at the most basic levels.”
Maybe Health IT is just too young an industry and it needs time to “grow up.” Don’t forget, five years ago, the wide majority of health care was on paper…
I disagree with much of what’s written here, but here’s one word, XML.
One of the many paradoxes of HITECH/ARRA is the cementing in place of the dinosaur systems. Some call this an unintended consequence, and some say it was a purposeful gift to the large HIT vendors that were struggling to just survive before the legislation.
These legacy systems were authored 10, 20, maybe 30 years ago and many are still in place. The authors had no idea what the standards and de facto standards would be in future — they did not have a crystal ball.
Back then, the HIT developers had not yet even heard of HTML just as the new programmers today couldn’t code COBOL if they had to.
Furthermore, the system authors and owners don’t have a duty (or the resources) to re-factor every time the landscape changes.
—
“but even the protocols themselves are widely debated within and between organizations”
Is the author talking about healthcare providers or IT providers? 🙂
Doesn’t anyone here remember the standardization that EHRs brought?
Many years ago I sat in on a sales presentation where the EHR vendor was selling the fact that by clicking certain things, specific paragraphs were built and shown as “typed” by the doc.
It was a great time saver for the doc.
The problem was two-fold: 1)HHS raised the BS flag on this, 2)Docs realized they couldn’t actually tell what was wrong with the patient from the standardized entries.
I believe it is possible to have a science that is not standardized. Though, maybe not by definition.
I’ve had two close friends in the last 6 months go from being told they have a life ending illness, to…nevermind, you are fine. They were being wrongly diagnosed.
If medicine is such a standardized science, why would it take up to 5 docs to figure out what is really wrong with somebody?
Of course, this doesn’t mean EHR datasets can’t be standardized, but if that was really what the government wanted, they would have pushed VistA on everyone.