There’s one thing Democrats and Republican politicians agree on when it comes to healthcare reform: digitizing healthcare records could be a godsend. When it works, it can improve patient care, reduce medical errors, and save billions of dollars a year — a crucial consideration given the threat that skyrocketing healthcare costs pose to our economic future.
But in an article entitled “The Dubious Promise of Digital Medicine,” Business Week pours a big bucket of cold water on this enthusiasm. To understand why computerizing healthcare records is running into trouble, we’ll take a look at the problem using the three assumptions about the economy I explored in the last few posts.
People Aren’t Calculators
In theory, Healthcare IT should be able to radically reduce the number of medical errors. For example, with computerized records nobody has to decipher a doctor’s handwriting. In practice, the track record is mixed.
The Joint Commission, a nonprofit group that inspects and accredits 15,000 health-care organizations, … issued a warning in December about problems with complex health-tech systems. It cited one U.S. pharmaceutical database that found 43,372 medication mistakes, or about 25% of the total reported in 2006, involved computer technology. The problems included flaws in data entry, inadequate software, and confusing screens.
As I’m sure you know from painful personal experience, a lot of software is built by geeks who don’t pay attention to how people actually think, let alone what it’s like to use the software in a hectic place like a hospital. And if software doesn’t work the way people think, it can actually increase errors.
Take the experience of Dr. Mark Del Beccaro, chief medical information officer Seattle Children’s Hospital.
Del Beccaro soon became troubled by incidents of children suffering medication overdoses despite alerts from the Cerner software. He asked the doctors involved whether they had seen the alerts onscreen. “They told me, ‘I get so many alerts, I click through [them],’ ” Del Beccaro says. “They do become mind-numbing.”
In principle, alerts are a very good idea. If someone’s tired or distracted, an alert could save someone from, say, ordering a very dangerous drug combination. But people only have so much room in their brains to handle alerts. If the people designing the software don’t take this into account, constant reminders could actually increase errors.
Organizations Aren’t Calculators
Why doesn’t the market take care of this? Healthcare IT vendors who do a better job of adapting to users’ real needs should beat the ones that don’t. In the real world this isn’t happening. Why? Organizations aren’t calculators — or at least not simple ones.
Hospitals are very complex organizations, often with their own organizational culture and ways of operating. That means that a healthcare IT system has to be flexible enough to fit how each hospital works. As BusinessWeek says, this creates “a fundamental tension”:
Info tech companies want to sell mass-produced software. But officials at large hospitals say such systems, once installed, require time-consuming and costly customization.
BusinessWeek cites a 2005 Pediatrics article on what can happen when the organizational needs of healthcare IT vendors and hospitals clash.
Digital technology slowed treatment in several ways, the researchers concluded. One example: Doctors and nurses in the intensive-care unit were accustomed to ordering medications and tests while a sick child was en route to the hospital. The Cerner system required that orders be submitted only when the patient arrived, costing crucial time.
Even “one of health technology’s best-known advocates,” Dr. James M. Walker isn’t happy with how this tension is playing out:
Vendors such as Epic, Walker says, sell relatively rudimentary electronic tools and expect hospitals and doctors to assure accuracy and safety. “This can be very tricky,” Walker adds. “A lot of us are trying to say: ‘Look, let’s slow down.’ ”
Why don’t healthcare IT vendors build systems that are more flexible? Here’s my guess: it’s harder to build than an inflexible system. If you’re going to build a flexible system, you can’t just take a look at the needs of one hospital (i.e., the first client you get). You have to study what goes on in a bunch of hospitals – and you can’t leave the details up to geeks, you’ve got to bring in folks with squishier, people-centered skills that many IT shops don’t value. You also need a smarter design to build a flexible system; you can’t just bang it out.
Then there’s the dirty little secret that if a system is too flexible, the vendor is leaving money on the table. A flexible, well-designed system may cost a lot less to install and customize – good for the hospital, not so good for the IT vendor.
And if an IT vendor’s system isn’t flexible, it can actually help the vendor gain more business – if you buy an inflexible package, you may have a hard time getting it to play nicely with another vendor’s software. Take what happened at Dr. Walker’s hospital. After spending $35 million buying and installing Epic Systems, they ran into a serious problem during a pilot project.
[Errors] began appearing at a rate of several a week in the hospital’s psychiatric unit. “The pharmacy would interpret an order as one drug at one dosage, and the patients were ordered the wrong medications at different dosages,” recalls Jean Adams, a nurse in charge of the IT team. Fortunately, astute staffers discovered the problem after a few weeks and began verifying the computer drug orders using the phone. Full implementation of the Epic system was put on hold. Adams says Geisinger traced the trouble to incompatibility between a common pharmacy database and Epic’s system.
Epic CEO Judith Faulkner says the episode at Geisinger, and similar incidents at other hospitals, taught her company that physician orders and pharmacy records cannot use distinct technologies. “It doesn’t work when you mix and match vendors,” Faulkner says. “It has to be one system, or it can be dangerous for patients.”
In theory, the costs of not locking out other vendors and the hard work of creating a flexible system could be outweighed by winning more contracts — after all, they’d be a better deal for most hospitals. But good luck making that argument in a healthcare IT company. Companies aren’t rational calculators, and selling that kind of gamble to the Top Brass is going to be tricky. And if all your competitors are facing the same issues, everybody may implicitly decide they’re better off not deciding to compete based on flexibility.
Politics Are an Inherent Part of the Market
Given these challenges – and the fact that Health IT can cost big bucks – it’s not too surprising that the feds are pushing hard to reshape the healthcare ecosystem to support Healthcare IT. Obama’s stimulus plan tried to solve the financial challenges by giving hospitals big incentives to computerize their records.
There’s also a stick: The federal government will cut Medicare reimbursement for hospitals and medical practices that don’t go electronic by 2015.
But when it comes to overcoming the nonfinancial challenges of Healthcare IT, the government hasn’t stepped in yet. For example:
When health technology fails for one medical provider, there is no central mechanism for reporting problems to others who use it. The federal government collects and disseminates this kind of information on drugs and medical devices. But tech contracts routinely bar medical providers from disclosing systemic flaws.
The U.S. Food & Drug Administration has been considering whether to regulate health technology in the manner it oversees medication and implants. That decision now falls to the Obama Administration, which faces opposition from industry groups arguing that additional red tape would impede adoption of helpful technology.
Instead, Healthcare IT vendors have been pushing for leaving oversight with a nonprofit called Certification Commission for Healthcare Information Technology, which has a heavy vendor presence and is chaired by a former CEO of a healthcare IT vendor. How tough are its standards?
Sharona Hoffman, a professor of law and bioethics at Case Western Reserve University in Cleveland, says CCHIT’s product testing, typically completed in a single day, isn’t rigorous enough. In an article last December in the Harvard Journal of Law & Technology, she and a co-author faulted the group for telling vendors the testing scenarios in advance and for not conducting ongoing monitoring
Got that? When the commission tests these aircraft carrier-sized systems that cost $35 million a pop, they usually test them for just one day.
Healthcare IT vendors might have a shot at convincing DC to back off. Why?
… red flags raised by doctors and researchers haven’t gotten much attention in Washington, in part because the health-tech industry has forged strong ties to the President, his top medical advisers, and Republican heavyweights such as Gingrich.
Take the new White House healthcare reform czar,Nancy-Ann DeParle.
[She] recently stepped down after eight years as a member of [healthcare IT vendor] Cerner’s board of directors. A former administrator of Medicare and Medicaid during the Clinton Administration, DeParle worked from 2006 through 2008 as a managing director at CCMP Capital Advisors, a private equity firm that invests in health-care businesses.
Similarly, healthcare IT vendor CEO Glenn Tullman, who used to work out with Obama, served on Obama’s campaign finance committee and is also tight with Gingrich:
Shortly after the stimulus became law two months ago, Tullman and Gingrich hosted a Webcast for thousands of hospital officials and doctors promoting the financial incentives. Since then, Tullman has worked with a client, the University of South Florida Health system in Tampa, to seek $15 million in stimulus money to hire 130 e-health “ambassadors” who would pass out free samples of Allscripts’ prescribing software to physicians. If the funding comes through, the $50,000-a-year representatives would receive a two-week training course from Allscripts, though the marketers otherwise are supposed to be independent of the company.
Conclusion
Sometimes Healthcare IT works really well – and when it does, it can be a godsend. The VA, for example, has saved millions of dollars through health IT. And my new doctors all work in practices that have been recently computerized where not only can they easily pull up my info but they can also electronically send my prescriptions to my pharmacist, which means that by the time I get to the pharmacist, the drugs are ready.
But making healthcare IT work well enough to drive down medical errors and health care costs overall won’t be easy. I’ll leave the last word to somebody stuck in the middle of this mess:
“Most big health IT projects have been clear disasters,” says Dr. David Kibbe, senior technology adviser to the American Academy of Family Physicians. “This [digital push] is a microcosm for health-care reform….Will the narrow special interests win out over the public good?”
