Doing The Right Thing

Having just returned from HIMSS, the Supershow of Health Information Technology, I'm trying to shake off the flashing lights, Health IT booth-turned-pub, the DeLorean silver car and other attention grabbers to see what the potential of $40B of our tax money to be spent on incentives means for improving the state of health care information technology in the country. The sad truth is not much. It's still really hard and really expensive to get clinical information from point A to B, (unless the same company is at both ends, and then it's just expensive) and to get well edited and timely information on mobile devices. These are basic fundamental needs that any other industry has, and yet the most curious thing to me is that there is no anger, no outrage, from the hospitals that are forking over the money, year after year. As a former CIO in two other industries, I know that until the clients start voting with their wallets, no change will be forthcoming.

Like any other industry where a lot of money is at stake, there are major companies that control a lot of the market. In our industry, these include well known names like GE, and, Siemens, and industry specific powerhouses like McKesson, Epic, Cerner and Allscripts. Historically, these companies have been selling primarily to hospitals, but with the new HiTech act of 2009, they are expanding their products to reach physicians as well.

It's been pretty clear for decades that these big companies' strategies include total ownership of the hospital's information. But health care never happens just in a hospital setting, and very infrequently in a single doctor's setting. There needs to be a simple, low cost way to get this information around--just like you can order on-line from multiple places and track your package on UPS, or go to a bank anywhere around the world and get cash.

So thought our government, when they spent 800+ pages detailing "Meaningful Use" for our country. To boil the law down to its essence, there are two parts to it; first, you must have an electronic system of record where patient information is kept securely in standard form. Second, you must make that information available electronically so a person or another system can use it. Pretty straightforward, right?

After two years of teeth gnashing and hand-wringing, led mostly by incumbent vendors, who led a charge to water down the requirements to meet them, I was eager to see our new, improved, interconnected world. As my grandfather used to say, "A leopard can't change its spots." What have these big companies done to help systems talk to each other? More of the same. It is still hard to communicate between systems, and if you insist on it, it will cost you a small fortune. Two software practices, bundling (making you buy a lot more than you wanted to get a specific feature), and forced upgrades seem to be the commercial methods in play if a hospital wants to achieve compliance.

As an experiment, I went to all six of the major vendors to see how we, a small company, could interface with their systems. This isn't a unique question, according to our government's tax roles, there are over 5000 smaller Health IT companies, devices, and system integrators that would like to make their systems work with the big systems.

"Oh yes, as soon as our mutual hospital client upgrades to our newest version, and then requests an integration specialist"--Upgrade cost? Over $10M dollars for a 500 bed hospital (assuming they already have paid tens of millions for the system). Three of the six gave a similar answer. Fine print--there are many other upgrades included in these new versions, but if you just want to be able to read/write HL7, you must upgrade)

"If our clients buy our whole system, they have no need to integrate with anything else." Really? Does your system links with PACs? "That's a special integration project." Labs? "Hospital labs are special integration projects." Mobile devices? "We have a Citrix client to display things on a browser." How much does your system cost? Anywhere from $30M to $250M dollars..."

Best answer, two of the six had a special HL7 services pack that was a separate module for purchase that didn't require a total system upgrade, (price not given), but the red tape, the information comes out standard in a pdf format, not machine format. Why not? "We believe our system should be the center of all the information for that hospital."

All these big companies have followed the letter of the law, not the spirit of the law. You still can't get a basic hospital system without spending tens of millions of dollars--going up to hundreds of millions for a full implementation. A hospital IT system is expensive, but these bundling and upgrade practices make simple interoperability very expensive.

And it's not like the costs of these systems are hidden costs--Moses Cone, a 500 bed hospital near us in Greensboro, chose the rip-and-replace strategy (taking out an old system and putting in a new one) had this in their newsletter...

"A team of more than 90 Moses Cone Health System employees will begin training in October on a comprehensive medical information system that will be built through a contract with Epic Systems Corp.

The Health System has committed more than $80 million for purchasing and installing the software and hardware, as well as more than $30 million in staffing costs related to the electronic health record over the next five years. The system, which will be in place in two to two and a half years, will result in integrated billing, registration and clinical software packages that seamlessly “talk” to each other."

To each other, yes, but not to anyone else.

Let's put this in some perspective. During that same time 2 1/2 year period, a 500 bed hospital will see about 250,000 emergency patients. About 25% of them will be uninsured and without primary care. At $80 per primary care visit, each patient could get to see a doctor 20 times for the same money. Ultimately, it's all the same pot of money a hospital has to utilize. Better continuity of care for uninsured patients, or a systems project that ultimately locks in the hospital to speak only with other physicians with the same system.

At Axial, we believe all health systems should speak to each other safely and securely at a fair price. To back that claim, we put our technology in the public domain. We are building a library of public "connectors" that will let any legacy system speak to another legacy system. I challenge hospital CIOs and CMIOs to demand open access to the information they've paid dearly for. I challenge anyone operating in Health IT to do the right thing. Charge for innovation and health improvement, not linking billing and clinical systems--something that every other industry did decades ago. We challenge the rest of the industry to follow our lead.

Build It and They Will Come

It is rare in the economic history of the United States that the Federal and State Governments moves faster than the private sector. But that's what seems to be happening in the field of Health IT, especially with respect to interoperability. You may well be familiar with some of the incentive money, but probably not with what is happening at the State level--and this could affect every American's ability to access information. So read on if you want to make sure your state does the right thing.

There are two big buckets of money in ARRA set aside for Health IT. The largest, by far ($36B gross, $19B net) is for hospitals and doctors to move from paper to electronic medical systems. These incentives come at the back-end of the Health Care provider's implementation project, in the form of multipliers to the reimbursement the practice receives from Medicare or Medicaid. Hospitals are gearing up to take advantage of these incentives, as it may mean somewhere between $2-$7M per hospital. However, the average doctor will receive $40,00-$60,000 over four years, not covering the cost of most implementations. Needless to say, the doctors are lukewarm to negative about this incentive package, and aren't rushing to the door to convert their practices.

There is a smaller bucket of ARRA money that doesn't get nearly as much press coverage, but may have a much bigger bang for the buck. That is money set aside for States to build Health Information Exchanges. These would be public "backbones" that will ultimately connect all the hospitals and physicians in each State. The announcement, made by Vice President Biden, and HHS Secretary Kathleen Sebelius on August 20th, set aside $564M for the States to build Health Information Exchanges. Because of the particulars of ARRA funding, States will have to respond and build quickly. The States submitted Letters of Intent on September 17th, and have to submit grant proposals by October 14th. Then, the Federal government will announce the grant recipients by December 1st. Having managed procurement for a major private institution, this is very fast turn-around indeed.

Here's the kicker. If the States want the Federal Government to pay their whole tab, then they must have their project completed within two years. After that time, State matching funds must apply to the project. Given the in-the-red status of most States, they will hurry, hurry, hurry.

That could be great, or it could be setting the table for failed implementations. A lot will depend on whether each state adopts four basic principles outlined below. If you want your state to provide a backbone for a health information exchange, feel free to cut and paste from this blog and send this to your State's Governor. Each state applying usually sets up some kind of Authority to manage the process (in my state, it is the NC Health and Wellness Trust Fund), but that Authority will be determined by the Governor.

1) Set up the State for easy Communication with the Federal Agencies at a low price


A state's Health Information Exchange's design should mirror the National Health Information Network’s open source CONNECT design. States should not recreate the wheel. Using the CONNECT design, the State will be able to seamlessly connect with Federal Agencies such as Medicare and Medicaid, the Department of Defense and the Veterans Administration. Additionally the flexible open source design keeps medical information resident in the existing systems within doctors and physicians’ practices and avoids the need for a statewide medical database--a fear for many citizens.


2) Keep Costs Low and Improve Interoperability by Mandating the Use of Data Standards by Companies involved with Health Care Information in the State


States should promote long-term sustainability by mandating that all companies doing business with the State, including laboratories, imaging centers, hospitals, and Physician offices make their results available in the ANSI-approved standards known as HL-7 (there are other relevant standards, as well). If a system is old or a vendor does not know how to do this, then they must put their interface technology in the public domain so that third parties can convert it to industry standards. We have learned from the 200 plus HIE's that were created and not sustained that the largest cost of both building and operating exchanges is due to proprietary, point-to-point non-standard connections between legacy hospital and lab systems.


3) Plan Personal Health Information to be the Centerpiece of the Exchange



A State should design its Exchange with the patient front and center. Currently, most HIEs are business-to-business, with only 2 HIEs making data available to the patient, though many have plans. We believe the key to getting long-term health care costs down is to educate patients. American consumers are very savvy, and they will learn what they need to in order to make good life decisions. They just need the information.

In addition to the broad principals, patient-centric records specifically will help patients with complex and chronic diseases make sure that all information is getting to every doctor and practitioner involved in their care. This is particularly useful for caretakers of sick children or aging parents.



4) Design For Public Health and Research Use By Building De-Identification Functionality into The Exchange


This same data can and should be repurposed for Public Policy Health Care, tracking of illness, and demographics. Often, Public Health Data comes from some other place or project that was financed independently of the Exchange (as most states have some kind of Public Health repository but may not have a Health Information Exchange). In addition to spotting key health rends like flus, the de-identification of data can be used to find clinical trial subjects, a costly and expensive proposition for Universities, Pharmaceutical companies, and Contract Research Organizations, ultimately bringing down the cost of getting new drugs and procedures to market.

Sometimes news that doesn't make headlines really matters. This is one of those times. You can make a difference in your state as well. If all 50 States followed these simple, logical design principles, we'd all be better off.