Computerizing health records

Health reform. Technological innovation. An increasing healthcare population. The world’s brightest minds applied to creativity. Add them all together and divide by limited funds, and you have yourself the current state of affairs within the majority of hospitals and healthcare organizations across the US. And with President Obama confirming that all US health records are indeed to become computerized, potentially within the next four years, CIOs and chief technology officers are frantically scouting the market for technology that can bring them up to scratch with where they need to be.

One of the people who has been doing precisely that is Kevin Williams, former CIO and Vice President of Information Systems for Triumph Healthcare in Houston, Texas. With an attitude that never fails to inspire and motivate, Williams understands the reality of converting an endless stack of paper into what is becoming commonly known as the electronic medical record (EMR). However, with so much inherent change, is the timescale truly viable?

"It's close to being feasible," offers Williams, "but it's wrought with some complexities that did not make their way into the legislative thinking. This would have been a great idea 25 or 30 years ago when the population was not quite as mobile as it is now. For example, if your hospital is up in Boston and has a nice EMR, but you get hurt in Dallas, you've got to go to a hospital there - so that nice EMR isn't going to do anything for the patient. Perhaps a competing idea would have been something like what I call a 'personal medical record', whereby the individual is the literal owner of the information.

"When we go to a hospital or to see a physician, we allow them to write test results and diagnoses into that record. That way, the next time we go somewhere we can grant another physician or hospital access to that information. That satisfies a couple of requirements. The first is that it gives us a true EMR; the second is that it allows it to become more logical and less physical. If you look at the way that it's proposed under health information technology (HIT) standards, the record will be physically housed at a particular hospital - but if the patient isn't there then it doesn't do any good."

Oversights

While other companies and industry experts are already half way down the acceptance road with EMRs, Williams' alternate perspective sheds new light on the matter, offering logical solutions that would undoubtedly solve more problems in the long term. However, the 'rules of reform' have already been cast, leaving little room for maneuverability. The main problem Williams envisions with trying to tackle the issue in such a strict timescale is that technology is moving with such speed that it is impossible to predict what will be available on the market within the next 18 months, let alone the next four years; it seems as though the assimilation of technology was missed by the sweeping hands of health reform.

"To give physicians, nurses and their providers the ability to assimilate that kind of information and technology in a country as large as the US is not feasible within the next five years," asserts Williams. "If you look deep into the legislation, EMR funding is stated as being for hospitals between 100 and 500 beds. That would get you Middle America, but it's like doing any kind of computer program application - the last 20 percent of the job is going to take 80 percent of the time."

Another kink in the health reform super-plan is that, in the grand scheme of things, Williams believes that far too much emphasis has been placed on EMRs. "My main disagreement with it," continues Williams, "is that there are too many players that have their own economic needs to take into consideration, so there are problems in two areas - the public health system and the individuals. You have to remember that we're obviously still a capitalistic system, so it's difficult in the role of a CIO, CMIO or VP of information systems to go to the management and say, 'We need to spend $5 million and it will benefit the public health sector'."

For Williams and his former organization - a long-term acute care (LTAC) and rehabilitation hospital - the biggest obstacle attached to this point is trying to find vendors that can satisfy a large majority of their specialist needs. Unlike areas such as OBGYN, where predictions can be made within certain parameters, LTAC deals with patients who have multiple diagnoses, so diagnostics are irrelevant, as the patient's problem is already understood. The job of those in LTAC is to ultimately get the patient back to their former glory.

"The issue that you run into is that, let's say a patient broke his leg and was in an acute care hospital and then went for physical therapy afterwards that involved some minor movement. If that patient also has some cardiac conditions and perhaps some respiratory ones as well, you can't say, 'We want you to go over here and do a mile on the treadmill,' because it's just not going to happen. You have to build those interdisciplinary teams and there isn't any one HIT vendor out there that is willing to take that approach.

"Part of the reason is that that specialized aspect of the healthcare continuum of care, where you're going to have really intensive care, doesn't represent enough money to make it worthwhile for some vendor to go out and take the risk and develop a product. They'd have to sell it to everybody in the market, and that's not going to happen."

Security

On the subject of risk, security has taken a leading role in recent discussions on EMR, and indeed is on the lips of CIOs nationwide. People want to be able to access data from a mobility standpoint; the bottom line remains that they want to be mobile, and they have to be mobile. However, nurses, physicians and healthcare staff in general have a very different relationship with technology than an accountant or administrative member might have.

"Rightfully so," affirms Williams. "Whenever there's need for some intervention, a clinician's going to drop everything. It's all about the patient at that point in time, so when they come back to what they were doing, they'll be thinking, 'Where did I put my iPhone?' The other thing, which doesn't come out as much, would be something like someone leaving a laptop in public; it's completely innocent, but potentially dangerous in the wrong hands.

"I'll give you an example that we've run into before. It's not unique to LTAC, but it gets exacerbated there. A patient comes into a hospital and fills out all their admission paperwork. Once we get all that in, we know we've got their work history, social security number, address and next of kin. Basically, what we have now is everything you need to apply for a credit card. You've got to make sure that that information doesn't fall into the wrong hands. One thing that's also unique about LTAC, and is a big potential problem, is that our patients are in hospital for 35 to 45 days. Those people are not going to be going to their mailbox any time soon, so all you need to do is get your hands on some of that information and go to apply for a credit card.

"The other piece of the puzzle is when you look at the country and what I refer to as 'third-tier cities' - places like Lima, Ohio and Mansfield, Ohio - that have maybe one or two hospitals, everybody know everybody. How do you keep the private aspects of someone's illness secure?" As it turns out, Williams only hits the tip of the iceberg when it comes to dealing with security issues; merely having the data is a potential security risk because as soon as it is released - through being printed or otherwise - it negates its role as secure content.

From this, Williams cites one of the biggest challenges faced by his former company as being the speed at which it could get information into the hands of clinicians - not an easy task when you have hospitals spread across different time zones with everybody needing information as early in the morning as possible. What then happens is the window to access and obtain the necessary information from the computer narrows very quickly. Understanding how to keep a system running 24 hours a day while releasing secure information without driving your staff into the ground is therefore pivotal. The answer? Getting the "talent" as Williams refers to it.

"It seems that you can get IT people, but IT people don't generally relate well to the users. To avoid this, we used a technique that, for lack of a better term, I'll call 'selective handholding'. I had an IT group that took care of most of the hardware, networking and security. I took care of the information systems people and paired them up with the users based on some common experiences that they had. I had one person that understood the intake process, so he worked with the intake Vice President of Marketing on referrals, as he was a programmer and a software guy.

"I had another software guy that took care of the accounting side of the house, as he could talk to the accountants and relate to them. I worked with the operators because I understood the operations of the hospital, so I worked with the CEO, COO, operational vice presidents and some of the physicians. It was then down to us, as information systems people, to talk to the IT people; otherwise you'd end up with the physicians and everybody else being frustrated. This was my idea, but I must admit that it was borrowed from other industries. For one, I noticed how Black & Decker had to rebuild itself to get back into some market share it had lost. One of the things I remember hearing the CEO say was that nobody wants to buy a drill - they just want a hole in the wall.

"When you start getting physicians and members of medical executive committees thinking like that - 'Did you want a coffee maker or did you want a cup of coffee?' - then you can offer them a better understanding of the situation relative to IT systems. We'd use the information systems people to talk to others about the cup of coffee. Then, we'd go and talk with the IT people about how hot the water should be, to continue the analogy. Essentially, we could discuss stuff that the end-users aren't interested in."

Williams also took a concept of three main principles that he wanted all information systems and IT people to understand fully. The first ensures that the automation process is fulfilled; in doing so, you have to know whether the process makes any sense, because it's easy to take a computer and "do something stupid faster", as Williams puts it. The second considers the time value of information; knowledge or information known and data disseminated at any given time is pivotally important to comprehend. The final principle concerns itself with the velocity of information. If a business' information is flowing slowly, then good decisions can't be made quickly; you have to match the decision-making process up with the velocity of that information.

The human element

For all the downfalls that Williams points out, if these three principles can become standardized, then the future of EMRs could be better than first predicted. "One of the things that an EMR will do that is very difficult to do in the situation as it exists right now is with the data - you can then look at the trends. That's one of the problems we have right now. When a patient comes in, we can sit down. We can do an EKG. We could run lab and blood tests to find out what the patient's condition is right now. But if you had an EMR where you had some historical pieces, you could check if there were any trends: How has their blood pressure been over the years, for example.

"On the flipside of that, one of the biggest potential dangers will be where people will try to make predictions and forecasts based on historical data. If you see someone's blood pressure going up, then quite possibly it could be that you may have some cardiac issues. Alternatively if you don't know enough about how that person lives, you could also just have someone who has been in a stressful situation, so that forecast wouldn't be valid." The biggest fear continues to be that, with all this accumulated data, an EMR could end up containing more personal information than patients even consciously know about themselves.

"Those things are going to come up three or four years down the road, as we start to implement these things," adds Williams. "The other thing that concerns me is that we'll start to raise a whole generation under these circumstances, so we've got their information all the way from birth. But what happens if a child's parents divorce? The mother wants things one way, the father another, and you've got all this historical data. What happens is that you start to get a lot of other mixed opinions into the data, and that can create problems.

"We see that a lot in on the LTAC side already, not so much with children, but more with the geriatric population. Then, for security purposes, we get into a child being 'granted custody' to one parent. Who now has custody of that child's medical record? There's plenty of little legal things that that are bound to emerge. A lot of the time marriages break up due to philosophical differences - when you throw a medical record into the mix, you're sure to end up with plenty of court settlements."

This rather frank and truthful perspective of Williams' leans towards a preliminary conclusion of how EMRs could be handled in the following months and years; for all the computer systems and technology in the world, the fundamental fact remains that the world of healthcare thrives on a balance of human compassion and knowledge. Of course EMRs can provide completely comprehensive and historical data with unmatched speed, but their success will come down to how that information is managed and secured - and ultimately how the human element is protected within an ever-growing network of computers.

Biography

Kevin Williams was formerly Vice President and CIO for Information Systems at Triumph Healthcare.

Triumph Healthcare

Based in Houston, Texas, with locations spanning the length and breadth of the country, Triumph Healthcare develops and operates long term acute care (LTAC) hospitals and specializes in providing services to medically complex patients for whom short term care hospitals are no longer appropriate. Patient stays in an LTAC hospital average over 25 days, however the length of stay can fluctuate dramatically depending on the patients individual plan of care. While Triumph Healthcare is a freestanding LTAC, others throughout the country operate as a "hospital within a hospital".

As LTAC hospitals focus on treating critically ill patients and those with complex medical conditions, the introduction of EMRs will have a specific and significant impact on LATCs as patient history undoubtedly plays a far greater role within the rehabilitation process; while the exact number of patients across Triumph Hospitals is unknown, it is predicted that EMRs will affect upwards of 50,000 patients per year.

LATC patients are almost always referred by physicians or medical caregivers and are admitted from short-term acute care hospitals once they have been identified for an extended hospital stay. The problem with allowing these types of people to remain in the general hospital setting is that they are set up to provide episodic care, so moving them to an LTAC facility becomes cost-effective while providing higher quality treatment and care. It is here that EMRs will have their greatest potential effect and ultimately where they will be tested the most.