Staying on Track

When is a national electronic health record system like a railroad? When it faces the same challenges in ensuring all the pieces fit together. Nina Schwenk, VP for Integration and Chair of Mayo Clinic’s IT Committee, explains.

“We can create more problems and increase the health risks and safety issues because we can’t specifically identify individuals”
-Nina Schwenk, Mayo Clinic

The election of an administration with a strong focus on healthcare reform can mean only one thing for Nina Schwenk: a faster, more dynamic system of patient care. President Obama’s proposals to dramatically change the way in which health information is managed require the development of innovative technologies to support them, and as Chair of Mayo Clinic’s IT Committee, Schwenk will be one of hundreds of IT heads in hospitals and associations across the US attempting to implement the administration’s proposed establishment of a National Health Insurance Exchange.

But at what stage is the technology, and can it realistically cope with an expansion of information flow whilst still keeping patient health records secure? As one of the largest healthcare organizations in the US, Mayo Clinic cares for a high number of patients in several locations, and coordinating an increased flow of information could present challenges for Schwenk and her team.

Making tracks
For President Obama to successfully achieve a transformation of the current patient medical health records (MHR), reform would need to take place across the entire US healthcare infrastructure. Given the time and funding needed for such a project, is it even possible to implement something of this size, in the current economic climate, even with the delayed passage of the economic stimulus package? The withdrawal of Tom Daschle as Health Secretary certainly seems to have slowed down the process.

Schwenk advises that how exactly the infrastructure is to be developed is critical, and the creation of a uniform mandate of standards will be the key to successfully achieving a federal EMR system. Even with the correct implementation of procedures, she notes the challenges that still remain: “We can create more problems and increase the health risks and safety issues because we can’t specifically identify individuals, as we do not have a universal or unique patient identifier in the United States,” she says.

Identification by name only cannot make a successful infrastructure. Without the creation of determining, specific factors, such as individual patient numbers, there will be a problem of patient differentiation. “If we can’t electronically exchange data, we won’t be able to make sure it’s the right person’s data we’re sending; a unique patient identifier is a key factor that the government will have to introduce. There will also be the need for the creation of a broad incentive to ensure the electronic system or clinical information data exchange follows certain guidelines.

“It’s similar to when we put the railroads in; if there wasn’t a common gauge, we wouldn’t have been able to get the railroad track from one side of the country to the other. We need something similar in healthcare information transfer, because without the standardization, we’re going to be hampered.”

Patient data
Mayo Clinic is one of the few organizations that has unified patient medical records in its paper format. “We began this journey in 1907,” Schwenk explains. “If you were a patient seen at any of the sites at Mayo Clinic, for example if you were seen by a cardiologist in one of the hospitals, you were provided with a paper record in a plastic jacket that followed you around everywhere. Such a system doesn’t exist even now in many organizations, but we’ve had that for a long time. So, the tradition for information in a consolidated, integrated manner has been there for a century.”

With patient data at Mayo Clinic already operating at an organized level for a substantial length of time, the blueprint for data handling has been set. The technologies needed to consolidate patient information into digital form have been in the development process for a long time prior to the election of Obama and his subsequent healthcare reforms.

“We began the journey of consolidating and accessing data in an electronic format in the early 1990s, and in 2009 we find ourselves fully digital in the clinic and in the hospital. As a physician, I can see a patient in my office and have access immediately to all their medical data that’s collected in Mayo Clinic’s records via the computer. All their lab data, their surgical data, their radiology data, their radiology images; their past information all in sequential format. And this is available for every single patient for me now, both outpatient and inpatient.”

As Schwenk explains, the implementation of an electronic system is beneficial not only in terms of access to medical data, but also in terms of the consolidation of every single patient record. Mayo Clinic’s main site is located in Rochester, Minnesota; its precise situation is in a cornfield, described by Schwenk as “pretty much the middle of nowhere.” However, such a setting does not limit the number of those that are cared for by the Clinic or its influence within healthcare.

Mayo Clinic is the primary employer for the town, and expands its operations outside of Minnesota to care for patients across the US, and even across the globe. “The vast majority of our patients come from our five-state or regional areas, and we do have extra regional clinics we are affiliated with; it’s called the Mayo Health System, and we now have the ability to electronically share data about those patients with those clinical sites as well,” she explains.

“If a patient is seen in Austin, Minnesota, the Austin physician, their primary care doctor in their community, is able to see all of the data on that patient that may have been provided at Mayo Clinic as a tertiary center. Likewise, I can see that data when I see the patient at Mayo Clinic or any of the health system sites that we are connected with; I can see the data that’s collected there as well. Also, if a patient saw their primary care provider two weeks ago and had tests done, when I see them here two weeks later, I have access to all of that data, so repetition and inefficiency in terms of repetitive testing is unlikely to occur.”

Digital benefits
The challenge that the Mayo Health System faces is not necessarily due to limitations within the organization, but rather to incompatibility between systems on a national level. As Schwenk explains, if patients come to the clinic from outside of the Mayo Health System, the lack of a federal uniform system of patient records slows down the processing of data.

“There’s not going to be an automatic sharing of the data because each individual organization is not connected. It’s still very much a manual process. We may fax it, we may give it to them in a paper format, or we can put it on a CD and give it to the patient before they go to their local site. But it doesn’t automatically populate it; somebody has to read it, decipher it and then change the formats to meet those of the electronic medical records of wherever they’re going to be treated.”

The biggest challenge facing Mayo Clinic is the same challenge facing Johns Hopkins Hospital, UCLA Medical Center and all other high ranking, innovative clinics: without the interoperability or the standards to be automatically exchanging that information, humans have to be intermediaries.

Schwenk explains that the two problems standing in the way of organizing patient data are two very different problems that must be dealt with separately and consecutively: computerizing all health records and then the sharing of this data between institutions. “These are two very different goals. If all health records are computerized, then that’s an advantage to the site that computerized them, but there’s a long gap between computerizing data and then being able to share that data.

“For example, we have other sites within Mayo Clinic, in Florida and Arizona. Because they have different electronic medical records, we still have to use other means to exchange data. Even within our own system, the standards and the interoperability aren’t there to be automatically shifting data across those lines, and this is often because of the differentiation in products from the different vendors who build the electronic medical records. Even if you give a computerized record to every physician in the United States, it doesn’t mean they’re going to be able to talk to each other, and to me that’s one of the biggest challenges.”

Privacy concerns
Schwenk notes the high value placed on the protection of individual privacy in the US as another potential problem factor. “Individuals don’t want their information shared and feel privacy is a human right, and are fighting a form of specifying identification via a number. The government has therefore put it on the back burner and hasn’t addressed it, and has gone so far as saying they will not address it. Again, if you look at in terms of patient safety and healthcare, we’re going to need to readdress it to see how can we get there from here.”

The issue of privacy is always going to be intrinsically linked with patient fear of information safety: with more things becoming electronic, there are going to be concerns as to the likelihood of access to that information. “Within Mayo Clinic, we deal with security in many ways,” Schwenk says. “We have policies to keep a check on who has access to that data, and the appropriateness of that access. Also, if we’re transmitting data, we ensure that we’re using appropriate encryptions and that there is enough network security so that somebody can’t hijack the data. The exact storing of the data is also kept in check.

“Having said this, you can put in better security barriers to information theft in the electronic format than in the paper format. But unfortunately, if they break down in the paper format, you’ve lost one person’s record; in the electronic one, once you get in you can steal multiple people’s records. The risk is greater, although the security measures are probably tighter around the electronic data than they ever were in the paper data.

“Monitoring levels of security remains most important: tracking data, knowing what data is being viewed by others. Electronically we can do that, in the paper records we couldn’t. Stringent policies are also needed to enforce it if there are breakages, as well as audit functions to make sure that we’re keeping on top of that. So there are two issues regarding privacy: one is hijacking or theft of the data, where technical security and policies need to be in place, and the second is confidentiality.

“The public is also concerned about the ability of insurance companies to access patient health records. What’s in my record makes a big difference. If I do a genetic profile and find out I have a huge risk of cancer in my forties, who will insure me? Or, what if I find I already have HIV in my blood test result, who will insure me? I think that becomes a real valid point, and certainly then becomes an employability issue if my employer has access to that data. Those are two other privacy issues that are going to need to be addressed by legislation as opposed to technical security.”

Personalized medicine
In his healthcare proposals, Obama has also pledged to promote public healthcare by requiring an increase in coverage of preventative services, such as cancer screenings. But, what will be the effect of personalized medicine on EHRs? Schwenk advises that Mayo Clinic has been in the business of delivering individualized healthcare since the last century. “When you saw a patient, you took their life circumstances into consideration, you looked at their financial situation to assess what they could afford in terms of healthcare, and you looked at what their specific disease pattern or multiplicity of diseases was. Then you formulated a treatment or a management plan.

“The difference today is that we cannot continue in that paradigm, it’s simply not scalable anymore because of the volume of information. The amount of data is so huge that I don’t think any one person or physician or brain can synthesize all the information that’s available and bring it to bear for that patient; this is the benefit of EHR. How do we understand or synthesize genomic data, which is incredibly complex? We need algorithms to even understand what the meaning is, and then how does that apply to this particular patient?

“Clinical decision support is very important. From viewing all of the patient’s data in electronic format, as a care provider, I receive alerts regarding drugs and dosages, according to the patient’s medical background. That’s the kind of individualized medicine that will help in the electronic era, so that’s one form of patient specific care.

“The other type is that once we start collecting this data in an integrated manner, we can go back and query it, viewing the treatment given to other patients within Mayo Clinic who were administered under the same disease. What were they given, and what were their outcomes? From comparing one patient’s genetic profile versus another’s, I can use that information to tailor the treatment for the patient in front of me. For that we need the capability of analyzing data in real time.”

It remains to be seen how the new administration will take those first steps to begin the transformation of the healthcare infrastructure, and whether the idea of computerizing all health records within five years is achievable. For Schwenk, it depends on how much money is invested.

“Dollars will get us there, but it’s also how the process flows and the buy-in. A two-physician clinic may ask why they should to put the money in to computerize their records. Well, here’s the reason: What does that EMR actually do? Is it just collecting patient data, or is it helping you manage and solve clinical decision problems? This is the next step, and I believe that is Obama’s agenda. Once the information is in a digital format you can start using clinical decision support, as reminders and alerts for other physicians.

“There are all of those pieces; that’s where the value will be. The computerizing, we can probably get there by brute force, by giving dollars to go ahead and do it, but the real value will be in improving the quality of the care, and just by merely putting computers in, we’re not going to solve the problem.”

Just like with the building of the national railroads, it may take some time to get there, but the results will change our country immeasurably for the better.

Nina Schwenk is VP for Integration and Chair of Mayo Clinic’s IT Committee.