Part one of a four-part series
ADVANCE: How would you characterize the progress made by health care organizations in electronically linking records across systems and with other organizations?
Just: Hospitals and IDNs have increased their recognition that patient-identity data needs to have high integrity. We don't have to convince customers that they have to have a clean MPI. Very rarely do we have to go down that path.
Where I think there is still a dearth of knowledge is in understanding how complicated the electronic data integration is within organizations. We see that most of the system integration occurring in most organizations is developed by an interface engineer and/or a systems analyst -- not necessarily someone who is the business process owner of that information and who has a broad enough perspective on it. Clinical departments frequently have their own departmental systems and they don't have a big-picture understanding as to how their system connects to other systems in the organization. At the same time, engineers don't understand what the data being interfaced means. So we see a big gap in the interface arena between clinical departments and the interface engineer.
For instance, a nurse in cardiology understands what her echocardiogram database is supposed to do. But she is not aware of what registration goes through or what HIM goes through, or what billing has to do, or how the clinical information from her system actually gets into an electronic medical record. And she doesn't realize the legal ramifications that might occur if her database gets out of synch with the electronic medical record.
ADVANCE: What are some of the commonalities you see among health care organizations that appear to be on the right track in their efforts to link records?
Just: There are a couple things that I've noticed because we've had the pleasure -- particularly over the past two or three years -- of working with organizations that I think are clearly in the lead.
One is that they have a multidisciplinary team approach. They recognize that it's not just an IT project. It's not just a nursing project. It's not just an HIM project.
I also see higher degrees of success when there's a good relationship between HIM and IT. When the HIM departments are involved, they tend to bring in other domains such as patient access or billing or nursing. IT will bring in the obvious people, such as the specific department that is a direct user of the system. If it's a clinical documentation project, then they sit down with nursing. But they don't necessarily think to bring in other domains that might get impacted like pharmacy or lab or HIM.
ADVANCE: Now we'll go to the opposite end of the spectrum with organizations that appear to be lagging. What could they be doing today to get on the right track, similarly to what the leading organizations are doing?
Just: Let's look at it from the perspective that you want to electronically link records across multiple systems. What do you need to do?
You start by doing an electronic analysis, using a sophisticated algorithm -- of which there are many on the market today -- of the master system. That's typically the hospital registration system or the clinic registration system. One of the first key steps is to analyze the system to see what types of data-integrity challenges you've got in the database. Otherwise, you don't know what you don't know. A good data analysis will tell you a lot about the quality and integrity of key pieces of data, not just a duplicate rate. We go in and look at the process to find out how patient data gets captured. Initially, we may find that the only people who can create a new medical record number, for instance, are registration people. But when we look deeper, we find lots of little pockets. "Oh, except for in the lab; they register specimens from Dr. Smith's office," someone tells us. "They do specimen registration themselves."
ADVANCE: It makes you wonder how often that happens.
Just: Exactly. And then you go talk to that department and you find out more individual exceptions. My favorite one occurs in the lab when specimens are collected from nursing homes. They get registered in the main registration system because that's the only way the organization can get billed. The lab staff don't think they're registering a "patient." But of course, the patient's name is entered into the registration system and the lab results are stored in the electronic record. I've had customers tell me that when they have someone who is going to donate blood or an organ, they register the donor, but that person doesn't get a medical record number. They call it a donor number, but it is a medical record number as far as the IT system is concerned.
So there's sometimes a lack of understanding of how all these systems really collect and store data and how they interoperate with other systems in the facility. The processes by which the systems are used are more critical than the specific system they're using.
ADVANCE: Health care organizations are so wrapped up in their day-to-day activities that they naturally don't have the perspective of someone coming in from the outside to look at processes and see where there might be holes.where there might be data being introduced that they hadn't thought about.
Just: That's true.
ADVANCE: Now let's look ahead to the future and talk about how technology will play into solutions for linking records.
Just: We will continue to see improved record-linking algorithms that will probably be incorporated into more of the clinical or departmental types of systems rather than in, for instance, a main registration area. So let's say you've got a registration system that is the master, and it is sending ADT transactions to the RIS or to a LIS, which is a very common scenario. The effectiveness of that record-linking is only going to be as good as the weakest system from the record-linking standpoint. So if the LIS is not very sophisticated in its interfacing program, then it doesn't matter what you have on the front end. For example, let's say the registration system is able to send an A34 (merge patient) transaction, but the lab system can't process it. The facility will still have to manually correct the lab system every time a duplicate record merge is performed in the registration system. Also, a lot of these systems don't keep up with the standards. For example, HL7 has released Version 3, but I don't know anyone who actually uses it. There are a lot of systems out there using variations of 2.1, 2.2or 2.3. This makes data integration very challenging.
Clinical and administrative business systems that are being interfaced with each other all have a little interface engine program inside them. They have to; otherwise, they couldn't receive an inbound transaction.
So I think that we'll probably see better technologies related to that.
We'll also see better analysis capabilities for business owners to actually help in the testing and the troubleshooting of interfaces in their systems and with other systems. Using cardiology as an example, for nurses to really participate effectively in the interface testing of ADT transactions or order transactions of results back to the main EHR system, they almost have to become HL7 experts. They are the only ones who will understand if that data point is in the right field in the message or whether the right translation of that value is occurring, or whether a transaction should be filtered coming into their system. That's not realistic.
ADVANCE: It doesn't sound like it.
Just: They depend on the engineer to do his thing, but the engineers don't understand the data. I think you'll see better audit systems, for lack of a better term, to help put the interface transaction into something that a user can read.
ADVANCE: Who would be looking at the audit system?
Just: HIM is really in perfect position to do that because of the cross-training that we have in health information. The HIM person is not a physical therapist, for example, but he or she knows what a PT has to document in a record. The HIM person has a medical terminology background so he or she can make sense out of a lot of the data sent in these messages. The audit system will be used as a tool by the HIM data-quality team or data-integrity team.
ADVANCE: Outside of strict technology, what other factors would be in play as organizations strive to do a better job of linking records across systems and organizations?
Just: Think about the globalization of health care services. We have people who go to India for open-heart surgery. There's a lot of outsourcing of various functions to other countries.
I think globalization is going to have a huge impact on how a lot of this stuff plays out.
From a tactical standpoint, I look for a transition in HIM, moving our staff and our professionals to more of the data-integrity, data-quality, interoperability support type of role, rather than being file clerks who sort papers and put them into paper charts. We know there's still a phenomenal amount of that going on out there. But I think there will be a transition within the HIM profession to move us more into that data-quality type of position.
Lastly, the staff training and QA processes in place to ensure high-quality data collection and transmission is occurring and is of paramount importance. Each organization needs to get the most out of the electronic record system in which they've invested and paying attention to the people and processes part of the implementation is critical -- as well as continuous monitoring and improvement processes.
ADVANCE: If health care organizations are able to make this transition of moving staff to more of a data-integrity role, how will that benefit the organization?
Just: I'm a firm believer in striving to do things right the first time. Certainly in the provision of health care, it can be deadly if it's not done the right way the first time. The more we can get information to providers and clinicians in a way that is accurate and timely -- and in a form that fits in their workflow -- all of that is going to improve health care.
ADVANCE: Thank you, Beth, for taking time to answer our questions. We look forward to the next installment of our Q&A series, which will appear in this position of the ADVANCE Web site in February 2009.
Beth Just can be reached at firstname.lastname@example.org; www.justassociates.com.