|
Facilities thrive on control. They monitor the master patient index (MPI), review charts and audit databases--all to nip potential problems in the bud. Now, facilities must learn how to trust. Health information exchange (HIE) is imminent, and the move to make records more available also leaves data more vulnerable to contamination, duplication and mix-ups. Individual participants can prepare all they want, but bad data can snarl the exchange.
"There's still not a great amount of awareness that if the core data integrity is not really good, then you're going to spend a lot of money trying to make things exchange well and they won't," said Beth Just, MBA, RHIA, president and CEO of Just Associates, Aurora, CO.
Discussions about HIE often focus on the business or technical aspects, she said, but problems often lie in the data itself.
Patient Identification
Maintaining a clean MPI is a constant concern, but it will play an even greater role as hospitals and physician offices increasingly share data. Duplicate records and overlaps are common; on average, they affect about 2-7 percent of records at a mid-size facility, Just said. The error multiplies when HIE is factored in. "It gets very complicated when hospital A has a duplicate for a patient, and hospital B also does," she said. "Then you really have four records in the enterprise MPI for that person."
MPI problems stem partly from the lack of standardized patient identification numbers. Each facility has its own unique set, so HIE participants will need to map record numbers to allow data exchange--and that means having enough data fields to make those links.
"A lot of records and therefore transactions that would be sent into the HIE are missing critical pieces of information, such as Social Security number (SSN), patient's birthplace, mother's maiden name or other data points that in aggregation help to uniquely identify that patient," Just said.
When data is insufficient, employees struggle to match records to the correct patient and often resort to creating a new record for the information. But that introduces duplicates.
Facilities can avoid patient identification headaches by establishing a community-wide MPI, according to Glenn Keet, president, Axolotl, San Jose, CA. The database acts as a crosswalk between different sets of identification numbers, translating, for example, a hospital's record number into the correct ID at a physician's office, or vice-versa. Some HIE systems automatically recognize duplicate patient records as they are added to the network and will merge them into a single record.
MPI staff can help prevent patient identification issues by running analyses on databases. Just recommends looking at specific data points, not just duplicate rates, to determine the strength of data being sent out. "How many times do you have default information in the SSN or date of birth field?" she asked.
Privacy and Security Constraints
State regulations and patient consent pose another challenge to data quality in HIE. Privacy and security requirements vary by state; some block data regarding behavioral health, HIV and treatment centers. Information on minors may also be withheld, according to Merlene Rodgers, RHIA, vice president of Axolotl.
Some HIE networks also give patients the ability to opt out or permit only certain physicians to access their data. Patients may prefer to have the final say, but their decisions could affect the level of exchange, and physicians could be left wondering why they can't locate a chart. "Is the reason the patient isn't coming up in the search because the patient opted out, or the data wasn't entered as accurately as it could be?" Just noted.
To avoid such confusion, HIE participants should educate patients about their options, Keet said, especially when the default setting would block physician access.
Data Integrity
They may be confident in their own data, but hospitals and physicians participating in HIE will need to rely on each other, as well as the exchange system, to ensure records are accurate. Most hospitals share data with other hospitals on a view-only basis, according to Keet, but physician offices that accept information often integrate it into their EHR--and that's where data integrity is most at risk. When physicians order a lab, for example, the results must be reformatted and manipulated to match the practice's EHR import specifications. It's imperative that data translation doesn't omit or change any clinical information. "There is the concern by the sending hospital system that results are accurately represented in the target applications in the physician practice, which are outside the hospital's control," Keet noted.
Participants can take safeguards by following standards and guidelines, but when a single lab must make data compatible with 25 or so different EHR systems, the chains are even harder to keep from tangling.
No matter how clean the exchange, it's useless if the original data is flawed. Just advises hospitals and physicians to beware the weakest link. "You could have one participating organization in an HIE that could totally pollute the HIE database with a lot of data integrity problems, and the whole system is going to be blamed by the physicians for that particular problem," she said.
Hospitals that only view data from another facility, rather than integrating it, may have fewer data integrity issues, but they'll still have to worry about the validity of information. Debate over the legal health record includes whether providers should be held accountable for decisions made based on another facility's inaccurate data. As a precaution, HIE participants should do their due diligence, Just said.
Keet agreed. "There's always the threat of garbage in and garbage out," he said. "If the source system is bad, there's always the risk the outcome of that will be bad."
At the same time, hospitals and physicians shouldn't abandon an HIE because of another's poor data. For every bad egg, there are plenty of good ones offering high quality data. "You may be absent patient data from the weakest link, but that doesn't bring down the whole HIE, nor its value to providers," Keet explained.
Correcting Errors
In the event of an error, whether it's a patient mix-up or poor data leaking in to the HIE, there are steps a facility can take to halt a ripple effect. As soon as the error is identified, the facility should notify the HIE and send an amended copy of the record, merge details or whatever information is needed to correct the problem. Community-wide audit logs track who accessed the document, whether they printed it and if they forwarded it to anyone else, which can help identify facilities that need to be specifically notified about the error, Rodgers said. Plus, the logs help HIE participants meet accounting of disclosure regulations.
Hospitals and physicians can avoid problems in the first place by maintaining open communication. Data integrity committees should have ongoing meetings to keep everyone abreast of changes, such as when a facility alters its naming conventions. That's where HIM professionals can play a key role, Rodgers said. "The crux is to be involved," she said. "There's a lot of hospital activity that goes on with IT departments, and HIM professionals should ask questions and provide their expertise."
Cheryl McEvoy is an assistant editor with ADVANCE.
|