Vol. 20 • Issue 1 • Page 12
Call it Murphy’s Law of EHRs: facilities plan for emergency downtime, but on the rare occasion, things still go wrong. Staff scrambles for paper while IT personnel reboot and reload-and it’s all at the worst time. “When you need it most and have the most users on the system, that’s when it’s going to happen,” said Christine Rys, RHIA, HIM consultant for Cardone Record Services and Midwest Medical Records Association.
That may sound like an exaggeration, but Fletcher Allen Health Care in Burlington, VT, faced the unfortunate reality last August, when a tree fell onto a major city power line, causing a surge. In another twist of fate, the uninterrupted power supply had two dead batteries, so it failed to block the surge. The EHR was down for 8 hours, forcing staff to document on paper. When the system was restored, they had some cleaning up to do.
Recovery comes with a sigh of relief, but problems continue long after the system’s back up. Here, ADVANCE offers tips on how to get floating papers, disparate records and other remnants of EHR downtime in check.
The Downtime Procedure:
Reverting to Paper
The Aftermath:
Temporary forms need to be merged with the permanent EHR.
How to deal:
When paper records are created during a downtime, facilities have two options for adding it to the EHR-entering data by hand or scanning it. When deciding which method to use, consider the end user, advised Scot Silverstein, MD, a medical informatics consultant and adjunct professor of health care informatics and IT at the College of Information Science and Technology, Drexel University, Philadelphia. If the ultimate goal is to help physicians care for patients, then scanned images could suffice. If in-house scanning is available, it’s also faster to scan than manually enter data into the EHR. “Scanning a boatload of documents is generally a very rapid thing,” Dr. Silverstein said. “The important thing is to make sure they get into the right patient file.”
Scanning also tends to be cost-effective, Rys added.
On the other hand, if coding and charge capture are of greater importance, facilities may want to take the time to manually enter records, especially if coders aren’t familiar with the paper forms. “The problem with that is you run into the risk of inaccurate transcription,” Dr. Silverstein cautioned. So physicians would need to check any transcribed forms to ensure accuracy and protect against bad billing or litigation issues.
Fletcher Allen Health Care did a combination of scanning and backloading. Ongoing and future orders were manually entered into the EHR, along with the patient’s most recent vital signs. “We really set it to be the things that for the next 8 hours somebody caring for the patient would need to see,” said Sandra Dalton, RN, chief nursing officer, Fletcher Allen Health Care.
Completed orders and less pressing records were held in a paper file and scanned at discharge. In the end, whether scanned or entered, all records eventually made it into the EHR.
The biggest challenge, Dalton said, was managing the additional workload. The system was down from 9 a.m. to 5 p.m. at a facility that sees about 350 inpatients per day. That added up to a lot of paperwork. “People in many departments worked extra [hours] to recover,” she said, “It was probably 11 p.m. before we were totally recovered.”
Even after those paper records are integrated, facilities should still expect to keep them on hand-and that means finding a place to store them. “What we find is our folks usually keep their documents 12 months after [scanning],” said John King, chief operating officer, EvriChart, White Sulphur Springs, WV. “That usually gives them time for any requests or pulls to come around where they might discover a mistake.”
The Downtime Procedure:
Using Ancillary or Redundant Systems
The Aftermath:
Coding and registration may have kept up during the downtime, but now their records don’t match
those in the EHR.
How to deal:
“If you have a lot of interfaces, that’s where the trick comes in,” said Chuck Podesta, senior vice president and chief information officer at Fletcher Allen.
The hospital’s registration and billing are on different systems than their EHR. It let them continue business as usual while IT staff worked on the EHR, but when the EHR came up, it hadn’t logged any of that activity. Clinicians had to wait while the interfaces were turned on and information from ancillary systems was replayed into the EHR-a process that took hours. “The interface engine could only handle so many transactions per minute, so you basically had to sit and wait,” Podesta recalled.
He advises facilities to turn interfaces on “as early as possible” following a downtime.
When recovering from a downtime, consistency is crucial, Dr. Silverstein said. The legal health record becomes more defined, and facilities will be held accountable for the information housed in EHRs and other systems. “You really want to make sure there’s not going to be discrepancies in electronic versions of the accounting of what happened,” Dr. Silverstein cautioned.
When the EHR is restored, don’t assume data will fall into place. Check the integrity of information entered immediately before the downtime and immediately after the system is back up, ensuring records are sent to the right location and arrive there intact.
The same goes for backup or redundant systems, Dr. Silverstein added. They’re helpful tools that keep staff from being totally in the dark, but they should be used on a view-only basis. Never attempt to alter records in a backup system, or you could wind up in an unfavorable situation under litigation. “There are too many chances to go astray,” Dr. Silverstein said.
What you should check, however, is that the feed to those backup systems works once the main EHR is up-and-running again.
The Downtime Procedure:
Communicating With Staff
The Aftermath:
Employees were told about the downtime; now you need to explain when it’s safe to use the system and how to prevent a repeat incident.
How to deal:
When Fletcher Allen’s EHR came back up, Dalton and Podesta were careful about when to let each department access the system. Pharmacists got the green light first, which gave them a chance to enter paper orders so physicians would have an accurate account of medications for their patients. “We could have allowed people in there sooner, and they would have loved to be,” explained Dalton. “But the decision we made was if the record isn’t totally updated with everything that has been entered in related systems-such as lab or radiology-in the last 4 or 5 hours, we don’t want people in the EHR because they’re not going to know what’s missing,”
Once all systems were go, the facility used e-mails and overhead paging to inform employees when they could log on. EHR support staff also made rounds. “They paired up and went to each unit, made sure everybody was signed on and there weren’t any hardware or printing issues,” Podesta said.
In the days following the downtime, Dalton and Podesta also gathered about 40 representatives from clinical, technical and facility staff to conduct a root-cause analysis of the incident. It was an important part of recovery, and one that stemmed new policies and procedures-including a new communications plan that ties disaster recovery/business continuity to the organizational incident command center. “We purposely went through a process to learn [lessons] while it was fresh in everybody’s minds,” Dalton said.
It All Depends
Downtime experiences are unique and can differ among facilities, from the records affected to the point-person in charge, which makes recovery even more difficult to plan. Scanning or backloading data, verifying interfaces and communicating with staff all depend on the systems each facility has in place. The best way to prepare, everyone agreed, is to talk with vendors, define procedures and hope to come out on top. “The best strategy,” Dr. Silverstein said, “is to not let [downtime] happen-as much as humanly possible.”
Cheryl McEvoy is an assistant editor with ADVANCE.
