Home  |   Subscribe  |   Resources  |   Reprints  |   Writers' Guidelines

November 9, 2009

Dealing With Downtime — How to Survive If Your EHR System Fails
By Lindsey Getz
For The Record
Vol. 21 No. 21 P. 16

When a tree fell on an electrical line and Fletcher Allen Health Care suffered a power failure this past August, there wasn’t initial cause for concern. After all, the Burlington, Vt.-based facility’s new $57 million EHR system had an uninterrupted power supply (UPS) system to provide backup. However, a few bad batteries made the UPS system useless, causing the EHR system to fail and forcing employees to put the facility’s “unplanned downtime plan” into effect. While the system was functioning again by noon, it was 5 pm before updates were verified and staff were allowed to access the system.

In June, Lawton Indian Hospital in Oklahoma suffered a similar incident when a power failure caused by wind damage denied staff access to the valuable information contained in its EHRs. As a result, the hospital was forced to resort to its old filing system.

These types of unanticipated situations are always a possibility and are the reason why it’s so critical to be prepared for downtime. First and foremost, says Ron Sterling, president of the consulting firm Sterling Solutions and author of Keys to EMR Success, it’s important to do what you can to avoid unplanned downtime in the first place. “The occurrence of unscheduled downtime can be disastrous for a patient and the facility,” he says. “The best planning for downtime includes mitigation strategies for eliminating or minimizing the possibility of such an event even occurring. Avoiding downtime means that backup and redundancies should be built into the various components of the system.”

Of course, unexpected events such as storm-related power failures can make downtime unavoidable. That’s why it’s important to plan ahead. Facilities should start by composing a list of systems and departments that would be impacted by downtime, says Kelly McLendon, RHIA, president of Health Information Xperts. “All computer systems need to be inventoried and evaluated for their impact in various areas,” he says. “That’s the first step. Once the HIM department has done that, they can put the appropriate policies and procedures into place to minimize that downtime.”

Debra M. Wolf, PhD, MSN, BSN, RN, an associate professor in the nursing department at Pennsylvania’s Slippery Rock University, adds that it’s also important to have a clear definition of downtime that can even be split into levels of significance—each of which should be independently defined. “You may have a level 1 downtime, level 2 downtime, and level 3 downtime,” she says. “A level 1 downtime might mean that part of the system is down, but the majority of the content is still accessible. Level 2 might be that the majority of the content is unavailable. And level 3 could mean that you have absolutely no access to the system. Once you define these levels, it’s important to create your response plan to each.”

Creating a Downtime Policy
A response plan requires creating a downtime policy that employees can follow in the event of both planned and unplanned downtimes. This policy should include clear instructions for continuing operations, as well as a recovery plan, says Sterling. “For scheduled downtime, there should be notification strategies and time frames spelled out within the policy,” he says. “The policy should include notification priorities and a notification confirmation of key facility management and staff. These notification strategies should employ e-mail, signage, and coordination with facility management to ensure that patients are covered and protected.”

A policy for unscheduled downtime should feature complete documentation of the recovery strategy and the resources available during the outage. “It should include prioritization of services and escalation strategies to deal with the various problems that may result from downtime,” says Sterling. “Key contacts and backup contacts for every relevant vendor, as well as management contacts, should be documented and verified on a periodic basis.”

At Fletcher Allen Health Care, the downtime policy involves reverting to the use of paper records. “This went smoothly because our staff still remembered how to document and write orders on paper, as it hadn’t been that long since we’d gone electronic,” says Sandra Dalton, senior vice president of patient care services and chief nursing officer. “The place that saw the most interruption was the operating room, and we spent extra time talking through each patient situation to make sure the doctors had all the information they needed to care for those patients.”

However, Wolf warns that as EHRs become more prevalent, familiarity with paper records will decrease. “What we’re finding is that a lot of young nurses are now training on EHRs and going directly to employing them, so they are not used to paper documentation,” she says. “If the downtime policy requires going back to paper records and that method doesn’t mirror how it was done on the EHR system, then they’re going to be lost. That’s why it’s important to be constantly refreshing staff on this method.”

What happens once the EHR is back up should also be part of an effective downtime policy. Prioritize processes and develop time frames for recovery, says Sterling, adding that the exact recovery options will depend on what record-keeping method was utilized during the downtime, what systems were operating, and what recovery resources are available at the facility. “In some cases, the paper records that were kept would be scanned into the EHR,” he notes. “In other situations, the EHR data accumulated during the failure period will have to be applied to the recovered production system. For example, a freestanding lab system may be operating during the EHR failure. After the EHR is back online, the lab tests performed and recorded during the failure will have to be added to the EHR.”

McLendon adds that some facilities have recovery plans that require paper records to be input manually. In such a plan, it’s imperative to designate who will be responsible for rekeying the information and in what time frame they are required to do so. What to do with the paper records once they are entered into the system must also be taken into account.

Training and Communication
Educating staff on how to handle EHR downtime will help ensure that operations run smoothly. “Most HIM departments have meetings on a regular basis,” says McLendon. “I suggest this type of training be a part of those meetings. You can create downtime exercises that staff can work on to make sure they are up-to-date on how to handle the situation. Downtime training can also be a part of any new employee orientation practice.”

Staff training should increase awareness of the downtime policy and define how to implement and access it, notes Bonnie B. Anton, RN, MN, electronic order set coordinator at the University of Pittsburgh Medical Center St. Margaret. This will prevent confusion when an event does occur. “Everyone should know their role during a downtime,” she emphasizes.

Considering that EHR downtime will affect every department, Wolf suggests that facilities form an interdisciplinary team. “This team should include people from IT, nursing, pharmacy, the physician group, and others,” she says. “Any department that is affected by downtime needs to have a representative on a downtime committee so that clear communication and coordination plans can be made. Each department cannot act on its own without being aware of how other departments are handling the situation. This team should meet on a regular basis to review the downtime policy.”

In the event of either planned or unplanned downtime, communication must be swift and properly targeted. It’s important that every institution determine the method that’s most effective in notifying staff, says Anton. Communicating planned downtime is easier, considering it can be done with plenty of advanced notice and while all power is still functioning. “This can be done through an e-mail that is sent to all staff several days prior to downtime and then repeatedly sent until the actual downtime occurs,” suggests Anton. “Or the message could even be communicated on a screensaver or as the opening screen in the EHR.”

For unplanned downtime, the electronic record’s help desk has an important role, according to Anton. “As soon as the help desk receives notice of issues related to the functioning of the EHR, they should follow specific procedures for notifying the appropriate individuals,” she says. “These individuals should each have a specific, defined role in both troubleshooting the problem as well as notifying staff of the EHR status.”

McLendon suggests that having “trees” established in advance, where a lead person for each department is notified and then responsible for notifying other key members, who in turn spread the word throughout the department, is an old-fashioned but effective method. “These days, so much is done electronically, through the computer system. But in the event of total power loss, that’s not always an option,” he says. “One hospital I work with uses a voice system that key members wear around their necks. Downtimes could be communicated this way.”

Dalton says that one of the biggest lessons learned when Fletcher Allen’s system went down was that a better communication method was necessary. “Because of the outage, our paging system and other methods of communication went down for a while,” she says. “It has made us rethink the way we would communicate in a future event like this.”

Although Fletcher Allen still had use of e-mail during the downtime, it proved to be a less-than-effective way of spreading the word—staff members were too busy caring for patients to have time to check their inboxes. “We decided in a case like this, where the pagers are also down, the best method of communication is the overhead system,” says Dalton. “We’re also talking about using little pop-up messages that can come up on all the computers. If these came up in all the departments, the word could also be spread that way. I think part of the solution is having multiple ways of communicating during downtime, not just relying on one method. That way you can be confident the message is getting out to everyone.”

Advanced planning and good communication will ensure that staff know how to react if downtime were to occur, as it did at Fletcher Allen. It may even prevent the situation in the first place. “Good, clear communication is necessary to prevent chaos and may even avert the problem,” says Wolf. “It’s important that staff becomes more comfortable with communicating EHR issues they’re experiencing, however minor they seem. For example, as soon as you start to experience any flaw in the system, it should be reported. Staff may think that the system seems slow one day, but not inform anyone. When the system goes down, all the signs were there, but nobody reported them.”

Be Prepared
The biggest fear tied to downtime is that chaos will ensue. In the end, the reaction of staff will impact how smoothly operations continue to run. Fletcher Allen’s staff steered clear of calamity by remaining clear headed and calm during the crisis. Dalton says key employees met every couple of hours throughout the day to ensure correct decisions were being made and operations were still progressing properly. Only one surgery—an elective procedure that the patient chose to postpone—was rescheduled. Although some tests may have been delayed and there may have been a few backups, there were no large-scale disruptions.

“Probably the biggest problem was that we were unable to print patient discharge instructions and we couldn’t get as much information together for [patients] as we normally might have,” recalls Dalton. “But we did place follow-up phone calls to everyone we sent home. Patients didn’t seem to be upset over what happened.”
Even if downtime can’t be prevented, good planning will make sure the situation is not disastrous. “The old Boy Scout motto of ‘be prepared’ is the simplest answer to preventing chaos and handling downtime effectively,” McLendon notes. “That means, if you haven’t already done so, getting a plan in place and taking all of the necessary steps in advance.”

 — Lindsey Getz is a freelance writer based in Royersford, Pa.