Improving Equipment Reliability Through Battery Management

The batteries that power an institution’s sometimes thousands of medical devices are perhaps not the sexiest part of a biomedical engineer’s profession. But they are an area where a fairly simple-to-implement program of measuring them and monitoring their results can maximize battery life and maintain more dependable equipment. Indeed, one facility that instituted such a program has seen solid savings with batteries actually lasting, in many cases, twice as long as the manufacturers recommend. The facility also enjoyed something that can be a rare commodity in hospital biomedical engineering departments: relative peace of mind. According to those at the University of Ottawa Heart Institute, Ottawa, Canada, there have been no repeats of the device failure—the result of a used-up battery—that sparked the battery-management program in the first place.

At Ottowa Heart’s battery-management-assessment station, batteries are treated like medical devices and are labeled with inspection dates.

The incident that brought batteries to his attention, reports Timothy J. Zakutney, MHSc, PEng, manager of biomedical engineering services for the Institute’s cardiovascular devices division, involved an infusion pump in the facility’s operating room (OR). The details do not mean much now, he points out, except that the event led the Institute to conduct a study to look at whether there was anything his department could do to improve its processes and avoid any such future “very, very dangerous situations.” The result, he happily reports, has been exactly what he had hoped for. And, he points out, it’s already been tested. “We had an incident several months ago while we were renovating our cath lab,” he comments. “There was a loss of power due to a construction mishap. We did a ‘postmortem’ and determined that not one piece of battery-operated equipment in the [intensive care unit (ICU)] failed. All the pumps and all the ventilators ran for the full hour it took us to bring in power from an adjacent area—and none of them failed from being on battery power.”

Of course, device manufacturers have tried to offer insight into the power in their products, but the information is not particularly useful as presented, Zakutney points out. The thermometer-like scale on the side of many of them is intended to graphically display how much juice is available. “But does that mean it’s good or bad?” he asks. “Does it mean that particular battery will last long enough for you or your nurses or patients to get where they want to go? The lack of specificity creates quite a bit of concern and confusion.” Treating batteries like devices is a key concept in Zakutney’s management program, and when you manage about 8,000 device files—which include 300–500 batteries, most of them lead-acid—concern and confusion are two of your worst enemies. “We are extremely busy,” he comments. “But we don’t have a huge amount of resources.”

Making a Plan
To start their battery-management program, Zakutney and colleague Mark J. Cleland, senior biomedical engineering technologist at Ottawa Heart, melded the technical description of battery capacity with the definition that means something to the nurses who most often use the devices: “For how long will this device operate?” Officially, Cleland explains, a battery’s capacity is established by the manufacturer and indicates its ability to deliver a specified current over a specified period of time—so, say, a 2 amp-hour battery should deliver 1 ampere of steady power for 2 hours, or 2 amps for 1 hour. “But when I talk to nursing about battery issues,” he adds, “I frequently interchange ‘capacity’ with ‘run time.’ That’s the function of its capacity and of what kind of current drain the equipment it powers makes on it.”

Rather than relying on manufacturers’ capacity determinations that do not apply after the unit has been used and reused, he adds, Ottawa Heart measures capacity by placing each battery on an analyzer. “Programs that simply replace batteries on manufacturers’ guidelines or after a specified time fail to provide a quantitative measure of the devices’ ability to perform,” he states. “As a result, some good-quality batteries are discarded and poor-quality batteries continue to be used.” Instead, his team focuses on the real-world life expectancy of the batteries it actually has on-site—which means data on how the device the battery will power operates and is used must be incorporated into the calculation of its power potential. “I compare a battery to a ladder,” he explains. “The top represents a fully charged battery. The remaining power can only scale the ladder so many times before you cut its life expectancy, so we encourage staff to maintain batteries in a plus-charge state. If it’s used, for example, in a transport situation, we ask them to plug it back in right away. If you only use the top few rungs of the ladder, you’ll get maybe 1,500 cycles out of it, rather than the 200 to 300 you get if you run it down over and over, extending its life from one year to as long as 3–5 years.”

Establishing Testing Protocols
To get information on how batteries are used in the real delivery of patient care—how long the devices they power actually have to run and how often they actually have to be drained more than the optimum amount, for example—Cleland and Zakutney talked to medical directors and the other primary users of the equipment. “When they tell us their demands,” he says, “we can start establishing our testing protocols to ensure that the batteries are going to do what they ask of them.” The program also takes into account how many batteries are employed in each device, whether they’re parallel, what the drain on the battery from the equipment is, and what the equipment’s end use will be. “We look at service manuals and at whether the device being powered is for life support, therapy, or diagnosis,” he reports, “and from that we establish a testing interval.”

The team also developed testing protocols around nursing’s window of opportunity when it comes to equipment failure. Zakutney’s department had experience with batteries in such poor condition that the time between the device indicating that it was low on power and the time it stopped working was not long enough to have a replacement put in place. He asked the nursing staff what their expectations were, and found they needed at least 3 hours of operation time between batteries because some patients have to be transferred to a cross-campus sister facility. “We looked at our data,” he reports, “and found that a complete transfer takes 182 minutes, so that’s the minimum value of run time we allow. We find we can achieve that if well-maintained batteries are kept at at least 40% capacity.” That changes, of course, if the device is being used for life support.

It is critical in a program like Zakutney’s to treat batteries like medical devices. When a new piece of equipment comes in—or one is sent to biomedical engineering for maintenance —the staff there logs the purchase date, if applicable; the purchase order number; the cost of the device; and the serial number. “We have the ability in our management-information system to monitor ‘parent-child’ relationships, so we keep track of which batteries are in which pieces of equipment,” he comments. That is especially important, he adds, because many manufacturers do not monitor the batteries that come in the devices they sell. “If a ventilator comes in, for example, we analyze its batteries before the equipment is deployed,” he says. “And we log all the assessment data in two places. It’s automatically entered on the battery-analyzer system, and we also transcribe it into our general records.”

That does sound like a lot of work. And Zakutney notes that he hears that “there just isn’t time in the day” all the time. “People look at the thousands and thousands of devices they’re responsible for in a large hospital, with, say, 40% using batteries, and they assume a battery-management program would add a lot to their workload.” Instead, he suggests, hesitant biomedical engineers should think in terms of “event time.” For example, think about how much time it adds when a pump arrives to be repaired. With an automated program like the one at Ottawa Heart, the management plan may tack an additional 5–10 minutes of staff time onto each battery repair. “In the grand scheme of things, that is not a significant amount,” he says. “And it could be less than that depending on the protocols your department uses—whether you replace batteries or reuse the one in the device.” The key, he stresses, is integrating the management process into the department’s day-to-day activities.

And, he adds, once the program is operational, testing can be performed on batteries that do not have to be put into service immediately. “You can amass a stock of batteries that are relatively newly conditioned,” he explains. “Say a new pump comes in. You can put in one of your tested and charged batteries so you know the condition of the device, then put the battery that came with the device in line for inspection later. Or, say, a broken pump comes up to biomedical engineering. Ordinarily, you’d repair it and then plug it in overnight to make sure it’s fully charged before it goes back down to the floor. Under our program, you could have a stockpile of fully charged batteries so you could replace the one in the defective device right away. That can dramatically reduce the downtime for, say, a pump being repaired.” His shop rarely does that, he notes, preferring to keep batteries and devices together. But facilities that opt for that tack can, over a year or so, “get a great handle on the condition of all the batteries they have stocked up.” They can also pare the time required to process the batteries’ information to 5 minutes or so apiece.

Cleland adds: “It doesn’t really add any significant amount of time to the testing time, but it is an extra step. But the information you get from that extra step is important.” In a test he conducted of battery capacity before coming to Ottawa Heart, Cleland’s former employer purchased a number of batteries for a specific service.¹ “We looked at 126,” he reports, “and 46 had less than 65% capacity, our acceptance threshold for newly purchased batteries. The first point we learned from that is if you don’t measure new batteries, you can’t be sure of what you’re getting.” That level of output capacity, he notes, would not have powered the defibrillators the batteries were intended to keep running—even though they were brand new. That hospital contacted the manufacturer and discussed possible problems and potential solutions. “We then reanalyzed the batteries,” he adds, “and saw dramatic improvement in those that had failed their incoming inspection.”

Elevated Confidence and Patient Safety
Applying that lesson to Ottawa Heart’s battery-management program has spawned a level of success that really impresses James A. Robblee, BSc, MBA, MD, FRCP(C), chief of Ottawa Heart’s Division of Cardiac Anesthesiology. He’s the one who initially raised the question of battery dependability after that incident where one failed during a patient transport from the OR to the ICU. “Initially, the battery-management program was the initiative of one of the biomedical engineers in our organization,” he says. “But it has now become actually embedded within the functioning of the department. From that point of view, it’s almost a seamless part of the job there. It’s one of their functions, just like everything else they’re doing. There has been no problem with buy-in whatsoever.”

And, he adds, nurses were “quite interested” in the program right from the outset. “I asked around after the failure incident and found that possibly undependable batteries are a regular occurrence,” he says. “Nurses named six other patients who’d been potentially compromised as a result of problems with infusion pumps that could have been tied to battery failure. And when we did the patient-safety review that initiated the whole process, there was extensive feedback from nursing. So they were instantly interested in the battery-management program.” It has, he adds, raised awareness on their part and has added momentum to Ottawa Heart’s whole patient-safety initiative.

Today, Zakutney notes, his department holds regular in-services for nursing that include details on the battery-management program, and contributes to the critical-care newsletter with reminders that equipment must be plugged in as often as possible. “We’ve found staff have great confidence in our equipment,” he comments. “Any rumors related to devices failing because of battery malfunction have been eliminated. In fact, confidence in our equipment has risen to the point where, sometimes when we get a pump from another hospital, nurses will refuse to use it because they’re not confident it will be in good working order. But they know our devices are.”

Russell Jackson is a contributing writer for 24x7.

Reference
1. Cleland MJ, Maloney JP, Rowe BH. The effects of lead sulfate on new sealed lead acid batteries. J Emer Med. 2000;18:305–309.