By K. Richard Douglas
We live in an information age where data on any particular subject is plentiful and, often, obtainable. Analyzing that data, can allow it to have a meaningful utility in order to combat crime, find medical solutions to disease or determining a stock to trade.
Big data can also be used by advertisers or retailers to monitor consumers’ search trends or spending habits to create targeted advertising campaigns.
Predictive analytics, using crime trend data, has been used to reduce crime. The city of Memphis Police Department worked with the University of Memphis’ Department of Criminology and Criminal Justice using predictive analytics software created by IBM. The data included the characteristics of various offenders, data from patrols and incident reports, with specific information about time of day and day of the week.
This data was overlaid onto charts and maps that made it useful to the police. The results helped the Memphis police department determine where to place patrols and when to do it. The use of the accumulated data, and the ability to analyze it accurately, resulted in a 75 percent decrease in carjackings and a 67 percent decrease in business robberies.
In health care, big data can help identify a state of deterioration in a patient and alert clinicians by identifying the onset of adverse events. Just like evidence-based medicine, that can guide a physician in a treatment regimen, big data can allow for more specific direction in care to head off a decline in health.
Tests done on blood, urine or tissue samples (in vitro diagnostics) can provide data to the data sets used in the emergency department or intensive care unit, that are combined with other data derived from patient monitors and other sources. This combined data may improve patient outcomes after it is parsed by analytics software and through machine learning and presented to attending doctors.
With the advent of electronic health records, more data became available to those providing treatment. Combined with real-time data from physiologic monitors and other devices that provide feedback, this data can dictate the treatment plan that a physician follows.
Delving into Data in HTM
If big datasets can lead to fewer carjackings, then with the resourcefulness of those in the HTM profession, it can certainly find application to medical device management.
In the HTM department, the ability to pull data and scrutinize it allows for many tasks to be completed. The goal of using big data in the HTM department is the hope that it can promote cost savings, efficiency, positive patient outcomes and reduced downtime. The ways that extrapolated data can do this in the biomed department are many.
“Data can be used in so many ways; technicians can use data to troubleshoot malfunctions which can speed up the repair process and reduce downtime. Data helps with part identification and can help the technicians find the best price on parts,” says James Swandol, BSM, CBET, manager of Healthcare Technology Management at Baylor Scott and White in McKinney, Texas, citing just a few common uses.
“HTM professionals live and breathe the use of data to drive decision making. For any equipment management program, HTM departments need to know what equipment is being repaired, how much time is spent repairing it and repair costs. With this information, HTM managers can best determine how to run their departments,” says Angelique Dawkins, clinical engineer at Baylor Scott and White Health in Dallas, Texas.
She says to promote cost savings, they can examine return on investment for vendor service contracts and compare that to in-house repairs. If equipment is down too often, they can pull repair records to quantify that downtime. And for problem equipment, they can compare the repair and downtime cost to the cost of replacing that equipment altogether.
“Meanwhile, technicians can use data to keep track of their workload. This can include their open work orders, response time to service calls, and completion percentage for preventive maintenance,” Dawkins says.
In addition to tracking important metrics, scrutinizing data can help the budget and bring some science to parts selection and sourcing.
“Knowledge is power for HTM professionals. Here’s why: having data on what you are spending and where provides the opportunity to find cost and time savings. Having data on the quality of your suppliers can help determine which suppliers you can most rely on and identify trends over time,” says Will Burgman, senior director of strategy and operations at PartsSource in Aurora, Ohio.
“Knowing what fails and how often can impact your buying decisions to positively impact patient outcomes and better uptime,” Burgman adds.
That same use of data can allow for informed decisions for capital purchases, according to Aaron Goryl, GM, Healthcare Technology Management (HTM) and On-Demand Development, U.S. and Canada at GE Healthcare.
“Every HTM program, like a fingerprint, is unique which means there is no one-size-fits-all solution. Medical equipment serving high-patient throughput areas is going to impact patient outcome, hospital revenue and end-user satisfaction in a different way than that same piece of equipment with half the utilization,” he says.
“Having strong data metrics that can drill down to expense and utilization of a single device allows the type of nuanced approach that both addresses downtime where it is most critical, and save cost when cost-out initiatives invite hard decisions. That same utilization and cost data empowers HTM programs to have a strong voice and evidence-based positions when it comes to capital planning,” Goryl adds.
Matthew Clark, MBA, CHTM, clinical engineer in the clinical engineering department at Advocate Health Care in Downers Grove, Illinois, says that the use of data takes the guesswork out of decisions in the HTM department.
“Using data is the difference between an intuitive decision and an informed decision. Intuitive decision making can work when determining a strategy, but when you create action plans, you really should have data to see if you are in the right place initially and if your actions are moving things in the right direction,” he says.
Clark says that if you are going to try to identify cost savings, you need to find an area where your spending is higher than it should be and then, after you have implemented a cost-savings initiative, you should check the spend to see if you were effective.
“Efficiency initiatives can be identified by looking at where a lot of maintenance time is spent. After which, you should check the same numbers to see if you reduced the time spent. With these initiatives, you also must look at other information, such as your downtime or any clinical metrics that may also be impacted by changing how the equipment is managed. By focusing on only one metric you can solve one problem while inadvertently creating another,” he says.
The availability of data can also help answer several questions for the HTM department according to Burgman.
“Is my team spending time on the high-impact areas? How well is my team purchasing,” he asks. “Where can I save more by purchasing better/smarter and are quality trade-offs in the secondary market commensurate with cost savings?”
Correlating Data Sets within CMMS
Using existing or add-on features with CMMS can make the most of available data.
Swandol says that most, if not all, CMMS databases allow for storage of documents and [that] is how data in conjunction with CMMS can help the HTM professional.
“By having all the data in one place, that is easily accessible, the HTM professional is not having to go search all over the Internet or spending hours on the phone with vendors to request documents. Instead the technician has the data at their fingertips, saving time and limiting down time of equipment,” he says.
“At Advocate Health Care, we have built a dashboard with our support operations department that combines different data sets to visualize trends in different areas. We are looking at the results of satisfaction surveys (patient and customer), equipment reliability, safety measures and cost measures,” Clark says.
“Trends are shown to identify potential correlations between things like cost and downtime, or reliability and customer satisfaction. There is a lean A3 form that is built into the dashboard that allows action plans to be built using the collected data and show the trends as the plan is carried out. Seeing the various areas in one place so that trends can be seen side by side really gives a lot of insight into whether a strategy is working or not,” he adds.
Dawkins says that many CMMS have built-in reports that come with the system, but also offer the ability for HTM professionals to customize their own. These can be run on an ad-hoc basis or sent out on a recurring schedule.
“Some CMMS even have the ability to create dashboards that pull data in real time. These tools allow HTM professionals to get an accurate and up-to-date picture of the state of their programs,” she adds.
Trends in Usage
While more and more information ends up digitized and accessible, the use of big data will continue to improve the decision process for purchases, maintenance and resource allocation.
“As more medical equipment is being integrated into the electronic medical record, more data is collected than ever before. We have begun to see requests to help assess how the information is collected, trended and reviewed,” Clark says.
“We also see more requests to compare devices and their results as they are fed into the EMR. The variation between devices becomes more apparent when data is reviewed side by side in a collection of data. This has led to more standardization in devices used and how they are used in the clinical environment. Naturally, this standardization of devices introduces the opportunity for efficiencies in how those devices are managed,” Clark adds.
Burgman says that his employer has actively utilized big data going back to 2008, which includes data on 2.5 million product requests, tracking 20-plus key attributes across each of those transactions. Making use of that data, they were able to extrapolate several findings.
“Every clinical site has opportunities for additional savings – this is likely because few are set up to handle the long tail of low-volume, low-dollar orders. Facility size or amount of spend has little to do with how well a site purchases,” he says.
“ ‘Best-in-class’ HTM programs require consistent processes and real-time decision making support based on data and insight. HTMs spend the majority of their time sourcing products under $1,000. Most organizations are spending a tremendous amount of time and resources on the lowest dollar items. The sheer volume of low-dollar transactions can overwhelm manual, paper-based or otherwise inefficient processes,” Burgman adds.
As a result of this data analysis, he says that reallocating HTM expertise from sourcing low-dollar, infrequent purchases to higher-value activities like fixing important equipment is the key to impact. Having the data to understand your spend, time and medical equipment quality can increase capacity without adding headcount, creates opportunity for cost savings and quality improvement and ultimately increases uptime for an organization.
The use of datasets also allows for tracking equipment status and helps pinpoint training needs.
“Currently, we use datasets to help determine what equipment should be placed on an AEM program. We also use data to find trends with equipment malfunctions, tracking operator errors, misuse or abnormal damage. This data helps us determine not only what training is needed for the HTM professional but also what training clinical staff could benefit from,” Swandol says.
A continuing trend will be the ability of analytics software to allow HTM departments to visualize the status of equipment and cyber threats.
“Data visualization will continue to be important to HTM professionals everywhere. A CMMS (or program interfacing with the CMMS) that can provide dashboards is an asset to any department. By giving a quick visual snapshot of the state of the program, it streamlines reporting and increases operational efficiency,” Dawkins says.
“Medical device cybersecurity is also a hot topic for HTM. As medical devices become more connected with health care networks, using data to prevent intrusions will become increasingly important. This data can come from monitoring devices on the network, and establishing a baseline for device activity. By analyzing this data, suspicious activity can be flagged and stopped. This promotes patient safety, protects their data and promotes device uptime,” she adds.
The ability to process an accumulation of information efficiently and transform it into a meaningful format to glean knowledge, cost savings and other benefits is still in its early stages. There will be more ways to utilize datasets and more available datasets in the future. HTM departments will continue to find ways to analyze and parse data to achieve more cost savings, efficiency, positive patient outcomes and reduced downtime.