At a Glance
- There are two primary factors for analyzing the cost of medical device integration: the total cost of ownership and gains in clinician efficiency.
- When calculating the long-term total cost of ownership, hospitals should consider the original acquisition cost, ongoing support costs, and also look at the incremental cost of adding a new device.
- In determining clinical efficiency, recent analyses show that medical device integration alleviates the initial negative impact of implementing an electronic health record and improves clinician efficiency by reducing clinicians' overall documentation activities.
Implementing an electronic health record (EHR) is no small feat, as EHR implementations are usually accompanied by a notable decrease in clinical efficiency. A recent study by the University of California, Davis, estimates that such implementations result in an initial reduction in productivity of 25 to 33 percent (Merrill, M., "Study: EMRs' Effect on Docs' Productivity Depends on Needs, Workflow," Healthcare IT News, Dec. 16, 2010).
How can hospitals rebound, or salvage clinical efficiency, after an EHR change? The good news is that productivity will naturally inch its way back up through clinician training and adjustment.
The even better news is that integrating medical devices with the EHR can not only alleviate the initial negative impact of the EHR implementation, but also take clinician efficiency to levels not experienced prior to the implementation.
Improving Clinical Efficiency
Medical devices generate incredible amounts of patient data, which must eventually be moved into a centralized information system, or EHR. In some instances, the process is manual. Clinicians acquire patient data from a device, write it down on paper, and then enter the data into the EHR later in multiple-patient batches.
By integrating, or connecting, these medical devices directly to the EHR, a hospital can automate the process of transferring device data to the EHR. Clinicians simply review the data digitally and transfer the data to the EHR, thereby avoiding time-consuming processes of dealing with paper, hunting down patient charts, and deciphering illegible handwriting.
Naturally, this automation reduces the clinician's documentation workload significantly. For example, a study of the effects of medical device integration at the University of Alabama at Birmingham Health System found that this approach reduced the time required for nurses to verify data from four minutes, which included manual data entry, to 20 seconds. The study was cited in a presentation to the New Jersey Chapter of the Health Information and Management Systems Society (HIMSS) by Marilyn Hailperin and Judith Chan ("Developing a Strategy for Integrating Medical Device Data with Clinical Information Systems," October 2009). The automation was estimated to save the hospital 86,000 nursing hours per year. The finding suggests that medical device integration can save a 150-bed hospital more than 2,000 nursing hours per year.
Clinicians tend to respond favorably to such improvements in clinical efficiency, to the extent that they will occasionally pressure CIOs to undertake medical device integration.
Such was the case at Cooper University Hospital, a health services, medical education, and clinical research provider in southern New Jersey. In 2010, as chief technology officer Paul Shenenberger prepared to launch a new EHR in the 10-floor patient pavilion, nurses added a last-minute IT initiative to the mix.
"We got pressure from the nurses," says Shenenberger. "They wanted us to implement device integration at the same time."
Shenenberger and the board at Cooper University Hospital capitulated. Today, Shenenberger says the integration of medical devices with the EHR is the single largest nursing satisfier out of all the IT projects he has ever been involved with. As far as nurses are concerned, the medical device integration is an inseparable part of the EHR initiative.
At MetroSouth Medical Center, a leader in cardiovascular primary care located near Chicago, integrating the organization's medical devices with the EHR saved nurses in the cardiac recovery unit approximately two hours per shift. MetroSouth was able to use the extra time to increase direct care to patients.
The benefits of increased nursing time devoted to direct care are well documented. For example, a brief issued by the National Foundation for American Policy found that "greater registered nurse hours spent on direct patient care were associated with decreased risk of hospital-related death and shorter lengths of stay" (Anderson, S., "Deadly Consequences: The Hidden Impact of America's Nursing Shortage," National Foundation for American Policy, September 2007).
Using High-Quality Data at the Point of Care
Just as direct care at the bedside improves patient outcomes, so too does data availability at the bedside. When clinicians have access to patient data at the point of care via the hospital's EHR, they can use the data to make better-informed decisions related to care and safety.
But for this premise to work, the data in the EHR must be correct and timely. In other words, the data have to be of high quality.
Research shows, though, that EHR data are often incorrect. One study looked at 623 sets of vital signs recorded at a 20-bed, cardiac step-down unit in Florida and found errors in 14.9 percent of the records. If the data in the EHR are wrong-or even if data are right but hours old-then the EHR is nothing more than a gateway to bad information.
Why do EHRs contain so much faulty data? If a hospital's medical devices are not integrated with the EHR, a great amount of patient data is hand-keyed into the EHR. In effect, the data chain is the nurse. People are relatively error-prone data chains, and entry errors should be expected. Likewise, association errors occur when patient data are entered but associated to the wrong patient.
Human data chains also are slow because busy clinicians often enter patient data into the EHR in end-of-shift batches. At that point, data are already hours old. Factors such as the hospital's staffing levels and the clinician's personal process can affect the data's transcribed latency.
In a study undertaken to assess the speed of its data processes, Wise Regional Health System (WRHS) in Decatur, Texas, found that an average of 12 hours passed from the time a patient monitor generated data to validation of the data in the EHR. Following device integration, that average dropped to just two hours.
Medical device integration also improves data latency. In their presentation to the New Jersey Chapter of HIMSS, for example, Hailperin and Chan cite the case of St. John's Medical Center in Jackson, Wyo., which was able to increase its vital sign charting from every 15 minutes to every five minutes as a result of such integration.
Because the integration allows data from medical devices to be channeled directly into the EHR, transcription errors and hard-to-read handwriting are avoided. This is one important way in which medical device integration improves data quality. As a result of the automation at WRHS, the organization's nurses and physicians estimated that data quality increased from 5 to 9 on a 10-point scale.
With accurate, real-time data in the EHR, clinicians at WRHS and other integrated hospitals across the nation are able to make better-informed decisions at the point of care. When medical devices are integrated with an EHR, the EHR becomes not only a repository of data, but also an accurate and timely-and meaningful-tool for clinical decision making.
Analyzing the Cost of Device Integration
Undoubtedly, integrating medical devices with an EHR transforms the way patient data are collected and documented. Federal provisions under the Health Information Technology for Economic and Clinical Health Act alleviate some of the financial burden in investing in such an undertaking. Still, hospitals should weigh such investments against capital and operational budgets. There are two primary factors for analyzing the cost of device integration: total cost of ownership and gains in clinician efficiency.
Total cost of ownership. Capital acquisition costs of any technology required for device integration are generally tied to the number of hospital beds. Costs per bed vary, of course, and are dependent on the nature of the integration solution. A study conducted in late 2005 by Indianapolis-based MindGent Healthcare Services, however, found the average cost of a device integration project to be $6,700 per bed, with an ROI period of 15.9 months (Addressing the Issues of Nursing Shortages and Patient Safety Through Bio-Medical Device Integration [BMDI], December 2005).
When calculating the long-term total cost of ownership, hospitals should consider the original implementation cost as well as ongoing support costs. In addition, it's important to look at the incremental cost of adding new devices, as the number of new medical devices grows rapidly.
Clinical efficiency. Ironically, as clinicians shift from paper to electronic records, they initially spend more time inputting information into the EHR than they did before the implementation. As clinicians become more acquainted with the new EHR, their productivity eventually increases. Recent ROI analyses and field studies support the previous assertion that medical device integration not only alleviates the initial negative impact of implementing an EHR system, but also improves clinician efficiency down the line.
Once clinicians are comfortable with a new EHR, then the integration of the medical device with the EHR begins to reduce their overall documentation activities. Because hospitals expend a large portion of their budgets on clinical staff, it is important to remove any obstacles that would keep clinicians from focusing on direct patient care.
Recent ROI analyses and field studies indicate that integrating medical devices with the EHR improves clinician efficiency. The aforementioned MindGent study estimates that a 150-bed hospital could save more than 2,000 nursing hours a year through such an effort. Another study projects a reduction in charting time of up to 50 percent for support staff charting and 20 percent for physicians (Rausch, T.L., and Judd, T.M., "The Development of an Interoperable Roadmap for Medical Devices," Conference Proceedings: Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2006).
As the possibilities for integrating medical devices with a hospital's EHR continue to increase, hospitals will continue to find creative and meaningful ways to use and leverage their patient data.
For example, Cooper University Hospital has effectively integrated dialysis machine data into its EHR. Data can now be reviewed in conjunction with vital signs data, making it one of the only hospitals in the United States to accomplish such an integration.
Mary Jo Cimino, RN, BSN, CCRN, clinical director of adult critical care, pushed for the integration of the dialysis machines. Cimino, whose biggest concern was patient safety, explains the integration enabled nephrologists to check on patients from any location.
Cimino notes that clinicians are now "able to discuss their cases together because they can see real-time dialysis information on their computers." She adds, "We can go back to the chart at any time and do any kind of research we want, which is so important in kidney care. Right now, it's very hard for most hospitals because they aren't able to readily put together the patients' vital signs with the dialysis data."
With her nurses recording top satisfaction scores on the National Database of Nursing Quality Indicators and almost no staff turnover, Cimino believes Cooper University Hospital is ahead of the curve. "Technology should free you up, not turn you into data collectors," she says.
There is little doubt that EHR data will continue to fuel research efforts at academic medical centers. "As an academic medical center, we wanted to be able to collect more granular data for our research database than we would need for clinical purposes," says Cooper's Paul Shenenberger.
At Jefferson Regional Medical Center in Pine Bluff, Ark., data integration resulted in data sharing that has improved patient safety. "Our integration allows us to create a custom report that shares patient data with our rapid response team," explains Leah Wright, director of clinical informatics at Jefferson Regional. "The better the report they receive, the more valuable they can be when they arrive to help the patient."
It's important to note, too, that accurate, real-time data in the EHR is the cornerstone of future advances in clinical decision support systems. These software-based systems link health observations with health knowledge. Imagine if, at the point of care, a clinician could review a chart that integrated a patient's symptoms, medical history, family history, genetics, historical and geographical trends, and published clinical data. The sky is the limit when it comes to clinical decision support systems, but it all starts with high-quality data in the EHR.
Peter Witonsky is president and chief sales officer, iSirona, Panama City, Fla. (firstname.lastname@example.org).
Evaluating Device Connectivity Solutions
Despite the stimulus-based incentives and the hospital-wide benefits of medical device/electronic health record (EHR) integration, many hospitals hesitate to move forward with such an initiative. Implementing connectivity involves change, effort, and costs. With multiple solutions on the market, it is helpful to consider the "Four Cs" when evaluating a potential connectivity solution.
Coverage. Does the solution work in all clinical environments, including bedside and mobile devices? Will it collect all the data, regardless of patient location? Location-based connectivity solutions rely on location information or bed numbers to associate a patient to a device. Patient-based solutions allow devices to be associated to a patient regardless of location.
Compatibility. Many point-of-care devices are designed to run as standalone networks. Connectivity for these devices is achieved in several ways. In assessing connectivity options, ask whether the solution will leverage existing investments in hardware, such as laptops and workstations-on-wheels. Some solutions require new, single-use hardware investments, which can be costly and impractical.
Confidence. Will the solution ensure data are documented to the correct patient through bar code scanning or radio frequency identification? How does the solution deal with information before it is charted? For clinicians to make decisions with confidence, they should know the EHR is always complete, accurate, and current.
Costs. Consider implementation costs, proprietary (single-use) hardware requirements, implementation and training time, ease of use, and scalability. Also assess whether the solution can embed into the existing clinical information system, which can shorten the learning curve for clinical staff, saving time and money.
Promoting hospitalwide acceptance of the endeavor also is important. Placing the right people in the project as soon as possible is important. Include clinicians, IT, and engineering in the plan, and be open to asking other departments as well.
Publication Date: Wednesday, August 01, 2012