July / August 2009
Biomedical Device Integration:
The Impact on Clinicians at the Point of Care
Hospitals are constantly evaluating new technologies that promise to improve the quality of care, create a safer care environment, and improve both clinical and operational efficiency. If, however, technologies are evaluated solely on their individual merits, they may introduce problems to the clinician’s complex work environment. When new technologies are deployed, there is often some impact to the clinical workflow. Even minor changes to clinical workflow can be disruptive—such as asking the nursing staff to perform (and remember to perform) just one or two additional steps—which may lead nurses to find workarounds that defeat the safety advantages of the technology.
Consider deployment of the electronic medical record (EMR), for example. While the EMR is essential to patient care and safety within a hospital, its deployment can actually add additional steps to the clinician’s workflow. This is especially true with the nursing documentation module of the EMR because, without automated biomedical device integration, not only do clinicians have to manually record vital signs, they also have to enter those vitals into the patient’s electronic chart manually. Even worse, this manual entry often takes place after the clinician completes his or her shift on the floor. So the technology is essential to patient safety, but without biomedical device integration, it may make the nurse’s workflow more complex.
Biomedical device integration provides a way to simplify workflow and actually enhance the value of the EMR and its adoption by clinical staff. When deployed, vital signs data are automatically sent from biomedical devices directly to the nursing documentation system. The clinician can focus more on direct patient care by quickly validating the vital signs data in the EMR and submitting them to the patient’s record. Elimination of manual entry increases clinical satisfaction and improves patient care and safety.
Current methods of device integration are evolving to meet changing needs and to address safety issues. Specifically, most integration solutions on the market today rely on an association of biomedical device data to a physical location such as a patient bed or room. Location-based association limits deployment to higher acuity environments where there is a fixed bed location, such as the ICU. Hospitals, however, need device integration across the enterprise, including lower acuity environments such as med-surg. And, regardless of the environment, data from more classes of devices, such as smart IV pumps and wireless devices, require integration. The result is that association of the patient to the device is essential to the future of biomedical device integration. The questions a hospital needs to consider are, “What technologies will be available to achieve positive patient association,” and, “How will those technologies affect workflow and patient care at my hospital?”
Biomedical Device Connectivity Workflow
Biomedical device connectivity is the automation of workflow through the integration of medical devices and information systems, with the goal of improving the safety and effectiveness of patient care. One of the key components of this definition is workflow automation, and optimizing the workflow is critical to the adoption of systems that automate connectivity. Workflow, however, becomes a challenge because the nurse must, 1) interact with and connect more medical devices at each patient’s bedside, 2) integrate more data from these devices including vitals and alarms, 3) send data from wireless devices and/or devices such as IV pumps that have no fixed location and therefore require association with the correct patient, and 4) integrate devices at other departments in the hospital.
Industry workgroups have reported on specific recommendations relating to medical device data capture. In a report on healthcare information capture (Consensus Workgroup, 2002), recommendations are listed that describe steps that can be taken to increase the safety aspects of device data collection and electronic documentation. The report highlights a key issue related to patient ID:
When data is required electronically from patient-connected medical devices, there should be some way to internally support a means for unique identification of the patient.
Additionally, the same report makes specific recommendations such as:
Data collection networks should be designed to ensure that patient identification information is coupled with patient data as close to the device as possible. If patient identification can be associated with a given network access point or point of care (POC), then maintaining this location or “port” information may be sufficient, with the patient ID added later. However, this too can be error prone when patients change location.
Patient Association and Workflow
The concept of context awareness in healthcare is relatively new. According to Dey et al. (2001), context is defined as:
Any information that can be used to characterize the situation of entities (i.e. whether a person, place or object) that are considered relevant to the interaction between a user and an application, including the user and the application themselves. Context is typically the location identity and state of people, groups and computational and physical objects.
When implementing biomedical device connectivity, the concept of context can be applied to medical devices to enable awareness as to who is the correct patient. In today’s patient care environment, patients and medical devices are both highly mobile, partly due to the use of wireless technology in medical devices. Most vendors’ integration solutions on the market today rely on an association of a biomedical device to a physical location such as a patient bed or room. But this is in the process of changing, as technology to manage positive patient association evolves.
Data integrated to an EMR application (i.e., nursing documentation) cannot always rely on an association of a medical device to a bed because some devices can be moved to a new patient at virtually any time. This results in the possibility of patient data being stored in the wrong patient record in the EMR. Hospitals cannot accept this risk, and therefore solutions must ensure that all data and alarms are positively associated with a specific patient identifier. The Joint Commission’s number one National Patient Safety Goal (NPSG) for 2009 is to improve the accuracy of patient identification. The goal states that patient identification should not be performed using a room or location-based identifier.
Most mobile medical devices and systems used for direct patient care have no patient or clinician context. Without a mechanism for establishing patient context, these devices lack awareness about who the patient is (patient ID) or where those devices are at any given time (location awareness or context). A confirmed patient ID is required to dynamically establish patient context through the process of associating medical devices to a specific patient. A vendor-neutral and open approach to managing patient context and positive patient ID enables data and alarms produced by these medical devices to be safely and effectively integrated to existing applications. This contributes to hospitals achieving their goals for integration and interoperability across their healthcare enterprise.
Similar to data integration, medical device alarms and life-critical events must also be associated with a specific patient identifier. Caregivers must respond in real time to life-critical events—such as cardiac alarms generated by physiologic monitors—and without a positive association of a medical device to patient identifier, an assigned caregiver cannot be notified in real-time about a life-critical event, and event information cannot be recorded safely.
Positive patient association (PPA) is necessary for successful biomedical device integration across the enterprise. As a result of the Joint Commission’s NPSGs, PPA has become a top priority for hospitals and device integration providers. However, it is essential that hospitals carefully consider what type of technology they choose to achieve patient association because technology adoption in a clinical environment is heavily dependent on the technology successfully integrating into the clinical workflow. Ideally the technology should enhance the clinical workflow and improve patient safety. Too often, however, technologies actually do the exact opposite, delivering inefficient workflows thereby decreasing employee satisfaction and potentially impacting patient care and safety.
Consider barcoded medication administration (BCMA) and IV pumps, for example; the five rights are essential to patient safety, but the number of steps a clinician has to take to barcode scan every pump, every medication, the patient, and themselves to achieve these five rights is fundamentally inefficient and prone to workarounds (Koppel et al., 2008). Workarounds often include bypassing the intended safety features of the technology. Of course, patient safety outweighs this inefficiency, but when considering new technologies, it is essential that hospitals review the options carefully to see if there is a technology that could potentially meet both the technical and workflow efficiency objectives.
Also bear in mind that when actively managing PPA, the ability to address the disassociation function and related clinical workflow is required. The disassociation function is crucial to the overall safety and efficiency of the patient care environment.
The following is a review of some current technologies that are either available or will emerge from device integration providers in the near future to address the automation of positive patient association and clinical workflow. It includes a brief description of the technology and then provides a list of questions the hospital might want to pose to vendors as they evaluate how the technology will affect workflow. It is assumed that using a fully manual process and workflow is not a viable option for clinicians, and therefore, manual association has not been considered.
Barcode & PDA-Based Association
This technology refers to nurse-centric solutions that assist with point-of-care tests and procedures such as medication administration, phlebotomy, blood transfusions, specimen collection, and mother’s milk matching. Nurses are provided with a handheld PDA-style mobile device that runs an application that prompts the nurse to perform workflow steps in a specific order. To ensure safety, the most important step in the process is the positive identification of the patient, which typically is performed via a barcode scan of the patient’s wristband.
Things to Consider: Does the nurse want to carry another device? How many steps will be required to scan the device(s), the patient, the caregiver, and then perform the association on the PDA? Is this a realistic workflow? Is the screen resolution on a PDA device optimized for the aging nursing population?
Medical Device-Based Association
In the past, with a lack of viable solutions for managing PPA, vendors had no alternative but to add an association capability to their medical devices. Some products such as IV pumps, include a barcode scanning solution for matching the patient ID to the specific medical device. In this case, the nurse can perform the patient association but only for that particular IV pump. The problem arises when the clinician wants to administer other non-infused medications, collect blood or tissue samples, and then must deal with a different solution and/or barcode scanner for PPA.
Things to Consider: How many different barcode scanners and different barcode technologies are required at the bedside? Can there be a universal solution or technology for managing PPA at the bedside?
Some applications assist the clinician with a partial automation of the association process. Often an application-assisted method of association is nothing other than a variant of a manual process performed on a computer. For example, in some electronic medical record (EMR) applications, there is feature that assists the clinician in establishing the association between a patient and a specific medical device. As a means of partially automating the workflow, the EMR application will have a drop down list with a list of medical devices that represent candidates for association.
The problem arises when hospitals standardize on one vendor and have hundreds of similar devices; the way to help manage the complexity is to provide a drop-down list in the application with all of the devices listed. This becomes error prone and almost unusable when the drop-down list grows to hundreds of choices.
Things to Consider: Is the end-user application designed and optimized to deal with the complexity of PPA?
Many point-of-care solutions leverage the hospital’s Admit, Discharge, and Transfer (ADT) system to assist with the process of PPA. Solutions that solely leverage the ADT system to identify the patient are prone to errors and promote inefficient clinical workflow. Solutions that do not first check identifying patient information such as the patient’s barcode at the bedside, rely on a query of the ADT system to determine what patient the ADT system thinks is admitted to the bed. There are two fundamental problems with this approach.
The first problem is that ADT systems are notoriously inaccurate at any given time and therefore contribute to the potential to identify the wrong patient. The second problem is that even when clinicians have direct control over correcting the inaccurate patient scenario, this introduces additional workflow steps and adds potential errors. In some hospitals, the ADT system can be inaccurate as much as 30% of the time or more. This means that clinicians could be asked to perform additional workflow steps in almost one out of three cases where the patient must be identified using this ADT-based method.
Things to Consider: Using the ADT system to assist in the PPA process is acceptable and preferred; but only when the clinician first positively identifies the patient using an auto-ID technology, such as barcode or RFID.
Real-time location system (RTLS) technology can be applied to devices and people. Virtually any type of asset, including biomedical devices, can be tracked using a variety of auto-identification technologies such as active RFID, passive RFID, ultrasound, infrared, and others. Patients and clinicians can also be tracked across a hospital using RTLS technology.
Things to Consider: Can RTLS be used to effectively manage positive patient association and disassociation? Can the RTLS system be used to determine location down to a guaranteed room, bed, or patient level of accuracy?
Conclusion and Recommendations
As your hospital begins to assess requirements for new technologies, such as positive patient association and biomedical device integration, look beyond the core technology. Create a formal process for evaluating the impact on nursing workflow at the point of care. Recognize where technologies are converging and assess where those technologies and clinical workflow overlap.
Conduct a comprehensive readiness assessment using the following guidelines and suggestions.
- Assess the role of active RFID versus passive RFID. Understand what problem each technology addresses and the limitations of each technology. Understand where each technology overlaps and potentially where each technology reaches its practical limitations and where another technology can take over (i.e. active RFID used for asset tracking and passive RFID for positive patient association at the point if care). Understand that there is no one technology that can solve all problems.
- Assess medical device purchases with specific scrutiny for how the devices will fit into a point-of-care PPA process and workflow. Assess how devices will be tracked, as assets and as part of the PPA process.
- Assess how positive patient ID will be implemented across your hospital and across multiple applications and clinical processes including the impact on biomedical device connectivity, medication administration, blood transfusions, phlebotomy, specimen, mother’s milk matching to babies, etc.
- Assess the impact on point-of-care workflow as each technology is phased in.
- Assess how medical devices and objects will be tagged. Will RFID tags supplement barcodes? Will both passive and active RFID tags be provisioned on devices at the same time? If not, it will take duplicate efforts to locate and tag all medical devices multiple times.
- Assess a strategy for patient ID wristbands. Determine if it makes sense to have a barcode and RFID tag combined on one patient wristband?
Biomedical Device Integration Solution Features
- How does the solution handle wireless and mobile device integration?
- How does the solution handle smart pump integration?
- How many live implementations and reference sites does the solution provider have?
- Are there any additional features such as safety alerts that the solution can provide and how are these delivered?
- Does the solution meet the requirements for a Medical Device Data System (MDDS) (FDA, 2008) and have an FDA 510k?
Brian McAlpine joined Capsule Tech, Inc. in 2008 as director of strategic products. Over the past 20 years, McAlpine has led several efforts to develop innovative medical device connectivity solutions. He has extensive experience with all aspects of medical device integration and healthcare IT. Before coming to Capsule, McAlpine was director of product management and marketing at Emergin. He holds a bachelor of science degree in electrical engineering from the University of Lowell in Lowell, Massachusetts. McAlpine may be contacted at email@example.com.
Bricon-Souf, N., & Newman, C. R. (2007, January). Context awareness in health care: A review. International Journal of Medical Informatics, 76, 2–12. Available at http://www2.chi.unsw.edu.a/pubs/souf_newman_07.pdf
Consensus Workgroup on Health Information Capture and Report Generation. (2002, June). Healthcare documentation: A report on information capture and report generation. Boston, MA: The Medical Records Institute.
Dey, A., Abowd, G., Salber, D. (2001). A conceptual framework and toolkit for supporting the rapid prototyping of context-aware applications in special issue on context-aware computing. International Journal of Human Computer Interaction, 16(2–4), 97–166.
Food and Drug Administration, HHS. (2008, Februrary 8). General Hospital and Personal Use Devices; Reclassification of Medical Device Data System. Federal Register, 73(27), 7498–7503. Available at http://www.fda.gov/OHRMS/DOCKETS/98fr/E8-2325.htm
The Joint Commission. National Patient Safety Goals and Universal Protocol for Hospitals, http://www.jointcommission.org/Standards/FieldReviews/fr_npsg_hap.htm
Koppel, R., Wetterneck, T., Telles, J. L., & Karsh, B. T. (2008). Workarounds to barcode medication administration systems: Their occurrences, causes, and threats to patient safety. Journal of the American Medical Informatics Association, 15(4), 408–423.