High-risk patients, who require respiratory support or monitoring, are found not only in high-acuity areas. Identifying the patients who are vulnerable to deterioration is critical to patient safety measures that emphasise timely interventions over costly emergency rescues. Ubiquitous data in a predetermined clinical workday demands streamlined workflow through device interoperability, transparency, algorithmic trends, rule-based theorems, and data standardisation. These are the necessary tools for today’s clinical toolbox. Without Meaningful Device Integration (MDI) in the contemporary complex hospital setting, safe patient care is relinquished to a second priority, as the clinician attempts to place all the disparate vital data points together, first.
Continuous clinical surveillance applies advanced analytics to real-time medical device data. Surveillance is evaluated against previous benchmark indices from the Electronic Health Record (EHR) and is used to intercept adverse events and prevent costly care escalations. For this, continuous streams of data is captured through MDI from multiple sources, such as ventilators, physiological bedside monitors, livestreaming ECG waveforms, even patient alerts from the bed itself. These data streams combined with retrospective data stored in EHRs and then filtered through an intelligent, rules-based engine that uncovers clinically relevant trends and deviations in the patient’s condition. This evaluation is presented in a significant way for health systems to integrate real-time patient safety into clinical workflow.
There are seven attributes to using clinical surveillance in the MDI model; attention, timeliness, recognition, intuition, analysis, action, collaboration1. This rigorous clinical approach to delivery of safe patient care can only be implemented if the technology at hand is integrated for the purpose of drawing meaningful data. In other words, the action from the clinician is only as good as the data at hand.
Continuous clinical surveillance, advanced analytics and timely intervention based upon evolving trends in a patient’s condition is attainable in today’s hospital care setting.
MDI and Continuous Surveillance
Continuous surveillance is an integrated, systematic, goal-directed process in which clinicians apply streaming live patient data, bench-marked against the average hospital EHR disclosed data to prompt real-time decision-making based upon evolving patient trends.
One of the objectives of analytics is to surveil (a reconnaissance quietly taking place in the background) of seemingly unrelated sources of data to prevent the onset of an adverse event that would not normally be visible by observing a single parameter or multiple parameters individually. Predictive models and rules-based algorithms using multiple sources of data can help clinicians anticipate adverse events more reliably than data from a single source, bringing the unique clinical picture into focus. It is a realtime/prospective approach versus a retrospective response to patient care.
Data collection and analysis is further enhanced when both the benchmarked data and the smart alarm signals are placed in the analytical equation. The strength in the newly combined algorithm facilitates patient care management and clinical workflow.
However, creating the environment to facilitate continuous surveillance and intervention is challenged by the fact that medical device data is often isolated, with each siloed device having unique communication protocols, physical connections, update rates, and terminology.
Data from MDIs helps in decision making, in the standard patient care management processes, such as charting, and proactive continuous surveillance. Enabling medical devices for continuous surveillance requires three primary capabilities which should be included in any MDI initiative.
Assured delivery of data. To support continuous surveillance, the communication pathway from the bedside medical device to the recipient must guarantee delivery of the data within a specified time frame. In order to guarantee delivery, the system must continuously monitor that communication pathway and report any delivery impediments.
Two-way communication. This capability ensures that data delivery and verification does not impede or otherwise interfere with the medical device operation. This is of particular importance when exploring external control of medical devices or when alarm data are communicated per continuous surveillance.
Ensuring data integrity. Algorithms cleared for continuous surveillance or data interpretation must be validated for all intended operational scenarios of the medical device. Data security, hostile attacks on data, medical device, denial of service, and ransom ware all have the potential to impact data integrity and these requirements must be identified and detailed through specific scenarios and validated through testing.
Clinical Value from Device Integration
Clinical surveillance utilises multivariate, continuous, real-time data from multiple monitoring devices; applies advanced analytics to provide a quantitative and qualitative estimate of a patient’s condition over time; and communicates clinically relevant alerts to the appropriate clinician.
The EHR in the Medical Device Integration model
While the widespread adoption of EHRs has largely resolved the challenges of independent data capture and have mitigated issues related to clinicians’ access to critical information, its attributes lack real-time trend capability. The central role the EHR plays in the day-to-day clinical operations practically requires that peripheral technologies—ranging from medical devices and telehealth to financial and administrative solutions—integrate with the system. Although EHRs were never designed to accommodate real-time data analytics, investing in a proactive MDI solution makes these systems an ideal hub for continuous surveillance.
Clinical surveillance compliments hospitals’ EHR capabilities by providing bi-directional integration features through middleware that resides in the space between point-of-care devices and the EHR. For example, a clinical surveillance solution can capture historical data from the EHR and correlate it with real-time streaming data from streaming devices, including heart rate, heart rhythm variation, oxygenation levels and blood pressure. The combination of high-fidelity data with multivariate, EHR information provides a holistic and complete source of objective information on a patient that can be used for prediction and clinical decision making prospectively.
Analytics in Medical Device Integration
The transformative power of MDI mitigates the shortcomings of conventional monitoring practices, including alarm fatigue, significant monitoring gaps, and data delivery delays.
Analytics based on multiple sources of physiological data can help offset the problem of alarm fatigue by filtering out false or artifact signals that typically invade the high-fidelity data at the core of continuous surveillance. The use of data for display and analysis, predictive analytics, rules-based algorithms, as well as the ability to process data collected at the point of care to create new information also drives data collection rates. The ability to retrieve data at variable rates, including at the sub-seconds level, requires the vendor to demonstrate that it has mitigated the risk associated with communicating higher frequency data for alarms and analysis—even patient monitoring and intervention.
This raises critical questions that pertain to patient safety and the level of risk assumed by the hospital. How do patient documentation needs differ from real-time patient intervention needs? What is real-time data flow and what is not? Because data used for real-time intervention, like clinical alarms, impact patient safety, any delay in their delivery to the correct individuals can have deleterious effects. Thus, it is important to understand the implications of requirements on data delivery latency, response and integrity2.
The Future of Medical Device Integration
Universal medical device standards won’t happen overnight, though it has been interesting to note manufacturer’s slow migration to a more standardised approach. Logistics and practicality rule the day in a world with steep costs in investment, development, acquisition, and regulation. This reinforces the need to have a comprehensive and forward-looking approach to selecting an MDI and middleware provider that can support the technical and clinical needs of your healthcare organisation6. A broader clinical view of the future solution is to include laboratory blood work, radiological findings, and pharmaceutical interventions, into the device integration model, to provide a total operational framework for the clinician.
Data Delivery, Communication and Integrity
“High Reliability: The Path to Zero Patient Harm”3, presents a root challenge that lies with frequent clinical interruptions affecting the consistency and reliability of patient care. Siloed data makes it difficult for the clinician to draw meaningful conclusions for a patient care plan. Complexity of disparate devices warrants consistent integration of the right information to be delivered to the right person at the right time.
How can hospital leaders expect clinical providers to meet patients’ needs and perform their work proficiently if the system provides only fragmented, episodic data to work with? Clinicians must have real-time, continuous data flow and analysis for accurate trending to occur. Surveillance infers that no clinical anomaly channels the acknowledgement/intervention filter; in this case, the resources are knowledge and technology4. To support active patient monitoring and verified delivery of data, the communication pathway from the bedside medical device(s) to the recipient must guarantee delivery of the data within a specified time frame.
In order to guarantee delivery, the system must continuously monitor that communication pathway and report when data are impeded or otherwise delayed beyond a maximum acceptable limit on latency and throughput.
Communication through Middleware
Two-way communication of data ensures that data delivery and verification does not impede or otherwise interfere with the medical device operation. This is of importance when exploring external control of medical devices or when alarm data are communicated per active patient.
In middleware systems cleared for active patient monitoring, the ability to transform the data is possible. Algorithms for performing transformations, calculation of tertiary results, and otherwise interpreting data must pass muster and be validated for all intended operational scenarios of the medical device, including failure modes. Data security, hostile attacks on data, medical device, and denial of service, and ransomware all have the potential to impact data integrity and these requirements must be fleshed out through specific scenarios and validated through testing5.
For hospitals and health systems, especially those that are breaking ground on a net-new MDI program, the formidable task list that comes with any MDI initiative requires the input and expertise of a project team, which ideally, should be comprised of leadership from myriad departments, including IT, networking, facilities, clinical staff, and biomedical engineering.
Jahrsdoerfer, M. (2019). Clinical Surveillance. A Concept Analysis: Leveraging real-time data and advanced analytics to anticipate patient deterioration. Bringing Theory into practice. https://www.himss.org/library/clinical-surveillance-concept-analysisleveraging-real-time-data-and-advancedanalytics-anticipate
DuPree, E. (2016). High reliability: the path to zero harm. Healthcare Executive (Jan/Feb), 66-69. http://www.jcrinc.com/assets/1/7/PathToZeroHarm.pdf