Why the difference between continuous surveillance and patient monitoring matters

The prevention of hospital-acquired illnesses (HAI) has come under greater regulatory scrutiny from the Centers for Medicare & Medicaid Services (CMS). Most notably, the Hospital-Acquired Conditions Reduction Program (HACRP) penalizes hospitals for high rates of such HAIs as ventilator-associated injuries, preventable readmissions, and sepsis.¹

Enhancing the safety and outcomes of patients and avoiding costly penalties begins with comprehensive continuous clinical surveillance practices. Now, clinical leaders reading this may push back with an argument that, yes, some reforms could be made at the margins, but such practices already exist within their health systems.

Here is where we get into an issue of semantics. In a forthcoming concept analysis to be published by the Healthcare Information and Management Systems Society’s Online Journal of Nursing Informatics (OJNI), I argue that clinicians often use the terms monitoring and surveillance interchangeably.

“In today’s acute-care environment, multiple internal and external factors give clinical surveillance characteristics distinct from those other terms and have changed how nurses apply critical thinking and decision-making to patient care and safety.”²

Defining attributes of the "clinical surveillance" concept I found frequently referenced in literature include

1. Attention
2. Timeliness
3. Recognition
4. Intuition
5. Analysis
6. Action
7. Collaboration

While monitoring may have one or more of these attributes, clinical surveillance requires all seven. The core of surveillance is real-time attention, recognition and analysis that allows for action and collaboration. It is a paradigm shift from a retrospective clinical response to prospective anticipation and planning.

This prospective shift may take one of two forms:

1. ARTIFICIAL INTELLIGENCE (AI), where the machine (software capability of the monitor) learns the physiological trends for a particular patient and notifies the clinician when a trend takes a different trajectory, or
2. A RULES-BASED phenomenon, where a human (clinical expert) sets the rules or combination of rules for which the clinician must be notified, therefore eliminating extraneous non-actionable alarms

The distinction between patient monitoring and continuous clinical surveillance is more than a matter of language. Understanding and applying this knowledge in clinical practice could potentially support patient safety initiatives and compliance with regulatory requirements.

Perception versus reality

An analysis that featured interviews with more than 30 clinical informatics executives uncovered a common perception that because patients in critical care units were attached to bedside monitoring devices, clinical surveillance was an established standard of care.³ However, Giuliano notes that “[s]urveillance and monitoring each represent a distinct process within patient care. Monitoring involves observation, measurement, and recording of physiological parameters, while surveillance is a systematic, goal-directed process based on early detection of signs of change, interpretation of the clinical implications of such changes, and initiation of rapid, appropriate interventions.”⁴ This author takes Giuliano’s definition a giant step further, evidenced by the seven clinical surveillance attributes (listed in the previous section) that must be present within a clinical scenario in order to label it as true clinical surveillance.

Adversely, when characterizing what clinical surveillance is not, we examine what patient monitoring actually is. Patient monitoring is both fragmented and episodic, capturing a patient’s condition in ways that are potentially dangerously narrow and incomplete. Most patient monitoring practices involve vital sign spot checks and responses to notifications sent from individual physiologic devices. Malkary notes that within a “MED-SURG environment, nurses conduct episodic monitoring several times per day or on an as-needed basis. This represents a gross under-sampling of what is going on with the patient, which could result in missing subtle changes in the patient’s condition.”⁵

Additionally, patient monitoring inherently assumes that an HAI, such as opioid-induced respiratory depression (OIRD), will be caught during the narrow windows of time that a clinician is visually observing a patient; in truth, spots checks can leave patients unmonitored 96% of the time.⁶ Even if a clinical team member (or patient) were to catch deterioration, the danger is active, present and likely requires emergency rescue or escalation to an intensive care unit.

A review of 357 claims from the Anesthesia Closed Claims Project database involving incidents which occurred between 1990 and 2009 revealed that in 42% of confrmed OIRD events, “the interval between the last nursing assessment and the detection of respiratory depression was less than two hours, and in 16 [%] of the cases, it was within 15 minutes.”⁷

In contrast, continuous clinical surveillance allows the clinical team to see the “forest for the trees” through real-time, continuous data flow from multiple sources that can be filtered and intelligently analyzed for significant trends and prospective intervention. There also is growing evidence that continuous clinical surveillance facilitates interventions long before a life-threatening event occurs.⁸

The significant difference rests in the acquisition and analysis of essential patient data. Whereas patient monitoring is episodic and littered with potentially consequential gaps, continuous clinical surveillance is ongoing, comprehensive and assimilated for prospective clinical decision-making.

The power of prediction

Patient monitoring depends on a team of clinicians, working as individuals, to observe the state of the patient’s health at a particular moment in time, and leans heavily on the (often technical) threshold violations of individual devices. By contrast, clinical surveillance is team-based, allowing multiple caregivers to assess a holistic portrait of multiple patients from a centralized location or via mobile alarm notifications. Because continuous clinical surveillance relies on multivariate rules to correlate data; identify clinically relevant temporal trends, sustained conditions, reoccurrences, and combinatorial indications; and create new early warning alarms, clinical team members potentially can quickly recognize and respond to signs of distress before the patient’s health is compromised.

“Data collection and analysis are further enhanced when including methods for disseminating, analyzing and distributing these data. These features facilitate better patient care management and clinical workfow by allowing patients to be monitored remotely.” ⁹

Conclusion