Surveillance: A Strategy for Improving Patient Safety in Acute and Critical Care Units

Medical errors and adverse events are not uncommon in acute and critically ill patients.^6,7,10,11  Hospitalized patients in the United States are estimated to have more than 1 million injuries as a result of medical errors and adverse events each year.¹¹  Many of these preventable errors and adverse events result in severe disability or death.¹² 

Reducing errors and adverse events depends on accurate and timely evaluation of patients and intervention, which includes assessing the patient’s risk of both preventable and nonpreventable complications developing. The process of evaluating the patient and assessing risk is often challenging because of the nature of an acutely or critically ill patient’s illness, combined with the hectic hospital environment. As stated about the acute and critical care environment, “There are too many data for one person to process effectively.”¹³ 

Surveillance has gained renewed attention in recent years with increasing recognition of the nurse’s role in patient safety. Nursing surveillance can decrease adverse outcomes for patients not only by decreasing the number of adverse events, but also by preventing and recovering medical errors. That is, surveillance can be used to determine a patient’s risk of an adverse event developing and to identify the early stages of a potentially serious adverse event. An event becomes adverse when it results in unintended harm to a patient by an act of omission or commission rather than by the patient’s condition or underlying disease.² 

The surveillance process can also be used to identify medical errors, defined as the failure of a planned action to be completed as intended or use of the wrong plan to achieve an aim (see sidebar). Medical errors also include errors of omission, that is, an unplanned action that should have been completed.³  Although the surveillance process has theoretical benefits for assessing risk, recognizing errors, and reducing adverse events, no studies to date have evaluated the impact of bedside nursing surveillance on patients’ outcomes.

On the other hand, practicing nurses in both critical care and emergency department settings have reported using surveillance as an important strategy to promote patients’ safety by identifying medical errors and preventing adverse outcomes for patients.^6–10  Most studies on risk assessment in hospitalized patients have been at the organizational or unit level, for example, surveillance efforts related to nosocomial infection.¹⁷  Similarly, organizational characteristics that enhanced nursing surveillance, as measured by the Hospital Nurse Surveillance Capacity Profile, were significantly associated with higher quality care and fewer adverse events.¹⁸ 

The effect of nursing surveillance on patients’ outcomes has been indirectly examined in several studies conducted to evaluate the impact of nurse staffing on patients’ outcomes.^14,15  The results of these studies suggest that nursing surveillance significantly influences the occurrence of adverse events. For example, nurses were suggested to play a critical role in “rescuing” patients from complications that could result in death.^14,15  Indeed, a higher proportion of hours of care provided by registered nurses was associated with lower rates of “failure to rescue” patients from death due to pneumonia, sepsis, shock, cardiac arrest, and bleeding in the upper part of the gastrointestinal tract.¹⁴  Similarly, nurses played a role in decreasing patients’ mortality rates, suggesting that “nurses constitute the surveillance system for early detection of complications and are in the best position to initiate actions that minimize negative outcomes for patients.”^4 (p1617)

Nursing surveillance and risk assessment were not measured directly in these studies,^4,14,15  but the association of higher “failure-to-rescue” rates and more adverse events with fewer hours of nursing care suggests that evaluation of patients and nursing interventions were insufficient to prevent adverse outcomes. The Agency for Healthcare Research and Quality has quantified the failure to rescue definition as: Deaths per 1000 discharges with complications potentially resulting from care in patients age 18 to 74.¹⁹ 

Although nurse staffing ratios and hours of care have been widely studied, the relationship between hours of care and patients’ outcomes is complex and cannot be attributed to any 1 variable.²⁰  A variety of factors, including organizational characteristics and a nurse’s surveillance skills are likely to affect the nurse’s ability to prevent adverse events.

The high acuity of critically ill patients, coupled with the complexity of the clinical environment, demands implementing and evaluating innovative approaches to providing safe patient care. The lack of a systematic and purposeful approach to evaluating patients and the environment is inefficient, ineffective, and costly. Effective surveillance processes could help bedside nurses and organizations to efficiently and accurately identify patients who are at risk for potential adverse events and medical errors.

Surveillance is a complex intervention that is principally distinguished from monitoring by its intent and scope (Table 1). Monitoring of patients is an essential, but not the only, component of surveillance, which is a markedly more involved process. Surveillance is a systematic and goal-directed process focused on early identification of risk and the need for intervention. This process includes identifying at-risk patients, promptly identifying potential adverse events, as well as preventing and recovering (identifying, interrupting, and correcting) medical errors. The monitoring process centers largely on observing, measuring, and recording physiological parameters such as vital signs and laboratory values. In contrast, surveillance data include not only physiological data but also the integration of information from family members, members of the health care team, medical databases, and clinical support systems.

Surveillance includes both collecting and analyzing data. The nurse collects relevant data over time, which is then interpreted in the context of the “big picture.” Information obtained during the surveillance process is synthesized and shared with other members of the interdisciplinary health care team.

Surveillance demands that the critical care nurse selectively attend to both patient and environmental factors, in an appropriate sequence, and at the correct time. When surveillance is effective, the nurse will pick up early changes in the patient’s condition or omissions in the treatment plan. If surveillance is ineffective, a nurse may miss a trend in a relevant sign or symptom; for example, that early sepsis is developing in a patient.

In the case study (see shaded box), the nurse was conscientiously monitoring and recording physiological parameters but did not recognize that the patient’s history and vital signs (in particular the low mean arterial pressure and increased body temperature) were consistent with the potential for sepsis developing.²¹  Early recognition of a complication such as sepsis is needed for timely treatment leading to good outcomes for the patient.^21,22  In addition, failure to survey the environment resulted in the failure to detect an error in the amount of oxygen being administered to the patient and to register the husband’s concern about changes in his wife’s condition.

Critical care nurses at the bedside are uniquely positioned to decrease the occurrence of both preventable and nonpreventable adverse events that may adversely affect patients’ outcomes.²²  Examples of preventable adverse events include medication and other treatment errors, equipment errors, and procedure-related errors. In the case study, a preventable error was the patient’s incorrect FIO₂. An example of a nonpreventable adverse event in the case study was sepsis occurring after an unpredictable injury.

The patient safety movement has emphasized reducing preventable adverse events arising from medical errors. Preventable adverse advents were traditionally addressed by focusing on individual mistakes; errors and adverse outcomes are now generally accepted to result not solely from human error, but also from system failure.³  In particular, preventable adverse events are believed to result from latent, systemic deficiencies in complex care-delivery systems.^23,24 

The nurse’s role in preventing medical errors from adversely affecting patients has been described in several models that recognize the systemic and human factors leading to situations that may become adverse events.^16,25  One model, adapted from the Eindhoven Model of error recovery,²⁶  holds that the key to averting adverse outcomes for patients is the use of “defense” mechanisms that recognize and interrupt dangerous situations¹⁶  (see Figure). An example of such a defense mechanism in clinical settings is nurses’ systematic surveillance of patients. Thus, effective surveillance may serve as both a defense mechanism to block dangerous situations and a strategy to recover errors.

Much attention in the bedside patient safety movement has focused on efforts to implement structures and processes that ultimately improve the surveillance process. These structures and processes include surveillance checklists, interdisciplinary rounds, clinical information systems, and clinical decision-support systems. Each example represents an organizational or system factor that could prevent the development of a dangerous situation as described in the modified Eindhoven Model.¹⁶  In addition, using these structures and processes allows nurses at the point of care multiple opportunities to recover errors before they harm a patient.¹⁶  See Table 2 for examples of ways to integrate surveillance activities into routine nursing responsibilities.

Surveillance checklists may guide bedside nurses in the surveillance process and cue them to risk factors associated with certain high-frequency complications (eg, postoperative complications). These checklists would include evidence-based risk factors that could lead to adverse events such as sepsis, pneumonia, and bleeding in the upper part of the gastrointestinal tract. In fact, using checklists effectively decreases the occurrence of common, costly, and potentially lethal adverse events such as catheter-related bloodstream infections in the ICU.²⁷ 

Medication surveillance checklists, such as those that use a head-to-toe or systems-based approach to collecting a medication history would help nurses by offering cues to avoid adverse events associated with omitting important medications. Indeed, using these checklists has been suggested to help teach nursing students and to cue even experienced nurses to commonly omitted medications, such as eye drops and inhalers.²⁸  (See Table 3 for an example of a checklist for change-of-shift report.)

Nurses conduct surveillance by actively participating in rounds with the interdisciplinary team. These rounds allow nurses and other team members to disseminate information they have gathered and analyzed, thus contributing information and their unique perspective. One unique benefit of bedside rounds is performing real-time surveillance of both patients and their environment. In addition, bedside rounds increase the opportunity to involve patients and their family members in surveillance. Although bedside rounds are perceived as time-consuming, they take no more time than “sit down” or other rounding formats.^29,30  Family-centered rounds, common in pediatric units, are perceived by members of the health care team as fostering communication among team members and increasing family members’ involvement and understanding.²⁹ 

Current methods of delivering patient care rely heavily on humans recognizing and interrupting incidents that could harm patients rather than using processes that systematically reduce dangerous situations from happening in the first place.^16,24,25  High-risk patients in high-risk environments such as ICUs require implementation of mechanisms to identify risk factors and alert clinicians to possible adverse events. These mechanisms could be categorized according to the adapted Eindhoven Model as human and technical factors that prevent errors from reaching the patient or system defense mechanisms that prevent a dangerous situation from reaching the patient.¹⁶ 

Clinical information systems/ clinical decision-support systems are examples of information-technology systems that can assist critical care nurses in the surveillance process.^29,31,32  Clinical decision-support systems are designed to aid clinical decision making by allowing clinicians to match the characteristics of an individual patient to a computerized clinical knowledge base and by presenting patient-specific assessments or recommendations to the clinician or patient for a decision. These technologies gather and present clinical information in an organized way, providing the nurse with a more efficient way to analyze data. Clinical support systems also aid the surveillance process by providing information on patient-specific safety issues such as allergies.

Effective nursing surveillance can be hindered by many systemic and human factors. Systemic factors include poor staffing, inadequate resources, poor collaboration among members of the health care team, failure to include the patient and the patient’s family in decision making, and underuse of existing technology. Human factors include nurses’ education and skills to perform the surveillance process. Nurses, particularly novice nurses, may be so overwhelmed with tasks and time constraints that they cannot attend to the “big picture,” which effective surveillance requires.³³ 

Resources such as time and access to data are critical to performing comprehensive assessments. Nurses who are distracted, interrupted, or responsible for nonnursing functions are unlikely to have adequate time to devote to the surveillance process.³⁴  Collaboration with other members of the health care team, including the patient and the patient’s family, influence patients’ outcomes, potentially via the impact of collaboration on effective surveillance.³⁵  Unit structure or policies that inhibit collaboration, such as restrictions on family visiting or the absence of nurses from daily rounds, may seriously impede the surveillance process and adversely affect patients’ outcomes. The effects of rounds and other unit structures and processes have not been well studied.

Surveillance depends on nurses being able to access data from multiple sources. Nurses can be supported in conducting surveillance by many existing technologies, such as bedside monitors and computerized information systems with trending functions. However, bedside nurses may be unaware of such capabilities or not know how to use them for maximal benefit. This failure to leverage available technologies can serve as a barrier to an otherwise effective surveillance process.

To optimize current surveillance practice, research is needed to describe effective surveillance processes and to evaluate the impact of effective surveillance on patients’ outcomes, including the prevention of adverse events and medical errors. Insight into effective surveillance will require both qualitative and quantitative studies on the processes used by nurses as they systematically evaluate and analyze data on patients and their environment.

Research on surveillance will require innovative approaches that allow more precision than current methods such as observing or videotaping surveillance processes. Eye-tracking is a current technology that has been used successfully in engineering and other sciences, for example, to study behaviors of automobile drivers during risky traffic situations.³⁶ 

Eye-tracking technology allows researchers to track nurses’ actual eye movements during the surveillance process to identify the data to which a nurse selectively attends and the sequence in which the nurse attends to those data. Understanding the selective attention processes used by nurses can provide unprecedented insight into the surveillance process. Eye-tracking technology has been used in a simulated clinical setting to evaluate nurses’ ability to recognize errors in patient identification.³⁷ 

Surveillance is an important strategy used by nurses to improve patients’ outcomes, particularly their safety. Patients in critical and acute care settings are at risk for medical errors and adverse events, making effective surveillance even more imperative in these settings.⁶  Although surveillance has been identified as an important nursing intervention to prevent medical error and adverse outcomes, no studies have described effective surveillance patterns or the effect of surveillance on patients’ outcomes. Thus, no evidence base is available to guide nurses in performing this critical and potentially lifesaving process.

Nurses play a pivotal role in influencing patients’ outcomes, prompting recommendations by the Institute of Medicine to redesign the nursing work environment.²³  This call for a redesign focuses specifically on reducing errors and adverse events associated with the failure of effective surveillance.^23,24  The Institute of Medicine’s report calls for the implementation of safety “bundles” that include components such as work processes and organizational culture. The interventions described here, such as checklists and interdisciplinary rounding, are examples of these components that could improve surveillance and ultimately increase patients’ safety.

The authors acknowledge Claire Baldwin for her editorial assistance.