TL;DR
This paper investigates the complex nature of registration errors in emergency departments (EDs), revealing that these errors, often stemming from misidentification, are not isolated incidents but rather systemic issues arising from a confluence of factors including high workload, time pressure, and the use of "workarounds." Understanding these dynamics is crucial for developing effective information technology and policy solutions to improve patient safety in these high-stakes environments.
What Problem Does This Paper Address?
The emergency department (ED) is a uniquely chaotic and high-pressure environment where timely and accurate patient identification is paramount for safe and effective care. However, the process of registering patients in the ED is fraught with potential errors, particularly those related to misidentification. These errors can range from assigning multiple medical record numbers to a single patient to completely misattributing patient information. While the consequences of such errors can be severe, leading to incorrect diagnoses, treatments, or delayed care, they are often under-reported, especially when caught before an adverse event occurs. This paper aims to delve into the "nature and occurrence" of these registration errors, exploring their root causes within the complex ED workflow and the interplay of human factors, technology, and organizational pressures.
Methods
The researchers employed a qualitative, ethnographic approach to study ED registration. This involved:
- Observation: Two researchers (one a physician not involved in patient care) spent three months observing patient registration processes in a large urban ED. Observations focused on group behaviors, cognitive workflows, information transfer, and decision-making.
- Interviews: Semi-structured, open-ended interviews were conducted with ED registration staff to understand their experiences, coping mechanisms, and perspectives on policies, procedures, and the challenges they face.
- Data Analysis: A basic workflow model of ED registration was synthesized. Each identified error was analyzed to trace its origin and pinpoint where the ideal workflow broke down. This was supplemented by reviewing ED registration policies and procedures and analyzing interview data to understand workarounds and shortcuts. The analysis was guided by a modified "grounded theory" approach, aiming to develop a theory inductively from the data.
Datasets
Not discussed in the paper. The study relies on observational and interview data collected directly from the study site.
Key Insights and Ideas
- Registration as a "Natural Laboratory for Error": The ED's inherent complexity, high patient volume, and time pressures create an environment ripe for errors, with registration being a critical, yet vulnerable, first point of contact.
- Misidentification is a Major Culprit: The study highlights misidentification as a recurring and significant problem, manifesting as multiple medical record numbers, wrong patient identification, or incorrect addresses.
- "Workarounds" and Shortcuts are Prevalent: To cope with the immense workload and time constraints, ED registration staff often resort to "workarounds" – strategies that bypass standard procedures, which, while improving efficiency in the short term, significantly increase the risk of errors.
- Systemic Failures, Not Just Individual Mistakes: Errors are rarely due to a single individual's lapse. Instead, they emerge from a breakdown in the overall system, often a result of the cognitive load placed on individuals, compounded by technological limitations and organizational pressures. This aligns with the "Swiss cheese" model of error.
- Technology Can Be a Double-Edged Sword: While IT solutions are often proposed to fix these issues, the study suggests that generic systems may be ill-suited for the dynamic ED environment. The computer system itself, if not user-friendly or too rigid, can actively contribute to errors.
- Near-Misses are Common and Informative: The study identified four cases of "near-misses" – errors caught before they caused harm. These incidents, though not resulting in adverse events, provide crucial insights into the system's vulnerabilities.
Key Results
The study identified four distinct classes of registration errors, all of which were eventually recognized and rectified by existing "filters" within the system, preventing adverse patient outcomes. These near-misses included:
- Case 1 & 2: Patients being assigned multiple medical record numbers due to incomplete or incorrect initial identification information, leading to a lack of access to prior medical history.
- Case 3: A patient's address not being updated during a subsequent visit, highlighting a failure to reconfirm demographic information to save time.
- Case 4: Two patients with similar characteristics being brought in simultaneously, resulting in their identities being mismatched on charts and wristbands due to rushed registration and a failure to follow safety protocols.
Limitations and Future Work
Limitations:
- Limited Scope of Observation: Observations were conducted only during morning and afternoon shifts, potentially missing error patterns that might occur during other times.
- Focus on One Site: The study was conducted at a single urban academic tertiary care hospital, which may limit the generalizability of findings to other types of healthcare settings.
- Under-reporting of Errors: The authors acknowledge that misidentification incidents are highly under-reported, meaning the true prevalence of these errors may be higher than captured.
Future Work:
- Improving Technology Usability: The study strongly advocates for improving the usability of registration computer systems and providing better staff training to make them more "user-friendly" and less prone to facilitating errors.
- Addressing Workload and Staffing: The paper emphasizes the need to address the intense work environment and high patient-to-staff ratios in ED registration. Increasing registration personnel could reduce workload and promote adherence to procedures.
- Bedside Registration: Implementing wireless computers for bedside registration is suggested as a way to help registration personnel focus on one patient at a time, potentially reducing errors.
- Usability Studies: Conducting further usability studies on clinical information systems is recommended to fully understand and address the nature and extent of problems related to technology in healthcare settings.
- Policy-Level Interventions: The authors call for policy-level attention to these issues, particularly regarding staffing levels in ED registration.
Summary updated