Clinical decision-making by the emergency department resident physicians for critically ill patients

Abstract

The application of main methodologies for clinical decision-making by residents in emergency medical practice was assessed, and issues in this area were investigated. The treatments provided to 2 611 critical patients by the Peking Union Medical College Hospital of were analyzed by independent investigators who evaluated the main clinical decision-making processes applied by the hospital residents. The application of decision-making strategies by PG1 and PG3 groups, which means the residents in first year and the third year, were compared. The patients were treated according to pattern recognition (43.0%), hypothetico-deductive reasoning (23.4%), event-driven models (19.3%), and rule-using algorithms (5.9%). A significant difference was found between PG1 and PG3 groups (χ²=498.01, P<0.001). Pattern recognition and hypothetic-deductive methods were the most common techniques applied by emergency physicians in evaluating critically ill patients. The decision-making processes applied by junior and senior residents were significantly different, although neither group adequately applied rule-using algorithms. Inclusion of clinical decision-making in medical curricula is needed to improve decision-making in critical care.

Keywords

clinical decision-making / emergency medicine / critically ill patient / resident / methodology

Introduction

Emergency medicine is one of the disciplines that most frequently employ clinical decision-making techniques. The ability to make rapid clinical decisions is essential in the practice of emergency medicine [1]. However, a recent investigation performed in another country has suggested that emergency physicians (EPs) seldom receive formal training in clinical decision-making [2]. Experienced EPs often make complex clinical decisions based on little medical information within a limited time. They seldom summarize and evaluate their clinical decision-making pathway. Furthermore, there is a paucity of literature involving effective clinical decision-making in chaotic emergency departments or its efficacy when combined with multiple patient management tasks. The current study aims to investigate the status quo of the application of common methodologies for clinical decisions made by residents in emergency medical practice and to analyze the differences between residents with different levels of training. It attempts to elucidate the application of clinical decision-making techniques in the practice of emergency medicine to improve clinical decisions of residents, thereby, reducing errors.

Materials and methods

Study design

The current study, a descriptive and retrospective research, was approved by the Human Studies Committee of the Peking Union Medical College.

Study setting and population

The present study was conducted at the Emergency Department of Peking Union Medical College Hospital, an 1800-bed urban, academic medical center. The annual emergency department (ED) visits in this hospital cover approximately 130000 patients, with a general admission rate of 3.6%. The ED has one 17-bed Critical Care Unit, two 30-bed Observation Rooms, and at most 20 stretchers can be found in the hallway.

Data were collected from January 2009 to January 2010. Of the 3 072 critically ill patients received in the Critical Care Unit at the ED, 1 413 were males and 1 659 were females. The patients were aged 5 to 99 years, with an average age of 58.35±17.42 years. A total of 301 patients died, reflecting a mortality rate of 9.80%.

Enrollment criteria

All patients with Emergency Severity Index of level 1 or 2 were admitted to our Critical Care Unit [3]. Patients in the Critical Care Unit were included in the current study if their symptoms were consistent with common emergency chief complaints defined by Rosen’s [1]. A total of 2 611 patients were enrolled, which represented 85% of the patients in the Critical Care Unit during the study period.

Study protocol

A team of three independent investigators, which consisted of an emergency medicine attending physician, a specialist from the emergency management, and a psychologist, evaluated the diagnoses and treatment of the enrolled patients during the first 3 h in the critical care unit. Clinical decision-making process and the accuracy of the methods applied were also evaluated.

The following indices were the four principal applied clinical decision-making methods [1,4-6]: (1) Pattern-recognition. Pattern recognition requires the memorization of a critical number of facts. A fact or concept is a group of objects, symbols, or events that are grouped together with common characteristics referred to by a collective name (e.g. nephrolithiasis and myocardial infarction); (2) Rule-using algorithms. Solutions to familiar problems using this level of decision-making are governed by previously memorized rules of the “if X, then Y” variety; (3) Hypothetico-deductive process. This process involves the generation, evaluation, refinement, and verification of hypothesis. In this strategy, the EP must create new solutions using previously gained knowledge to determine an answer to a difficult situation; and (4) Naturalistic or event-driven models. When presented with an unstable patient, certain therapeutic actions are necessary to stabilize the patient long before the cause of the instability is known. In this process, the EP often applies a strategy to rule out the worst-case scenario.

Grouping

Residents who work in the ED critical care unit were grouped based on their level of post-graduate emergency medicine training. The junior resident group (PG1, n = 16) was defined as resident physicians who received less than 1 year of training in emergency medicine. The senior resident group (PG3, n = 32) was defined as resident physicians who have received 3 or more years of post-graduate training in emergency medicine. The two resident groups were comparable in terms of age, gender, and educational background. Residents that fall in between the two levels of training or were rotating in other areas of the emergency department were not included in the study. All resident physicians were unaware of the present study.

Data analysis

Data were analyzed using SPSS 17.0 for Windows (SPSS Inc., Chicago, IL, USA). Descriptive statistics were used to report the frequency of clinical decision-making methodology. The difference of clinical decision-making process between the PG1 and PG3 group were compared using the χ2-test with a significance level (α) of 0.05 (2-tailed).

Results

Demographic data of the study population

The number of cases of critically ill patients who were enrolled in the present study is listed in Table 1. The patients were classified by symptoms consistent with common emergency chief complaints, as defined by Rosen’s [1].

General situation of the applied clinical decision-making methods

The four main clinical decision-making methods applied during the treatment of the enrolled patients were as follows: pattern recognition, 43.0%; rule-using algorithms, 5.9%; hypothetico-deduction, 23.4%; and event driven 19.3%. Other clinical decision-making methods that ruled out most of the dangerous situations and so on comprised 8.4%.

Comparison of applied clinical decision-making methods between the two groups

Considerable difference was found between the clinical decision-making methodologies applied by the PG1 group compared with the PG3 group (χ² = 498.01, P < 0.001) (Table 2). Further analysis indicated a significant difference in the application of pattern recognition and hypothetico-deductive processes between the two groups (P < 0.001). However, no difference was found in the application of rule-using algorithm and event-driven processes between the 2 groups.

Errors in emergency clinical decision

A total of 214 (8.2%) errors were noted in the clinical decisions made during the treatment of the 2 611 patients. The 10 most common types of clinical decision-making errors are listed in Table 3.

Discussion

EPs are often faced with many clinical decisions, and they must make these decisions within a short time frame based on limited information. Additionally, EPs usually manage 5 to 7 patients simultaneously in the overcrowded and stressful environment of EDs [1,2]. EPs at academic hospitals are expected to multitask in 10% to 20% of their working time [7,8]. Limited time frame, insufficient patient information, and uncertainty in the states of patients (stable or unstable) contribute to the need for rapid clinical decision-making in a high-risk environment. Thus, the need to learn and establish methods to reduce this risk with effective clinical decision-making is apparent.

However, few EPs summarize and survey their clinical decision-making process. Even experienced attending physicians often experience difficulty in teaching clinical decision methodology to residents and medical students. Methodology for clinical decision-making is not integrated in the Core Curriculum identified by Society for Academic Emergency Medicine[9], and there is currently limited literature on emergency clinical decision-making.

Application of methodology for clinical decision in the ED

Pattern recognition and hypothetico-deduction were the two most commonly applied clinical decision-making methods in the practice of ED residents in a typical Chinese academic center. Rule-using algorithms, which were mainly applied in patients with heart attacks, were only applied in 5.9% of patients in the present survey.

Although clinical decision-making through rule-using process include algorithms, clinical pathways, and heuristics, this technique only accounted for a small percentage of ED practice in the current study. Recent literature suggests that these methodologies are increasingly being applied in other countries [9-11]. The advantages of rule-using algorithms include saving time, reduction in physician anxiety, and improving work efficiency.

Experience of ED changes the behavior of clinical decision-making

Significant differences in clinical decision-making practice exist between the PG1 and PG3 groups. The senior residents were more likely to apply pattern recognition, whereas the junior residents more frequently employed hypothetico-deduction. Early in their practice, EPs should apply higher hierarchy of the clinical diagnostic decision-making, such as hypothetico-deduction [1]. Clinical decision-making choices may be influenced by a physician’s educational background and training levels [10]. With their accumulated clinical experience, the senior residents applied more frequently pattern recognition and event-driven methods, which require little conscious reasoning. Through pattern recognition, algorithm, and clinical pathway decision-making methods, EPs may treat the agnogenic patients rapidly without mental rigor, particularly during night-watch and weekend duties [1,2].

Need to improve clinical decision-making skills of the ED residents

The ED residents need to be trained further in the following aspects to improve their emergency clinical decision-making skills:

1. Utilization of hypothetico-deductive methods. This technique allows the resident to modify and improve continually hypothesis diagnoses to perform essential emergency functions and deductively diagnose the patient.

2. Application of heuristics, a rule-using clinical decision-making method, known as the best method of emergency clinical decision-making [1]. EPs should be familiar with algorithms for diverse acute conditions, especially in non-ideal care conditions [11]. The algorithms can reflect the diagnosis and treatment strategy, reduce misdiagnosis, and reveal life-threatening situations. However, no resident in our ED could successfully define nor apply heuristics. The present data also showed that the rule-using algorithm was seldom applied by emergency residents (5.9%), regardless of the experience of the resident (χ² = 2.91, P = 0.08).

3. Use of current computer and network technology. A previous study [12] has indicated that utilizing PDAs is more convenient for clinical diagnosis and treatment as well as frequently influences the doctors’ choice of clinical decision-making. Patients also feel more comfortable when a physician refers to a PDA rather than a textbook for diagnosis information. The database of network intoxicant, medicine, and diagnostic prompting system could reduce the errors committed by junior residents caused by imprudence [13].

4. Most important of all, ED residents should avoid incorrect clinical decisions by identifying sources of error and bias. They must be willing to admit and discuss mistakes. In Table 3, we list the 10 most common types of errors in the clinical decision-making pathway. Most errors resulted from a fault in the clinical inference process, as reported by Croskerry and Redelmeier [13-15].

Conclusions

The present study indicates that pattern recognition and hypothetico-deductive algorithms are the main methodologies applied for clinical decision-making by EPs in the evaluation of critically ill patients. Despite the significant differences between the methods applied by junior and senior residents, rule-using algorithm process, deemed as the most appropriate method for critical care, is rarely applied. Inclusion of decision-making in the medical curricula has the potential to improve decision-making in critical care and is recommended to be seriously considered.

References

[1] Chapman DM, Char DM, Aubin CD. Clinical Decision Making. In: Marx JA. Rosen's Emergency Medicine: Concepts and Clinical Practice. 6th ed. Mosby, 2005: 125-133

[2] Sandhu H, Carpenter C, Freeman K, Nabors SG, Olson A. Clinical decisionmaking: opening the black box of cognitive reasoning. Ann Emerg Med 2006; 48(6): 713-719

[3] Agency for Healthcare Research and Quality. U.S. Department of Health and Human Services. Emergency Severity Index, Version 4: Implementation handbook. 2005

[4] Wears RL. Comments on “Clinical decision-making: an emergency medicine perspective”. Acad Emerg Med 2000; 7(4): 411-412, author reply 412-414

[5] Kovacs G, Croskerry P. Clinical decision-making: an emergency medicine perspective. Acad Emerg Med 1999; 6(9): 947-952

[6] Croskerry P. Achieving quality in clinical decision-making: cognitive strategies and detection of bias. Acad Emerg Med 2002; 9(11): 1184-1204

[7] Atzema C, Bandiera G, Schull MJ, Coon TP, Milling TJ Jr. Emergency department crowding: the effect on resident education. Ann Emerg Med 2005; 45(3): 276-281

[8] Chisholm CD, Collison EK, Nelson DR, Cordell WH. Emergency department workplace interruptions: are emergency physicians “interrupt-driven” and “multitasking”? Acad Emerg Med 2000; 7(11): 1239-1243

[9] Xu TD, Pei V,Yu XZ, Smith J, Ma S, Wang Z. Developing a standardized curriculum for emergency medicine residency training:an international comparison and resources guide. Chin J Emerg Med 2006; (12): 1147-1148 (in Chinese)

[10] Ghafouri HB, Shokraneh F, Saidi H, Jokar A. How do Iranian emergency doctors decide? Clinical decision-making processes in practice. Emerg Med J. 2011<month>Apr</month><day>21</day>. [Epub ahead of print]

[11] Hall KK, Schenkel SM, Hirshon JM, Xiao Y, Noskin GA. Incidence and types of non-ideal care events in an emergency department. Qual Saf Health Care 2010; 19(Suppl 3): i20-i25

[12] Rudkin SE, Langdorf MI, Macias D, Oman JA, Kazzi AA. Personal digital assistants change management more often than paper texts and foster patient confidence. Eur J Emerg Med 2006; 13(2): 92-96

[13] Croskerry P. The importance of cognitive errors in diagnosis and strategies to prevent them. Acad Emerg Med 2003; 78: 1-6

[14] Redelmeier DA. Improving patient care. The cognitive psychology of missed diagnoses. Ann Intern Med 2005; 142(2): 115-120

[15] Croskerry P, Nimmo GR. Better clinical decision-making and reducing diagnostic error. J R Coll Physicians Edinb 2011; 41(2): 155-162

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg