ORIGINAL CONTRIBUTION
The Association Between Emergency Department Crowding and the Disposition of Patients With Transient Ischemic Attack or Minor Stroke Maxim Ben-Yakov, MDCM, Moira K. Kapral, MSc, MD, FRCPC, Jiming Fang, PhD, Shudong Li, MSc, Marian J. Vermeulen, MHSc, and Michael J. Schull, MSc, MD, FRCPC
Abstract Background: Emergency department (ED) crowding has been associated with adverse events, including short-term death and hospitalization among discharged patients. The mechanisms are poorly understood, but may include altered physician decision-making about ED discharge of higher-risk patients. One example is patients with transient ischemic attack (TIA) and minor stroke, who are at high risk of subsequent stroke. While hospitalization is frequently recommended, little consensus exists on which patients require admission. Objectives: The authors sought to determine the association of ED crowding with the disposition of patients with minor stroke or TIA. Methods: This was a retrospective cohort study of prospectively collected data from the Registry of the Canadian Stroke Network at 12 EDs in Ontario, Canada, between 2003 and 2008, linked to administrative health databases. A hierarchical logistic regression model was used to determine the association between crowding at the time the patient was seen in the ED (defined as mean ED length of stay) and patient disposition (admission/discharge), after adjusting for patient and hospital-level variables.
From the Division of Emergency Medicine, Department of Medicine, University of Toronto (MB, MJS), and the Department of Emergency Medicine Sick Kids Hospital (MB), Toronto, Ontario; the Division of General Internal Medicine, University Health Network, Institute for Clinical Evaluative Sciences, Institute for Health Policy, Management and Evaluation, Department of Medicine, University of Toronto, Toronto, Ontario, and the Canadian Stroke Network, Ottawa (MKK), Ontario; the Division of Emergency Medicine, Department of Medicine, University of Toronto, Sunnybrook Research Institute, Institute for Health Policy, Management and Evaluation, University of Toronto, and the Institute for Clinical Evaluative Sciences (MJS, MV, SL, JF), Toronto, Ontario; and the Clinical Epidemiology Unit, Sunnybrook Health Sciences Centre (MV), Toronto, Ontario, Canada. Received September 10, 2014; revision received February 13 and May 6, 2015; accepted May 25, 2015. Presented at the Society for Academic Emergency Medicine Annual Meeting, Atlanta, GA, May 2013; and the Association of Emergency Physicians Annual Meeting, Vancouver, BC, June 2013. The Registry of Canadian Stroke Network/Ontario Stroke Registry: This work was supported by an operating grant from the Canadian Stroke Network. The Registry of the Canadian Stroke Network is funded by an operating grant from the Ontario Ministry of Health and Long-Term Care. The Institute for Clinical Evaluative Sciences is supported by an operating grant from the Ontario Ministry of Health and Long-Term Care. The results and conclusions are those of the authors and are not attributed to any of the sponsoring or funding agencies. The funding agencies had no role in the design or conduct of the study or the collection, management, analysis, or interpretation of the data. The manuscript was reviewed and approved by the publications committee of the Registry of the Canadian Stroke Network. Institute for Clinical Evaluative Sciences: This study was supported by the Institute for Clinical Evaluative Sciences, a nonprofit research institute funded by the Ontario Ministry of Health and Long-Term Care. The opinions, results and conclusions reported in this paper are those of the authors and are independent from the funding sources. No endorsement by Institute for Clinical Evaluative Sciences or the Ontario Ministry of Health and Long-Term Care is intended or should be inferred. MKK is supported by a Career Investigator Award from the Heart and Stroke Foundation of Ontario, Ontario Provincial Office. MJS is supported by funding from the Applied Chair in Health Services and Policy Research from the Canadian Institutes of Health Research. The authors have no additional relevant financial information or potential conflicts to disclose. Supervising Editor: Bema Bonsu, MD. Address for correspondence and reprints: Maxim Ben-Yakov, MDCM; e-mails:
[email protected] or
[email protected].
© 2015 by the Society for Academic Emergency Medicine doi: 10.1111/acem.12766
ISSN 1069-6563 PII ISSN 1069-6563583
1145 1145
1146
Ben-Yakov et al. • ED CROWDING AND THE DISPOSITION OF PATIENTS WITH TIA Results: The study cohort included 9,759 patients (4,607 with TIA and 5,152 with minor stroke); 49.5% were discharged from the ED. The mean (SD) age of study patients was 70.78 (13.40) years, with 52.9% being male, 37.3% arriving by emergency medical services, and 92.3% triaged as emergent or urgent. Greater severity of ED crowding was associated with a lower likelihood of discharge, regardless of ED size. Conclusions: These results suggest that crowding may influence clinical decision-making in the disposition of patients with TIA or minor stroke and that, as crowding worsens, the likelihood of hospitalization increases. ACADEMIC EMERGENCY MEDICINE 2015;22:1145–1154 © 2015 by the Society for Academic Emergency Medicine
E
mergency department (ED) crowding is a concern for health care systems in many countries1–4 and affects clinically important outcomes, such as time to thrombolysis in acute myocardial infarction, rates of medical errors, patient satisfaction, and mortality.5–14 The stress of crowding may influence the physicians’ clinical decision-making process. For example, a key cause of ED crowding is difficulty accessing inpatient beds, a situation known as “access block,”1,2,5,15,16 which could affect physicians’ decisions to admit or discharge patients from the ED. A recent study of pediatric ED patients with asthma and gastroenteritis found that crowding was associated with a lower likelihood of admission.17 However, the mechanisms by which crowding affects disposition decisions for higher risk conditions in adults is poorly studied. To address this question, we chose to examine transient ischemic attack (TIA) and minor stroke, as these conditions are associated with a high risk of subsequent adverse events. In addition, although routinely encountered by emergency physicians (EPs), these conditions lack standardized ED disposition guidelines.18–22 Expedited disposition decisions may influence outcomes; previous studies have found that nonadmitted TIA patients are less likely to receive effective evidencebased therapies.23–29 However, it is unknown whether ED crowding influences disposition decisions for patients with TIA or minor stroke. Our objective was to study the association between ED crowding (acuity-standardized mean ED length of stay [LOS]) and the ED decision to discharge patients with TIA or minor stroke. Our secondary objective was to evaluate subsequent adverse events (readmission, stroke, or death) among those discharged from the ED. Our hypothesis was that patients presenting during periods of crowding would be more likely to be discharged than those presenting during periods without crowding. METHODS Study Design We conducted a retrospective analysis of Ontarians with acute minor stroke or TIA seen in the EDs of any of the 12 regional stroke centers between July 1, 2003, and March 31, 2008. Ethics approval for the study was provided by the Sunnybrook Health Sciences Research Ethics Board and by the research ethics board of each participating site.
Study Setting and Population Designated stroke centers in Ontario, Canada, are a part of regular acute care hospitals with EDs that treat all comers but have specific skills and resources for stroke care. These centers are generally tertiary care centers with stroke care expertise and the capacity to administer thrombolysis and perform neurosurgery. Most stroke patients presenting to EDs at stroke centers are assessed and managed by EPs, with neurology consultants available if required. Most Ontario emergency medical services systems have standing protocols for ambulances to bring suspected acute stroke patients directly to stroke center EDs based on specific clinical criteria. Ontario residents have universal access to hospital care, physicians’ services, and postdischarge health services. The Registry of the Canadian Stroke Network/Ontario Stroke Registry (OSR) collects detailed clinical information on all consecutive acute stroke patients seen at regional stroke centers.30 Data on patients seen in stroke center EDs with stroke or TIA are entered into the OSR regardless of disposition and whether they are ultimately managed by EPs or neurologists. Chart abstraction is performed by trained neurology research nurses, and validation by duplicate chart abstraction has shown excellent agreement for key variables including age, sex, and stroke type.31 The registry provides information on key clinical variables such as presenting symptoms, stroke severity, initial investigations (imaging and electrocardiogram [ECG]), consultations, and final dispositions, as well as processes of care such as referral to a stroke prevention clinic at discharge. The registry is prescribed under provincial privacy legislation so that chart abstraction can be performed without individual patient consent.30 The registry data is housed at the Institute for Clinical Evaluative Sciences (Toronto, ON) where the data are linked to population-based administrative databases using unique anonymized patient identifiers. For this study, we linked the stroke registry to the Canadian Institute for Health Information Discharge Abstract Database32 to identify readmissions, the National Ambulatory Care Reporting System33 to identify initial ED visits, and the Registered Persons Database34 to identify deaths. These data sets were linked at the individual patient level using unique encoded identifiers and analyzed at the Institute for Clinical Evaluative Sciences. We included patients who had final diagnoses of TIA or minor ischemic stroke (defined as a Canadian
ACADEMIC EMERGENCY MEDICINE • October 2015, Vol. 22, No. 10 • www.aemj.org
Neurological Scale [CNS] score of 9 or greater) on their first ED visits. Stroke severity is measured using the validated CNS, which measures level of consciousness, aphasia, orientation, and motor strength and has a maximum score of 11.5; scores of 8 and above generally indicate a “minor stroke.”35 The CNS is a simple tool for the evaluation and monitoring of acute stroke patients— similar to the National Institutes of Health Stroke Scale (NIHSS). There is a simple linear relation of CNS to NIHSS (NIHSS = 23 2 9 CNS).36 The CNS uses 10 domains, which include level of consciousness; orientation; speech; and motor function of the face, arm, and leg. It has been shown to be a valid and reliable tool for the evaluation of acute stroke patients and has been tested against other stroke scores.36–38 We further categorized TIA patients as high risk when they were documented as having weakness or aphasia and presence of any of the following: diabetes mellitus, hypertension, or age greater than 60 years. Those with minor strokes were grouped into those with higher CNS scores (>10.5 to 11.5) versus lower CNS scores (>9 to 10.5). We excluded patients with moderate or severe strokes as these patients are usually admitted for inpatient management and investigation. We also excluded patients who were transferred from another ED or hospital and all patients under 18 years of age. If a patient appeared more than once in the stroke registry, only the first visit was included as the index visit in our analysis. Study Protocol The exposure of interest was ED crowding at the time the patient presented to the ED. This was calculated as the mean ED LOS in hours for patients of similar acuity seen in the same ED on the same shift as the index patient.39 This measure has previously been shown to be associated with subsequent risk of death and hospitalization among discharged ED patients12 and is considered the criterion standard for the measurement of throughput and crowding.40 Acuity was grouped according to the Canadian Triage Acuity Scale (CTAS) as high (level 1 = resuscitation, level 2 = emergent, or level 3 = urgent) or low (level 4 = less urgent or level 5 = non-urgent). CTAS is a countrywide triaging system, based on the Australian Triage Scale and akin to the Emergency Severity Index. It has been shown to be reliable and valid at predicting the need for admission, LOS, and the need for urgent intervention.41 All Ontario EDs use the CTAS, with common training protocols for triage nurses.42 We calculated mean ED LOS among patients of similar triage acuity since they are usually triaged to the same treatment zones in the ED and tend to experience similar delays in the receipt of investigations and treatments. If no similarly triaged patients presented during the shift, LOS was calculated based on patients of the same CTAS category grouping in the same ED on the previous day. An ED shift was defined as an 8-hour interval divided into day (8 AM to 4 PM), evening (4 PM to midnight), and night (midnight to 8 AM), mean ED LOS was grouped as 0 to
