a b s t r a c t

Background: The diagnostic cascade in the emergency department (ED) has not been fully elucidated. Aim

To inspect whether the usage of consulting medicine and imaging contributes to hospital outcomes. We also pro- pose a theoretical model for better understanding the diagnostic cascade of needless medical testing.

Methods: A two-level study was conducted. The first local phase was a retrospective archive study that was con- ducted between 2014 and 2017 in a tertiary hospital. We extracted the number of requested imaging and con- sultations for each patient, and main time lags were calculated. The second-phase (January-April 2018) was conducted on a national level. We examined 22 hospitals with the emphasis on five hospital outcomes: recurrent ED visits (within 24 h), median waiting time at the ED, rate of early discharge at the same day and day after and percentage of patients spending N4 h in the ED. Results: A 5% upsurge in CT scans was observed (p = .032), and a significant increase was found in the number of consultations (14%, p = .002). On a national level, a linear regression model found that the proportion of patients discharged from the hospital on the same day and on the day after and the proportion of patients staying in the ED N 4 h predict ED recurrence visits within 24 h (p = .025; R² = 46.3%).

Conclusions: Upsurge in resource usage in the ED leads to a diagnostic cascade of health consumption. Further study is necessary to examine the proposed model in a global scale.

(C) 2018

All authors attest to meeting the ICMJE.org authorship criteria. No financial support was received for this research.

There is no conflict of interest.

Introduction

For over two decades emergency departments (ED’s) have been plagued with poor quality of care outcomes [1]. ED crowding is a prev- alent and important issue concerning hospitals all over the world [2,3]. Reasons for ED overcrowding that are relevant for this research are in- creasED patient volume [4], increasingly complex patients [5], non- urgent use of ED and “access block” [4,6]. Access block is a frequently re- lated concept to ED overcrowding [3,6,7]. It is a term defined as the in- ability of patients in the ED to gain access to in-patient beds in a certain time lag or the time it takes to be admitted, transferred or discharged.

? The research has not been presented.

* Corresponding author at: Rambam Health Care Campus & School of Nursing, Haifa, University, Israel.

E-mail address: [email protected] (M. Saban).

Access block leads to ED overcrowding and results in poor outcomes [7]. Simultaneously, over the years, health consumer behavior has changed and become more socially aware [8,9]. One major reason for this phenomenon is extensive waiting time for consulting medicine [10]. This behavior has led health consumers to seek alternative means of treatment, i.e. shorter queues found in the ED. [11,12]

Non-urgent ED visits are common and comprise 37% (Range 8%- 62%) of all ED visits [13,14]. Several reasons for this phenomenon are convenience (i.e. ease of use), dissatisfaction from community alterna- tives, poor access to consulting medicine and imaging outside the ED, in part due to long waiting times [13,15]. Over the years, ED consultants became an integral part of the ED staff [10,12,16], Taking part in admit- ting patients along with ED emergency specialists [16]. These consul- tants comprise mainly of general surgeons, Orthopedic surgeons and internists, which cover most of ED visits [17], while working in collabo- ration with other specialists, such as neurologists, psychiatrists, urolo- gists etc. A systematic review included five papers that inspected consultation proportion at ED and found variance of 20% to 40% for all patients [18]. The rate of consultations in the ED has increased substan- tially to a point where it affects Length of Stay (LOS) [17]. However,

https://doi.org/10.1016/j.ajem.2018.12.017

0735-6757/(C) 2018

1706 M. Saban et al. / American Journal of Emergency Medicine 37 (2019) 1705-1708

none of these studies portrayed the time trends of consultation rates. Concurrently, diagnostic imaging, especially computed tomography (CT) scans, are common and have increased substantially over the years in the ED setting [19]. Overall rate of imaging studies has risen by N60% [20]. Broder and his colleagues showed that the rates for vari- ous types of CT‘s has increased by 51%-463% over a period of 5 years²¹. A study published previously by Yoon and his colleagues pointed out the positive correlation between imaging and LOS at ED.

[17] To support this publication, recent studies found that 40% -60% of the diagnostic tests that are performed during hospitalization are un- necessary [22,23]. Needless tests can cause patient discomfort, Patient harm, increase Healthcare costs and also lead to false positive results which initiate a cascade of other tests, a phenomenon known as a diag- nostic cascade [24].

However, this diagnostic cascade was not evaluated in the ED setting and no updated data is available to show the current incidence state of diagnostic imaging in a time ED overcrowding becomes a worldwide phenomenon. Furthermore, implantation of intervention programs to reduce overuse of health resources such as “choosing wisely” are not common in the ED setting.

Thus, the aim of this study is to inspect whether the usage of consult- ing medicine and imaging influences waiting times and access block in the ED, as well as to emphasize the urgent need for implementation of resource optimization programs in the ED worldwide. We also formu- late a theoretical model that illustrates the diagnostic cascade of need- less medical testing in the ED.

Methods

Study design and setting.

A two-level study was conducted. The first local phase is a retrospec- tive quantitative archived study that was performed from January 2014 to April 2018 in a tertiary hospital. The hospital is a 1000-bed academic hospital serving over 2 million residents of Northern Israel. The ED is based on three main medical sectors (orthopedic, surgical and internal) defined as case managers and are regarded as ED staff. The ED staff ro- tate daily and are under the supervision of their department. During early morning and late evening hours there are three ED specialist that serve as internal specialists. Most of the patients (75.4%) are exam- ined and processed directly by these departments, whereas internal physicians examine majority of patients (41.8%), surgical physicians ex- amine 14.8% of the patients and the remaining (19.1%) are examined by orthopedic physicians.

In other cases (24.6%) patients arrive with a specified case such as: signs of stroke, single localized burn, nose trauma, corneal foreign body, Vaginal bleeding and with a direct referral from a specialist physi- cian (neurologist, plastic surgeon, ear, nose, and throat (ENT), optome- trist and gynecologists). Physicians such as: urologists, ENTs, rheumatologists, cardiologists, hematologists, gynecologists, plastic sur- geons, optometrists, cardiac and thoracic surgeons, oral and maxillofa- cial surgeons, psychiatrists, neurosurgeon, oncologists, nephrologists and dermatologists, are regarded as consulting physicians and can assist ED staff in patient care. Such collaboration is referred as consulting medicine.

Data collection.

Data was extracted from the electronic medical record (EMR). The ED contains 105 beds and about 100,000 patients N18 years old are treated per year on average.

In the second phase we examined the effect of the common use of consultation and imaging on a national level (n = 22 hospitals), be- tween 2011 and 2016. We included only tertiary and secondary hospi- tals with annual ED visits above 55,000 patients. We focused on five outcomes: recurrent ED visits (within 24 h), median waiting time at the ED, percentage of patients spending N4 h at the ED and the rate of early discharge (on the same day and the day after). Charlson score was assessed as a covariate. The Charlson Comorbidity Index has been

a useful tool for health researchers in their effort to measure comorbid disease status or case mix in health care databases [25,26].

Methods and Measurements.

The study sample for Phase 1 consisted of 392,919 patient files who were retrieved from the ED EMR during the study period. The data was collected retrospectively in January-April 2018 using a data sheet by the first and the second authors. For each patient we collected the following number of consultations and imaging tests ordered (X-ray, US, CT), and calculated the rate changes during the years.

ED physician staff (internal, surgical and orthopedic) can admit pa- tients independently of consultant involvement only to their respective departments. If they wish to admit patients to other departments it re- quires a consultation with the relative physician.

In addition, for each patient we assessed the time to triage, time to physician, time to decision, total time in ED, mean waiting time, hospi- talization rate and in-hospital length of stay. In the second phase, we ex- amined the phenomenon on a national level among 22 hospitals to find out whether the upsurge in consultations and imaging affects the afore- mentioned five outcomes. Data was collected from the national registry during 2011-2016. For each hospital we collected the ED visit rate, re- current ED visits (within 24 h), median waiting time at the ED, percent- age of patients spending N4 h at the ED and short hospitalization rate (discharge at the same day or discharge the day after). Due to the influ- ence of morbidity on hospitalization rate we also collected data from the national registry in the scope of the Charlson score.

Data Management and Statistical Analysis.

Descriptive statistical analyses were performed for ranges, means, and standard deviations for continuous variables and frequencies and percentages for categorical variables. Chi square test or One-way ANOVA were used when appropriate to compare between means, Pear- son analysis was performed to examine the following correlations: number of patients and waiting time; number of patients and number of imaging tests; number of patients and number of consultants; num- ber of consultations and hospitalization rates. Two Linear regression models were performed to predict ED waiting time and hospitalization rate by number of consultations and imaging tests. Poisson regression was performed to predict LOS by imaging tests and number of consulta- tions. The level of significance for all statistical analysis was 5%. The data analysis was performed using the Statistical Package for Health & Wel- fare Science for Windows (SPSS, version 23.0, Chicago, IL, USA).

Results

During the first study period, the number of patients admitted to the ED increased by 15%. Admission rate is on a rising trend (2014-32% whereas at 2017 it stands on 42%), while ED admission rate in the local hospital is the highest in the country compared to the national rate which stands at 22%.

We observed a significant statistical decrease in time to triage (from 26.0 +- 1.81 in 2014 to 12.0 +- 2.89 min in 2017, p b .001) and time to

physician (from 81.0 +- 12.6 in 2014 to 65.4 +- 18.0 min in 2017, p b

.001). In contrast, time to final decision and total time in ED increased from 249.6 +- 41.9 min in 2014 to 264.0 +- 52.2 in 2017 and 558 +-

410.4 min in 2014 to 594 +- 430.8 in 2017; respectively (Table 1). Data of resource consumption from 2014 to 2017 showed a signifi-

cant increase in the number of consultations. The number of consulta- tions has risen from 117,109 in 2014 to 149,724 in 2017 (+14%; p b

.001). We also observed an increase in all types of CT scans (5%, p =

.032). We observed an increase (5%, p = .032) in head, spinal, chest, ab- domen and pelvic CT, as well as total body complete tomography in trauma cases. The number of X-ray tests decreased significantly (11%, p = .021), whereas no significant change was found in US tests (Table 2).

A strong positive correlation was found between number of consul- tations and CT scans to total ED waiting time (r = 0.99, p <=0.001; r = 0.97, p <=0.001; respectively). We also found strong positive correlation

M. Saban et al. / American Journal of Emergency Medicine 37 (2019) 1705-1708 1707

between CT scans order in ED in hospitalization rate (r = 0.942, p =

.051; Fig. 1), as well as between number of consultations and hospitali- zation rate (r = 0.935, p = .045; Fig. 2). The inpatient bed availability has not changed during the study period.

On a national level, we found an increase of 16% in ED visits during the study period (2012-2016). On average, 120,100 patients are admit- ted to the ED annually, while the busiest ED treated 234,500 patients (hospital 1). The recurrence of ED visits within 24 h raised from 3% in 2012 to 3.55% in 2016, and the median waiting time at the ED was

2.74 h (0.29 h increase). The national percentage of patients who spent more than four hours in the ED has significantly increased (from 23.7% in 2012 to 33.4% in 2016, p b .01). During 2012 and 2016, the pro- portion of discharge at the same day and the day after was 10.71% (+0.29%) and 34.50% (+1.65%) respectively. No significant differences were found in the Charlson index between the hospitals during the years.

On a national level, a linear regression model found that the percent of patient discharge from hospital at the same day and the day after, and the percent of patient stay at the ED N 4 h predict significantly ED recur- rent visits within 24 h (p = .025; R² = 46.3%).

Discussion

The upsurge of needless medical testing in ED’s is an important uni- versal issue. This study examined whether the usage of consulting med- icine and imaging contributes to ED boarding and access block. Although we observed a significant decrease in times for the initial as- sessment, the time for final decision and total waiting time increased in- tensely. These trends are universal and waiting times in EDs are a considerable problem in many countries [25,26]. The main explanation may be due to the noticeable increase in CT imaging tests and consulta- tions, with most ED examinations were aimed to verify the diagnosis and prepare the patient for admission. As evidence, most patients who underwent CT scans were hospitalized (90%).

Previous studies suggested that there has been a tremendous in- crease in utilization of advanced imaging despite little evidence to sug- gest associated improvement in patient outcomes [21,27]. For example, Kanzaria and his colleagues claim that over-imaging is a systemic prob- lem, that led to over diagnosis which in turn led to overtreatment resulting in substantial harm [27]. A recent study of 5 million fee-for- service Medicare beneficiaries from USA found that increased regional CT risk was associated with a higher nephrectomy risk (r = 0.38; 95% CI, 0.28-0.47) [28].

The etiology of this increase is well investigated, considered as mul- tifactorial and may include malpractice fear, desire for diagnostic cer- tainty, lack of or inconsistent clinical Decision instruments and their poor adoption, inadequate training in evidenced-based medicine, slow knowledge translation, perverse financial incentives, requests of con- sulting and referring physicians, and increased practice intensity and overall complexity of emergency care [27,29-31].

The current literature shows that the use of consultations is pres- ent in 20-60% of all ED visits [32,33]. We also found positive correla- tion between CT scans ordered in the ED and hospitalization rate, as well as between the number of consultations and hospitalization rate. Previous studies identified that diagnostic tests and consultations have a major effect on admission characteristics. Waiting time to see a physician, diagnostic imaging and laboratory tests were associated with longer LOS [32-34]. A different research found that admitted pa- tients receiving consultations in the ED had a longer overall LOS com- pared to discharged patients. Linear regression modelling revealed that multiple consultations were significantly associated with longer LOS [35]. Other studies also showed the impact of consultation deci- sion times on LOS and the elongation of LOS due to multiple consulta- tions [36,37].

In contrast, increase in availability of consulting medicine found to be effective for the ED setting by reducing 30-day in-hospital mortality

and prediction of survival [38]. Another study found that with reduced consultant availability there are worsened outcomes at the ED, and sug- gested that the increasing patient volume with current ED medical and nursing staff requires more consultants, both in numbers and availabil- ity [39]. Hospitalization is more likely for patients receiving consulta- tion, although this is highly confounded by illness severity. However, our results did not find any changes while controlling illness severity for the Charlson score.

Considering the previous data we acquired and demonstrated, we propose the following theoretical model (Fig. 3). The model illustrates the diagnostic cascade of needless medical testing in the ED. Rise in ED visits lead to a combined effect of needless medical testing and de- fensive medicine which is characterized by an increase in consultations and imaging tests (with an added synergistic effect). In this scenario the likelihood of Pathological findings is increased. Due to high variability of the findings further consultations and imaging tests might be ordered. The aforementioned leads to a prolonged stay in the ED and therefore an upsurge in ED boarding and access block. The hospitalization rate is influenced by the rise of patients waiting to be hospitalized. Rise in hos- pitalization rates, little to no increase in Hospital beds and access block in the ED forces short hospitalizations. In the patient’s eyes short hospi- talizations are deemed incomplete and as such lead to recurrent visits to the ED.

Limitation.

This study has several limitations. The first limitation is a meth- odological one; most of the second phase data has been taken from preexisting public records and did not undergo data supervision by us. However, the data did undergo supervision with ordered and su- pervised methodology as well as random checkup of over 30% of the collected information. In addition, we do not have information re- garding similar patients who did not undergo CT or consultation. It is possible to compare patient ED stays between those who had ei- ther CT scans or consult from previous years. However, systemic re- view found that hospitalization for patients receiving consultations in the ED indicate a high percentage of admissions among consulted patients. The third limitation would be that not every examination (imaging or consultation) has been examined to ensure it they are indeed justified. On the other hand, the rise demonstrated in the study, mainly in the CT and consultation aspect is relevant beyond any doubt and therefore requires a thorough analysis. The fourth limitation is that a retrospective study does not allow the long- term follow up of the patient’s outcome due to needless medical test- ing. Nevertheless, the study is a one of a kind, mostly innovative and offers a theoretical model to examine this phenomenon in a larger scale in the long-term.

Conclusion

In recent years resource usage in the health care departments has in- creased dramatically. Our two-level ED study demonstrated the in- creased usage of needless medical testing leading to exhaustion of Hospital resources, prolonged times for medical decisions, patient dis- charge and patient stay at the ED. The theoretical model we propose il- lustrates the diagnostic cascade of needless medical testing and can be useful in emphasizing and hopefully resolving this phenomenon as well as assist achieving clinical certainty on a global scale.

Author contributions

All authors interpreted the data and edited and approved the final article. MS and HB drafted and conceived the study. MS, HP and TS de- signed the intervention. MS, TS, HP, RS and AD analyzed the data, de- signed the study and performed data collection. MS and HB take responsibility for the paper as a whole.

1708 M. Saban et al. / American Journal of Emergency Medicine 37 (2019) 1705-1708

References

1. Bernstein SL, Aronsky D, Duseja R, et al. The effect of emergency department crowding on clinically oriented outcomes. Acad Emerg Med 2009;16(1):1-10. https://doi.org/10.1111/j.1553-2712.2008.00295.x.
2. Dunn R. Reduced access block causes shorter emergency department waiting times: an historical control observational study. Emerg Med Australas 2003;15(3):232-8. https://doi.org/10.1046/j.1442-2026.2003.00441.x.
3. Chan SS, Cheung N, Graham C, Rainer T. Strategies and solutions to alleviate access block and overcrowding in emergency departments. Hong Kong Med J June 2015. https://doi.org/10.12809/hkmj144399.
4. Crawford K, Morphet J, Jones T, Innes K, Griffiths D, Williams A. Initiatives to reduce overcrowding and access block in Australian emergency departments: a literature review. Collegian 2014;21(4):359-66. https://doi.org/10.1016/J.COLEGN.2013.09. 005.
5. Trzeciak S, Rivers EP. Emergency department overcrowding in the United States: an emerging threat to patient safety and public health. Emerg Med J 2003;20(5):402-5. https://doi.org/10.1136/EMJ.20.5.402.
6. Fatovich DM, Nagree Y, Sprivulis P. Access block causes emergency department overcrowding and Ambulance diversion in Perth. Western Australia Emerg Med J 2005;22(5):351-4. https://doi.org/10.1136/emj.2004.018002.
7. Hong Y-C, Chou M-H, Liu EH, et al. The effect of Prolonged ED stay on outcome in patients with necrotizing fasciitis. Am J Emerg Med 2009;27(4):385-90. https:// doi.org/10.1016/J.AJEM.2008.03.010.
8. McMullan M. Patients using the internet to obtain health information: how this af- fects the patient-health professional relationship. Patient Educ Couns 2006;63 (1-2):24-8. https://doi.org/10.1016/J.PEC.2005.10.006.
9. Hibbard JH, Mahoney E. Toward a theory of patient and consumer activation. Patient Educ Couns 2010;78(3):377-81. https://doi.org/10.1016/J.PEC.2009.12.015.
10. Soong C, High S, Morgan MW, Ovens H. A novel approach to improving emergency department consultant Response times. BMJ Qual Saf 2013;22(4):299-305. https:// doi.org/10.1136/bmjqs-2012-001503.
11. Lee F, Teck K, Hospital P, Travers JP, Lee FCY. Avoiding prolonged waiting time during busy periods in the emergency department: Is there a role for the senior emergency Physician in triage? , vol. 13Lippincott Williams & Wilkins; 2006https://www. researchgate.net/publication/6706339 [Accessed October 8, 2018].
12. Terris J, Leman P, O’Connor N, Wood R. Making an IMPACT on emergency depart- ment flow: improving patient processing assisted by consultant at triage. Emerg Med J 2004;21(5):537-41. https://doi.org/10.1136/emj.2002.003913.
13. Uscher-Pines L, Pines J, Kellermann A, Gillen E, Mehrotra A. Emergency department visits for nonurgent conditions: systematic literature review. Am J Manag Care 2013; 19(1):47-59http://www.ncbi.nlm.nih.gov/pubmed/23379744. [Accessed October 8, 2018].
14. Hsia RY, Friedman AB, Niedzwiecki M. Urgent care needs among nonurgent visits to the emergency department. JAMA Intern Med 2016;176(6):852. https://doi.org/10. 1001/jamainternmed.2016.0878.
15. Honigman LS, Wiler JL, Rooks S, Ginde AA. National Study of non-urgent emergency department visits and associatED resource utilization. West J Emerg Med 2013;14 (6):609-16. https://doi.org/10.5811/westjem.2013.5.16112.
16. Kee R, Knott JC, Dreyfus S, Lederman R, Milton S, Joe K. One hundred tasks an hour: an observational study of emergency department consultant activities. Emerg Med Australas 2012;24(3):294-302. https://doi.org/10.1111/j.1742-6723.2012.01540.x.
17. Yoon P, Steiner I, Reinhardt G. Analysis of factors influencing length of stay in the emergency department. CJEM 2003;5(03):155-61. https://doi.org/10.1017/ S1481803500006539.
18. Lee RS, Woods R, Bullard M, Holroyd BR, Rowe BH. Consultations in the emergency department: a systematic review of the literature. Emerg Med J 2008;25(1):4-9. https://doi.org/10.1136/emj.2007.051631.
19. Kosydar K, Drozd M, Guz W, Samojedny A, Ralowska M, Kosydar K. Computed to- mography in the emergency department. Polish J Radiol 2014;79(Supplement): 23-7. https://doi.org/10.12659/PJR.890550.
20. Rao VM, Levin DC, Parker L, Frangos AJ, Sunshine JH. Trends in utilization rates of the various imaging modalities in emergency departments: Nationwide Medicare data from 2000 to 2008. J Am Coll Radiol 2011;8(10):706-9. https://doi.org/10.1016/J. JACR.2011.04.004.
21. Broder J, Warshauer DM. Increasing utilization of computed tomography in the adult emergency department, 2000-2005. Emerg Radiol 2006;13(1):25-30. https://doi. org/10.1007/s10140-006-0493-9.
22. Rao VM, Levin DC. The overuse of diagnostic imaging and the choosing wisely initia- tive. Ann Intern Med 2012;157(8):574. https://doi.org/10.7326/0003-4819-157-8- 201210160-00535.
23. Koch C, Roberts K, Petruccelli C, Morgan DJ. The frequency of unnecessary testing in hospitalized patients. Am J Med 2018;131(5):500-3. https://doi.org/10.1016/j. amjmed.2017.11.025.
24. Nguyen LT, Sullivan CT, Makam AN. The diagnostic Cascade of incidental findings. JAMA Intern Med 2015;175(7):1089. https://doi.org/10.1001/jamainternmed.2015. 1520.
25. Guttmann A, Schull MJ, Vermeulen MJ, Stukel TA. Association between waiting times and short term mortality and hospital admission after departure from emergency department: population based cohort study from Ontario, Canada. BMJ 2011;342: d2983. https://doi.org/10.1136/BMJ.D2983.
26. Carter EJ, Pouch SM, Larson EL. The relationship between emergency department crowding and patient outcomes: a systematic review. J Nurs Scholarsh 2014;46 (2):106-15. https://doi.org/10.1111/jnu.12055.
27. Kanzaria HK, Hoffman JR, Probst MA, Caloyeras JP, Berry SH, Brook RH. Emergency physician perceptions of medically unnecessary advanced. Diagn Imaging 2014. https://doi.org/10.1111/acem.12625.
28. Welch HG, Skinner JS, Schroeck FR, Zhou W, Black WC. Regional variation of com- puted tomographic imaging in the United States and the risk of nephrectomy. JAMA Intern Med 2018;178(2):221. https://doi.org/10.1001/jamainternmed.2017. 7508.
29. Westphalen AC, Hsia RY, Maselli JH, Wang R, Gonzales R. Radiological imaging of pa- tients with suspected urinary tract stones: National Trends, diagnoses, and predic- tors. Acad Emerg Med 2011;18(7):699-707. https://doi.org/10.1111/j.1553-2712. 2011.01103.x.
30. Goldstone A, Bushnell A. Does diagnosis change as a result of repeat renal colic com- puted tomography scan in patients with a history of Kidney stones? Am J Emerg Med 2010;28(3):291-5. https://doi.org/10.1016/J.AJEM.2008.11.024.
31. Pines JM. Trends in the rates of radiography use and important diagnoses in emer- gency department patients with abdominal pain. Med Care 2009;47(7):782-6. https://doi.org/10.1097/MLR.0b013e31819748e9.
32. Yoon P, Steiner I, Reinhardt G. Analysis of factors influencing length of stay in the emergency department. CJEM 2003;5(03):155-61. https://doi.org/10.1017/ S1481803500006539.
33. Rhodes SM, Patanwala AE, Cremer JK, et al. Predictors of prolonged length of stay and adverse events among older adults with behavioral health-related emergency department visits: a systematic medical record review. J Emerg Med 2016;50(1): 143-52. https://doi.org/10.1016/J.JEMERMED.2015.06.073.
34. Lee RS, Woods R, Bullard M, Holroyd BR, Rowe BH. Consultations in the emergency department: a systematic review of the literature. Emerg Med J 2008;25(1):4-9. https://doi.org/10.1136/emj.2007.051631.
35. Salehi L, Phalpher P, Valani R, et al. Emergency department boarding: a descriptive analysis and measurement of impact on outcomes. CJEM April 2018:1-9. https:// doi.org/10.1017/cem.2018.18.
36. Brick C, Lowes J, Lovstrom L, et al. The impact of consultation on length of stay in ter- tiary care emergency departments. Emerg Med J 2014;31(2):134-8. https://doi.org/ 10.1136/emermed-2012-201908.
37. Lee J, Kirschner J, Pawa S, Wiener DE, Newman DH, Shah K. Computed tomography use in the adult emergency Department of an Academic Urban Hospital from 2001 to 2007. Ann Emerg Med 2010;56(6):591-596.e1. https://doi.org/10.1016/J. ANNEMERGMED.2010.05.027.
38. Conway R, O’Riordan D, Silke B. Consultant volume, as an outcome determinant, in emergency medical admissions. QJM 2013;106(9):831-7. https://doi.org/10.1093/ qjmed/hct119.
39. Sanders JL, Raja AS, Hasegawa K, et al. Decline in consultant availability in Massachu- setts emergency departments: 2005 to 2014. Ann Emerg Med 2016;68(4):461-6. https://doi.org/10.1016/j.annemergmed.2016.06.013.