The Journal of Emergency Medicine, Vol. 42, No. 1, pp. 69–73, 2012Copyright � 2012 Elsevier Inc.

Printed in the USA. All rights reserved0736-4679/$ - see front matter

doi:10.1016/j.jemermed.2011.03.022

Presented at thAnnual Meeting,

RECEIVED: 2 JulACCEPTED: 20 M

Education

THE IMPACT OF EMERGENCY DEPARTMENT OVERCROWDINGON RESIDENT EDUCATION

Simon A. Mahler, MD, Jeannie R. McCartney, MD, Thomas K. Swoboda, MD, Lauren Yorek, MD,and Thomas C. Arnold, MD

Department of Emergency Medicine, Louisiana State University Health Sciences Center, Shreveport, LouisianaReprint Address: Simon A. Mahler, MD, Department of Emergency Medicine, Louisiana State University Health Sciences Center-Shreveport,

1501 Kings Hwy, Shreveport, LA 71130

, Abstract—Background: Few studies have evaluated theeffect of Emergency Department (ED) overcrowding onresident education. Objectives: To determine the impactof ED overcrowding on Emergency Medicine (EM) resi-dent education. Materials and Methods: A prospectivecross-sectional study was performed from March to May2009. Second- and third-year EM residents, blinded tothe research objective, completed a questionnaire at theend of each shift. Residents were asked to evaluate the ed-ucational quality of each shift using a 10-point Likert scale.Number of patients seen and procedures completed wererecorded. Responses were divided into ED overcrowding(group O) and non-ED overcrowding (group N) groups.ED overcrowding was defined as >2 h of ambulance diver-sion per shift. Questionnaire responses were comparedusing Mann–Whitney U tests. Number of patients and pro-cedures were compared using unpaired T-tests. Results:During the study period, 125 questionnaires were com-pleted; 54 in group O and 71 in group N. For group O,the median educational value score was 8 (interquartilerange [IQR] 7–10), compared to 8 (IQR 8–10) for group N(p = 0.24). Mean number of patients seen in group O was12.3 (95% confidence interval [CI] 11.4–13.2), comparedto 13.9 (95% CI 12.7–15) in group N (p = 0.034). In groupO, mean number of procedures was 0.9 (95% CI 0.6–1.2),compared to 1.3 (95% CI 1–1.6) in group N (p = 0.047).Conclusions: During overcrowding, EM residents saw

e Society of Academic Emergency MedicinePhoenix, Arizona, June 2010.

y 2010; FINAL SUBMISSION RECEIVED: 6 Septembearch 2011

69

fewer patients and performed fewer procedures. However,there was no significant difference in resident perceptionof educational value during times of overcrowding vs.non-overcrowding. � 2012 Elsevier Inc.

, Keywords—emergency department; overcrowding;resident education

INTRODUCTION

Emergency Department (ED) overcrowding has been calleda crisis in EmergencyMedicine (EM) (1). ED overcrowdingoccurs when the demand for services overwhelms availableresources. Symptoms of ED overcrowding include boardingadmitted patients, diverting ambulances, and caring forpatients on stretchers in hallways. As evidence to the mag-nitude of this problem, the Institute of Medicine hasestimated that over 90% of EDs are affected by overcrowd-ing (1). However, overcrowding is most severe in academiccenters and urban hospitals, many of which support EMresidency training programs (1–3).

Several studies have demonstrated that ED overcrowd-ing is associated with increased morbidity and mortality(4–13). However, the impact of ED overcrowdingon EM resident education is largely unknown. Severalauthors have hypothesized that overcrowding increasesED attendings’ clinical and administrative workload,

r 2010;

Table 1. Questionnaire Response Rates by Second- andThird-Year Residents for Overcrowding andNon-overcrowding Shifts

Overcrowding Non-overcrowding Total

Second-year residents 31 (41%) 44 (59%) 75Third-year residents 23 (46%) 27 (54%) 50All residents 54 (43%) 71 (57%) 125

70 S. A. Mahler et al.

resulting in decreased quality and time available forteaching (14–16).

At present, only one published peer-reviewed study as-sessing the relationship between overcrowding and EMresident education exists. Pines et al. assessed the qualityof teaching for residents and medical students during in-dividual patient encounters (17). This study failed to finda correlation between the quality of education and EDovercrowding. Unfortunately, the study was powered todetect only very large differences in educational qualityand did not measure the number of patients seen or num-ber of procedures completed. Smaller, possibly importanteducational differences may have been detected if thestudy had been adequately powered. Therefore, theobjective of this study is to determine the impact of EDovercrowding on EM resident education as measuredby number of patients seen, number of procedurescompleted, and resident perception.

MATERIALS AND METHODS

A prospective cross-sectional study was performed overa 3-month period from March to May 2009 at LouisianaState University Health Sciences Center Shreveport(LSUHSC-S). LSUHSC-S is a tertiary care, level Itrauma center, and an academic center. The ED has an an-nual volume of 60,000 patient visits per year and treatsprimarily uninsured or underinsured patients. It is alsohome to a 1- to 3-year EM residency program with 7 res-idents per year.

Second- and third-year EM residents, blinded to the re-search objectives, completed a questionnaire at the end ofeach shift. Residents were asked to evaluate: the qualityof supervision and instruction during procedures, qualityof teaching from the faculty during the shift, and the over-all educational value of each shift. Answers were given ona 10-point Likert scale (with 1 = very poor and 10 =outstanding/superior). The number of patients cared forand procedures completed for each shift were recordedby self-report on the questionnaire. ED faculty and staffwere also blinded, with the exception of faculty involvedin the planning of the study (authors SM and TS). Com-pleted questionnaires were placed into a receptacle thatwas accessed by study investigators only.

Questionnaire responses were divided into ED over-crowding (group O) and non-ED overcrowding (groupN) groups based on whether overcrowding took placeduring the shift in which the resident completed thequestionnaire. ED overcrowding was defined by >2 hof ambulance diversion during a shift. Ambulance diver-sion has been previously validated as a reliable measureof ED overcrowding (2,12,18).

Questionnaire responses answered on a Likert scalewere compared using Mann–Whitney U tests. Number of

patients seen and procedures performed were comparedusing unpaired t-tests. The study was powered to detecta 1.0 difference on the Likert scale at an overcrowdingrate of 40%, power of 80%, and alpha of 0.05. Statisticalanalysis was preformed with SPSS 11.0 (SPSS Inc,Chicago, IL) for Windows.

RESULTS

During the 3-month study period from March 2009 toMay 2009, second- and third-year residents worked a totalof 236 shifts. Questionnaires were completed for 125shifts, a response rate of 53% (125/236). Group O had54 questionnaires and group N had 71. Second-yearresidents completed 75 questionnaires, compared to 50by third-year residents (Table 1).

The median ambulance diversion time per shift forGroup O was 5 h (interquartile range [IQR] 3–7) com-pared to 0 h (IQR 0) for group N (p < 0.001). For groupO, the median score for the overall educational value ofa shift was 8 (IQR 7–10), compared to 8 (IQR 8–10)for group N (p = 0.24). Mean number of patients caredfor per shift in group Owas 12.3 (95% confidence interval[CI] 11.4–13.2), compared to 13.9 (95% CI 12.7–15) ingroup N (p = 0.034). In group O, mean number of proce-dures completed per shift was 0.9 (95%CI 0.6–1.2), com-pared to 1.3 (95% CI 1–1.6) in group N (p = 0.047). Therewas also no significant difference in the quality of super-vision and instruction during procedures or the quality ofteaching from the faculty during the shift. Results aresummarized in Table 2.

DISCUSSION

Due to a high prevalence of overcrowding in academic cen-ters and urban hospitals with EM residency trainingprograms, several authors have hypothesized that over-crowding decreases the quality and time available for teach-ing in the ED (1–3,14–16). In this study, the residents’perception of the overall educational value of each shiftdid not differ significantly based on overcrowding.However, during times of overcrowding, residents caredfor fewer patients and completed fewer procedures.Although the residents’ did not feel that shifts performedduring overcrowded situations were educationally less

Table 2. Ambulance Diversion Time: Questionnaire Responses, Number of Patients Seen, and Number of ProceduresCompleted for Overcrowding and Non-overcrowding Shifts

Overcrowding Non-overcrowding p Value

Ambulance diversion time (h/shift)Median (IQR) 5 (3–7) 0 (0) <0.001

Teaching qualityMedian (IQR) 8 (7–10) 8 (7–10) 0.63

Instruction during proceduresMedian (IQR) 10 (8–10) 10 (8–10) 0.72

Overall educational valueMedian (IQR) 8 (7–10) 8 (8–10) 0.24

Patients seenMean (95% CI) 12.3 (11.4–13.2) 13.9 (12.7–15) 0.034

Procedures completedMean (95% CI) 0.9 (0.6–1.2) 1.3 (1–1.6) 0.047

IQR = interquartile range; CI = confidence interval.

ED Overcrowding and Resident Education 71

valuable, it is clear that they received less experience inpatient care and procedures. Therefore, it is possible thatresident perception may not correlate with the ‘‘true’’educational value of a shift.

It is unclear why residents would find shifts to be ofequal educational value despite caring for fewer patientsand performing fewer procedures. It is possible that anynegative impact from overcrowding on resident percep-tion was offset by other factors. During times of over-crowding, when large numbers of admitted patients areboarded in the ED, faculty and residents may have fewerdirect patient care responsibilities. This would permitmore time for teaching, reading, or other downtime activ-ities and thereby improve residents’ perception of educa-tional value.

Experiences gained from the number of proceduresand number of patients cared for over a wide variety ofchief complaints is clearly an important aspect of residenteducation. Adequate patient volume is required to honethe skills of multitasking essential to the community prac-tice of EM. Adequate procedural experience is requiredfor procedural competency. The impact of overcrowdingover the course of a 3- or 4-year EM residency remainsunknown. However, if the relationship from this studyof decreased patient care and procedural experiencepersists throughout residency, the cumulative effects ofovercrowding could be highly detrimental. It is possiblethat residency training programs with high overcrowdingrates may graduate residents with less hands-on experi-ence than programs with lower overcrowding rates.

The results of this study agree with previously pub-lished studies. Pines et al., in a recent study addressingthe educational impact of overcrowding, failed to finda correlation between ED overcrowding and teachingquality, as measured by resident and medical studentopinion (17). Another study, published in abstract formonly, also failed to find a relationship between overcrowd-ing and educational value (19). A third study, by Kelly

et al., although not focused on overcrowding, failed toshow a relationship between resident perception of edu-cational value and times of high or low clinical workload in the ED (20). However, all of these studies usedteaching scores and surveys to measure resident percep-tion of educational value without measuring the numberof patients cared for or number of procedures completed.

Other important differences exist between this studyand the study completed by Pines et al. The Pines et al.study was designed to assess the quality of individualpatient encounters, rather than the educational qualityof an ED shift. It also included EM residents, non-EMresidents, and medical students; unlike our study, whichincluded only second- and third-year EM residents. Thestudy had a large number of admitted patient encounters(49% of encounters), and it was unclear if residents andstudents participating were blinded to the purpose ofthe study.

Limitations

This study is limited by small sample size, surveyresponse rate <60%, and single-center design. The find-ings at LSUHSC-S may not be generalizable to otherEM residency programs. Future studies should considera multicenter design. Although residents were assuredthat completion of their questionnaires was anonymousand confidential, it is possible that residents feared giv-ing low scores. This may be responsible for high scoresreceived in both groups and for the lack of difference be-tween groups. Different faculty can also contribute tothe residents’ perception of the educational value ofa shift. Individual faculty members may have hadmore shifts in group O or N. This variable was not ac-counted for in our analysis. The number of patientscared for and the number of procedures completedwere recorded on the questionnaire by self-reporting.Self-reporting of data can lead to unintentional and

72 S. A. Mahler et al.

intentional reporting bias/recall bias. It is likely that atthe end of a shift, a patient cared for or procedure com-pleted at the beginning or middle of the shift may havebeen forgotten and not reported. In addition, althoughthe questionnaires were anonymous, residents mayhave intentionally over-counted patients or procedures,because being ‘‘highly productive’’ is often equatedwith excellence in EM residencies.

The questionnaire response rate was lower for over-crowding shifts than non-overcrowding shifts. Althoughthe cause of this difference is unknown, it representsa possible source of bias. It is also possible that residentswere less likely to complete questionnaires after verybusy shifts, which may have biased our results. In addi-tion, resident perception, which was measured in thisstudy and prior studies, may not correlate with the‘‘true’’ educational value of a shift. Future studies shouldlook at less subjective measures of educational value.

CONCLUSIONS

During shifts with ED overcrowding, EM residents sawfewer patients and performed fewer procedures. However,residents’ opinion of the educational value of shifts did notdiffer significantly based on overcrowding. The long-termeffects of overcrowding on EM resident education remainunknown. Future studies should assess educational value,number of patients cared for, and number of procedurescompleted over a longer time period. In addition, futurestudies should measure educational value objectively,rather than relying on resident opinion.

REFERENCES

1. Institute of Medicine, Committee on the Future of Emergency Carein the United States Health System. Hospital-based emergency careat the breaking point. Washington, DC: National Academies Press;2007.

2. Schull MJ, Slaughter PM, Redelmeier DA. Urban emergencydepartment overcrowding: defining the problem and eliminatingmisconceptions. CJEM 2002;4:76–83.

3. Weiss SJ, Derlet R, Arndahl J, et al. Estimating the degree ofemergency department overcrowding in academic medical centers:

results of the National ED Overcrowding Study (NEDOCS). AcadEmerg Med 2004;11:38–50.

4. Barrett TW, Schriger DL. Annals of Emergency Medicine JournalClub. Emergency department crowding is associated with poorcare for patients with severe pain. Ann Emerg Med 2008;51:6–7.

5. Bernstein SL, Aronsky D, Duseja R, et al. The effect of emergencydepartment crowding on clinically oriented outcomes. Acad EmergMed 2009;16:1–10.

6. Kulstad EB, Kelley KM. Overcrowding is associated with delays inpercutaneous coronary intervention for acute myocardial infarction.Int J Emerg Med 2009;2:149–54.

7. Kulstad EB, Sikka R, Sweis RT, et al. ED overcrowding is associ-ated with an increased frequency of medication errors. Am J EmergMed 2010;28:304–9.

8. Miro O, AntonioMT, Jimenez S, et al. Decreased health care qualityassociated with emergency department overcrowding. Eur J EmergMed 1999;6:105–7.

9. Pines JM, Localio AR, Hollander JE, et al. The impact of emer-gency department crowding measures on time to antibiotics forpatients with community-acquired pneumonia. Ann Emerg Med2007;50:510–6.

10. Pines JM, Hollander JE. Emergency department crowding is associ-ated with poor care for patients with severe pain. Ann Emerg Med2008;51:1–5.

11. Richardson DB. Increase in patient mortality at 10 days associatedwith emergency department overcrowding. Med J Aust 2006;184:213–6.

12. SchullMJ, VermeulenM, Slaughter G, et al. Emergency departmentcrowding and thrombolysis delays in acute myocardial infarction.Ann Emerg Med 2004;44:577–85.

13. Sprivulis PC, Da Silva JA, Jacobs IG, et al. The association betweenhospital overcrowding and mortality among patients admitted viaWestern Australian emergency departments. Med J Aust 2006;184:208–12.

14. Sabbah S. The impact of hospital overcrowding on postgraduateeducation: an emergency medicine resident’s perspective throughthe lens of CanMEDS. CJEM 2009;11:247–9.

15. Aldeen AZ, Gisondi MA. Bedside teaching in the emergencydepartment. Acad Emerg Med 2006;13:860–6.

16. Atzema C, Bandiera G, Schull MJ. Emergency department crowd-ing: the effect on resident education. Ann Emerg Med 2005;45:276–81.

17. Pines JM, Prabhu A, McCusker CM, et al. The effect of ED crowd-ing on education. Am J Emerg Med 2010;28:217–20.

18. Shenoi RP, Ma L, Jones J, et al. Ambulance diversion as a proxyfor emergency department crowding: the effect on pediatricmortality in a metropolitan area. Acad Emerg Med 2009;16:116–23.

19. Hoxhaj S, Mosely MG, Fisher A, et al. Resident education does notcorrelate with the degree of emergency department crowding. AnnEmerg Med 2004;44:S77.

20. Kelly SP, Shapiro N, Woodruff M, et al. The effects of clinicalworkload on teaching in the emergency department. Acad EmergMed 2007;14:526–31.

ED Overcrowding and Resident Education 73

ARTICLE SUMMARY

1. Why is this topic important?The impact of Emergency Department (ED) over-

crowding on resident education is largely unknown.2. What does this study attempt to show?

The objective of this study is to determine the effect ofED overcrowding on Emergency Medicine (EM) residenteducation.3. What are the key findings?

During shifts with ED overcrowding, EM residents sawfewer patients and performed fewer procedures. However,residents’ opinion of the educational value of shifts didnot differ significantly based on overcrowding.4. How is resident education impacted?

The long-term effects of overcrowding on EM residenteducation remain unknown. Further research is necessary,and future studies should assess educational value, num-ber of patients cared for, and number of procedures com-pleted over a longer time period. In addition, futurestudies should measure educational value objectively,rather than relying on resident opinion.