Emergency. 2017; 5 (1): e53 OR I G I N A L RE S E A RC H Determinants of Prolonged Length of Stay in the Emer- gency Department; a Cross-sectional Study Seyed Mohammad Hosseininejad1, Hamed Aminiahidashti2∗, Seyede Masoume Pashaei2, Iraj Goli Khatir2, Seyed Hosein Montazer2, Farzad Bozorgi3, Fahime Mahmoudi4 1. Department of Emergency Medicine, Diabetes research center, Faculty of Medicine, Mazandaran University of Medical Sciences, Sari, Iran. 2. Department of Emergency Medicine, Faculty of Medicine, Mazandaran University of Medical Sciences, Sari, Iran. 3. Emergency Department, Orthopedic research center, Faculty of Medicine, Mazandaran University of Medical Sciences, Sari, Iran. 4. Faculty of Medicine, Mazandaran University of Medical Sciences, Sari, Iran. Received: August 2016; Accepted: November 2016; Published online: 18 January 2017 Abstract: Introduction: Timeliness has been considered as a key domain in quality of emergency department (ED) care and delay in care providing is influential determinants of patient’s outcomes. The present study, aimed to evalu- ate the determinants of prolonged ED length of stay (LOS). Methods: In this cross-sectional study, using adopted version of the latest form for external evaluation and accreditation of EDs introduced by Iranian Ministry of Health, determinants of prolonged LOS were evaluated in the ED of an educational Hospital. Using SPSS 11, multivariate binary logistic regression was applied to estimate adjusted odds ratios (OR) for determining factors associated with prolonged LOS. Results: 162 (10.2%) cases with prolonged LOS were detected. Based on uni- variate analysis, female gender (OR: 1.42, 95% CI: 1.14-1.75, p = 0.001), older age (OR: 1.05, 95% CI: 1.02-1.08, p < 0.0001), admission on evening shifts (OR: 4.0; 95% CI: 1.84-8.68, p < 0.001), triage level I (OR: 1.76, 95% CI: 1.21-2.57, p = 0.003), lack of insurance support (OR: 1.56, 95% CI: 1.12-2.19, p = 0.010), higher number of ordered para-clinical tests (OR: 1.23, 95% CI: 1.11-1.37, p = 0.016), and disposition time > 6 hours (OR, 0.13, p < 0.0001), were significant risk factors of prolonged LOS. Conclusion: Older age, lack of insurance support, disposition time > 6 hours due to complexity of patients’ complaint, and the necessity of repeated para-clinical measures were the most important reasons for failed provision of timely services. From the view point of ED personnel, a small part of prolonged LOS in ED was concerned with defective ED workflow, while, the most important cause of such delays was the delayed response of the consultancy services. Keywords: Emergency service, hospital; length of stay; quality of health care; risk factors; Iran © Copyright (2017) Shahid Beheshti University of Medical Sciences Cite this article as: Hosseininejad SM, Aminiahidashti H, Pashaei S-M, Goli Khatir I, Montazer SH, Bozorgi F, Mahmoudi F. Determinants of Prolonged Length of Stay in the Emergency Department; a Cross-sectional Study. Emergency. 2017; 5(1): e53. 1. Introduction E mergency department (ED) crowding is a chronic health challenge worldwide (1). This growing chal- lenge could results in curtailed and dysfunctional emergency activities (2, 3). It has bidirectional synergic asso- ciation with delayed emergency care (2). Both crowding and delays in ED cares are influential determinants of patient’s outcome (4). Therefore, they affect ED performance, mainly ∗Corresponding Author: Hamed Aminiahidashti; Department of Emergency Medicine, Imam Khomeini Hospital, Amir Mazandarani Bolivar, Sari, Iran. Tel: +989113540546; Email: hamedaminiahidashti@yahoo.com regarding timeliness (5). Timeliness has been considered as a key domain in quality of emergency care. By timely we mean that EDs should continuously move toward “reducing waits and sometimes harmful delays for both those who receive and those who give care”(5). To achieve acceptable timeli- ness and high quality emergency care, the period from pa- tient’s arrival to discharge could be segmented, determinants of each segment should be identified, and then evidence- based interventions may be introduced. In recent years, several interventions including employing emergency spe- cialists, holding formal interdisciplinary team-work training programs, use of triage systems, fast-track units, and maxi- mum length of stay (LOS) rules, e.g. 4-hour rule in the UK, This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: www.jemerg.com M. Hosseininejad et al. 2 have been introduced to reduce waits and delays in ED (6- 9). In Iran, population aging has resulted in a rising number of admissions to EDs (3, 10). Accordingly, improvement of performance in EDs is an urgent challenge. Recently, a set of rules have been introduced by Iranian Ministry of Health and Medical Education (IMOH), regarding waiting times or LOS in ED, and EDs are encouraged to achieve these targets (11). In addition, Iranian researchers have focused on the LOS and shown that the average LOS of patients admitted to teaching EDs has been much more than the maximum targeted LOS (LOS<6 hours), introduced by IMOH as a safe LOS (11, 12). However, a growing body of evidence demonstrated that fo- cusing on a set of pre-specified time rules could lead to un- intended detracts from clinical priorities and, consequently, attenuation of patient-centeredness and poor outcomes (13- 15). In spite of dozens of reports on ED performance eval- uation from Iran, data on the causes of lower ED perfor- mance and, specially, prolonged LOS are scarce. Therefore, this study aimed to evaluate determinants of prolonged LOS in emergency department. 2. Methods 2.1. Study design and setting In this cross-sectional study, using adopted version of the latest form for external evaluation and accreditation of EDs introduced by IMOH, determinants of prolonged LOS were evaluated in ED of Imam Khomeini educational Hospital, Sari, Iran, during three months from November 2014 to February 2015. Applied procedures and data collection methods in this study were approved by the ethics committee of Mazandaran University of Medical Sciences (ethical ap- proval number: 854). Authors adhered to the all ethical prin- ciples of Helsinki declaration and confidentiality of patients’ records. 2.2. Participants Patients who had stayed in the ED for more than 6 hours were considered as cases of prolonged LOS and were enrolled to the study using a sequential convenience sampling. Eligible patients were selected from the lists of all patients admitted to the ED, every 60 hours. LOS was calculated according to the following equation: LOS = time of medical record review - time of admission. Imam Khomeini Hospital is a teaching hospital affiliated to Mazandaran University of Medical Sci- ences. It is the most equipped hospital throughout Mazan- daran province, and consequently, emergency patients are admitted to its ED, either by pre-hospital emergency teams, by themselves or their caregivers, or by other hospitals or clinics. This ED has been administered by an experienced emergency specialist and benefited from highly competent nurses and staff. According to guidelines provided by IMOH, the emergency severity index (ESI) is used for patients’ triage in this ED. 2.3. Data gathering Data gathering was done using an adopted version of the latest form for external evaluation and accreditation of EDs at teaching hospitals introduced by IMOH. The adoption process was done in a team including emergency special- ists and highly competent ED staff. The first version of this checklist was used in two pilot studies and some corrections were made according to field experiences. The final version was designed in three sections including the demographic and background section, a section for indices of timely care, and a section for measuring causes of prolonged LOS. Its face validity was confirmed by a team of experts in emer- gency medicine. Regarding its reliability, we estimated Cron- bach’s alpha as 77.0, which revealed an acceptable reliability. The data were extracted by retrospective review of medical records, interview with the patient’s doctor, and also super- visor nurses at the time of presence of the patient in the ED. First and second sections of the study checklist were com- pleted according to medical records, while causes of pro- longed LOS were determined by in-depth interviews. In- terviewees were asked to select causes of prolonged LOS of each patient according to the study checklist. A prolonged LOS could be considered owing to more than one cause. The number of prolonged LOS assigned to each cause was counted and then categorized during a review session by the research team. Data were collected by three medical students who were trained through a series of educational sessions. During the first session, the study checklist was introduced to students and completed for selected patients. In the next role playing sessions, they were asked to complete the check- lists for several complicated patients who were purposefully assigned. Then, the completed checklists were discussed and students’ competency was assessed by the principal investi- gator, SMH. The data were reviewed by our principal investi- gator in weekly sessions. Missing and inconsistent data were specified and corrected according to medical records or ad hoc interviews, if applicable. 2.4. Statistical Analysis Data were computerized and analyzed using statistical pack- age for social sciences (SPSS) version 11.0. Data was cleaned and prepared according to recommended procedures (16) and descriptive statistics was applied to describe the data. Variable reduction was conducted using univariate statisti- cal tests, considering P value ≤ 0.25 (17). Then, multivariate binary logistic regression was applied to estimate adjusted odds ratios (OR) and its 95% confidence intervals (CI) for as- sociated factors with prolonged LOS. In this step, a P-value less than 0.05 was considered as statistically significant. This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: www.jemerg.com 3 Emergency. 2017; 5 (1): e53 Figure 1: Reasons for visiting the emergency department among patients with prolonged length of stay (N = 162). Table 1: Baseline characteristics of patients with prolonged length of stay in the emergency department Variable N (%) Gender Male 71(43.8) Female 91(56.2) Transferred by Ambulance 130(80.2) Caregivers 26(16.1) Referred 6(3.7) Insurance support Yes 141(87.0) No 21(13.0) Triage level* I 2(1.2) II 7(4.3) III 127(78.4) IV 12(7.4) Missing 14(8.6) Admission work shift Night 88(54.3) Evening 40(24.7) Morning 34(21.0) *: Based on emergency severity score (ESI). 3. Results 3.1. Baseline characteristics 1581 patients were admitted to the ED during the study pe- riod. 162 (10.2%) cases of prolonged LOS with the mean age of 58.5 ± 20.2 (Range: 12-98) years were detected (56.2% fe- male). Table 1 shows the baseline characteristics of these 162 cases. 80.2% patients had been brought to the ED by care- givers and 78.4% had triage level of III. Figure 1 displays the reasons for visiting ED among patients with prolonged LOS. Mean waiting time from arrival to the first nursing visit was 2.85 ± 1.75 minutes (1 – 8), while mean time before the first visit by a doctor was 3.4 ± 2.7 minutes (1 – 11). 153 (94.4%) cases had disposition order within the first 6 hours of admis- sion. 3.2. Determinants of prolonged LOS Based on univariate analysis, female gender (OR: 1.42, 95% CI: 1.14-1.75, p = 0.001), older age (OR: 1.05, 95% CI: 1.02- 1.08, p < 0.0001), admission on evening shifts (OR: 4.0; 95% CI: 1.84-8.68, p < 0.001), triage level I (OR: 1.76, 95% CI: 1.21- 2.57, p = 0.003), lack of insurance support (OR: 1.56, 95% CI: 1.12-2.19, p = 0.010), higher number of ordered para-clinical tests (OR: 1.23, 95% CI: 1.11-1.37, p = 0.016), and disposition time > 6 hours (OR, 0.13, p < 0.0001), were significant risk fac- tors of prolonged LOS. Patients had a 5% higher risk of pro- longed LOS for every 5-year increase in their age. Based on the results of multivariate analysis, older age (p = 0.019), lack of insurance support (p = 0.024), disposition time > 6 hours (p = 0.003), and higher number of ordered para-clinical tests (p = 0.029) were significantly associated with prolonged LOS (Table 2). 3.3. Causes of prolonged LOS based on ED per- sonnel’s view point According to ED physicians and supervisor nurses, causes of prolonged LOS in these 162 cases could be categorized into ED-related factors, poor cooperation of other departments, and factors outside of the hospital (Table 3). 4. Discussion According to the findings of the present study, older age, lack of insurance support, disposition time > 6 hours, and higher number of ordered para-clinical tests were among the most important determinants of prolonged LOS. On the view point of ED personnel poor cooperation of other departments in providing proper consultancy and patients’ disposition, as well as some factors inside the ED, such as delayed consult request, complicated cases, untimely admission, and crowd- ing, were among the most frequent causes of prolonged LOS in the studied ED. More precisely, according to the medi- cal and nursing staff reports, the cause of prolonged LOS, in many cases, is the delayed response by the departments with which the ED maintains interactions in order to pro- vide proper services. Meanwhile, 6 percent of the prolonged LOS was caused as a result of defects in the ED’s workflow and ED crowding. However, this may be due to the bias of interviewees (18). Yet, considering the fact that they were asked to be honest in their statements, this bias is unlikely to be significant. Although, at first glance, the quantita- tive results differed from the causes stated by the medical and nursing staff, when we look closer, the two have largely This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: www.jemerg.com M. Hosseininejad et al. 4 Table 2: Determinants of prolonged length of stay in the emergency department based on multivariate analysis Factors Crude OR Adjusted OR P Older age 1.05(1.02-1.08) 1.10(1.02-1.20) 0.019 Having insurance support 0.64(0.46-0.89) 0.29(0.10-0.85) 0.024 Disposition time ≤ 6 hours 0.13(0.02-0.99) 0.03(0.003-0.30) 0.003 Higher Number of para-clinical tests 1.23(1.11-1.37) 1.3(1.03-1.64) 0.029 Odds ratios (OR) were presented with 95% confidence interval. Table 3: Sources of prolonged length of stay based on the emergency department personnel’s view point Factors Number (%)1 Inside the emergency department (n = 20) Delayed request for consult 15 (75.0) Complicated cases 5 (25.0) Untimely admission 16 (80.0) Crowding 2 (10.0) Poor cooperation of other departments2 (n=114 ) Gastroenterology 21 (18.4) Respiratory 12 (10.5) Nephrology 9 (7.8) Surgery 9 (7.8) Orthopedic 3 (2.6) Internal medicine 35 (30.7) Oncology 7 (6.1) Endocrinology 12 (10.5) Neurosurgery 7 (6.1) Cardiovascular 6 (5.3) Outside the hospital (n=28 ) Outside consult3 23 (82.1) Imaging4 17 (60.7) 1: Length of stay could be prolonged due to more than onesource. 2: For patients disposition. 3: Consulting with department outside the hospital such as neurology, infectious diseases, cardiovascular, and toxicology. 4: Computed tomography scan (CT), magnetic resonanceimaging (MRI), and ultrasonography. confirmed each other. For instance, older age can increase the risk of co-morbidity and complexity of clinical decision- making (19), and may, eventually, delay the response by the consultancy services. Also, lack of insurance coverage can indicate lower socioeconomic status, increasing the risk of co-morbidity and complexity of clinical decision-making (20, 21), and eventually, delaying the response by the consultancy services. As for patients in need of repeated paraclinical mea- sures, the consultancy services couldn’t provide proper con- sultation prior to receiving the respective results, hence, the delayed response by the mentioned services. On the other hand, a part of the delayed response of the consultancy ser- vices can be associated with the reasons other than what had entered the quantitative analysis. It is possible, for in- stance, that defective capabilities of the staff in constructive interdisciplinary interaction (22), crowding of public medical centers while implementing the Healthcare Reform Initiative (23), technical problems of the employed equipment in par- aclinical centers, and shortage of the required materials in such centers, had seriously influenced the delay of different services for timely response to the ED. It is, therefore, sug- gested that the reasons for delayed response of the consul- tancy services be examined in a study. Our results showed that the LOS of a few patients has been prolonged due to ED crowding. This might indicate the insufficiency of the num- ber of the staff and their adequate skills in providing timely services. That being the case, establishing a new ED in the study site is not necessary for the time being, however, de- signing proper interventions to obviate ED crowding can def- initely prove helpful. Yet, considering the aging population in Iran (10), the effect of older age in prolonged LOS can be a warning for an increase in ED crowding (24), followed by increased ratio of patients with prolonged LOS in near fu- ture. Although the generalizability of our results can be in- fluenced by regional differences concerning the staff’s capa- bilities for interdisciplinary interaction and cooperation (25), regional infrastructures and development, population care patterns, and patient distribution, but since Sari, is among the fairly developed Iranian cities, we hope that the results of this study are generalizable to most Iranian EDs, especially This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: www.jemerg.com 5 Emergency. 2017; 5 (1): e53 those situated in areas with populations below one million. In any case, further studies can reveal the generalizability of our findings. 5. Limitation Unfortunately, we couldn’t manage to conduct prospective follow-up of the patients admitted to the ED to determine the outcome of patients with prolonged LOS, as well as, their dis- tribution by time of discharge. Considering the conditions of the patients and also the nature and limitations of the study, we failed to contact the patients to collect more data. We also couldn’t examine, with adequate accuracy, the contributing factors and causes of disposition after 6 hours from admis- sion. We believe these limitations can be objectives for fur- ther studies. 6. Conclusion Older age, lack of insurance support, disposition time > 6 hours due to complexity of patients’ complaint, and the ne- cessity of repeated para-clinical measures were the most im- portant reasons for failed provision of timely services. On the view point of ED personnel, a small part of prolong ED length of stay was concerned with defective ED workflow, while, the most important cause of such delays was the de- layed response of the consultancy services. 7. Appendix 7.1. Acknowledgements: The authors wish to thank the staff of Emergency Depart- ment of Imam Khomeini Hospital of Sari for their valuable contribution in data gathering. 7.2. Author contribution: All authors passed four criteria for authorship contribution based on recommendations of the International Committee of Medical Journal Editors. 7.3. Funding: This study was funded by Mazandaran University of Medical Sciences. 7.4. Conflict of interest: None. References 1. Higginson I. Emergency department crowding. Emer- gency medicine journal. 2012;29(6):437-43. 2. Hoot NR, Aronsky D. Systematic review of emergency de- partment crowding: causes, effects, and solutions. An- nals of emergency medicine. 2008;52(2):126-36. e1. 3. Pines JM, Mullins PM, Cooper JK, Feng LB, Roth KE. National trends in emergency department use, care patterns, and quality of care of older adults in the United States. Journal of the American Geriatrics Society. 2013;61(1):12-7. 4. Pines JM, Hollander JE. Emergency department crowd- ing is associated with poor care for patients with severe pain. Annals of emergency medicine. 2008;51(1):1-5. 5. Reid PP, Compton WD, Grossman JH, Fanjiang G. Build- ing a better delivery system: a new engineering/health care partnership: National Academies Press; 2005. 6. Mason S, Weber EJ, Coster J, Freeman J, Locker T. Time patients spend in the emergency department: England’s 4-hour rule–a case of hitting the target but missing the point? Annals of emergency medicine. 2012;59(5):341-9. 7. Dinh M, Walker A, Parameswaran A, Enright N. Evalu- ating the quality of care delivered by an emergency de- partment fast track unit with both nurse practitioners and doctors. Australasian Emergency Nursing Journal. 2012;15(4):188-94. 8. Hallas P, Ekelund U, Bjornsen LP, Brabrand M. Hoping for a domino effect: a new specialty in Sweden is a breath of fresh air for the development of Scandinavian emergency medicine. Scandinavian journal of trauma, resuscitation and emergency medicine. 2013;21(1):1. 9. Goransson KE, Ehrenberg A, Ehnfors M. Triage in emer- gency departments: national survey. Journal of clinical nursing. 2005;14(9):1067-74. 10. Noroozian M. The elderly population in iran: an ever growing concern in the health system. Iranian journal of psychiatry and behavioral sciences. 2012;6(2):1-6. 11. Mohammad A, Kor EM, Mahmoodi M. The Effect of Time-to-Provider, Left-without-Treatment and Length- of-Stay on Patient Satisfaction in Training Hospitals’ Emergency Department, Iran. Iranian journal of public health. 2015;44(10):1411. 12. Golaghaie F, Sarmadian H, Rafiie M, Nejat N. A study on waiting time and length of stay of attendants to emer- gency department of Vali-e-Asr Hospital, Arak-Iran. Arak Medical University Journal. 2008;11(2):74-83. 13. Clark K, Normile LB. Influence of time-to-interventions for emergency department critical care patients on hospital mortality. Journal of Emergency Nursing. 2007;33(1):6-13. 14. Vermeulen MJ, Guttmann A, Stukel TA, Kachra A, Sivilotti ML, Rowe BH, et al. Are reductions in emergency de- partment length of stay associated with improvements in quality of care? A difference-in-differences analysis. BMJ quality & safety. 2015:bmjqs-2015-004189. This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: www.jemerg.com M. Hosseininejad et al. 6 15. Liew D, Liew D, Kennedy MP. Emergency depart- ment length of stay independently predicts excess in- patient length of stay. Medical Journal of Australia. 2003;179(10):524-7. 16. Molavi-Vardajani H, Haghdoost AA, Shahravan A, Rad M. Cleansing and preparation of data for statistical analysis: A step necessary in oral health sciences research. Journal of Oral Health and Oral Epidemiology. 2016;5(4). 17. Kleinbaum DG, Kupper LL, Nizam A, Rosenberg ES. Ap- plied regression analysis and other multivariable meth- ods: Nelson Education; 2013. 18. Jones P, Schimanski K. The four hour target to reduce emergency department ‘waiting time’: a systematic re- view of clinical outcomes. Emergency Medicine Australa- sia. 2010;22(5):391-8. 19. Bortnick AE, Epps KC, Selzer F, Anwaruddin S, Marro- quin OC, Srinivas V, et al. Five-year follow-up of patients treated for coronary artery disease in the face of an in- creasing burden of co-morbidity and disease complexity (from the NHLBI Dynamic Registry). The American jour- nal of cardiology. 2014;113(4):573-9. 20. Nejhad ZH, Vardanjani HM, Abolhasani F, Hadipour M, Sheikhzadeh K. Relative effect of socio-economic status on the health-related quality of life in type 2 diabetic pa- tients in Iran. Diabetes & Metabolic Syndrome: Clinical Research & Reviews. 2013;7(4):187-90. 21. Penson DF, Stoddard ML, Pasta DJ, Lubeck DP, Flanders SC, Litwin MS. The association between socioeconomic status, health insurance coverage, and quality of life in men with prostate cancer. Journal of clinical epidemiol- ogy. 2001;54(4):350-8. 22. Morey JC, Simon R, Jay GD, Wears RL, Salisbury M, Dukes KA, et al. Error reduction and performance improvement in the emergency department through formal teamwork training: evaluation results of the MedTeams project. Health services research. 2002;37(6):1553-81. 23. Hashemi B, Baratloo A, Forouzafar MM, Motamedi M, Tarkhorani M. Patient satisfaction before and after ex- ecuting health sector evolution plan. Iranian Journal of Emergency Medicine. 2015;2(3):127-33. 24. Naughton C, Drennan J, Treacy P, Fealy G, Kilkenny M, Johnson F, et al. The role of health and non-health- related factors in repeat emergency department visits in an elderly urban population. Emergency medicine jour- nal. 2010;27(9):683-7. 25. Moss JE, Flower CL, Houghton LM, Moss DL, Nielsen DA, Taylor DM. A multidisciplinary Care Coordination Team improves emergency department discharge planning practice. Medical Journal of Australia. 2002;177(8):435-9. This open-access article distributed under the terms of the Creative Commons Attribution NonCommercial 3.0 License (CC BY-NC 3.0). Downloaded from: www.jemerg.com Introduction Methods Results Discussion Limitation Conclusion Appendix References