- Email: [email protected]
Facilitating emergency hospital evacuation through uniform discharge criteria
YAJEM-56393; No of Pages 4 American Journal of Emergency Medicine xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Facilitating emergency hospital evacuation through uniform discharge criteria☆,☆☆ Sandra Keret, M.EM a,b, Meital Nahari, M.EM b,c, Ofer Merin, MD d, Limor Aharonson-Daniel, PhD b,e, Sara Goldberg, MPH d, Bruria Adini, PhD a,b,⁎ a
Ministry of Health, Jerusalem, Israel Department of Emergency Medicine, Leon & Mathilde Recanati School of Community Health Professions, Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer Sheva, Israel Hadassah Medical Center, Jerusalem, Israel d Shaare Zedek Medical Center, Jerusalem, Israel e Prepared Center for Emergency Response Research, Ben-Gurion University of the Negev, Beer Sheva, Israel b c
a r t i c l e
i n f o
Article history: Received 8 October 2016 Received in revised form 29 December 2016 Accepted 29 December 2016 Available online xxxx
a b s t r a c t Background: Though hospitals' operational continuity is crucial, full institutional evacuation may at times be unavoidable. The study's objective was to establish criteria for discharge of patients during complete emergency evacuation and compare scope of patients suitable for discharge pre/post implementation of criteria. Basic procedures: Standards for patient discharge during an evacuation were developed based on literature and disaster managers. The standards were reviewed in a two-round Delphi process. All hospitals in Israel were requested to identify inpatients' that could be released home during institutional evacuation. Potential discharges were compared in 2013–2014, before and after formulation of discharge criteria. Main findings: Consensus exceeding 80% was obtained for four out of five criteria after two Delphi cycles. Average projected discharge rate before and after formulation of criteria was 34.2% and 42.9%, respectively (p b 0.001). Variance in potential dischargeable patients was 31-fold less in 2014 than in 2013 (MST = 8,452 versus MST = 264,366, respectively; p b 0.001). Differences were found between small, medium and large hospitals in mean rate of dischargeable patients: 52.1%, 41.5% and 42.2%, respectively (p = 0.001). Principle conclusions: The study's findings enable to forecast the extent of patients that may be released home during full emergency evacuation of a hospital; thereby facilitating preparedness of contingency plans. © 2016 Elsevier Inc. All rights reserved.
1. Introduction Hospitals play a vital role in emergencies and ordinary times alike, providing a critical framework that caters to the community's needs and promotes its sense of security [1,2]. While ensuring operational continuity is of paramount importance, at times evacuation of a medical facility may be unavoidable – due to fire, earthquake, terror incident, or any other situation that places patients and personnel at risk [1,3-6]. Numerous hospital evacuations have been described in the literature. In the years 1971–1999, 275 medical centers were evacuated due to one of the following causes: fire (23%), hurricane (14%), earthquake (9%), flooding (6%), collapse of infrastructure (5%), and internal or external leakage of hazardous materials (18% and 4% respectively) [7]. Most probably, additional hospitals were evacuated at the described ☆ This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. ☆☆ There are no conflicts of interest for any of the authors. ⁎ Corresponding author at: Disaster Management & Injury Prevention Dept., School of Public Health, Sackler Faculty of Medicine, Tel-Aviv University, Ramat Aviv, Tel Aviv 6997801, Israel. E-mail address: [email protected] (B. Adini).
period resulting from various emergency scenarios, but due to lack of press coverage or other means of reporting, they were not included in the documented statistics. In 1994, following the earthquake at Northridge, eight hospitals were evacuated in Los Angeles [8]. More than 20 hospitals had to be evacuated in the United States in 2005 in the wake of Hurricane Katrina [2]. In 2010 a forest fire necessitated the evacuation of an entire psychiatric hospital in Israel [9] and threatened an additional facility, which was prepared for evacuation. Several medical institutions were evacuated in New York in 2012 following Hurricane Sandy, among them some leading public hospitals, despite the early warning that was issued which enabled implementation of early emergency preparedness measures [4]. Hospitals operate continuously 24/7. Occupancy tends to be high, and hospital populations comprise of vulnerable and sensitive patients [3] as well as extensive medical and support staff. Planning the evacuation of a medical center is thus a complex process [1,3-6] in which logistic challenges have to be addressed [1-2,9-10]. When a disaster dictates complete evacuation of a hospital, some of the patients can be released directly to the community. A study of the evacuation of 62 hospitals in Los Angeles found that 22–30% of the patients could be discharged home [2], while the rest needed to be
http://dx.doi.org/10.1016/j.ajem.2016.12.071 0735-6757/© 2016 Elsevier Inc. All rights reserved.
Please cite this article as: Keret S, et al, Facilitating emergency hospital evacuation through uniform discharge criteria, American Journal of Emergency Medicine (2016), http://dx.doi.org/10.1016/j.ajem.2016.12.071
2
S. Keret et al. / American Journal of Emergency Medicine xxx (2016) xxx–xxx
relocated to other hospitals. In some cases the decision to discharge patients to their homes in the framework of the emergency evacuation was guided based on defined criteria, e.g. the patients were fit to be immediately released to the community or were already slated for discharge within 24 to 72 h from the time of evacuation [11]. Nonetheless, clear and consistent criteria for patient evacuation, including the distinction of the destination of the evacuation, is still missing [4, 12]. Discharging patients directly to the community, rather than evacuating them to alternate medical facilities, may significantly decrease the burden [1]. Thus, there is great value in identifying elements that may facilitate the release of patients to their homes. In view of the assumption that defined criteria [12] may increase the number of hospitalized patients that can be discharged home during an emergency hospital evacuation, a comprehensive study was undertaken. The aim was to establish criteria regarding discharge of hospital patients to home during complete evacuation of a medical facility and to compare the number of patients deemed suitable for discharge prior to and after definition of these criteria. 2. Methods A survey of the scientific literature published in English from 1990 to 2013 was conducted through PubMed and Cochrane search engines using the following keywords: hospital evacuation, emergency health care facility, surge capacity, triage, discharge criteria, and evacuation standards. The survey was continued throughout the study, to include additional articles that were published up to 2015 inclusive. Standards governing the release of hospitalized patients to the community in situations of full hospital evacuation were defined based both on the literature review and on the recommendations of the Israeli National Advisory Committee on Hospital Evacuation. In the framework of a two-round Delphi process, these standards were disseminated to 36 content experts, 20 of them members of the Committee; 16 were experts in the field of disaster management or hospital directors. The content experts were asked to express agreement or disagreement regarding each of the standards and to propose additional standards for patient discharge. Agreement between experts was defined as consensus in excess of 80%. Following the approval of the Ministry of Health, all 26 general hospitals in Israel were requested to evaluate the needs of their inpatients and identify those that could be released to the community in a situation requiring full evacuation of the institution. The evaluations took into account the volumes of inpatients (occupancy rates of the Israeli hospitals ranged in the study years from 78% to 125%, with an average of 95%) and average length of stay (4.3 days) [13]. Two cycles of such evaluations were conducted in the general hospitals during the years 2013–2014. The first round of mapping was conducted in 2013 prior to the establishment of uniform criteria for patient release to their homes and relied on the personal judgment of the staff of each hospital department. A second round of mapping was performed in 2014, following the distribution of uniform criteria for patient release; thus, the hospital teams used the newly developed standards for identifying patients that could be discharged to the community. The patients found suitable for discharge to the community were collected in an Excel spreadsheet listing the overall number of hospitalized patients and the number of patients that can be discharged. The rates of inpatient discharge in the simulated evacuation as determined in 2013 and 2014 – before and after the formulation of the standards – were compared for hospitals with matching characteristics. Hospitals were divided into three categories according to size: 1) large, with N 700 beds (N = 6);2) medium, with 400–700 beds (N = 8); 3) small hospitals with fewer than 400 beds (N = 5). They were also classed according to location: urban (N = 10) or rural (N = 9). The different hospital departments were grouped together in five healthcare divisions as follows: pediatric, obstetrics/gynecology (OB/Gyn), surgery,
internal medicine and intensive care. The data were analyzed using SPSS software. Chi square analysis was used to examine relationships between hospitals' characteristics and the percentage of inpatients dischargeable to the community. One-way ANOVA was used to examine the variability in discharge before and after the establishment of the discharge standards. 3. Results 3.1. Formulation of criteria for releasing patients to the community and testing them using the Delphi technique Based on the literature review, three standards were formulated directing the discharge of patients to their homes. The standards were then forwarded to 36 content experts for validation. The response rate was 67% (N = 24), and agreement of over 80% was found in regard to each of the three standards. In this first round, the content experts proposed two additional standards to be incorporated in the medical directives regarding inpatients to be discharged home during hospital evacuations. In view of the high rate of agreement obtained in the first round, the second Delphi round focused only on the newly proposed standards. Among the 18 content experts that responded to the second cycle, consensus exceeding 80% was obtained for only one of the two standards surveyed. The standards disseminated in the Delphi rounds and the degrees of consensus regarding their incorporation in the discharge criteria are detailed in Table 1. 3.2. Percentages of patients who can be discharged during evacuation of a hospital In the first cycle of evaluating the hospitals' inpatients, conducted in February–March 2013, results were obtained from 19 hospitals (73% response rate). In the second cycle, conducted in February–March 2014, data were received from 20 of the hospitals (77% response rate). In order to conduct paired comparisons, only data from the 19 hospitals that responded in both cycles was used in the analysis. The percentage of patients slated for discharge in the event of a hospital evacuation as assessed in 2013 in the first round (prior to formulation of the standards) was compared with the percentage as assessed in 2014 in the second round (after formulation of the standards). The average projected discharge rate before and after formulation of the standards was 34.2% and 42.9%, respectively (p b 0.001). The variance in the data provided by the different hospitals regarding their ability to discharge patients before and after the establishment of the standards (in the two separate years) was analyzed using one way ANOVA. The variance in the rate of dischargeable patients was 31-fold less in 2014 than in 2013 (MST = 8452 versus MST = 264 366, respectively; p b 0.001). The reduction in variance was particularly evident in hospitals in
Table 1 Degree of consensus regarding standards for discharge of hospitalized patients to home after each of two Delphi cycles.
Criterion
1st Delphi round, % agreement (N = 24)
Patient slated for discharge on same day 100 Candidate for discharge within 24–48 h, no 87.5 further treatment needed Patients whose treatment can be interrupted for a 87.5 brief perioda Postpartum mother/newborn baby 8–12 h after normal delivery Postpartum mother 24 h after vaginal delivery without complications
2nd Delphi round, % agreement (N = 18)
38.8 88.8
a Including chronic patients, patients admitted for elective surgery, testing, or in-patient evaluation.
Please cite this article as: Keret S, et al, Facilitating emergency hospital evacuation through uniform discharge criteria, American Journal of Emergency Medicine (2016), http://dx.doi.org/10.1016/j.ajem.2016.12.071
S. Keret et al. / American Journal of Emergency Medicine xxx (2016) xxx–xxx
which the release of patients to the community was exceptionally low or high in 2013. For example, the rate of potential patient discharge in a hospital which had reported a very low ability to release patients in 2013 increased significantly, namely from 8.6% to 38.8%. Similarly, the rate of potential patient discharge in a hospital that had reported a very high ability to discharge patients in 2013, 67.5%, dropped to 49.1% in 2014. The rates of patients discharged from the entire patient population of hospitals in 2013 and 2014 are presented in Fig. 1. 3.3. Percentage of dischargeable patients relative to hospital characteristics Chi square analysis was used to examine the relationship between the size of the hospital and the percentage of inpatients dischargeable to the community. Marked differences were found between small, medium and large hospitals in mean rate of dischargeable patients: 52.1%, 41.5% and 42.2%, respectively (p = 0.001). Chi square analysis was also applied to examine the link between location and the rate of dischargeable patients. A clear-cut difference (p = 0.001) was found between the projected discharge rates from rural and urban hospitals (48.6% and 41.3%, respectively). 4. Discussion Hospitals are vulnerable to both natural and man-made disasters. Damage to the hospital structure or an event that endangers the population of the hospital may compromise its operational continuity to the point where full evacuation to alternative facilities becomes unavoidable [1-6]. Hospitals shelter a fragile patient population with special needs, together with the medical and paramedical personnel looking after them, and moreover must cope with continuous high occupancy [1,3,9]. Full evacuation of a hospital is a complex operation that requires extensive logistic capabilities on the part of all elements involved – evacuating hospital, medical facilities receiving the evacuees, transportation needs to evacuate patients, community health services responsible for supporting the evacuees, and emergency health services that relocate the patients to alternative facilities [10,14]. Standards governing patient discharge criteria were defined in this study with a view to facilitate the process of deciding which patients could be released to the community, even while needing further treatment. It was established that adoption of these uniform discharge standards results in an increase in the percentage of patients deemed dischargeable. In the course of the study itself the percentage of dischargeable patients was observed to increase from 34.2% before introduction of the standards to 42.9% subsequent to their adoption. The
R
90
a
80
t
70
e i n
60 50
pre-standards discharge rate supports findings of an earlier study conducted by Zoraster indicating that the proportion of discharges (in the absence of any criteria governing release) was of the order of 22%– 30% [2]. This increase in release rates is significant as it promotes channeling of resources – necessarily restricted in an emergency situation – to the hospital population that needs those most [1,10]. It should though be noted that the higher discharge rates may decrease the resources that will need to be invested by the hospitals but simultaneously, will probably challenge the out-of-hospital medical services such as primary care, social work and additional medical or welfare services. Using agreed criteria not only contributes to higher rates of patient discharge to home, it also reduces variance between hospitals, which in turn enables more reliable evaluation of a hospital's requirements when faced with the need to evacuate its entire population [6]. It can be learned from the study's results that the decreased variance after implementation of the discharge criteria is mainly derived from change that was noted in hospitals that presented extreme rates of potential discharge (b 30% or N 60%). Four hospitals that reported ability to release b30% of the patients and two hospitals that reported ability to release over 60% of the patients prior to implementation of the discharge criteria, most significantly modified their forecasted discharge rates following the use of the standard discharge criteria. Utilization of clear-cut standards saves time, an important consideration in view of the fact that most evacuations are characterized by some degree of urgency. Relying on checklists and clearly delineated criteria reduces uncertainty and promotes efficiency [4,12], though exercise of personal judgment is still required. Another advantage of reliance on clear standards is that it promotes release of those patients who are indeed fit for discharge and by the same token precludes discharge of patients whose condition is likely to grow worse. Rates of patient discharge from small hospitals tended to be high compared with medium and large institutions. It is reasonable to presume that the medical illnesses of patients routinely treated in medium and large hospitals tend to be more complicated than those of patients in small facilities [15], which might explain their inability to discharge as many patients. It was also found that rates of discharge from rural hospitals were higher as compared with facilities situated in urban centers. Perhaps owing to their greater remoteness, in normal times rural hospitals in the periphery maintain longer lengths of stay [16] possibly delaying the release of patients before it can be ascertained that their condition is stable; during an emergency, these patients may then be discharged home more rapidly. Utilization of the discharge standards may help hospital directors and Ministry of Health officials prepare appropriate contingency plans for the evacuation of medical institutions. In an actual crisis, the availability of such plans may assist authorities in coping effectively with complex situations, facilitate decision-making, and implement policies in the shortest possible time. It should be mentioned that although the standards governing patient discharge in the study were intended to foster uniformity, divergences may still occur between the various medical agents/authorities in regard to the decision making process [15]. 5. Limitations
40 30
%
3
20
10 Before setting 0 standards After setting standards
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19
Hospitals
Fig. 1. Rate of discharge of hospitalized patients to the community in 2013 and 2014, before and after establishment of discharge standards.
The response rate in the Delphi cycles was relatively low (67% in the initial cycle) and even lower in the second cycle (50%), thus there may be bias among the responses. Nonetheless, the testing of the adopted criteria across a national healthcare system enables to significantly review their effectiveness and increase the generalizability of the findings. 6. Conclusions Four criteria for discharging hospitalized patients to the community were developed and validated, in the aim of facilitating an efficient
Please cite this article as: Keret S, et al, Facilitating emergency hospital evacuation through uniform discharge criteria, American Journal of Emergency Medicine (2016), http://dx.doi.org/10.1016/j.ajem.2016.12.071
4
S. Keret et al. / American Journal of Emergency Medicine xxx (2016) xxx–xxx
evacuation of a hospital during crisis. The formulation and use of standard criteria tends to increase the number of patients deemed dischargeable to home and helps narrow the variance between hospitals. The rate of patients released to the community was found in the study to be higher after adoption of these criteria. The percentage of dischargeable patients was higher for smaller hospitals as compared with medium and large hospitals, and also for rural hospitals situated in the periphery as compared with urban facilities. In light of the natural disasters and other emergency events that occur worldwide time and again, advance planning of hospital evacuations is critically important. The findings of this study make it possible to forecast the extent of patients that may be released to the community during an event that necessitates full evacuation of a hospital during an emergency. In view of the study's results, it is strongly recommended that hospital administrators and ministries of health adopt the discharge criteria and discuss them with the health care providers, as the need to evacuate a hospital will in most cases catch us without any early warning.
Author contribution statement MO & AB conceived the study and designed the methodology. KS & NM conducted the study, while AB and ADL supervised its conduct. KS & NM analyzed the data and drafted the manuscript. AB, MO, ADL and GS reviewed the manuscript and contributed substantially to its revision. AB takes responsibility for the paper as a whole.
References [1] Hicks J, Glick R. A meta-analysis of hospital evacuations: overcoming barriers to effective planning. J Healthc Risk Manag 2015;34(3):26–36. [2] Zoraster RM, Amara R, Fruhwirth K. Transportation resource requirements for hospital evacuation. Am J Disaster Med 2011;6(3):173–86. [3] Bish DR, Agca E, Glick R. Decision support for hospital evacuation and emergency response. Ann Oper Res 2014;221:89–106. [4] Powell T, Hanfling D, Gostin LO. Emergency preparedness and public health - the lessons of Hurricane Sandy. JAMA 2012;308(24):2569–70. [5] Liu X, Liu Y, Zhang L, et al. Mass aeromedical evacuation of patients in an emergency: experience following the 2010 Yushu earthquake. J Emerg Med 2013;45(6):865–71. [6] Golmohammadi D, Shimshak D. Estimation of the evacuation time in an emergency situation in hospitals. Comput Ind Eng 2011;61:1256–67. [7] Petinaux B, Yadav K. Patient-driven resource planning of a health care facility evacuation. Prehosp Disaster Med 2013;28(2):120–6. [8] Schultz CH, Koenig KL, Lewis RJ. Implications of hospital evacuation after the Northridge, California, earthquake. N Engl J Med 2003;348(14):1349–55. [9] Kreinin A, Shakera T, Sheinkman A, et al. Evacuation of a mental health center during a forest fire in Israel. Disaster Med Public Health Prep 2014;8(4):288–92. [10] Wabo N, Ortenwall P, Khoram-Manesh A. Hospital evacuation; planning, assessment, performance and evaluation. J Acute Dis 2012;58–64. [11] Davis DP, Poste JC, Hicks T, et al. Hospital bed surge capacity in the event of a masscasualty incident. Prehosp Disaster Med 2005;20(3):169–76. [12] Institute of Medicine. Crisis standards of care: a systems framework for catastrophic disaster response. Washington, DC: National Academies Press; 2012. [13] OECD Library. Lengths of hospital stay. Available at: http://www.oecd-ilibrary.org/ social-issues-migration-health/average-length-of-stay-acute-care_l-o-s-acutecaretable-en. (Accessed December 11, 2015). [14] Tayfur E, Gaaffe K. A model for allocating resources during hospital evacuation. Comput Ind Eng 2009;57:1313–23. [15] Adini B, Laor D, Cohen C, et al. Decision to evacuate a hospital during an emergency: the safe way or the leader's way? J Public Health Policy 2012;33(2):257–68. [16] EL Bitar YF, Illingworth KD, Scaife SL, et al. Hospital length of stay following primary total knee arthroplasty: data from the nationwide inpatient sample database. J Arthroplasty 2015;30(10):1710–5.
Please cite this article as: Keret S, et al, Facilitating emergency hospital evacuation through uniform discharge criteria, American Journal of Emergency Medicine (2016), http://dx.doi.org/10.1016/j.ajem.2016.12.071
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Facilitating emergency hospital evacuation through uniform discharge criteria☆,☆☆ Sandra Keret, M.EM a,b, Meital Nahari, M.EM b,c, Ofer Merin, MD d, Limor Aharonson-Daniel, PhD b,e, Sara Goldberg, MPH d, Bruria Adini, PhD a,b,⁎ a
Ministry of Health, Jerusalem, Israel Department of Emergency Medicine, Leon & Mathilde Recanati School of Community Health Professions, Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer Sheva, Israel Hadassah Medical Center, Jerusalem, Israel d Shaare Zedek Medical Center, Jerusalem, Israel e Prepared Center for Emergency Response Research, Ben-Gurion University of the Negev, Beer Sheva, Israel b c
a r t i c l e
i n f o
Article history: Received 8 October 2016 Received in revised form 29 December 2016 Accepted 29 December 2016 Available online xxxx
a b s t r a c t Background: Though hospitals' operational continuity is crucial, full institutional evacuation may at times be unavoidable. The study's objective was to establish criteria for discharge of patients during complete emergency evacuation and compare scope of patients suitable for discharge pre/post implementation of criteria. Basic procedures: Standards for patient discharge during an evacuation were developed based on literature and disaster managers. The standards were reviewed in a two-round Delphi process. All hospitals in Israel were requested to identify inpatients' that could be released home during institutional evacuation. Potential discharges were compared in 2013–2014, before and after formulation of discharge criteria. Main findings: Consensus exceeding 80% was obtained for four out of five criteria after two Delphi cycles. Average projected discharge rate before and after formulation of criteria was 34.2% and 42.9%, respectively (p b 0.001). Variance in potential dischargeable patients was 31-fold less in 2014 than in 2013 (MST = 8,452 versus MST = 264,366, respectively; p b 0.001). Differences were found between small, medium and large hospitals in mean rate of dischargeable patients: 52.1%, 41.5% and 42.2%, respectively (p = 0.001). Principle conclusions: The study's findings enable to forecast the extent of patients that may be released home during full emergency evacuation of a hospital; thereby facilitating preparedness of contingency plans. © 2016 Elsevier Inc. All rights reserved.
1. Introduction Hospitals play a vital role in emergencies and ordinary times alike, providing a critical framework that caters to the community's needs and promotes its sense of security [1,2]. While ensuring operational continuity is of paramount importance, at times evacuation of a medical facility may be unavoidable – due to fire, earthquake, terror incident, or any other situation that places patients and personnel at risk [1,3-6]. Numerous hospital evacuations have been described in the literature. In the years 1971–1999, 275 medical centers were evacuated due to one of the following causes: fire (23%), hurricane (14%), earthquake (9%), flooding (6%), collapse of infrastructure (5%), and internal or external leakage of hazardous materials (18% and 4% respectively) [7]. Most probably, additional hospitals were evacuated at the described ☆ This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. ☆☆ There are no conflicts of interest for any of the authors. ⁎ Corresponding author at: Disaster Management & Injury Prevention Dept., School of Public Health, Sackler Faculty of Medicine, Tel-Aviv University, Ramat Aviv, Tel Aviv 6997801, Israel. E-mail address: [email protected] (B. Adini).
period resulting from various emergency scenarios, but due to lack of press coverage or other means of reporting, they were not included in the documented statistics. In 1994, following the earthquake at Northridge, eight hospitals were evacuated in Los Angeles [8]. More than 20 hospitals had to be evacuated in the United States in 2005 in the wake of Hurricane Katrina [2]. In 2010 a forest fire necessitated the evacuation of an entire psychiatric hospital in Israel [9] and threatened an additional facility, which was prepared for evacuation. Several medical institutions were evacuated in New York in 2012 following Hurricane Sandy, among them some leading public hospitals, despite the early warning that was issued which enabled implementation of early emergency preparedness measures [4]. Hospitals operate continuously 24/7. Occupancy tends to be high, and hospital populations comprise of vulnerable and sensitive patients [3] as well as extensive medical and support staff. Planning the evacuation of a medical center is thus a complex process [1,3-6] in which logistic challenges have to be addressed [1-2,9-10]. When a disaster dictates complete evacuation of a hospital, some of the patients can be released directly to the community. A study of the evacuation of 62 hospitals in Los Angeles found that 22–30% of the patients could be discharged home [2], while the rest needed to be
http://dx.doi.org/10.1016/j.ajem.2016.12.071 0735-6757/© 2016 Elsevier Inc. All rights reserved.
Please cite this article as: Keret S, et al, Facilitating emergency hospital evacuation through uniform discharge criteria, American Journal of Emergency Medicine (2016), http://dx.doi.org/10.1016/j.ajem.2016.12.071
2
S. Keret et al. / American Journal of Emergency Medicine xxx (2016) xxx–xxx
relocated to other hospitals. In some cases the decision to discharge patients to their homes in the framework of the emergency evacuation was guided based on defined criteria, e.g. the patients were fit to be immediately released to the community or were already slated for discharge within 24 to 72 h from the time of evacuation [11]. Nonetheless, clear and consistent criteria for patient evacuation, including the distinction of the destination of the evacuation, is still missing [4, 12]. Discharging patients directly to the community, rather than evacuating them to alternate medical facilities, may significantly decrease the burden [1]. Thus, there is great value in identifying elements that may facilitate the release of patients to their homes. In view of the assumption that defined criteria [12] may increase the number of hospitalized patients that can be discharged home during an emergency hospital evacuation, a comprehensive study was undertaken. The aim was to establish criteria regarding discharge of hospital patients to home during complete evacuation of a medical facility and to compare the number of patients deemed suitable for discharge prior to and after definition of these criteria. 2. Methods A survey of the scientific literature published in English from 1990 to 2013 was conducted through PubMed and Cochrane search engines using the following keywords: hospital evacuation, emergency health care facility, surge capacity, triage, discharge criteria, and evacuation standards. The survey was continued throughout the study, to include additional articles that were published up to 2015 inclusive. Standards governing the release of hospitalized patients to the community in situations of full hospital evacuation were defined based both on the literature review and on the recommendations of the Israeli National Advisory Committee on Hospital Evacuation. In the framework of a two-round Delphi process, these standards were disseminated to 36 content experts, 20 of them members of the Committee; 16 were experts in the field of disaster management or hospital directors. The content experts were asked to express agreement or disagreement regarding each of the standards and to propose additional standards for patient discharge. Agreement between experts was defined as consensus in excess of 80%. Following the approval of the Ministry of Health, all 26 general hospitals in Israel were requested to evaluate the needs of their inpatients and identify those that could be released to the community in a situation requiring full evacuation of the institution. The evaluations took into account the volumes of inpatients (occupancy rates of the Israeli hospitals ranged in the study years from 78% to 125%, with an average of 95%) and average length of stay (4.3 days) [13]. Two cycles of such evaluations were conducted in the general hospitals during the years 2013–2014. The first round of mapping was conducted in 2013 prior to the establishment of uniform criteria for patient release to their homes and relied on the personal judgment of the staff of each hospital department. A second round of mapping was performed in 2014, following the distribution of uniform criteria for patient release; thus, the hospital teams used the newly developed standards for identifying patients that could be discharged to the community. The patients found suitable for discharge to the community were collected in an Excel spreadsheet listing the overall number of hospitalized patients and the number of patients that can be discharged. The rates of inpatient discharge in the simulated evacuation as determined in 2013 and 2014 – before and after the formulation of the standards – were compared for hospitals with matching characteristics. Hospitals were divided into three categories according to size: 1) large, with N 700 beds (N = 6);2) medium, with 400–700 beds (N = 8); 3) small hospitals with fewer than 400 beds (N = 5). They were also classed according to location: urban (N = 10) or rural (N = 9). The different hospital departments were grouped together in five healthcare divisions as follows: pediatric, obstetrics/gynecology (OB/Gyn), surgery,
internal medicine and intensive care. The data were analyzed using SPSS software. Chi square analysis was used to examine relationships between hospitals' characteristics and the percentage of inpatients dischargeable to the community. One-way ANOVA was used to examine the variability in discharge before and after the establishment of the discharge standards. 3. Results 3.1. Formulation of criteria for releasing patients to the community and testing them using the Delphi technique Based on the literature review, three standards were formulated directing the discharge of patients to their homes. The standards were then forwarded to 36 content experts for validation. The response rate was 67% (N = 24), and agreement of over 80% was found in regard to each of the three standards. In this first round, the content experts proposed two additional standards to be incorporated in the medical directives regarding inpatients to be discharged home during hospital evacuations. In view of the high rate of agreement obtained in the first round, the second Delphi round focused only on the newly proposed standards. Among the 18 content experts that responded to the second cycle, consensus exceeding 80% was obtained for only one of the two standards surveyed. The standards disseminated in the Delphi rounds and the degrees of consensus regarding their incorporation in the discharge criteria are detailed in Table 1. 3.2. Percentages of patients who can be discharged during evacuation of a hospital In the first cycle of evaluating the hospitals' inpatients, conducted in February–March 2013, results were obtained from 19 hospitals (73% response rate). In the second cycle, conducted in February–March 2014, data were received from 20 of the hospitals (77% response rate). In order to conduct paired comparisons, only data from the 19 hospitals that responded in both cycles was used in the analysis. The percentage of patients slated for discharge in the event of a hospital evacuation as assessed in 2013 in the first round (prior to formulation of the standards) was compared with the percentage as assessed in 2014 in the second round (after formulation of the standards). The average projected discharge rate before and after formulation of the standards was 34.2% and 42.9%, respectively (p b 0.001). The variance in the data provided by the different hospitals regarding their ability to discharge patients before and after the establishment of the standards (in the two separate years) was analyzed using one way ANOVA. The variance in the rate of dischargeable patients was 31-fold less in 2014 than in 2013 (MST = 8452 versus MST = 264 366, respectively; p b 0.001). The reduction in variance was particularly evident in hospitals in
Table 1 Degree of consensus regarding standards for discharge of hospitalized patients to home after each of two Delphi cycles.
Criterion
1st Delphi round, % agreement (N = 24)
Patient slated for discharge on same day 100 Candidate for discharge within 24–48 h, no 87.5 further treatment needed Patients whose treatment can be interrupted for a 87.5 brief perioda Postpartum mother/newborn baby 8–12 h after normal delivery Postpartum mother 24 h after vaginal delivery without complications
2nd Delphi round, % agreement (N = 18)
38.8 88.8
a Including chronic patients, patients admitted for elective surgery, testing, or in-patient evaluation.
Please cite this article as: Keret S, et al, Facilitating emergency hospital evacuation through uniform discharge criteria, American Journal of Emergency Medicine (2016), http://dx.doi.org/10.1016/j.ajem.2016.12.071
S. Keret et al. / American Journal of Emergency Medicine xxx (2016) xxx–xxx
which the release of patients to the community was exceptionally low or high in 2013. For example, the rate of potential patient discharge in a hospital which had reported a very low ability to release patients in 2013 increased significantly, namely from 8.6% to 38.8%. Similarly, the rate of potential patient discharge in a hospital that had reported a very high ability to discharge patients in 2013, 67.5%, dropped to 49.1% in 2014. The rates of patients discharged from the entire patient population of hospitals in 2013 and 2014 are presented in Fig. 1. 3.3. Percentage of dischargeable patients relative to hospital characteristics Chi square analysis was used to examine the relationship between the size of the hospital and the percentage of inpatients dischargeable to the community. Marked differences were found between small, medium and large hospitals in mean rate of dischargeable patients: 52.1%, 41.5% and 42.2%, respectively (p = 0.001). Chi square analysis was also applied to examine the link between location and the rate of dischargeable patients. A clear-cut difference (p = 0.001) was found between the projected discharge rates from rural and urban hospitals (48.6% and 41.3%, respectively). 4. Discussion Hospitals are vulnerable to both natural and man-made disasters. Damage to the hospital structure or an event that endangers the population of the hospital may compromise its operational continuity to the point where full evacuation to alternative facilities becomes unavoidable [1-6]. Hospitals shelter a fragile patient population with special needs, together with the medical and paramedical personnel looking after them, and moreover must cope with continuous high occupancy [1,3,9]. Full evacuation of a hospital is a complex operation that requires extensive logistic capabilities on the part of all elements involved – evacuating hospital, medical facilities receiving the evacuees, transportation needs to evacuate patients, community health services responsible for supporting the evacuees, and emergency health services that relocate the patients to alternative facilities [10,14]. Standards governing patient discharge criteria were defined in this study with a view to facilitate the process of deciding which patients could be released to the community, even while needing further treatment. It was established that adoption of these uniform discharge standards results in an increase in the percentage of patients deemed dischargeable. In the course of the study itself the percentage of dischargeable patients was observed to increase from 34.2% before introduction of the standards to 42.9% subsequent to their adoption. The
R
90
a
80
t
70
e i n
60 50
pre-standards discharge rate supports findings of an earlier study conducted by Zoraster indicating that the proportion of discharges (in the absence of any criteria governing release) was of the order of 22%– 30% [2]. This increase in release rates is significant as it promotes channeling of resources – necessarily restricted in an emergency situation – to the hospital population that needs those most [1,10]. It should though be noted that the higher discharge rates may decrease the resources that will need to be invested by the hospitals but simultaneously, will probably challenge the out-of-hospital medical services such as primary care, social work and additional medical or welfare services. Using agreed criteria not only contributes to higher rates of patient discharge to home, it also reduces variance between hospitals, which in turn enables more reliable evaluation of a hospital's requirements when faced with the need to evacuate its entire population [6]. It can be learned from the study's results that the decreased variance after implementation of the discharge criteria is mainly derived from change that was noted in hospitals that presented extreme rates of potential discharge (b 30% or N 60%). Four hospitals that reported ability to release b30% of the patients and two hospitals that reported ability to release over 60% of the patients prior to implementation of the discharge criteria, most significantly modified their forecasted discharge rates following the use of the standard discharge criteria. Utilization of clear-cut standards saves time, an important consideration in view of the fact that most evacuations are characterized by some degree of urgency. Relying on checklists and clearly delineated criteria reduces uncertainty and promotes efficiency [4,12], though exercise of personal judgment is still required. Another advantage of reliance on clear standards is that it promotes release of those patients who are indeed fit for discharge and by the same token precludes discharge of patients whose condition is likely to grow worse. Rates of patient discharge from small hospitals tended to be high compared with medium and large institutions. It is reasonable to presume that the medical illnesses of patients routinely treated in medium and large hospitals tend to be more complicated than those of patients in small facilities [15], which might explain their inability to discharge as many patients. It was also found that rates of discharge from rural hospitals were higher as compared with facilities situated in urban centers. Perhaps owing to their greater remoteness, in normal times rural hospitals in the periphery maintain longer lengths of stay [16] possibly delaying the release of patients before it can be ascertained that their condition is stable; during an emergency, these patients may then be discharged home more rapidly. Utilization of the discharge standards may help hospital directors and Ministry of Health officials prepare appropriate contingency plans for the evacuation of medical institutions. In an actual crisis, the availability of such plans may assist authorities in coping effectively with complex situations, facilitate decision-making, and implement policies in the shortest possible time. It should be mentioned that although the standards governing patient discharge in the study were intended to foster uniformity, divergences may still occur between the various medical agents/authorities in regard to the decision making process [15]. 5. Limitations
40 30
%
3
20
10 Before setting 0 standards After setting standards
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19
Hospitals
Fig. 1. Rate of discharge of hospitalized patients to the community in 2013 and 2014, before and after establishment of discharge standards.
The response rate in the Delphi cycles was relatively low (67% in the initial cycle) and even lower in the second cycle (50%), thus there may be bias among the responses. Nonetheless, the testing of the adopted criteria across a national healthcare system enables to significantly review their effectiveness and increase the generalizability of the findings. 6. Conclusions Four criteria for discharging hospitalized patients to the community were developed and validated, in the aim of facilitating an efficient
Please cite this article as: Keret S, et al, Facilitating emergency hospital evacuation through uniform discharge criteria, American Journal of Emergency Medicine (2016), http://dx.doi.org/10.1016/j.ajem.2016.12.071
4
S. Keret et al. / American Journal of Emergency Medicine xxx (2016) xxx–xxx
evacuation of a hospital during crisis. The formulation and use of standard criteria tends to increase the number of patients deemed dischargeable to home and helps narrow the variance between hospitals. The rate of patients released to the community was found in the study to be higher after adoption of these criteria. The percentage of dischargeable patients was higher for smaller hospitals as compared with medium and large hospitals, and also for rural hospitals situated in the periphery as compared with urban facilities. In light of the natural disasters and other emergency events that occur worldwide time and again, advance planning of hospital evacuations is critically important. The findings of this study make it possible to forecast the extent of patients that may be released to the community during an event that necessitates full evacuation of a hospital during an emergency. In view of the study's results, it is strongly recommended that hospital administrators and ministries of health adopt the discharge criteria and discuss them with the health care providers, as the need to evacuate a hospital will in most cases catch us without any early warning.
Author contribution statement MO & AB conceived the study and designed the methodology. KS & NM conducted the study, while AB and ADL supervised its conduct. KS & NM analyzed the data and drafted the manuscript. AB, MO, ADL and GS reviewed the manuscript and contributed substantially to its revision. AB takes responsibility for the paper as a whole.
References [1] Hicks J, Glick R. A meta-analysis of hospital evacuations: overcoming barriers to effective planning. J Healthc Risk Manag 2015;34(3):26–36. [2] Zoraster RM, Amara R, Fruhwirth K. Transportation resource requirements for hospital evacuation. Am J Disaster Med 2011;6(3):173–86. [3] Bish DR, Agca E, Glick R. Decision support for hospital evacuation and emergency response. Ann Oper Res 2014;221:89–106. [4] Powell T, Hanfling D, Gostin LO. Emergency preparedness and public health - the lessons of Hurricane Sandy. JAMA 2012;308(24):2569–70. [5] Liu X, Liu Y, Zhang L, et al. Mass aeromedical evacuation of patients in an emergency: experience following the 2010 Yushu earthquake. J Emerg Med 2013;45(6):865–71. [6] Golmohammadi D, Shimshak D. Estimation of the evacuation time in an emergency situation in hospitals. Comput Ind Eng 2011;61:1256–67. [7] Petinaux B, Yadav K. Patient-driven resource planning of a health care facility evacuation. Prehosp Disaster Med 2013;28(2):120–6. [8] Schultz CH, Koenig KL, Lewis RJ. Implications of hospital evacuation after the Northridge, California, earthquake. N Engl J Med 2003;348(14):1349–55. [9] Kreinin A, Shakera T, Sheinkman A, et al. Evacuation of a mental health center during a forest fire in Israel. Disaster Med Public Health Prep 2014;8(4):288–92. [10] Wabo N, Ortenwall P, Khoram-Manesh A. Hospital evacuation; planning, assessment, performance and evaluation. J Acute Dis 2012;58–64. [11] Davis DP, Poste JC, Hicks T, et al. Hospital bed surge capacity in the event of a masscasualty incident. Prehosp Disaster Med 2005;20(3):169–76. [12] Institute of Medicine. Crisis standards of care: a systems framework for catastrophic disaster response. Washington, DC: National Academies Press; 2012. [13] OECD Library. Lengths of hospital stay. Available at: http://www.oecd-ilibrary.org/ social-issues-migration-health/average-length-of-stay-acute-care_l-o-s-acutecaretable-en. (Accessed December 11, 2015). [14] Tayfur E, Gaaffe K. A model for allocating resources during hospital evacuation. Comput Ind Eng 2009;57:1313–23. [15] Adini B, Laor D, Cohen C, et al. Decision to evacuate a hospital during an emergency: the safe way or the leader's way? J Public Health Policy 2012;33(2):257–68. [16] EL Bitar YF, Illingworth KD, Scaife SL, et al. Hospital length of stay following primary total knee arthroplasty: data from the nationwide inpatient sample database. J Arthroplasty 2015;30(10):1710–5.
Please cite this article as: Keret S, et al, Facilitating emergency hospital evacuation through uniform discharge criteria, American Journal of Emergency Medicine (2016), http://dx.doi.org/10.1016/j.ajem.2016.12.071