Hostname: page-component-6898cfcb55-kwdml Total loading time: 0 Render date: 2026-11-12T00:11:22.454Z Has data issue: false hasContentIssue false
  • English
  • Français

Time Pattern of Presentation of Victims of High-Speed Passenger Ferry Mass Casualty Incidents to the Emergency Department

Published online by Cambridge University Press:  18 September 2024

Rex P.K. LamOpen the ORCID record for Rex P.K. Lam[Opens in a new window],
Ronald T.M. Wong,
Eric H.Y. Lau,
K.W. Wong,
Arthur C.K. Cheung,
V.K. Chaang,
L. Chen,
T.C. Tsang,
T.K. Chan and
Peter P.Y. Chee
...Show all authors
Rex P.K. Lam*
Affiliation:
Department of Emergency Medicine, School of Clinical Medicine, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, ChinaAccident and Emergency Department, Queen Mary Hospital, Hong Kong Special Administrative Region, ChinaAccident and Emergency Department, Pamela Youde Nethersole Eastern Hospital, Hong Kong Special Administrative Region, China
Ronald T.M. Wong
Affiliation:
Department of Emergency Medicine, School of Clinical Medicine, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, China
Eric H.Y. Lau
Affiliation:
School of Public Health, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, China
K.W. Wong
Affiliation:
Department of Emergency Medicine, School of Clinical Medicine, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, ChinaAccident and Emergency Department, Pamela Youde Nethersole Eastern Hospital, Hong Kong Special Administrative Region, China
Arthur C.K. Cheung
Affiliation:
Department of Emergency Medicine, School of Clinical Medicine, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, China
V.K. Chaang
Affiliation:
Department of Emergency Medicine, School of Clinical Medicine, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, China
L. Chen
Affiliation:
Department of Emergency Medicine, School of Clinical Medicine, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, China
T.C. Tsang
Affiliation:
Accident and Emergency Department, Queen Mary Hospital, Hong Kong Special Administrative Region, China
T.K. Chan
Affiliation:
Accident and Emergency Department, Ruttonjee Hospital, Hong Kong Special Administrative Region, China
Peter P.Y. Chee
Affiliation:
Accident and Emergency Department, St John Hospital, Hong Kong Special Administrative Region, China
Frank H.F. Ko
Affiliation:
Accident and Emergency Department, Queen Elizabeth Hospital, Hong Kong Special Administrative Region, China
C.S. Leung
Affiliation:
Accident and Emergency Department, Princess Margaret Hospital, Hong Kong Special Administrative Region, China
S.M. Yang
Affiliation:
Accident and Emergency Department, Kwong Wah Hospital, Hong Kong Special Administrative Region, China
*
Corresponding author: Rex P.K. Lam; Email: lampkrex@hku.hk
Rights & Permissions [Opens in a new window]

Abstract

Objective

Mass Casualty Incidents (MCIs) involving high-speed passenger ferries (HSPFs) may result in the dual-wave phenomenon, in which the emergency department (ED) is overwhelmed by an initial wave of minor injuries, followed by a second wave of more seriously injured victims. This study aimed to characterize the time pattern of ED presentation of victims in such accidents in Hong Kong.

Methods

All HSPF MCIs from 2005 to 2015 were reviewed retrospectively, with the time interval from accident to ED registration determined for each victim. Multivariable linear regression was used to identify independent factors associated with the time of ED presentation after the accidents.

Results

Eight MCIs involving 492 victims were identified. Victims with an Injury Severity Score (ISS) ≥ 9 had a significantly shorter median time interval compared to those with minor injuries. An ISS ≥ 9 and evacuation by emergency service vessels were associated with a shorter delay in ED arrival, whereas ship sinking, accident at nighttime, and a longer linear distance between the accident and receiving ED were associated with a longer delay.

Conclusion

The dual-wave phenomenon was not present in HSPF MCIs. Early communication is the key to ensure early resource mobilisation and a well-timed response.

Keywords

disaster planningmass casualty incidentstime-to-treatmentemergency medical services

Information

Type
Original Research
Information
Disaster Medicine and Public Health Preparedness , Volume 18 , 2024, e123
Creative Commons
Creative Common License - CCCreative Common License - BYCreative Common License - NCCreative Common License - SA
This is an Open Access article, distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike licence (http://creativecommons.org/licenses/by-nc-sa/4.0), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the same Creative Commons licence is used to distribute the re-used or adapted article and the original article is properly cited. The written permission of Cambridge University Press must be obtained prior to any commercial use.
Copyright
© The Author(s), 2024. Published by Cambridge University Press on behalf of Society for Disaster Medicine and Public Health, Inc

Introduction

Maritime disasters involving high-speed passenger ferries (HSPFs), though not common, can lead to mass casualties. High-speed collisions result in blunt trauma similar to road vehicle crash injuries.Reference Lam, Wong and Lau 1 Sinking vessels can cause fatalities due to drowning. Rescue is often challenging due to the remoteness of the accident from the shore, large numbers of victims per accident, limited deck space, and on-board medical resources, as well as unfavorable weather, and rough sea conditions.Reference Eischen, Hogan and Burstein 2 Previous reports in literature have highlighted several important problems in such rescues, including difficulty in victim evacuation from the vessels, insufficient communication, and lack of cooperation between responding agencies.Reference Lockey, Purcell-Jones and Davies 3 , Reference Ryu, Yeom and Jeong 4 Emergency departments (EDs) are at the receiving end of the chain of rescue. It is important to mobilize appropriate personnel and resources in a timely manner when responding to such accidents, which often occur with little or no warning.

However, ED response to HSPF accidents may be impeded by a phenomenon called ‘the dual-wave phenomenon.’ This phenomenon refers to a disaster response situation where the ED is overwhelmed by an initial wave of minor injuries, followed by a second wave of victims of more serious injuries. This has been mentioned in many textbooks and articles written on disaster management,Reference Hogan, Burstein, Hogan and Burstein 5 , Reference Waeckerle 6 and included in simulation models of hospital response to mass casualty incidents (MCIs).Reference TariVerdi, Miller-Hooks and Kirsch 7 In the context of HSPF MCI, such a phenomenon was postulated to be possible, given the limited access to the victims on-board, and the difficulty in transporting seriously injured victims out of damaged vessels. However, studies on HSPF MCI are lacking.

There is a need to understand the time pattern of ED presentations of victims, the time intervals between the accidents, and various key clinical activities such as X-ray examination, hospital admission, intensive care unit (ICU) admission, and surgical operations, to inform hospital disaster planning for HSPF MCIs, which requires a concerted response across different clinical departments. This study was conducted to characterize the time pattern of ED presentation of victims, and various key clinical activities after the accidents, and to identify independent factors associated with the time of ED presentation after the accidents. The hypothesis was that the dual-wave phenomenon might be present in such accidents.

Methods

This was a retrospective study on the victims of HSPF MCIs received by the public EDs of 7 major hospitals around Victoria Harbor in Hong Kong from January 1, 2005, to December 31, 2015. These 7 hospitals comprised 3 trauma centers: Queen Mary Hospital, Queen Elizabeth Hospital, and Princess Margaret Hospital; and 4 district hospitals: Ruttonjee Hospital, Pamela Youde Nethersole Eastern Hospital, St John Hospital, and Kwong Wah Hospital. Ethical approval was obtained from the institutional review boards of all participating hospitals (HKU/HA HKW IRB UW16-318, HKEC-2016-110, KC/KE-17-0010/ER-2, KW/EX-17-025[108-04]) prior to the commencement of the study. Patient consent was waived because data were collected and analyzed anonymously.

An MCI was defined as an incident involving 8 or more casualties, as commonly adopted by all government rescue agencies and public EDs in Hong Kong.Reference Shum 8 HSPF accidents that involved fewer than 8 casualties were excluded because they could be routinely managed by most EDs without mobilizing extra personnel and resources. For the definition of HSPF, the local Marine Department classification was followed when screening for accidents for inclusion in our study. 9

Data were obtained from the following sources:

  1. 1) Details of the accidents, including the date, time, and Global Positioning System (GPS) coordinates, as well as vessels involved, circumstances, and number of occupants on board, were sourced from the investigation reports published by the Marine Department.

  2. 2) Clinical records of the victims involved in the selected MCIs were extracted from electronic databases, including the Disaster Group of the Accident and Emergency Information System (AEIS), which were activated in the event of an MCI in the receiving hospitals, and the Clinical Management System of the study hospitals.

  3. 3) Autopsy reports, where available, were accessed from the coroner’s court for victims who died from the included accidents.

When analyzing the time pattern of ED presentation, the time interval (in hours) from the official time of each accident, as stated in the Marine Department investigation report, and the ED registration time of each victim was calculated. For victims who attended more than 1 ED or an ED multiple times after the accident, only the first ED registration was counted. The time intervals between the time of accident and time of various clinical activities, including the first X-ray examination, hospital admission, and intensive care unit (ICU) admission, as well as surgical operations, were also calculated. The number of events was recorded at 15-minute intervals based on the recommendations made by the Academy for Emergency Management and Disaster Medicine (EMDM Academy) Consensus Group.Reference Debacker, Hubloue and Dhondt 10 Victims who presented more than 8 hours after the accident, many of whom attended the ED several days after the accidents for minor injuries, were excluded due to their limited impact on the initial emergency response, and also because their time data would skew the overall analysis (data not shown).

On ED arrival, all victims were triaged by nurses according to the prevailing Hong Kong Accident and Emergency Triage Guidelines uniformly adopted by all public EDs. The 5 categories are 1 (critical); 2 (emergent); 3 (urgent); 4 (semi-urgent); and 5 (non-urgent) based on the severity of the presenting condition and stability of vital signs.Reference Fan and Leung 11 For each incident, the proportion of victims in each triage category was reported. Severity of injury was determined based on the clinical notes, radiology reports, operative records, and autopsy reports where available. A trained coder assigned Abbreviated Injury Scale (AIS) to each injury by body region on a 6-point scale according to the AIS 2015 Dictionary: AIS 1 (minor); AIS 2 (moderate); AIS 3 (serious); AIS 4 (severe); AIS 5 (critical); and AIS 6 (nonsurvivable). 12 The Injury Severity Score (ISS) was calculated by selecting the 3 ISS regions with the highest AIS and summing their squared AIS. An ISS of 75 was allotted if any AIS was 6. Any injury of undetermined severity was allotted an AIS of 9 and an ISS of 99.Reference Baker, O’Neill and Haddon 13 Victims with an ISS 99 were excluded from analysis.

In this study, an ISS ≥ 9 was used as the cut-off point to identify patients with moderate or major trauma.Reference Copes, Champion and Sacco 14 Victims with an ISS < 9 were considered as having minor injuries. This cut-off point has been validated in large trauma registries involving different ethnicity, gender, and age group.Reference Bolorunduro, Villegas and Oyetunji 15 , Reference Palmer 16 An ISS ≥ 9 is associated with a higher mortalityReference Bolorunduro, Villegas and Oyetunji 15 and a longer hospital length of stay.Reference Palmer 16

Analysis

Missing values were not imputed. The characteristics of victims and the time pattern of ED presentation were analyzed using descriptive statistics. The GPS coordinates of the accident locations and the 7 receiving hospitals were plotted on Google Maps and Google Earth (Google, Inc., Mountain View, CA, USA). The geographic distance between the accident sites and receiving EDs of individual victims was calculated based on their GPS coordinates using the Haversine Formula,Footnote 1 which determines the great-circle distance between 2 points over the earth’s surface. The median time of ED registration after the accident between victims with moderate or major trauma (ISS ≥ 9) and those with minor injuries (ISS < 9) was compared using the Mann–Whitney U test.

Multivariable linear regression was then performed to identify independent factors associated with the time of ED presentation after the accidents, taking into account patient factors, such as age and severity of injury (ISS ≥ 9 vs ISS < 9); and circumstantial factors, including the linear distance between the accident site and receiving ED, daytime or night-time occurrence of the accident, whether there was sinking of a vessel, and whether there was vessel evacuation and transportation by emergency service vessels on the spot of accident. We did not enter visibility into the regression model because of the significant collinearity of visibility and night-time occurrence of accident (Spearman correlation = 1.0, P < 0.001). In practice, the time of accident occurrence is more easily available to the receiving EDs. Also, the weather conditions varied considerably across different accidents, and it was difficult to dichotomize them. Therefore, we did not enter weather conditions into the regression model.

Statistical analysis was performed using Statistical Package for the Social Sciences Statistics for Windows, version 23.0 (IBM Corp., Armonk, NY, USA). A P value of < 0.05 was considered statistically significant.

Results

Over the study period, 8 HSPF MCIs were identified. The details of the accidents are shown in Table 1. Figure 1 shows the location of the accidents and receiving hospitals. All accidents were caused by collision of a HSPF with another vessel or object in the sea. In most incidents, the damaged vessels were able to return under their own power to the shore, where victims were transported to various hospitals nearest to the shore gathering point by ground and air transport. In 2 of the accidents (Incidents 4 and 8), evacuation from the damaged vessels was needed and the victims were transported by emergency service vessels or nearby civilian vessels to the shore, then by ground transport to the receiving hospitals. One accident (Incident 5) caused the sinking of a ferry and drowned many victims on board. All victims were rescued by nearby civilian and emergency service vessels and transported to the shore. The median distance between the accident site and the receiving hospital was 12.3 km (IQR 9.4 – 22.9 km).

Table 1.

Details of the mass casualty incidents that involved high-speed passenger ferry

Abbreviations: HSPF, high-speed passenger ferry

Figure 1.

The locations of the mass-casualty incidents that involved high-speed passenger ferries and the locations of the receiving study hospitals, respectively.

A total of 492 victims presented to the EDs of the study hospitals within 8 hours of the accidents, with the number of victims per incident ranging from 15 to 115. Most victims were triaged as semi-urgent (Table 2). The median age of the victims was 43.0 years (range 2–85 years) and there was no gender preponderance. The characteristics of the victims are presented in Table 3. Most injuries were minor and the median ISS for the whole cohort was 1.0 (IQR 1.0 – 2.0). None had an ISS of 99 (undetermined severity of injury). Forty-three (43) victims (8.7%) had moderate or major trauma (ISS ≥ 9), and most of the victims (79.3%) required an X-ray. The proportions of victims who required hospitalization, ICU admission, and surgical operation were 20.1%, 3.0%, and 3.7%, respectively. Eleven victims (2.2%) died, mostly due to drowning.

Table 2.

The number and triage categories of victims presenting to ED within 8 hours of accident

Table 3.

Characteristics of victims

Abbreviations: AIS, Abbreviated Injury Scale; ICU, Intensive Care Unit; IQR, Interquartile Range; ISS, Injury Severity Score

Overall, the median time interval between accident and ED registration was 3.4 hours (IQR 2.8 – 4.2 hours). The distribution of time interval from accident to ED registration of victims with an ISS ≥ 9 and ISS < 9 is shown in Figure 2. Comparing victims with an ISS ≥ 9 to those with a lower ISS, the median time interval from accident to ED registration (2.8 vs 3.5 hours, P = 0.007) was significantly shorter for those with moderate or major trauma. Figure 3 and Supplementary Table 1 show respective time intervals stratified by the severity of injury and individual accidents. In Incidents 4 and 5, the median time intervals to ED registration were significantly shorter for victims with moderate or major trauma. Although the median time interval appeared to be longer for those with ISS ≥ 9 in incident 8, the difference did not reach statistical significance.

Figure 2.

Distribution of time interval from accident to ED registration of victims with ISS < 9 and ISS ≥ 9.

Figure 3.

Distribution of time interval from accident to ED registration of victims stratified by accident and severity of injury.

The distribution of time interval from accident to various clinical activities is shown in Figure 4. The median time intervals between the accident and X-ray, hospital admission, ICU admission, and surgical operation were 4.4 hours (IQR 3.7 – 5.4 hours), 5.5 hours (IQR 4.4 –6.9 hours), 4.4 hours (IQR 3.0 – 19.6 hours), and 27.5 hours (IQR 6.7 – 68.2 hours), respectively. Counting from the time of ED registration, the corresponding figures were 0.6 hours (0.4 - 1.1 hours), 1.8 hours (1.1 - 2.3 hours), 2.2 hours (1.1 - 14.2 hours), and 22.6 hours (3.2 - 63.2 hours), respectively.

Figure 4.

Distribution of time interval from accident to various clinical activities.

Multivariable linear regression showed that a longer time interval from accident to ED registration was significantly associated with the sinking of a vessel (P < 0.001), accident at nighttime (P < 0.001), as well as a longer linear distance between the accident site and receiving ED (P < 0.001). An ISS ≥ 9 (P = 0.04) and vessel evacuation and transportation by emergency service vessels on the spot of accident (P < 0.001) were associated with a significantly shorter time interval. Patient’s age was not significantly associated with the time interval after adjusting for the circumstantial factors and severity of injury (Table 4). The adjusted R-squared of the regression model was 0.3, indicating that it could explain 30% of the variance in the time interval from accident to ED registration.

Table 4.

Multi-variable linear regression model in predicting the time interval between the accident and ED registration

Abbreviations: CI, confidence interval; ISS, Injury Severity Score

Discussion

Accidents involving HSPFs have been reported in Canada, 17 Japan, 18 Spain,Reference Louro, Vazquez and de la Campa 19 and the USA. 20 , 21 When compared with conventional ocean-going vessels, high-speed vessels are more liable to collision, grounding, and other contact events.Reference Antão and Soares 22 Analysis of past accidents has shown that the key causes are human factors, such as failure to maintain a proper lookout because of overconfidence of bridge personnel,Reference Louro, Vazquez and de la Campa 19 and failure to maintain a safe speed due to pressure to keep to schedule.Reference Louro, Vazquez and de la Campa 19 , Reference Antão and Soares 22 Though many of these accidents have resulted in mass casualties, the literature still lacks studies evaluating emergency hospital response to such accidents with the aim to inform disaster planning.

This study showed a time lag of 2 to 3 hours between the accident and ED arrival of the victims across different accidents. This finding is consistent with that reported for similar accidents in the literature. Lockey et al reported that it took 1 hour and 17 minutes to evacuate 300 passengers from a listing catamaran, despite no injuries being sustained during the collision between the catamaran and a rock.Reference Lockey, Purcell-Jones and Davies 3 In a marine disaster where a high-speed passenger ship collided with a whale in South Korea, the field triage and patient transportation process lasted for 3 hours after the collision.Reference Ryu, Yeom and Jeong 4

The implication of such a considerable time lag is that most receiving EDs should have adequate time in mobilizing extra staff and resources to cope with the surge in demand, provided that information about the number of victims and their severity of injury is clear soon after the accident. Given the even longer time intervals between the accident and X-ray examination, hospital admission, ICU admission, and surgical operations; receiving hospitals should have enough time to increase capacity in key downstream clinical departments to avoid creating bottlenecks after ED arrival of the victims.

In Hong Kong, all public EDs are under the administration of the Hospital Authority. As a result of this, in the event of an MCI, patient transfer to different public hospitals is coordinated by the Hospital Authority’s Head Office Major Incident Control Centre (MICC). On request by the scene commander, an ED in the catchment area of a particular MCI can dispatch an emergency physician to play the role of medical control officer to coordinate patient transfer to different public EDs from the scene. In some MCIs, emergency medical teams are dispatched to provide field triage and emergency treatment to casualties. 9 In most scenarios in this study, field triage and patient diversion to different hospitals were performed only when victims reached the gathering point on the shore, not on the involved vessels. Earlier communication between the ferry crew, rescue agencies, MICC, and the receiving hospitals, while the damaged vessel is returning to the pier or the evacuation is still in progress, would help the receiving hospitals to mobilize personnel and resources earlier.

Understandably, assessment of passenger condition is not the only consideration immediately after ferry collision. There are many competing priorities on board, such as assessment of water ingress, possibility of abandoning ship, and risk of oil leak and environmental contamination, as well as notification to the authorities, communication with other affected vessels, etc. In order to quickly assess and communicate passenger condition, the use of a simple field triage algorithm modified for laypersons can be considered. The Simple Triage and Rapid Treatment (START) algorithm is a field triage tool commonly used by rescue agencies and emergency medical teams for disaster in Hong Kong and in many other countries. The START triage ratings correlate well with the ISS.Reference Hong, Sierzenski and Bollinger 23 The algorithm can also be simplified further for quick application by non-medical personnel with little training.Reference Badiali, Giugni and Marcis 24 Instead of applying the full algorithm, ferry crew can be trained to quickly identify the ‘walking wounded,’ i.e., those with minor injuries, and count the rest of the passengers as having more serious injuries. The quickest way to identify the ‘walking wounded’ is to ask the victims whether they can walk. This does not require much medical training of the ferry crew but can provide invaluable information on the scale of accidents to rescue agencies and receiving hospitals using a common language. This algorithm can be included in the emergency response protocol onboard and should be tested and refined in simulation exercise before adoption.

The results of this study also showed that the dual-wave phenomenon was not present in the MCIs that involved HSPFs. Victims with moderate or major trauma (ISS ≥ 9) had a significantly shorter median time interval between accident and ED registration as compared to those with minor injuries. Even after adjustment for circumstantial factors, multivariable linear regression showed that these patients arrived 0.37 hours (i.e. 22.2 minutes) earlier than victims with minor injury. These findings suggest that the pre-hospital triage system was effective in prioritizing care to victims with more serious injuries. The absence of the dual-wave phenomenon in HSPF accidents can also be explained by the fact that none of the victims required prolonged extrication from the damaged vessels, and self-transportation from the accident site, which would otherwise have resulted in earlier presentation of victims of mild injuries, was not seen in HSPF accidents because almost all victims were brought to the shore triage point by rescue agencies. However, it is noteworthy that a few victims with moderate or major trauma continued to arrive at EDs up to 6 hours after the accidents.

Multivariable linear regression showed that the sinking of a vessel in an accident and the night-time occurrence of accidents were associated with 0.98 hours (i.e. 58.8 minutes) and 0.67 hours (i.e. 40.2 minutes) delay in ED arrival, respectively. Both factors increase the difficulty of rescue, and it is not surprising to see such delays. Interestingly, despite a significant association, each 1 km increase in the linear distance between the accident, and receiving ED was only associated with 0.08 hours (i.e. 4.8 minutes) of delay in ED arrival, indicating circumstantial factors are more important than physical distance in affecting the time interval between accident and ED arrival. It is important to note that the regression model showed that victims who were evacuated by emergency service vessels on the spot of the accident arrived at the ED 0.92 hours earlier. This can be explained by the fact that in accidents where the damaged ferries returned to the pier under their own power after collision (thus obviating the need for vessel evacuation and transportation by emergency service vessels), they travelled at a very slow speed, which prolonged the time interval between the accident, and ED arrival. It is important to also note that vessel evacuation and transportation by emergency service vessels on the spot of the accident in the sea carries significant risks. Its association with a shorter time interval should not be viewed as a simple solution to shorten the time delay. Other circumferential factors, such as the conditions of the damaged vessel, weather, and sea conditions, should also be factored in when making such a decision. Selective evacuation of the mostly seriously injured victims on the spot is a potential strategy to balance the benefit of shorter delay and evacuation risks.

Limitations

This study had a few limitations. First, information about the pre-hospital phase of the rescue, such as the field triage category of individual victims, was not accessible. These factors might have influenced the time pattern of presentation. Second, the actual distance between the accident site and receiving hospitals was not traceable. We could only estimate the distance based on GPS coordinates and we believe the calculated distance is the best available surrogate for the actual distance. Third, the factors in the multivariable linear regression model could only explain 30% of the variance in the time interval between accident and ED registration. Many pre-hospital factors, such as the accident site to the port time, waiting time for ground or air-transport at the port, and ground or air transport time of individual victims were not accessible. Fourth, the findings were derived from MCIs that involved HSPFs in Hong Kong waters, with collisions as the major form of the accidents. They might not be generalizable to other places due to regional variations in maritime traffic density, regulation of marine vessels, and disaster response systems. The findings might also not be applicable to other forms of HSPF accidents, such as fire and grounding.

Despite these limitations, this study provides useful real-world data for disaster planning of surge capacity in response to HSPF MCI in the urban setting. To make the response plan more adaptive to different scenarios, further studies comparing different forms of HSPF accidents and MCIs that involve HSPF versus other types of passenger vessels are warranted. Different time points along the chain of emergency response before the hospital arrival of individual victims should also be captured in future studies in order to identify potential targets to shorten the time delay. Calculation of distance between the accidents and receiving hospitals would best be based on data from the nautical chart of the damaged and rescued vessels.

Conclusions

Given the time lag of 2 to 3 hours between the accident and the arrival of victims, EDs should have enough time to mobilize extra resources to cope with the anticipated surge in demand. The dual-wave phenomenon was not present in HSPF accidents, but ED may need to reserve resources for more seriously injured victims who present later during rescue. Our data showed that an ISS ≥ 9 and evacuation by emergency service vessels were associated with a shorter delay in ED arrival, whereas ship sinking, accident at nighttime, and a longer linear distance between the accident, and receiving ED were associated with a longer delay. Early communication between the ferry crew, rescue agencies, and receiving EDs is the key to ensure early mobilization and well-timed response to seriously injured victims arriving at different time points after the accident.

Abbreviations

AIS

Abbreviated Injury Scale

AEIS

Accident & Emergency Information System

CI

Confidence Interval

ED

Emergency Department

EMDM Academy

Academy for Emergency Management and Disaster Medicine

GPS

Global Positioning System

HSPF

High-speed Passenger Ferry

ICU

Intensive Care Unit

IQR

Interquartile Range

ISS

Injury Severity Score

MCI

Mass Casualty Incident

MICC

Major Incident Control Centre

N/A

Not Applicable

START Algorithm

Simple Triage and Rapid Treatment Algorithm

Supplementary material

The supplementary material for this article can be found at http://doi.org/10.1017/dmp.2024.90.

Acknowledgements

This study would not have been possible without the assistance of Karen Chan PhD and Jimmy Leung PhD of the Queen Elizabeth Hospital in retrieving medical records. It would have been impossible without the help of the Marine Department and the Coroner’s Court in providing the details of accidents and autopsy reports. We are also grateful to Ms. Jenny Tsang for her help in trauma coding.

Authors contribution

RPKL conceived and designed the study and developed the method. WRTM, WKW, CACK, and TTC, as well as CTK, CPPY, KFHF, and LCS, retrieved medical records, performed chart review, and collected data. YSM also contributed. WRTM performed trauma coding. RPKL and EHYL analyzed and interpreted the data. EHYL, CVK and CL plotted the figures. RPKL drafted the article. All authors contributed substantially to its revision and provided final approval. PRKL takes the responsibility for the paper.

Funding statement

No grant, funds, or financial support has been received for conducting this study.

Ethical standard

The authors declare that they have no conflict of interest.

References

Lam, RPK, Wong, RTM, Lau, EHY, et al. Injury patterns of mass casualty incidents involving high-speed passenger ferries presenting to accident and emergency departments in Hong Kong: a retrospective review. Injury. 2020;51(2):252–259. doi:10.1016/j.injury.2019.12.001CrossRefGoogle Scholar
Eischen, J. Maritime disasters. In: Hogan, DE and Burstein, JL. Disaster medicine. 2nd ed. Lippincott Williams & Wilkins; 2007.Google Scholar
Lockey, D, Purcell-Jones, G, Davies, CT, et al. Injuries sustained during major evacuation of the high-speed catamaran St Malo off Jersey. Injury. 1997;28(3):187–190. doi:10.1016/s0020-1383(96)00190-8CrossRefGoogle ScholarPubMed
Ryu, JH, Yeom, SR, Jeong, JW, et al. Characteristics and triage of a maritime disaster: an accidental passenger ship collision in Korea. Eur J Emerg Med. 2010;17(3):177–180. doi:10.1097/MEJ.0b013e328330f452CrossRefGoogle ScholarPubMed
Hogan, DE, Burstein, JL. Basic perspectives on disaster. In: Hogan, DE and Burstein, JL. Disaster medicine. 2nd ed. Lippincott Williams & Wilkins; 2007.Google Scholar
Waeckerle, JF. Disaster planning and response. N Engl J Med. 1991;324(12):815–821. doi:10.1056/NEJM199103213241206Google ScholarPubMed
TariVerdi, M, Miller-Hooks, E, Kirsch, T. Strategies for improved hospital response to mass casualty incidents. Disaster Med Public Health Prep. 2018;12(6):778–790. doi:10.1017/dmp.2018.4CrossRefGoogle ScholarPubMed
Shum, KL. Prehospital management of casualties in disaster. The Hong Kong Med Diary. 2015;20(7):7–10.Google Scholar
Hong Kong e-Legislation. Merchant shipping (safety)(high speed craft) regulation. Cap 369, section 107. Hong Kong e-Legislation; 1999.Google Scholar
Debacker, M, Hubloue, I, Dhondt, E, et al. Utstein-style template for uniform data reporting of acute medical response in disasters. PLoS Curr. 2012; doi:10.1371/4f6cf3e8df15a.CrossRefGoogle Scholar
Fan, MM, Leung, LP. Validation of the Hong Kong accident and emergency triage guidelines. Hong Kong Med J. 2013 Jun;19(3):198–202. doi:10.12809/hkmj133900Google Scholar
Association for the Advancement of Automotive Medicine. The abbreviated injury scale 2015 revision. Des Plaines, IL: Association for the Advancement of Automotive Medicine; 2016.Google Scholar
Baker, SP, O’Neill, B, Haddon, W Jr et al. The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma. 1974;14(3):187–196.CrossRefGoogle ScholarPubMed
Copes, WS, Champion, HR, Sacco, WJ, et al. The injury severity score revisited. J Trauma. 1988;28(1):69–77. doi:10.1097/00005373-198801000-00010CrossRefGoogle ScholarPubMed
Bolorunduro, OB, Villegas, C, Oyetunji, TA, et al. Validating the Injury Severity Score (ISS) in different populations: ISS predicts mortality better among Hispanics and females. J Surg Res. 2011;166(1):40–44. doi:10.1016/j.jss.2010.04.012.CrossRefGoogle ScholarPubMed
Palmer, C. Major trauma and the injury severity score--where should we set the bar? Annu Proc Assoc Adv Automot Med. 2007;51:13–29.Google ScholarPubMed
CBC News. Ferry accidents in BC timeline. Published Jun 6, 2006. Accessed Oct 1, 2019. https://www.cbc.ca/news2/background/search_rescue/ferry_accidents_bc.htmlGoogle Scholar
BBC News. Japan ferry ‘hits whale’, injuring more than 80 on board. Published March 9, 2019. Accessed Oct 1, 2019. https://www.bbc.com/news/world-asia-47509017Google Scholar
Louro, J, Vazquez, P, de la Campa, RM. Accident risk factors in conventional and high speed ferry ships in Spain. JMR. 2012;9(1):3–8.Google Scholar
National Transportation Safety Board (NTSB). Confusion over engine mode caused New York City ferry crash. Published July 31, 2014. Accessed October 1, 2019. https://professionalmariner.com/ntsb-confusion-over-engine-mode-caused-new-york-city-ferry-crash/Google Scholar
Newsweek. Hyannis ferry crash: injuries reported as high-speed boat ploughs into jetty. Published June 17, 2017. Accessed October 1, 2019. https://www.newsweek.com/hyannis-ferry-crash-nantucket-marthas-vineyard-woods-hole-626949Google Scholar
Antão, P and Soares, CG. Causal factors in accidents of high-speed craft and conventional ocean-going vessels. Reliab Eng Syst Saf. 2008;93(9):1292–1304. doi:10.1016/j.ress.2007.07.010CrossRefGoogle Scholar
Hong, R, Sierzenski, PR, Bollinger, M, et al. Does the simple triage and rapid treatment method appropriately triage patients based on trauma injury severity score? Am J Disaster Med. 2008;3(5):265–271.Google ScholarPubMed
Badiali, S, Giugni, A, Marcis, L. Testing the START Triage Protocol: can it improve the ability of nonmedical personnel to better triage patients during disasters and mass casualties’ incidents? Disaster Med Public Health Prep. 2017;11(3):305–309. doi:10.1017/dmp.2016.151CrossRefGoogle ScholarPubMed
Figure 0

Table 1.Details of the mass casualty incidents that involved high-speed passenger ferry

Figure 1

Figure 1.The locations of the mass-casualty incidents that involved high-speed passenger ferries and the locations of the receiving study hospitals, respectively.

Figure 2

Table 2.The number and triage categories of victims presenting to ED within 8 hours of accident

Figure 3

Table 3.Characteristics of victims

Figure 4

Figure 2.Distribution of time interval from accident to ED registration of victims with ISS < 9 and ISS ≥ 9.

Figure 5

Figure 3.Distribution of time interval from accident to ED registration of victims stratified by accident and severity of injury.

Figure 6

Figure 4.Distribution of time interval from accident to various clinical activities.

Figure 7

Table 4.Multi-variable linear regression model in predicting the time interval between the accident and ED registration

Supplementary material: File

Lam et al. supplementary material

Lam et al. supplementary material
Download Lam et al. supplementary material(File)
File 14.5 KB