Welcome to my blog. I document my adventures in travel, style, and food. Hope you have a nice stay!

Aum Shinrikyo, Part 2: Points of Failure

Aum Shinrikyo, Part 2: Points of Failure

The following is the second in a series examining Aum Shinrikyo and the Tokyo subway attack of 1995. While the work is my own, it benefited greatly from input provided by my old colleague Scott Cormier. Thanks Scott. – (Brad Trefz, 19 April, 2016).

In our last post, we broke down the events in Tokyo on March 20, 1995. (You can read that here). In this post we’ll look at some of the points of failure in the response to those events, as well as lessons learned, and those not learned.  There were four primary points of failure, in this post we will examine the first three and address the last separately, in the third part of this series.

Four points of failure emerge from the events of March 1995:

  1. Agent Identification
  2. Decontamination/Personnel Protection
  3. Medical System Response
  4. Psychological casualties and the worried well

Many of the issues in medical system response during Tokyo primarily relate to the areas of decontamination, psychological casualties, and the worried well - which affect the medical system disproportionately, but cause issue in other areas as well. A few problems are unique to the medical system. The sections below discuss each of these.

1. Agent Identification

It was three hours before St. Luke’s confirmed its suspicions of organophosphate poisoning. The “confirmation” was unofficial as it came from media reports based on statements by the Tokyo Police Department (TPD). TPD knew about Aum Shinrikyo’s chemical and biological program and its production of sarin. However, it was several hours into the event before those with that knowledge communicated it beyond the investigative team. The Tokyo Fire Department (TFD) did not possess a capability in 1995 to identify or deal with sarin and relied on experts from the Japanese Ground Self Defense Force, who eventually carried out decontamination of the station following the attack.[i]

Since 1995, extensive efforts around the world have largely mitigated this problem. Many major metropolitan areas in the developed world now possess the capability to do chemical detection and preliminary analysis of chemicals. Technology has also advanced radically and instruments formerly confined to physical laboratories for chemical analysis like infrared or Raman spectroscopy are now handheld instruments ruggedized for field use. In addition, many developed nations now have specialized teams designated for response to CBRN terrorism. In the United States ,the Aum Shinrikyo attack lead to the formation of the U.S. National Guard’s Civil Support Team-Weapons of Mass Destruction program. Congressionally mandated in 1999, the Department of Defense specifically designed the teams to address the agent identification issue and they continue to carry out that mission in all fifty states and five territories.

2. Personnel Protection

As noted above, many responders and hospital workers did not wear the proper personal protective equipment (PPE) during their response to the attack. The attack significantly raised awareness of this issue. The TFD possessed adequate protection for the response. Fire turnout gear with self-contained breathing apparatus (SCBA) if worn with chemical resistant gloves is adequate for initial response to most chemical incidents.[ii]  The first fire department responders wore this protection until they determined the source was not a gas leak. Follow-on units did not deploy with such protection. In addition, a large number of police, EMTs, and others that responded, wore no PPE. Thus, many of the responders had little to protect them from contamination, agent vapors, or contaminated casualties.

Sarin volatilizes at relatively low temperatures. Contamination via aerosalization is the preferred goal for military chemical weapons that contain sarin or other chemical agents in liquid form. This is because liquid exposure resulting in absorption and inhalation of aerosolized liquid produces the most severe casualties. Sarin is a non-persistent agent because of its rapid and low temperature volatilization. As sarin evaporates, the vapor, if inhaled, can also produce casualties. In this case, severity of injury will depend on a variety of factors, primarily on concentration level of the vapors. Therefore, a confined space will have higher concentrations but ventilation can reduce the concentration to more manageable levels. In Tokyo, the concentration inside the trains was high while concentrations in the station were lower.

Vapor inhalation is a form of “internal contamination.” That requires medical treatment, not decontamination.  If no liquid externally contaminates the individual who inhaled the vapors, then that individual poses no cross-contamination threat. External contamination implies actual liquid agent on a surface. Individuals evacuating the trains through the pooled sarin liquid and a worker who handled one of the packages became externally contaminated. There was no external contamination on the majority of passengers in the stations, until they encountered a contaminated individual that managed to transfer the contamination to them. While emergency responders treat all items that emerge from a “hotzone” as contaminated for safety reasons, not everything that enters or exits a contaminated area becomes contaminated. The actual ratio of externally contaminated individuals versus those suffering inhalation injuries was presumably large. While cross contamination occurred between individuals, the majority of secondary casualties were also likely from inhalation, probably from proximity to externally contaminated casualties, like those in the St. Luke’s Chapel, or ambulance personnel.  

Figure   5   -    Initial responders outside a subway station during the attack. The man in the trench coat is standing at the subway entrance. (Source: Daily Mail, June 15, 2012)   [iii]

Figure 5 - Initial responders outside a subway station during the attack. The man in the trench coat is standing at the subway entrance. (Source: Daily Mail, June 15, 2012)[iii]


Many of the PPE issues evident in Tokyo no longer exist. In the years since 1995, training efforts and the issuance of PPE equipment to responders outside the fire service including police and EMS addressed these problems, though they may still occur in unprepared communities or in less developed nations. Many hospitals, due to recent outbreaks like Ebola and Avian Flu, as well as concerns over terrorism and other incidents, have also increased their stockpile of PPE, though the majority of it remains focused on biological threats rather than chemical or radiological threats, except in those areas were such threats are more likely/common.   

3. Decontamination

There was no decontamination at the incident scene during the Tokyo attack until special teams responded long after all casualties and passengers departed. The only decontamination of casualties known to have occurred, as discussed previously, happened some three hours after the initial event. That decontamination process removed any remaining clothing on casualties and washed them with a soap and water mix using existing bathing facilities within the hospital. Decontamination, despite not occurring in Tokyo, deserves a further examination here.

 The lack of decontamination during the Tokyo attack was widely absorbed as a lesson learned after the attack. Since 1995, hospitals, fire departments, and other elements have all expended considerable effort in the United States to improve their decontamination capabilities. The U.S. Department of Health estimated in 2008 that 99 percent of the nation’s 5,212 hospitals had emergency response plans for chemical incidents or attacks, which St. Luke’s and other Japanese hospitals lacked in 1995. An estimated 93.2 percent had plans for dealing with biological incidents (94.1 for epidemics), and 81.3 percent had plans for radiological incidents.[iv] In the same survey, 69.6 percent of hospitals reported addressing decontamination as part of their emergency drills.[v] Only 11 percent conducted no drills.[vi]

Decontamination of casualties can come in two forms, gross or detailed. While OSHA designates a lengthy and detailed ten to fifteen step process for hazardous materials operations, decontamination of mass casualties need not be so involved. Removal of outer clothing may remove the majority of contamination.[vii] If followed with a high-volume, low-pressure rinse, the likelihood of contamination leaving a decontamination site is low. For ambulatory patients, the majority in Tokyo, gross decontamination is possible with a single fire truck using a fog nozzle and an established procedure.

Unfortunately, the issue of decontamination is a contentious and unsettled one with significant amounts of debate. There are also a number of vendors selling advanced and expensive decontamination systems and multiple Federal and State decontamination teams with equally large footprints. These can include the U.S. Marine Corps Chemical and Biological Incident Response Force (CBIRF) and the U.S. National Guard’s chemical, biological, radiological, nuclear, and high yield explosive (CBRNE) enhanced response force packages (CERFP). Of course, these teams can take hours to days to deploy and unless pre-deployed for a large event are unlikely to be present to support a community in the initial response. Likewise, their systems, and the systems sold by many vendors take time to set up. Given most agents in contact with skin will be absorbed in four to fifteen minutes, these solutions are poor for all but preplanned events considered to be potential terrorist targets.

The majority of gross and mass decontamination solutions use the principle that “dilution is the solution to all pollution.” Outer clothing removal is critical, followed by a copious rinse. The use of decontamination solutions other than water are not recommended for skin under most conditions, though the use of regular soap can increase efficacy when washing contaminated casualties.[viii]

Many U.S. fire departments use a more basic and scalable solution, recommended here. This starts with the aforementioned pumper truck and hose with a fog nozzle while other firefighters in bunker gear, chemical resistant gloves and boots, and SCBAs direct ambulatory casualties through the wash instructing them to remove and pile their clothing. In turn, this expands to two trucks side by side and an aerial truck, increasing water volume and providing a lane for casualties. As additional department assets deploy, casualties use bags for clothing, firefighters shift to true OSHA Level B protective suits, and they establish a separate lane for non-ambulatory casualties, which may be one of the aforementioned vendor systems. 

This represents a “decontamination triage” that precedes medical triage. Casualties completing this gross decontamination are provided blankets and/or additional clothing and are then medically triaged, while non-ambulatory casualties go through their own two-stage triage, though they will typically be medically triaged before decontamination unlike ambulatory patients. For the most part, ambulatory casualties will be low evacuation priorities while non-ambulatory casualties will process for immediate evacuation to medical facilities due to their injuries, hence the medical prioritization for decontamination.


Figure 6 -  A Mass Decontamination set up with two pumper trucks and an aerial spray. (Source: Village of Palatine, Illinois)

Figure 6 - A Mass Decontamination set up with two pumper trucks and an aerial spray. (Source: Village of Palatine, Illinois)

This setup is the basic system most commonly found in jurisdictions with decontamination planning for mass casualties. It can vary from locality to locality and adjusts based on a variety of factors. For instance, northern states may need to plan for mass casualty decontamination from indoor facilities where cold weather presents unique challenges for gross decontamination. 

The key breakdown in most of these decontamination plans, procedures, and teams that developed since 1995 is that they fail to account for self-evacuation. As Tokyo shows, the majority of casualties self-evacuated. While a prepared and trained fire department might be able to establish scene control and hasty decontamination relatively quickly, response time is a factor. Sixty-one percent of structure fires in the United States have a response time of less than six minutes.[ix] Response time in this case runs from the “time of call to the arrival of the first apparatus.”[x]This suggests that basic primary single hose decontamination could begin within about ten to fifteen minutes from first call. In a mass casualty situation, many individuals could leave the scene in ten to fifteen minutes. Likewise, in a large incident or in a dispersed multiple location incident like that in Tokyo, scene control may take some time to establish. During the response to the Alfred P. Murrah Federal Building Bombing in Oklahoma City, Oklahoma, which preceded the Tokyo attack by a month, basic scene control did not exist until an hour into the response, after the site evacuated due to bomb scares and responders controlled who returned. In addition to numerous self-evacuations, bystanders also entered the site to assist, and one, a nurse, died due to injuries sustained in that effort.[xi]

Self-evacuating casualties arrive at hospitals and medical facilities, and not just those adjacent to the attack. Some portion of them will carry contamination. This is exactly what happened in Tokyo. Tokyo also received contaminated casualties by ambulance and those ambulances became contaminated. This is not a rare event either. Many trauma centers and emergency rooms (ER) in the United States, especially those near chemical facilities, have a story about a contaminated casualty entering the ER.[xii] These events occur when self-evacuating contaminated casualties enter the ER or when the fire department or EMS fails to notify the receiving hospital to expect a contaminated casualty. Contaminated and closed ERs can be the result. Both problems were manifest in Tokyo – self-reporting contaminated casualties and contaminated casualties arriving in ambulances. 

This is the area of greatest weakness in most medical systems. The while some hospitals are not equipped, trained, or prepared to conduct mass casualty decontamination, even more lack the ability to handle basic decontamination of a small number of casualties from industrial accidents or spills.[xiii] Likewise, they are unlikely to establish perimeter control during incidents. St. Luke’s, as noted, did not control access during the Tokyo attack. Even if a hospital establishes a decontamination station, if it does not control access to the hospital and directs patients to that station, contamination can still result. Some hospital emergency management plans still consider decontamination to be a “fire department” responsibility, though this has improved significantly in recent years. Of course, as Tokyo showed, the fire department and EMS are fully committed at the scene in a large CBRN event and are unlikely to have assets to devote to multiple hospitals where casualties might arrive. While a possible planning solution is to devote fire assets to hospitals in the event of a CBRN event, fire and hospital officials view the problem differently and such solutions are difficult to implement. As noted, just resolving the normal communication problem involving a single contaminated casualty would be a huge step forward in the majority of jurisdictions.  

Therefore, hospitals must be prepared to do decontamination using organic assets. Hospitals with decontamination capabilities usually possess a shower system somewhere near their emergency room for the purpose. These can often end up as storage, and most can only handle two patients every 15 minutes or so, if normal procedures are followed. This is clearly not adequate in a mass casualty scenario and it is not immediately scaleable. Since the solution in mass casualty situations is gross decontamination, the key for medical facilities is to establish site access control, direct casualties from the incident to an initial decontamination triage (similar to the aforementioned process conducted on scene), and then carry out medical triage on the backside of the decontamination line. Hospitals should also designate an alternate entrance to the facility for decontaminated casualties arriving from the incident site, and for those not associated with the CBRN event, and must have a procedure to identify those casualties as already decontaminated - a procedure established prior to the incident and familiar to all personnel.  Medical personnel should not run decontamination lines to the degree possible. Trained non-medical hospital staff should run scene control and decontamination to the extent possible, while medical personnel focus on treating casualties after decontamination.

The key elements of this process are actually quite simple and do not require extensive equipment, extended training, or expense. Many of these elements are “free” in that they only require dedication and time. Annual refreshers with a focus on “train the trainers” who can implement a mass decontamination through “Just-In-Time” training of volunteers and staff are essential.[xiv] These elements can be readily adapted into existing operating practices, and some, like site and access control, make sense for a number of reasons having nothing to do with CBRN events. These key elements are:

  • Site access control - First step in the event of a CBRN event at any hospital or medical facility should be to “lock the doors” and establish control over all arriving patients to prevent contaminating the hospital, staff, or other patients. Begin with the emergency department and expand outward until control extends over the entire campus.
  • Gross decontamination – The principles are the same as those used by the fire department – dilution is the solution to all pollution. Both the fire hoses and hydrants are already available at many hospitals. Training, fog nozzles, and a procedure are the only missing elements. Additional procedures for dealing with casualty clothing and PPE for those meeting potentially contaminated casualties are important, but not vital in a pinch, as “field expedient” measures can replace such niceties. Some hospitals have even adapted fire control and lawn sprinklers for use in an emergency. The goal is to have a decontamination plan that will work and is “good enough.” It need not be perfect. Any way casualties get from clothed to near naked and rinsed while not contaminating others is a functional solution.
  • Coordination with the scene – The lack of communication between the TPD, TFD, EMS, and St. Luke’s was a constant problem. This problem is not a rarity in other jurisdictions. Hospitals and medical facilities like urgent care clinics must be “in the loop” for all disaster planning, but especially for CBRN incidents. A CBRN response plan that does not address this issue is not adequate. As noted, addressing this issue in day-to-day operations is already an imperative in some jurisdictions. The more that this can be worked into regular operations the better it will work in a CBRN event.   

Next time we’ll look at the unique issue of the” Worried Well,” that presented a significant problem in Tokyo, and are a known factor in most CBRN events. Until then, thanks for visiting CBRNPro.net, please be sure to like our facebook page for all our latest updates.


[i] The Japanese Constitution forbids the establishment of an army. The Japanese Ground Self-Defense Force is more like the US National Guard in its basing and mission, but is an active force. Its CBRN experts also assisted in the response to the Fukishima disaster.

[ii] Turnout gear, also called bunker gear, and SCBA with chemical resistant boots and gloves offers protection roughly equivalent to Occupational Health and Safety Administration Level B PPE levels, which require a chemical resistant suit in place of the bunker gear. Several tests have shown that the “breakthrough” time from limited chemical exposure on bunker gear is long enough to afford basic protection for initial response if other means of protection are unavailable, though the issue is contentious and unsettled. 

[iii] Shari Miller, “Japanese police arrest final fugitive connected with 1995 deadly gas attack on subway,” Daily Mail, June 15, 2012, http://www.dailymail.co.uk/news/article-2159752/Katsuya-Takahashi-arrested-Aum-Shinrikyo-gas-attack-fugitive-caught-Japanese-police.html (accessed December 8, 2013).

[iv] Niska, Richard W. and Iris M. Shimizu, “Hospital Preparedness for Emergency Response: United States, 2008,” National Health Statistics Reports 37 (Hyattsville, MD: National Center for Health Statistics, 2011).

[v] Ibid.

[vi] Ibid.

[vii] This depends on the time of year and the clothing, among other factors. During summer or in warmer climates, clothing removal is less effective for the simple fact people tend to wear less of it when it is hot, thus increasing the amount of exposed skin vulnerable to agent absorption.

[viii] Bleach and HTB have a contact time longer than the absorption rate of skin and are contra-indicated on people. They are also caustic in high concentrations and can cause skin burns and inhalation damage in sensitive individuals. Water is best, soap and water is better. Studies have shown the efficacy of surfactants like dish soap as being the most effective to remove contamination. It is better to decontaminate runoff solutions and equipment using neutralization.

[ix] Department of Homeland Security, U.S. Fire Administration, National Fire Data Center, Structure Fire Response Times (Emmitsburg, MD: U.S. Fire Administration, 2006):1-4.

[x] Ibid.

[xi] For a full accounting of the Oklahoma bombing response, see The Oklahoma Department of Civil Emergency Management, After Action Report: Alfred P. Murrah Federal Building Bombing, 19 April 1995 in Oklahoma City, Oklahoma (Oklahoma City: Oklahoma Department of Central Services, Central Printing Division, 1995), http://www.ok.gov/OEM/documents/Bombing%20After %20Action%20Report.pdf (accessed December 2, 2013).

[xii] During training a few years ago with seven hospitals in one city, the author heard several of these stories similar to those heard in many other jurisdictions since. One of those examples is particularly appropriate. In that incident the casualty, a truck driver injured in a wreck of his tanker, brought to the designated hospital to receive contaminated casualties arrived grossly contaminated with chlorinated liquid. The fire department and ambulance crew both failed to notify the hospital the casualty had contamination until hospital staff and patients began to suffer from the fumes after the hospital placed the man in the middle of the ER, which had to be closed.

[xiii] This is the odd problem created by government actions post 9-11. Many large hospitals received training and funds from homeland security and other programs to acquire equipment and systems to conduct mass casualty decontamination, yet most still cannot execute small-scale containment and decontamination for a limited number of casualties, a problem noted during the recent Ebola outbreaks in the US. Even in hospitals where the capability for small scale decontamination exists, a lack of training and recognition still results in contaminated emergency rooms around the country.

[xiv] Due to high staff turnover rates in many medical facilities, hospitals cannot expect all of their staff to be trained for major events. Instead, medical facilities should focus on training key individuals across all shifts who can carry out leadership roles in the event of a major incident and provide on-the-spot training and instruction. This makes for a scalable and cost/time-efficient solution.


Aum Shinrikyo, Part 3: Worried Well

Aum Shinrikyo, Part 3: Worried Well

Aum Shinrikyo, Part 1: What really happened 21 years ago?

Aum Shinrikyo, Part 1: What really happened 21 years ago?