Health Care Facility Hazard and Vulnerability Analysis

In 1988 Jan deBoer and colleagues published the first attempt to mathematically score and classify a disaster, with the intent to create data to be used prospectively during the management of a calamity. They defined a disaster as “a destructive event that caused so many casualties that extraordinary mobilization of medical services was necessary.” In the proposed Medical Severity Index of Disasters, the parameters needed to quantify a disaster were the casualty load (number of casualties), the severity of incident (severity of injuries sustained), and the capacity of medical services. Twenty-six years later, the importance of determining the effects of a disaster on a health care facility (HCF) has heightened because HCFs have become industrial leaders in the community and, therefore, must be able to swiftly return to normal business functioning. Individual health care providers are acutely aware of the business side of their practice while at the bedside, but they are not cognizant of the ramifications that a disruption of normal HCF operations would have on the community. Business and industry emergency management principles by which HCFs can accommodate the clinical effects of a disaster are discussed in this chapter.

Disasters are events that cause significant enough damage to disrupt the normal activities or function of a community and overwhelm the local resources. What may be an easily handled event in a large urban city may be a disaster for a rural town. Although disasters are not predictable with any great accuracy, many consequences of disasters can be anticipated as part of a comprehensive emergency management plan that includes a hazard and vulnerability analysis (HVA). The HVA will help HCFs plan for these events and allow them to continue operating while assessing structural and operational damage, acquiring needed essentials, and protecting staff and patients. Much can be learned from business and industry with respect to preparedness.

Although not a new concept in business and industry, an HVA is a component in the development of a hospital disaster plan, as recognized since 2001 by The Joint Commission. The emergency management standard (EC.4.1) requires hospitals to identify specific procedures in response to a variety of disasters based on an HVA performed by the organization. The HVA will assist in the mitigation and preparedness of the HCF to respond to and then recover from a disaster. A hazard can be any threat that could cause injury, fatality, property, infrastructure, or environmental damage or impair operations. An HVA is a tool used by emergency management to screen for risk and plan for strategic use of potentially limited resources.

Historical perspective

Many events have affected HCFs in the past. The future is certain to exhibit challenges as we move into a more technologically advanced society challenged by geopolitical terrorist threats. Hospitals inherently have to be prepared for emergencies. Preparation traditionally was based on informal HVAs and was largely dependent on perceived issues. For example, hospitals in northern climates typically planned for adverse winter weather–related issues; hospitals in the South planned for hurricanes; and hospitals in Southern California planned for forest and wildland fires and earthquakes. Failure to consider all hazards when developing the facility emergency response plan was a flaw in the informal approach.

Hazard identification relies on the collection of potential emergencies that the HCF or operation could anticipate encountering. This list may be assembled by cause, by location, or by a combination of both criteria. Causes may be divided into the following categories for assessment purposes: naturally derived emergencies, technologically derived emergencies, and humanmade emergencies ( Box 23-1 ). HCFs experience two types of disasters: (1) those internal to the HCF, isolated to the confines of the HCF physical plant, and (2) those occurring external to the HCF that produce direct effects (casualties) and indirect effects (e.g., loss of electricity and supply due to damaged roads) to the HCF. In the past few years, several major disasters, natural external events, have affected HCFs: in 2005, Hurricane Katrina caused complete devastation of the primary hospital in New Orleans; in 2011, an F5 tornado devastated St. John’s Regional Medical Center in Joplin, Missouri; and in 2012, Hurricane Sandy forced the evacuation of three large New York City medical centers. Internal disasters are similar to those encountered by business and industry, which until recently were not formally considered in an HVA. Milsten’s exhaustive review of direct and indirect disasters that hospitals faced from 1977 to 1997 showed that external and internal disasters are not mutually exclusive. This chapter focuses on the HVA and events specific to internal disasters, those that directly occur within the confines of the HCF, and those that indirectly affect the HCF because of consequences of a disaster.

Box 23-1

Hazard Identification

Natural Events

•

Drought
•

Earthquake
•

Flood
•

Hurricane (cyclone, typhoon)
•

Landslide
•

Severe thunderstorm
•

Temperature extremes
•

Tidal wave
•

Tornado
•

Wildland fires
•

Windstorm
•

Snow/ice storms/blizzard
•

Volcanic ash
•

Meteor crashes
•

Infestation

Technology Event

•

Aircraft crash
•

Medical evacuation helicopter crash
•

Other aviation crash
•

Loss of medical gases
•

Air
•

Oxygen
•

Nitrogen
•

Nitrous oxide
•

Electrical/power shortage or failure
•

Loss of backup generator(s)
•

Fire: chemical, paper, wood, and other
•

Computer network disruption or loss
•

Loss of fire alarm/smoke detection
•

Loss of steam
•

Food contamination
•

Pneumatic tube disruption or loss
•

Food supply interruption
•

Loss or leak of potable water
•

Fixed facilities incidents
•

Loss of suction/vacuum
•

Loss of fuel oil supply or delivery
•

Elevator service disruption or loss
•

Hazardous material release
•

Structural failure
•

Natural gas/pipeline disruption
•

Noxious fumes
•

Sewer failure
•

HVAC failure
•

Loss of equipment requiring cooling
•

Patient/staff at risk
•

Loss of instrumentation (thermostat control/regulation)
•

Supply chain interruption
•

Labor dispute
•

Shortage of labor
•

Communication failure
•

Paging: internal and external
•

Emergency medical services or other radio
•

Internal HCF telephone
•

External telephone
•

Cellular phone
•

Satellite
•

Transportation disruption
•

Labor dispute
•

Roadway/highway incident/blockage

Human Events: With or Without Political, Terrorist, or Criminal Intent

•

Mass casualty incidents
•

Trauma
•

Civil disturbance
•

CBRNE *

* Chemical, biological, radiological, nuclear, or high-yield explosive.
•

Infectious disease
•

Foodborne illness
•

Abduction (infant, child, or adult)
•

Armed or threatening intruder
•

Bomb threat
•

Civil disorder
•

Forensic admissions
•

Hostage situation
•

Violent labor action
•

VIP visitor
•

Workplace violence

Current practice

Successful mitigation efforts and effective response plans are based on the best possible knowledge of the HCF’s vulnerability in terms of deficiencies in its capacity to provide services, physical weaknesses, and organizational shortcomings in responding to emergencies. The HVA should also highlight and identify strengths within personnel, processes, plans, and other attributes. Past successes during disasters should be revisited to learn best practices.

All HVAs have some degree of subjectivity in their findings because many assumptions are made with regard to the perceived risk and even the level of preparedness if hard data are not available. Use of a multidisciplinary team should be encouraged to ensure a holistic characterization of each hazard and to help minimize the inherent subjectivity of the analysis and skewed or erroneous results. The team should be led by someone familiar with the HVA process and consist of representatives from at least the following areas within the HCF:

•

Emergency management
•

Security/safety
•

Facilities (e.g., engineering, maintenance, information technology, and telecommunications)
•

Operations (e.g., nursing, medical staff, laboratory, and radiology)
•

Ancillary services (e.g., materials, food, housekeeping, and environmental services)
•

Administration
•

Finance/business

Community representatives, such as the local emergency manager, fire official, police official, and city manager, can also provide valuable input. Additional members, including the hospital administrator-at-large may be beneficial, as long as the group size remains manageable and consensus is achievable within a reasonable amount of time. Regularly scheduled meetings with a defined agenda and other business-related models will assist the completion and maintenance of the assignment of the HVA team.

Most HVA tools come preloaded with a listing of likely hazards that the developer believes the average HCF could face. It is important that the HVA team begin by reviewing the listing of hazards in the HVA tool to ensure that it is comprehensive and applicable to the facility(s). The HVA tool should address all possible events regardless of their likelihood. The first step is to “brainstorm” and determine all possible hazards. This can be accomplished with assistance from the county local emergency preparedness committee (LEPC) in conjunction with the Office of Emergency Management for both the county and state. The hazards are then classified into categories, as described in Box 23-1 .

Risk, or effect, relates to the threat a particular hazard has with respect to the effects on humans: safety of people (patients and staff); effects on property: structure(s) and property; and effects on business: the ability to continue operations. Each risk can be assigned a numerical value to allow for a comparison or relative risk. The three types of effects are averaged, and a score is assigned for each category. This will be important in the overall assessment.

Examples of each category include but are not limited to :

•

Effects on humans
- •
  
  Potential for injury or death to staff members
- •
  
  Potential for injury or death to visitors
- •
  
  Potential for injury, death, or adverse outcomes to patients
•

Effects on property
- •
  
  Damage to the facility (up to and including loss of the facility)
- •
  
  Loss of use of the facility
- •
  
  Loss of or damage to equipment and/or supplies
- •
  
  Costs associated with replacement/repair of the facility, equipment, or services
•

Effects on business
- •
  
  Business interruption
- •
  
  Unanticipated costs
- •
  
  Loss of revenue (from all causes)
- •
  
  Record keeping issues (e.g., loss of records, inability to access, and compromise of integrity)
- •
  
  Employees unable or unwilling to report for work
- •
  
  Patients unable to reach the facility
- •
  
  Damage to reputation
- •
  
  Fines, penalties, and legal costs
- •
  
  Future insurance premium increases

The degree of risk may be expressed as a numerical score or verbally with use of terms such as nonexistent, low, medium, high, and catastrophic. As a consideration, the HVA team may wish to add greater weight to hazards that occur without warning (e.g., tornado strike).

Probability relates to how likely an event is to occur at the facility or to affect the facility, based on proximity. This, too, can be assigned a numerical value and is best determined from historical data (e.g., a scale from 1 to 5, with 1 representing a low probability of occurrence and 5 a very high probability of occurrence). Looking back at historical data is critical in making an “educated guess” about the future. This is an assessment of the likelihood of a hazard or emergency occurring that is often described as improbable, low, medium, or high. Other related factors that may be helpful in assessing or describing probability include the following :

•

Frequency of occurrence: Obviously, the more frequent the occurrence, the higher the likelihood.
•

Location of the hazardous event and the region affected: Events that occur proximal to the HCF are more likely to directly (or indirectly) affect the facility, whereas events that occur at some distance may be less likely to affect the HCF.
•

Seasonal (or other cyclic) variations: Events that occur with some regularity may be presumed to be more probable. Commonplace examples include the occurrence of “influenza season” each fall through the winter, and drought and/or floods (location-dependent) associated with El Niño.

Where possible, probability should be based on objective data, such as historical archives, to learn of local disasters. Equipment failure rates or mean time between failure data should be available to the HVA team. Even maintenance records and expected length of service of equipment may lead to objective data that influence an HVA. Often, however, probability assessments are colored by the prior experiences of HVA team members and recent organizational memory.

Facility preparedness may be expressed explicitly in a separate category or integrated with another element (probability or risk). Intuitively, if the facility is well prepared to deal with an emergency, the effects of the emergency should be lessened. The presence of a preparedness component aids in tracking the organization’s preparedness efforts and is a means to decrement HVA scores as preparedness levels increase. Preparedness also should be reported to help determine the need for improvement in areas that have high risk and/or probability. Preparedness may be assigned a numerical value, or it simply may be a listing of what, if any, plans currently exist to address that particular event. It may also represent resources and the amount of them available (e.g., a lot, little, or none); resources can be subdivided into internal and external resources. The average of these two is the numerical value for preparedness.

Adding the numerical values of these three components (risk, probability, and preparedness) provides a value. Looking at probability versus effects graphically ( Figure 23-1 ), one would expect higher sums for those events that fall in the high-probability–highly effected areas and lower values for those events in the low-probability–lowly effected areas.