A Quantitative Public Health Risk Assessment Tool for Planning for At-Risk Populations

Abstract

The Drexel University Center for Public Health Readiness and Communication developed a quantitative public health risk assessment tool that has been used in Pennsylvania and across the United States. The tool assesses hazard impacts on health care and public health metrics and incorporates the planning needs of at-risk populations. The flexible platform using Microsoft Excel allows planners to customize hazard assessment and use jurisdiction-specific data. Findings from the risk assessment can guide planning and evaluate preparedness progress over time.

In 2012, the Center for Public Health Readiness and Communication at the Drexel University Dornsife School of Public Health developed a public health risk assessment tool (PHRAT) for the Pennsylvania Department of Health to identify risks and guide planning for emergencies that affect the health of the public.

PURPOSE

Planners from Pennsylvania and the Philadelphia metropolitan statistical area identified three priorities for the development of the PHRAT:

• 1.The determination of risk should be quantitative and measurable and reflect a systematic and objective assessment of an event’s impact on health.
• 2.The severity assessment should integrate the impact of hazards on at-risk populations.
• 3.The tool should focus on the work of public health agencies specifically and account for hazards that require prolonged recovery.

Additionally, the tool is intended to create a mechanism for a jurisdiction to track preparedness over time.

DESCRIPTION OF THE TOOL

The tool defines risk as the expectation of loss from a hazard¹ and calculates risk as the product of a hazard’s probability and severity, as measured by its impact across five domains: human health, health care services, inpatient health care infrastructure, community health, and public health services. Each domain contains a set of quantifiable metrics to measure the hazard’s impact on an outcome or service. Impact is compared with a baseline (e.g., the number of deaths or emergency medical service transports during an incident compared with the baseline for the community during a nondisaster period) and is scored on a four-point scale on the basis of the magnitude of the change. A composite risk score is produced for each hazard using weighted averages of scores across all five domains. Table 1 summarizes the metrics in each domain. The analysis is ideally conducted at the county or regional level; findings are aggregated for state-wide jurisdictional analysis using weighted averages on the basis of regional population size. Baseline data are collected for each county or region from a range of sources, including state and local departments of health, health care organizations, and interviews with local public health and emergency planners. Regional (multicounty) assets may be used as baseline data for counties that lack health care facilities; no regional capacity would be scored as zero.

TABLE 1—

Metrics Used for Hazard Impact Assessment by Domain: Public Health Risk Assessment Tool

Domain	Metric	Scoring Strategy
Human impact	Mortality	Deaths/d
	EMS Transports	Transports/d
	ED	ED visits/d
	Outpatient visits	Visits/d
	Trauma center injuries	Trauma center injuries/d
	Mental health	% population developing psychopathology after the incident (estimated on the basis of research data)
Health care service impact	Outpatient services	Primary care provider supply and demand
	ED	ED bed supply and demand
	Hospital beds	Hospital bed supply and demand
	Ancillary services	Pharmacist supply and demand
	Trauma units	Functioning OR supply and demand
	Mental health services	Mental health provider supply and demand
Inpatient health care facility infrastructure	Hospital personnel	Patient to nurse ratio
	Facility water supply	H without water—% affected beds
	Facility electricity	H without electricity—% affected beds
	Facility generator fuel supply	H of fuel required—% affected beds
	Hospital IT and communications	H without access to EMRs or IT systems
	Facility critical supplies	D without linen services
	Facility evacuation	% regional hospital beds evacuated
	Patient decontamination	No. of patients requiring decontamination (as % of regional ED capacity)
Community impact	Water supply	% population water outage or boil water
	Sanitation and sewage systems	% population sanitation system disrupted
	Public utilities	% population without electricity
	Transportation	Duration ≥ 1 major corridor closed
	Business continuity	% businesses closed
	Population displacement	No. of persons evacuated to or from jurisdiction
	Environmental contamination	Area requiring remediation or decontamination
Public health service impact	Personnel	Public health agency employee supply and demand
	Surveillance	Case reports requiring investigation or monitoring
	Mass care	Persons requiring sheltering or mass care
	MCM	% population requiring MCMs
	Laboratory services	Specimens processed/d
	Health communications	Personnel h/d devoted to information sharing with public or partners
	Fatality management	Morgue capacity supply and demand

Note. ED = emergency department; EMS = emergency medical service; IT = information technology; MCM = medical countermeasures; OR = operating room.

The PHRAT also calculates an adjusted risk, which modifies a hazard’s risk calculation on the basis of the additional planning necessary to ensure whole community access to emergency response resources for nine different at-risk populations (children aged < 18 years, elderly aged > 65 years, limited English proficiency communities, and those with hearing impairment, vision impairment, cognitive disability, chronic disease, or those in poverty). The adjusted risk is the product of the hazard’s original risk score and a population impact score, which incorporates the relative size of each at-risk population in the planning jurisdiction (using data from the US Census,² County Health Profiles,³ or the Center for Disease Control and Prevention’s Behavioral Risk Factor Surveillance System⁴) and an access planning score determined by whether the specific population requires any of the following for the hazard under consideration: (1) special emergency communication procedures, (2) special plans for sheltering and mass care, (3) special plans for evacuation, or (4) special plans for human services and medical management.⁵

Preparedness is also assessed quantitatively, using public health agency self-assessment of the capability performance measures in the Public Health Emergency Preparedness⁶ and Health Care Preparedness Program.⁷ Scores for each capability are averaged to create an overall preparedness score. The ratio of the adjusted risk and preparedness creates a planning priority score for each hazard. The current version of the PHRAT analyzes 20 common hazards and can be customized for use by different jurisdictions. Figure 1 depicts the framework and conceptual overview of the tool. The tool, a detailed user’s guide, and a 2016 summary report for Pennsylvania are supplements to this article (available at http://www.ajph.org).

FIGURE 1—

Overview of the Public Health Risk Assessment Tool

JURISDICTIONAL RISK ASSESSMENT

The tool was originally developed for a jurisdictional risk assessment in the Philadelphia metropolitan statistical area and its four participating states (Pennsylvania, Maryland, New Jersey, Delaware). It was used to update the Pennsylvania risk assessment in 2016 and a jurisdictional risk assessment for southeastern Pennsylvania in 2018. Since 2012, the tool has been publicly available through the Web sites of the Center for Public Health Readiness and Communication, the Pennsylvania Department of Health, and the Assistant Secretary for Preparedness and Response Technical Resources Assistance Center Information Exchange and has been used by jurisdictions outside the mid-Atlantic region, including Hawaii; Vermont; Maricopa County, Arizona; and Marin County, California.

IMPLEMENTATION

Public health planners in Pennsylvania and the Philadelphia metropolitan statistical area have used the PHRAT to identify the hazards that either pose great threat or require additional preparedness. They have also used the tool to extend hazard-specific planning for at-risk populations. For example, the Philadelphia Department of Public Health developed a plan for extreme heat that focused on high-risk populations affected by heat emergencies. The Bucks County Health Department developed a tool to overlay key response assets for selected hazards (e.g., shelters, points of dispensing, health care facilities, culturally focused community-based organizations) on heat maps of census tracts depicting the high-risk populations that would need those assets during disasters. Other jurisdictions have used the findings to convene regional or state-wide meetings with public safety, health care, and community partners and initiate collaborative planning for emergencies.

EVALUATION

In Pennsylvania, the tool has provided a useful framework over a six-year period to prioritize planning and plan exercises for specific hazards. The hazards that have occurred with the greatest frequency over the time frame and continue to challenge public health agencies in Pennsylvania are those with the highest planning priority scores: weather-related emergencies, flooding, utility interruptions, and local infectious disease outbreaks. Pandemic influenza emerges consistently as the highest-ranking hazard because of its significant impact across all the domains assessed by the PHRAT; its occurrence is both probable and potentially catastrophic. Notably, the PHRAT-identified planning priorities differ from hazards that emerge as important in hazard and vulnerability assessments conducted by state and local government emergency management agencies. These differences reflect the focus of this tool on health-specific effects and the response capabilities of public health agencies.

LIMITATIONS

There are challenges associated with using the PHRAT. Baseline data may be difficult to find and time intensive to collect, requiring one to two months of full-time effort. Estimated effects on the basis of data from actual events that have occurred elsewhere may not be easily extrapolated to a given jurisdiction or reflect a worst-case scenario, potentially overestimating risk. Conversely, because severity is calculated from an event’s impact across all five domains, the tool may underestimate risk for hazards that produce significant impact in only one or two domains (e.g., human health or health care services). Finally, the design of this tool relies exclusively on measured (or estimated) metrics for risk assessment. Thus, planners who use the tool must engage in discussions with emergency response partners and community stakeholders after the analysis to validate the findings and use them to ensure that preparedness and response planning aligns with their concerns and capabilities.

PUBLIC HEALTH SIGNIFICANCE

The PHRAT provides a flexible platform for planners to estimate the risks of public health emergencies in their jurisdiction. The use of measurable indicators to assess impact mitigates the influence of subjective opinion in the estimation of risk and provides benchmarks for public health agencies to improve preparedness efforts. The tool’s focus emphasizes the responsibilities of public health organizations. The integration of hazard-specific planning considerations for at-risk populations facilitates targeted response measures and can be used to promote collaborative planning with stakeholders.