Office of Public Health Emergency Preparedness (OPHEP)

Public Health Preparedness
Integrating Public Health and Hospital Preparedness Programs

Lois Davis, Jeanne Ringel, Sarah Cotton, Anne Griffin, Elizabeth Malcolm, Louis Mariano, Jennifer Pace, Karen Ricci, Molly Shea, Jeffrey Wasserman, James Zazzali

TR-317-DHHS

December 2005

Prepared for the U.S. Department of Health and Human Services Office of the Assistant Secretary for Public Health Emergency Preparedness

Contents

Preface
Figures
Tables
Summary
Acknowledgments
1. Introduction
    Objectives
    Organization of This Report
2. Analytic Approach
    Literature Review
    Survey Analysis
    Case Studies
        Selection of Candidate Models of Integration
        Data Collection
        Data Analysis
3. Conceptual Framework
    Introduction
    Discussion
4. Survey Results
    Coordination Between Local Health Departments and Hospitals
        Programming Mechanisms
        Feedback Mechanisms
        Changes in Coordination Between 2001 and 2003
    Discussion
    Limitations of the Survey Analysis
5. Case-Study Results
    Introduction
    Case-Study Results
        Leadership
        Differences in Mission and Culture
        Characteristics of the Organizations and the Environment
        Funding
    Discussion
    Limitations of the Case-Study Analysis
6. Conclusions
    Critical Success Factors for Coordination at the Local Level
    Recommendations for State- and Federal- Level Officials
    Summary and Future Steps
Appendix.  Case-Study Protocol
References
Copyright Information: Rand Corporation

Figures
Figure 3.1 Determinants of Coordination
Figure 3.2 Preparedness Activities and Types of Organizations that May Be Involved in Coordination

Tables
Table S.1.  Summary of Survey Results: Coordination Mechanisms Used in 2003
Table S.2. Percentage-Point Increase Between 2001 and 2003
Table 4.1 Programming Coordination Mechanisms Used by Hospitals and Local Health Departments in 2003
Table 4.2 Feedback Coordination Mechanisms Used by Hospitals and Local Health Departments in 2003
Table 4.3 Comparison of Coordination Mechanisms Used by Local Health Departments and Hospitals in 2001 and 2003
Table 5.1 Characteristics of Case Study Sites
Table 5.2 Funding Programs Participated in by Case-Study Sites

Preface

The September 11, 2001, terrorist attacks and the subsequent anthrax incidents and hoaxes underscored how important it is for medicine and public health to find new ways to collaborate with one another to address public health emergencies and to improve overall preparedness in their local areas. Indeed, encouraging linkages between local health departments and hospitals (as well as with other key stakeholders) has been an important component of the Centers for Disease Control and Prevention (CDC) and Health Resources and Services Administration (HRSA) cooperative agreements for public health and hospital emergency preparedness to more effectively address different aspects of preparedness.  Hospitals and public health departments have employed different approaches for doing so and have varied in their focus on such areas as patient treatment and infection control, quarantine and isolation procedures, disease surveillance and reporting, and risk communication. 

In this study, we examine different models for coordination that have been employed by local health departments and general acute-care hospitals to improve public health preparedness.  The study’s aims are to  (1) describe those aspects of public health preparedness on which local health departments and hospitals are focusing their coordination effort ; (2) examine how these relationships have changed over time (from pre-September 11, 2001, through fall 2003); (3) identify factors that facilitate or hinder coordination; (4) understand how funding affects coordination; (5) identify the mechanisms, policies, and procedures that have been demonstrated to be effective in coordinating local health departments’ and hospitals’ public health preparedness activities; and (6) suggest strategies to improve coordination between local health departments and hospitals and the integration of their preparedness activities.

These results will be of interest to the U.S. Department of Health and Human Services (HHS), as well as to public health and medical professionals responsible for improving public health preparedness and ensuring collaboration between key stakeholders at the state and local levels.

This work was prepared for the U.S. Department of Health and Human Services, Office of Public Health Emergency Preparedness.  The research was produced within the RAND Health Center for Domestic and International Health Security.  RAND Health is a division of the RAND Corporation. A profile of the Center, abstracts of its publications, and ordering information can be found at http://www.rand.org/health/centers/healthsecurity/. More information about RAND is available on our Web site at http://www.rand.org.

Summary

Medical and public health professionals have long been aware of the need to improve coordination between their communities.  The September 11, 2001, terrorist and subsequent anthrax attacks underscored the importance of finding new ways to collaborate in order to address public health emergencies and to improve overall preparedness.  The federal government’s bioterrorism funding is aimed specifically at encouraging such collaboration, and the cooperative agreements of the Centers for Disease Control and Prevention (CDC) and Health Resources and Services Administration (HRSA) have encouraged linkages between local health departments and hospitals and other key stakeholders to improve public health preparedness. 

Hospitals and public health departments have employed a variety of approaches to improving preparedness coordination.  In this study, we examined different models for coordination used by local health departments and local acute-care hospitals. Understanding which strategies have been successful allowed us to identify which aspects of public health preparedness lend themselves to being coordinated, which factors facilitate or hinder coordination, how barriers to coordination have been overcome, and how local health departments and hospitals are using the separate funding streams provided by CDC and HRSA for coordination. We developed a conceptual framework to identify factors that are likely to influence coordination between local public health departments and local hospitals.  This framework guided an examination of coordination activities undertaken in 2001 and 2003 and site visits to six locations, five of which were selected as successful models of public health and hospital coordination.

Anayltic Framework

We developed a three-pronged approach to explore the issues surrounding coordination between local health departments and hospitals. We began with a literature review to inform the development of a conceptual framework for thinking about coordination issues. This framework then guided the development of our analysis plan, which comprises two distinct components. First, to provide a broad national picture of coordination between local health departments and hospitals, we analyzed data from the Surveys of Federal Preparedness Programs for Combating Terrorism (see the Survey Analysis section in Chapter 1). Then, to delve more deeply into the factors that facilitate and hinder coordination, we conducted a series of site visits to counties in which coordination appeared to be taking place.

Conceptual Framework

Before evaluating the various coordination mechanisms that could be used, we first considered the particular tasks that would need to be coordinated. For example, a bioterrorist attack would require tasks related to identifying cases, treating cases, containing the outbreak, communicating with the public, etc.  Although the focus of this report is on hospital-public health coordination, some of the tasks that would need to be accomplished would require the involvement of additional organizations, such as law enforcement, emergency management organizations, emergency medical services, and other key stakeholders. The literature on organizational behavior argues that the characteristics of these tasks and the different organizations involved determine the coordination mechanisms that would be most effective.

Coordination can be conceptualized as the “conscious activity of assembling and synchronizing differentiated work efforts so that they function harmoniously in attainment of organizational objectives" (Longest and Klingensmith, 1994).  The basic notion involves bringing together in some fashion various aspects of an organization (i.e., people, departments, etc.) or different organizations, so that there is an exchange of information and/or resources in accomplishing a common task or meeting a common goal. 

Coordination mechanisms fall into two broad categories (Van de Ven, Delbecq, and Koenig, 1976): programming (impersonal, formal) mechanisms and feedback (personal, informal) mechanisms.  Programming mechanisms include development of pre-established plans, schedules, and forecasts; formalized rules, work policies, processes, and procedures (including clinical guidelines, pathways, and protocols); outcome standards (e.g., quality assurance); and standardized information and communications systems.  Roles are formally prescribed, and there is little, if any, room for discretion on the part of the organizational “actors” (i.e., the people doing the work).  Programming mechanisms are usually formal and institutionalized.

Less formal feedback mechanisms (i.e., personal or group mechanisms) include personal modes of coordination that allow individuals to respond to feedback from vertical forms of communication (e.g., supervisory staff) or horizontal or lateral forms of communication (e.g., coworkers and peers).  Group modes (such as teams) enable individuals to receive feedback from other group members.  Specific mechanisms of this type of coordination include the use of teams, committees, task forces, and work groups.

Key Findings

A Wide Array of Coordination Mechanisms Was Used in 2003

Our study considered a wide array of coordination mechanisms, including both programming mechanisms (e.g., plans, policies and procedures, standards, communication systems) and feedback mechanisms (e.g., person-to-person, group modes).  To understand which coordination mechanisms were used, we analyzed results from the 2001 and 2003 Surveys of Federal Preparedness Programs for Combating Terrorism, which were administered to hospitals and local public health departments by RAND on behalf of the Advisory Panel to Assess Domestic Response Capabilities for Terrorism Involving Weapons of Mass Destruction, otherwise known as the Gilmore Commission.  

The data used to provide a broad, national picture of coordination among local health departments and hospitals come from national Surveys of Federal Preparedness Programs for Combating Terrorism.

Table S.1 shows the proportion of hospitals and local health departments reporting use of various coordination mechanisms in 2003.

The results indicate that, two years following the September 11, 2001, terrorist attacks, the vast majority of both hospitals and local health departments had emergency response plans in place. These plans addressed a wide range of issues, including procedures for quarantine and isolation, as well as communication with other health responders (e.g., hospitals, medical providers, emergency medical services). Local health departments were ahead of hospitals in terms of integrating their plans with those of other local organizations.

Table S.1.  Summary of Survey Results: Coordination Mechanisms Used in 2003

NOTES:  Standard error of the percentage estimate is shown in parentheses.  This table summarizes data that are presented in greater detail in Tables 4.1 and 4.2. N/A = question not asked of this group.  

a Indicates 2001 data.

Coordination Generally Improved Between 2001 and 2003

As shown in Table S.2, the percentage of hospitals and local health departments participating in key coordination activities increased between 2001 and 2003.¹

Between 2001 and 2003, both hospitals and health departments appear to have increased their use of informal, or feedback, coordination mechanisms, such as joint training, participation in interagency task forces or working groups for disaster preparedness, and joint training activities for terrorism-related incidents with other local health organizations. Although no causal relationship has been shown, these results are consistent with the hypothesis that the influx of bioterrorism funding following the 9/11 attacks helped facilitate adoption of such informal coordination mechanisms. 

Table S.2. Percentage-Point Increase Between 2001 and 2003

(for those organizations responding in both years)

NOTE:  Standard error of the estimate is shown in parentheses. 

Case Studies Provide an In-Depth Look at the Critical Success Factors for Coordination

Our site visits provided several lessons about the critical success factors for coordination at the local level. Our study resulted in the following observations and recommendations, many of which require the involvement not only of hospitals and health departments, but of other stakeholders as well (e.g., law enforcement, emergency medical services).

Preexisting relationships. Prior relationships provide a framework from which to build coordination efforts for public health preparedness. For example, local health departments can use existing interagency groups to identify key stakeholders that need to be involved in preparedness planning.

An understanding by health departments and hospitals that coordination is mutually beneficial.  Health departments have an education, or outreach, role to play. Public health is relatively new to the area of emergency preparedness and response and, thus, needs to communicate its usefulness clearly to hospitals.

Strong, but flexible, leadership. Strong leadership drives the coordination process forward by providing a clear and shared statement of the mission and objectives. A flexible management style allows leaders to bridge differences in culture and priorities across the many stakeholder organizations. While personality plays a role in leadership quality, there are many aspects of successful leadership that can be learned. Therefore, health departments and hospitals could provide leadership training to those people charged with overseeing the coordination of public health preparedness activities.

A well-developed, facilitative process.  To be successful, the process of coordination must involve all stakeholders as equal partners in decision-making and be respectful of their opinions, concerns, and time. Health departments can provide some structure and guide the effort to ensure that progress is made.

Institutionalized coordination mechanisms. The use of more-formal coordination mechanisms (e.g., interagency work groups, written plans, communications systems) insulates a county against the inevitable changes in personnel that can hinder less-formal coordination mechanisms (e.g., those based on personal relationships).

Funding for successful coordination efforts. All sites noted that HRSA and CDC funding streams have helped increase coordination, in part by making public health a player in this arena and also by helping support formal coordination mechanisms.  However, inflexible restrictions regarding which agency can receive the funding, what the money can be used for, and the time frame for its use all caused widespread problems.  In addition, HRSA and CDC funding was sometimes received from the state late in the fiscal year, with little prior information on how much money to expect, forcing local officials to make rush decisions on how to use the money.

Recommendations for State and Federal Level Officials

It is important to recognize that public health preparedness is only one, albeit important, aspect of the overall emergency and disaster preparedness of a region.  In this analysis, we focused specifically on public health and the interface between hospitals and health departments; however, many aspects of public health preparedness (e.g., distribution of the SNS, quarantine plans) will require the involvement of nonhealth stakeholders (e.g., law enforcement, emergency medical services, etc.), as well.  Each stakeholder in this wider group is responsible for addressing different aspects of disaster preparedness and will bring its own funding sources, missions, organizational characteristics, and priorities to the planning table. Thus, it is important to understand that public health preparedness does not occur in a vacuum, but in the broader context of the overall disaster preparedness of a region.

Many of the insights gained from the survey and case-study analyses suggest ways in which HHS and state health departments can facilitate coordination at the local level between health departments and hospitals. More generally, many of the lessons learned from this analysis apply not just to hospitals and health departments but to interagency coordination as a whole.  Not unexpectedly, the majority of recommendations for the federal and state officials are related to the funding of preparedness activities through HRSA and CDC cooperative-agreement programs and other related programs, such as the Cities Readiness Initiative.

• Provide a clear statement of the mission and goals regarding public health preparedness. This is important at both the federal and state levels. Without such a statement, local health departments are unsure about where to focus their energy and resources.
• Make the CDC and HRSA cooperative-agreement programs more flexible in terms of:

• Coordinate the CDC and HRSA cooperative-agreement programs with those of other federal agencies, such as the Department of Homeland Security (DHS). In considering how to encourage coordination, HHS needs to think more broadly than a single cooperative agreement or cooperative-agreement program about how coordination can be funded and mandated, and needs to set requirements that standardize coordination requirements across multiple cooperative-agreement programs. More broadly, there is a need to assess how different funding streams and programs can be better coordinated across agencies (primarily HHS and DHS).
• Educate local health departments and hospitals about:

Acknowledgments

The authors would like to thank the local health organizations that have provided thoughtful feedback and information on their approach to and challenges encountered in coordinating public health preparedness at the local level. We also wish to thank Dr. Donald Goldmann, Dr. Allison Diamante, and Ms. Lara Lamprecht who reviewed and commented on drafts of this report.  Of course, any errors or omissions are the sole responsibility of the authors.

Chapter One. Introduction

In 1996, when the Committee on Medicine and Public Health was formed to examine collaboration between medicine and public health, focus groups composed of professionals from each of the two health disciplines revealed an interesting paradox in how they viewed one another and the intersection between their two disciplines (Lasker and the Committee on Medicine and Public Health, 1997):  They believed that the two health disciplines were interlinked, with public health professionals tending to consider medicine as an arm of public health and medical professionals viewing public health as a subspecialty of medicine.  However, the participants were unable to articulate the exact nature of this relationship.  Further, most focus group participants reported little or no experience in working with professionals or organizations from the other health discipline, and few could describe how the activities of the other health discipline were relevant to what they did.

At the time of the committee’s 1997 report, the authors commented that significant changes within the health care system required medicine and public health to find new opportunities for cross-sectoral collaboration (Lasker and the Committee on Medicine and Public Health, 1997).  Some of these key changes included

• the restructuring of the health care system, with individual practitioners being replaced by corporate entities
• the redefining of the governmental role of public health
• the severe fiscal constraints associated with managed care and governmental downsizing and privatization.

Combined, these forces were seen as compelling medicine and public health to reexamine their roles and to consider more cross-sectoral collaboration to meet these new challenges.  Yet, the committee noted that finding new ways to collaborate in such areas as managed care, community health, and indigent care would not be easy, given the differences between these two disciplines in how they viewed one another and their role in health care. 

The September 11, 2001, terrorist attacks and the subsequent anthrax incidents and hoaxes further underscored the importance of medicine and public health finding new ways to collaborate with one another to address public health emergencies and to improve their locality’s and region’s overall preparedness.  Indeed, the federal government’s bioterrorism funding is specifically aimed at encouraging such collaboration to improve public health preparedness. For example, an important component of the Centers for Disease Control and Prevention’s (CDC’s) and the Health and Research Services Administration’s (HRSA’s) cooperative agreements has been to encourage linkages between local health departments and hospitals (as well as with other key stakeholders) to more effectively address different aspects of public health preparedness. 

Although the funding is meant to encourage coordination, in practice that goal has not been fully realized. Hospitals and public health departments have employed different approaches for improving preparedness and have varied in their choices regarding on which areas of preparedness to focus (e.g., patient treatment, infection control, quarantine and isolation procedures, disease surveillance and reporting, or risk communication).  More generally, it appears that health departments and hospitals continue to lack an understanding of each other’s respective roles and, more important, the benefits of coordinating their preparedness efforts. For example, during the public health preparedness exercises RAND has conducted, we frequently found that public health departments were unaware of hospitals’ plans for responding to a bioterrorism (BT) event or other public health emergencies and that hospitals often do not include local health departments in their disaster exercises.  The fact that states and localities vary in how the CDC and HRSA cooperative-agreement funds are being managed further adds to this problem.

In this study, we examine different models for coordination or integration that have been employed by local health departments and general acute-care hospitals. We use the terms integration and coordination interchangeably in our discussion to mean the activity of assembling and synchronizing differentiated work efforts so that they function harmoniously in attainment of shared goals and objectives.  Understanding which strategies have been successfully undertaken to improve integration of health departments’ and hospitals’ preparedness efforts is important for several reasons: It allows us to identify (1) which aspects of public health preparedness lend themselves to being integrated; (2) which factors facilitate or hinder coordination; (3) which approaches to overcoming identified barriers to integration have been successful; (4) how local health departments and hospitals are funding these activities; and (5) most important, the successful strategies that may be implemented in other localities to improve public health and hospital coordination for preparedness.

In addition, the separate CDC and HRSA funding streams intended to promote integration may themselves present some structural constraints that may make integration more difficult to achieve.  We examine the strategies health departments and hospitals have employed to try to overcome funding barriers and offer suggestions on how the state and federal levels may help to remove barriers to integration and encourage the adoption of promising models of integration.  The study’s results also provide important insights for other localities attempting to improve public health and hospital coordination in order to achieve better integration of their preparedness programs.

Public health preparedness requires many different types of coordination activities to occur both within and across organizations. To mount an effective public health response, local health departments must work closely and seamlessly with many other organizations, including

• local hospitals and/or hospital councils
• medical providers
• the state health department and other health departments in the region, and possibly federal agencies, including the CDC
• first responders (e.g., emergency medical services, fire service, law enforcement)
• emergency management
• interagency disaster planning groups, as well as local and state homeland security task forces.

In this study, we focused specifically on interagency coordination between local health departments and hospitals. However, in our case-study interviews with health department and hospital staff, issues related to other coordination activities and partners inevitably arose. We incorporate those findings in our discussion to provide important background and context for understanding the health department-hospital interface.

Objectives

This report addresses six key issues of interest to the U.S. Department of Health and Human Services (HHS): (1) describe on which aspects of public health preparedness local health departments and hospitals are focusing their coordination efforts; (2) examine how these relationships have changed over time (from pre-September 11, 2001, through fall 2003); (3) identify factors that facilitate or hinder coordination; (4); understand how funding affects coordination; (5) identify the mechanisms, policies, and procedures that have been demonstrated to be effective in coordinating local health departments’ and hospitals’ public health preparedness activities; and (6) suggest strategies to improve the coordination between local health departments and hospitals and the integration of their preparedness activities.

To address these issues, we used a multipronged approach.  To help guide our analysis, we first developed a conceptual framework to identify factors that are likely to influence coordination between these organizations.  We then used the conceptual framework to inform the analysis of a 2003 survey of local health departments and general acute-care hospitals (both public and private) and compare changes in their coordination activities between 2001 and 2003. We also undertook a series of case studies to examine in-depth coordination in six localities, so that we could understand what factors facilitate and hinder coordination between hospitals and local health departments, and how funding influences coordination, and so that we could identify characteristics of successful examples of coordination.

Organization of this Report

In this report, we have organized the methods and results into five chapters.  Chapters Two and Three present the analytic approach and conceptual framework, respectively.  Chapter Four presents the survey results, which provide a broad national picture of coordination between local health departments and hospitals.  Chapter Five presents the case-study results, organized around four key areas: leadership, culture and mission, organizational characteristics and environment, and funding.  Chapter Six provides some conclusions and recommendations. The Appendix provides a copy of the interview protocol used for the case studies.

Chapter Two.  Analytic Approach

We developed a three-pronged approach to explore the issues surrounding coordination between local health departments and hospitals. We began with a literature review to inform the development of a conceptual framework for thinking about coordination issues. This framework then guided the development of our analysis plan, which comprises two distinct components. First, to provide a broad national picture of coordination between local health departments and hospitals, we analyzed data from the Surveys of Federal Preparedness Programs for Combating Terrorism (see the Survey Analysis section below). Then, to delve more deeply into the factors that facilitate and hinder coordination, we conducted a series of site visits to counties in which coordination appeared to be taking place. We discuss our research methodologies for each component of the study below.

Literature Review

We reviewed key documents and reports on terrorism preparedness and response that discuss the role of public health departments and hospitals. These included the Gilmore Commission Reports (Advisory Panel to Assess Domestic Response Capabilities for Terrorism Involving Weapons of Mass Destruction, 2001, 2003) and U.S. General Accounting Office (GAO; now Government Accountability Office) reports. Taken together, these reports provided critical background information and identified issues to address in the survey data analyses and site-visit interviews.

In addition to reviewing the reports, we conducted a bibliographic search using the following computerized databases: Social Services Abstracts, Social Sciences Abstracts, ABI Inform, and MEDLINE.  To search each of the databases, we developed the following set of key words:

Integration OR coordination OR collaboration

AND

Public health OR hospitals OR human services OR emergency preparedness

Upon completing the online searches, we retrieved all promising documents for closer review.

The reports and documents identified through the bibliographic searches informed the development of a conceptual framework for identifying organizational characteristics, task characteristics, and strategies associated with successful integration. In addition, the results of the literature review were used to develop the survey analysis plan and to identify topic areas and issues to include in the interview protocol for the site visits.

Survey Analysis

The data used to provide a broad, national picture of coordination among local health departments and hospitals come from national Surveys of Federal Preparedness Programs for Combating Terrorism, which were administered to hospitals and local public health departments by RAND on behalf of The Advisory Panel to Assess Domestic Response Capabilities for Terrorism Involving Weapons of Mass Destruction, otherwise known as the Gilmore Commission.² The first of these surveys was fielded in late summer 2001, just prior to the terrorist attacks of September 11, 2001. Subsequent waves of data collection on this longitudinal sample occurred annually through 2003.³  For this analysis, we focus primarily on the data from 2003, the most recent year available, but do provide information where possible on how coordination changed between 2001 and 2003.  All analyses of these data are weighted so that the results are nationally representative.  The 2003 sample includes 132 local health departments and 103 general acute-care hospitals.

The hospital and public health versions of these surveys include items that address the specific preparedness needs and activities of these organizations, as well as more-general preparedness questions applicable to a wider array of response organizations.  The content of these surveys comprises several topic areas, including planning, training, funding of preparedness activities, and support needs.  Embedded in each of these topic areas are questions pertaining to the organization’s coordination efforts both within the health community and with other organization types that would respond to acts of terrorism and other emergencies.  Details regarding the survey questionnaire have been published previously (Advisory Panel to Assess Domestic Response Capabilities for Terrorism Involving Weapons of Mass Destruction, 2001, 2003).

Our analyses of the survey data were primarily descriptive, reporting the proportion of hospitals and public health departments responding in a particular way to the questions of interest (e.g., the proportion of hospitals that report having an emergency response plan that addresses quarantine procedures, the proportion of local health departments [LHDs] that participates in an interagency emergency planning group).  For comparisons of proportions across the two groups, we used the appropriate statistical tests of the difference/equality of proportions, taking into account the correlation among samples as necessary.

One of the attributes of the survey is its longitudinal structure.  Having multiple waves of data allows us to look at changes over time in the overall level of preparedness and to measure progress toward integration and coordination of LHDs and hospitals.  Unfortunately, the survey questionnaires fielded in 2001 and 2003 differed to some extent, which limited our ability to make comparisons over time.  In some cases, questions asked in 2001 were dropped (or similarly, questions asked in 2003 were not asked in 2001), leaving us unable to make comparisons over time.  In other cases, similar questions were asked in 2001 and 2003, but they were not exactly the same.  In such cases, it is impossible to determine whether any change over time is real or merely reflects different responses to different question wording.

The results have been statistically adjusted to represent the entire population in each professional group (i.e., local health departments and general acute-care hospitals within the United States).  A detailed description of the weighting methodology used to derive these estimates has been published previously (Advisory Panel to Assess Domestic Response Capabilities for Terrorism Involving Weapons of Mass Destruction, 2001, 2003).

Case Studies

Selection of Candidate Models of Integration

Before beginning the selection process, the RAND team identified the following selection criteria:  (1) localities that national public health and hospital leadership identify as having innovative public health or hospital preparedness programs, (2) programs from different types of counties (e.g., large versus small, urban versus rural), and (3) programs from different regions of the country.

We used a multipronged approach to identify candidate sites for the case-study analysis.  As a preliminary step, data from the 2003 Survey of Federal Preparedness Programs for Combating Terrorism, administered to a sample of local health departments throughout the country, were used to identify those counties that appeared to be more actively engaged in addressing emergency preparedness in general and in which coordination among the health department and hospitals (as well as with other response organizations) appeared to be taking place.  The initial sample of 132 counties was narrowed to 14, based on whether the local health departments reported having response plans in place for hospital surge capacity, quarantine, Strategic National Stockpile (SNS) distribution, and smallpox.⁴ 

Among these 14 counties, we examined the coordination-related questions and ranked the counties according to the reported levels of coordination between local health departments and hospitals. This analysis enabled us to identify those counties within the survey sample in which local health departments appeared to be more engaged in coordinating their preparedness activities with local hospitals.  One of the attributes that these counties shared was being located in areas of the country that are more prone to major natural disasters (e.g., hurricanes, tornadoes, snowstorms, earthquakes), suggesting that such locations tend to foster an environment in which public health and local hospitals need to work together to be able to plan effectively for and respond to major disasters.  Thus, one might expect that having such coordination mechanisms and relationships already established might better position health departments and hospitals to jointly address public health preparedness.

The information from the survey data was combined with results from the HHS project’s previous year’s tasks on exemplary practices and input from the National Association of County and City Health Officials (NACCHO) and the Association of State and Territorial Health Officials (ASTHO).  This helped the study team to confirm which of the 14 counties merited closer examination and to identify other localities in which coordination between health departments and local hospitals appeared to be successful.  The HHS project advisory board’s input also was requested to help in the selection of localities in which site visits would be conducted.  In all, six sites were selected:  four counties, one rural region of a northeastern state, and one state itself.  Given the variation across states in how the public health system is organized, we had to modify our definition of sites to include not only counties but also regions and in one instance an entire state.

Five of the six sites for in-depth case studies were thought to be examples of locations where coordination between the local health department and hospitals to integrate public health preparedness efforts were taking place successfully.  To provide a contrasting view and to gain a better understanding of the barriers to coordination, we identified an additional site for which coordination between hospitals and the local health department appeared to be minimal. An analysis such as this, which draws from only six case studies, cannot by any means be called a representative sample.  We selected the sites to achieve geographic variation as well as variation in different types of counties (urban versus rural; large versus small) to illuminate the range of issues and challenges one might encounter and the strategies for overcoming those challenges that have or have not worked.

In those sites for which coordination efforts were identified as being successful, RAND worked with the locality to identify the specific public health preparedness programs (e.g., communications, hospital-surge capacity) within the county or region that had benefited the most from coordination efforts.

Data Collection

For the case studies, we conducted in-depth interviews with key actors who have been involved in the development or implementation of the preparedness programs selected for examination.  Interviewees included a wide range of personnel from local health departments and hospitals. At the health departments, interviewees generally included senior staff responsible for public health and hospital preparedness programs, such as the local health officer, other health department staff who have been involved in the aspects of the preparedness program of interest, and the public information officer.  At the hospitals, interviewees typically included infectious-disease officers, emergency department directors, laboratory directors, and public information officers.  We also interviewed, where appropriate, representatives from the office of emergency management and other participants in county or regional interagency task forces to gain their perspective regarding coordination issues.

We developed a standardized protocol to guide our key-actor interviews.  The protocol, which is in the Appendix, addressed the following issues:

• Perceptions regarding internal and external barriers to integration and strategies to address those barriers
• Areas considered as potential candidates for integration of preparedness programs and factors that led to the decision to focus on particular aspects of the programs
• Views regarding what functions and types of personnel facilitate the coordination or integration of program elements
• Resourcing of integration efforts, and how joint activities are funded
• Impacts of federal, state, or local cooperative-agreement requirements on the decision to undertake integration initiatives and ways in which requirements either facilitate or hinder such initiatives
• Support needs for improving coordination
• Suggestions regarding modifications to existing cooperative-agreement mechanisms or legislation to facilitate the use of funding, staff, or other resources in developing such initiatives.

In addition, we collected relevant documentation from each locality we visited, including descriptions of implemented programs, planning and lessons-learned documents, and reports describing local public health and hospital initiatives.  Written documentation was used to augment the information from the interviews and helped identify any discrepancies in what the interview team learned. 

Data Analysis

We provide a descriptive analysis of these qualitative data with the aim of identifying similarities and differences between sites with respect to the following domains:  strategies considered and implemented; factors that motivated, facilitated, or hindered integration; resourcing of coordination and integration activities; which aspects of their preparedness programs lend themselves to integration and which aspects do not; effects of cooperative-agreement requirements on integration initiatives; resourcing of integration activities; and support needs.

Chapter Three.  Conceptual Framework

Introduction

Our review of the literature aided in the development of a conceptual framework for understanding how and why hospitals and public health organizations coordinate their efforts in preparing for bioterrorism or other public health emergencies. The framework synthesizes the information from a number of different types of literature, such as organizational behavior that examine interagency coordination efforts. It provides a working definition of coordination, identifies the methods that organizations use to coordinate, and highlights the characteristics of the tasks and organizations that impact coordination efforts.  In turn, the conceptual framework guided our survey analyses and the development of our case-study interview protocol. In this chapter, we provide a high-level description of the conceptual framework.

Coordination can be conceptualized as the "conscious activity of assembling and synchronizing differentiated work efforts so that they function harmoniously in attainment of organizational objectives" (Longest and Klingensmith, 1994).  Other researchers have adopted a similar definition, but they include the notion that coordination reflects the extent to which organizational activities are “integrated” (Bloom and Alexander, 1982; Lawrence and Lorsch, 1967) and “linked together” (Van de Ven, Delbecq, and Koenig, 1976).  The basic notion involves bringing together in some fashion various aspects of an organization (i.e., people, departments) or different organizations, so that there is an exchange of information and/or resources in accomplishing a common task or meeting a common goal.

Depending on various characteristics of the tasks or goals at hand and the people involved, coordination can be accomplished in a variety of ways.  The different coordination mechanisms can be classified into two broad categories:  programming and/or impersonal mechanisms and feedback and/or personal and/or group mechanisms (March and Simon, 1958; Van de Ven, Delbecq, and Koenig, 1976).

Figure 3.1 shows the different determinants of coordination as they pertain to public health preparedness.

Figure 3.1 Determinants of Coordination

Programming or impersonal mechanisms of coordination include “…the use of preestablished plans, schedules, forecasts, formalized rules, policies and procedures, and standardized information and communication systems.”  (Van de Ven, Delbecq, and Koenig, 1976).  Roles are formally prescribed, and there is little, if any, room for discretion on the part of the organizational “actors” (i.e., the people doing the work).  Specific mechanisms of coordination include the use of standardization of work processes and procedures (including clinical guidelines, pathways, and protocols), adopting rules and schedules, and developing standards for work outcomes (e.g., quality assurance).

Coordination by feedback (i.e., personal or group mechanisms) is less formal and not as stringent or rigid as coordination by programming.  It involves interaction between individuals or groups. The use of teams, committees, task forces, and workgroups as well as of other forms of communication to facilitate the achievement of common goals are specific mechanisms of this type of coordination (Daft, 1995; Scott, 1992).

The characteristics of the tasks at hand will affect the mechanisms of coordination that are used.  The task characteristics that are relevant include the level of interdependence among the individuals or groups executing a particular set of tasks, and the amount of uncertainty involved in executing tasks (i.e., the degree of predictability in performing one’s job or how many exceptions are encountered), and the level of complexity involved (i.e., how many different pieces of information must be collected and processed to make a decision). 

In addition to the characteristics of the task at hand, the characteristics of the organizations that must do the coordinating also affect the types of coordination mechanisms used and, potentially, the success of the coordination effort.  For example, the size of the group performing the tasks is relevant when a task force or committee will be formed to perform a particular task.  Generally speaking, larger groups find it more difficult to coordinate themselves because of the wide array of perspectives that may be found in large groups.  In addition, the composition of the group performing the tasks is quite relevant to the ability of the group to coordinate its activities using feedback forms of coordination.  More professionally or occupationally diverse groups may experience difficulty coming to consensus because of differences in professional training and the ways they have been taught to approach problems and problem solving.  Differences in culture and mission between public health and private health organizations have been suggested as barriers to effective coordination across these groups (Lasker and the Committee on Medicine and Public Health, 1997).  Furthermore, groups that are more diverse along gender, age, and racial and ethnic lines may also experience challenges with feedback forms of coordination.  That said, more-diverse groups may develop more-creative solutions to problems they encounter.  With good facilitators and leadership, it is quite possible that differentiated groups can effectively manage this terrain.

Process is the other key factor that affects the success of coordination efforts.  A review of the literature in this area conducted by the GAO (1992) identified three general steps that are essential to successful coordination efforts:  gaining commitment, building consensus, and creating an effective administrative entity.  Gaining commitment, or “buy in,” from involved parties requires obtaining and maintaining political support from key officials within and above the involved agencies.  To obtain commitment and full participation, it may be useful to offer financial or other types of incentives to agency officials and service providers.  Building consensus across organizations regarding the client needs that will be filled, the goals of the program, and the process for achieving those goals facilitates successful coordination.  In addition, it is important that an effective administrative entity be created to oversee the coordination process.  To be effective, the administrative entity must have the authority to make changes necessary for implementation.  Without such authority, the process of coordination can become bogged down in the bureaucracy. 

The literature identifies the primary barriers to effective coordination as including differences across organizations in philosophy, culture, and mission; issues related to turf protection; incompatible management information systems; and the lack of sufficient long-term funding for coordination efforts (GAO, 1992; Edwards and Stern, 1998; Martinson, 1999).  The development of good working relationships and effective partnerships requires significant time and hard work.  Moreover, the work is not finished when the collaboration is in place.  The maintenance of such relationships requires the time and energy of all involved parties.  Therefore, for successful coordination to take place, a stable and long-term stream of resources must be available for the effort.  In the absence of such long-term resources, it may not be possible for coordination efforts to even be attempted.

Discussion

Figure 3.2 shows the types of public health activities that may require coordination between public health and hospitals, as well as with other stakeholders.   The types of activities public health may undertake to improve preparedness can range from disease surveillance, case reporting, ensuring hospital surge capacity, distribution of the SNS and mass prophylaxis, to workforce training, and development and coordination of communications plans with other stakeholders.  The experience of various localities in responding to disease outbreaks or public health emergencies, such as West Nile Virus or Severe Acute Respiratory Syndrome (SARS) have highlighted the problems that can arise from inadequate communication plans or lack of coordination between local and state health departments, hospitals, and local officials in communicating with the media or public (Stoto et al., 2005).

Figure 3.2 Preparedness Activities and Types of Organizations that May Be Involved in Coordination

The conceptual framework guided our analyses by identifying the types of mechanisms that are available to health departments and hospitals to coordinate these various activities.  In addition, the framework highlights the fact that different aspects of public health preparedness (whether it be distribution of the SNS, mass prophylaxis of a population, or isolation and quarantine, among others.) will differ according to the set of necessary tasks and the set of stakeholders that need to be involved.  As a result, we would expect to see different coordination mechanisms in place, depending on the aspect of public health preparedness on which a local community is focusing.  Similarly, the conceptual framework (Figure 3.1) indicates that characteristics of the organizations (in this case, health departments and hospitals) involved in coordination will also affect the coordination mechanisms used and, thus, will be important to examine in the case-study analysis.  Lastly, the conceptual framework provides us with guidance on what types of factors may facilitate or hinder coordination between hospitals and health departments, including funding issues, differences in organizational mission and priorities, turf battles, and variation in the commitment of organizational leadership, among other factors.

The conceptual framework illustrated here draws primarily on literature from organization behavior and theory. Nonetheless, our interpretation of the paradigm is relatively broad and is fully consistent with a number of different perspectives and issues raised in other disciplines. For example, sociologists and behavioral psychologists might emphasize group dynamics and effective communication as key determinants of coordination, which is completely consistent with the conceptual framework we have described here. Our framework suggests that the characteristics of the group and the organizations they represent would affect group dynamics and how successful feedback-type coordination will be. Similarly, for effective communication, the conceptual framework suggests that the characteristics of the task, as well as the characteristics of the people involved in coordination, would affect the level, quality, and need for personal or written communication. Therefore, although different disciplines may have different perspectives and potentially different models of how coordination within and across organizations work, we interpret the conceptual framework described here as broad enough to incorporate or speak to these different perspectives.

Chapter Four.  Survey Results

We used the conceptual framework developed in the preceding chapter to guide our analyses of the survey data for considering coordination issues between local public health departments and hospitals.  The framework divides coordination activities into programming mechanisms, which tend to be more formal and generally impersonal in their use of preestablished plans, formalized rules, memoranda of understanding, and communication systems; and feedback mechanisms, which are generally less formal and more personal, and include both group modes (e.g., work groups, committees, and task forces) and individual modes (e.g., communication between a supervisor and a member of her staff or between coworkers).

To operationalize this framework for analyses of the survey data, we identified the questions that related to coordination mechanisms and classified them as programming or feedback mechanisms.  We then estimated the proportion of local health departments and hospitals that report the use of each coordination mechanism.  The analyses described here are meant to provide a broad national picture of coordination at the local level between health departments and general acute-care hospitals. 

Coordination between Local Health Departments and Hospitals

Programming Mechanisms

The formal use of preestablished plans, formalized rules, memoranda of understanding, or communication systems by programming mechanisms mean that such mechanisms tend to be more institutionalized.  The 2003 nationwide survey of local health departments and hospitals (Advisory Panel, 2003) asked a number of questions about the use of such mechanisms for public health preparedness, including whether the organization had an emergency or disaster response plan, what elements it addressed, the existence of mutual-aid agreements, the maintenance of contact lists, and the existence of communication systems.  The proportion of hospitals and local health departments reporting the use of these coordination mechanisms is presented in Table 4.1.  While many of these items refer specifically to coordination between local health departments and hospitals, others refer more generally to the issue of coordination within and across groups.  Combined, these items provide a broad picture of the level or nature of coordination in these organizations.

The results indicate that, in 2003, the vast majority of both local health departments, 88 (s.e. 4)⁵ percent, and hospitals, 98 (s.e. 2) percent, had emergency response plans in place.  These plans addressed a wide range of different issues.  Among those organizations with plans, 94 (s.e. 3) percent of both hospitals and 93 (s.e. 3) percent of local health departments reported that the plan addressed communication with other health responders (e.g., hospitals, medical providers, emergency medical services).  The integration of these plans with other organizations varied to some extent.  Among hospitals with plans, 82 (s.e. 8) percent reported that their plan was integrated with other local response plans, 66 (s.e. 8) percent reported integration with their state’s emergency response plans, and 52 (s.e. 8) percent reported integration with federal response plans.  The pattern of responses was similar for local health departments, although they were more likely to report integration with state response plans than were hospitals.  Among both hospitals and health departments, their emergency response plans typically addressed procedures for quarantine (hospitals, 73 [s.e. 6] percent; health departments, 66 [s.e. 6] percent), isolation (hospital, 90 [s.e. 4] percent; health department, 63 [s.e. 6] percent), and coordination with agencies outside their jurisdiction (hospitals, 83 [s.e. 5] percent; health departments, 90 [s.e. 4] percent).

Table 4.1 Programming Coordination Mechanisms Used by Hospitals and Local Health Departments in 2003 ^a

NOTES:  Standard error (s.e.) of the estimate is shown in parentheses. Numbers in the table represent percentages unless otherwise indicated. na = question not asked of that group.

^a Although the results for hospitals and health departments are presented together in the table as a convenient display, it is important to note that the survey’s sampling plan was designed for comparisons within each organizational type, not necessarily for drawing comparisons across organizational types.  Given that the same sample of counties was used for both organizational types, comparisons across the organizational types would have to capture the correlation due to clustering at the county level.  In particular, the sample of hospitals was designed to be representative of general acute-care hospitals nationwide. However, although a hospital randomly selected within a county may not necessarily be representative of all general acute-care hospitals within that particular county, most counties have either a single or a few local health departments eligible for sampling.
^b CBRNE = chemical, biological, radiological, nuclear, or high-yield explosives. Ratings of plan adequacy for the type of event each individual organization deemed most important for preparation were used to compute the average.
^c Indicates that the data are from 2001, because the survey did not ask the question in 2003.

As part of the 2003 survey, local health departments and hospitals also were asked to identify for what type of event--chemical, biological, or radiological—it was most important for their organization to prepare.  Local health departments tended to select a biological event; hospitals tended to select a chemical event (with biological events being a close second).  With that event in mind, organizations were then asked to rate the adequacy of their emergency or disaster response plan for that particular type of event.  On average, using a scale of 1 to 5, where 1 is equal to “inadequate” and 5 equal to “excellent,” hospitals rated the adequacy of their plan slightly higher than did local health departments (3.4 [s.e. 0.1] for hospitals, 3.2 [s.e. 0.1] for local health departments) (the difference in mean scores was not statistically significant), but both types of organizations appeared to feel that there was room for improvement.

For local health departments, the survey included questions regarding how the department has worked with hospitals to develop a smallpox plan.  The estimates indicated that health departments’ coordination with hospitals on this issue appears, on the surface, to be widespread, with 90 (s.e. 3) percent reporting some activity (not shown in Table 4.1).  However, the depth of coordination appears to be somewhat lower, with only 43 (S.E. 6) percent of health departments reporting multiple coordination activities (not shown in Table 4.1).  As evidence, we found that 46 (s.e. 6) percent of health departments had helped vaccinate local hospital staff, 24 (s.e. 5) percent had designated a specific facility to serve as the smallpox treatment center, and only 13 (s.e. 4) percent had worked with hospitals to make plans for how patients could be treated at home.  The “other activity” response category (not shown) was the most common, at 49 percent.

While data for 2003 were not available for some questions, we found that, in 2001, other coordination mechanisms complementing emergency response plans appeared to be quite prevalent.  Among local health departments, 73 (s.e. 5) percent reported having a communications system in place to rapidly deliver health alerts to the major health care facilities in their area. In addition, from the 2001 survey, we found that 70 (s.e. 5) percent of health departments and 86 (s.e. 4) percent of hospitals reported having mutual-aid agreements in place at the local, state, and/or federal level.  We also found that, in 2001, the maintenance of contact lists was quite prevalent (hospitals, 83 [s.e. 4] percent; health departments, 86 (s.e. 4) percent), although we have neither information regarding whether those lists were updated on a regular basis--a factor that would certainly affect their usefulness in an emergency--nor information on which communications mechanisms health departments would utilize to contact those on the list.  In contrast to the high prevalence of contact lists in 2001, only 46 (s.e. 7) percent of local health departments maintained capability lists outlining the hospitals, clinics, laboratories, or other facilities that might be used in the event of a large-scale disaster.

Feedback Mechanisms

More-informal coordination tends to take place via feedback mechanisms, which include both individual and group modes of communication.  The survey data are best suited to providing information about the group-communication mechanisms, such as joint training efforts, committees, and working groups.  Among hospitals, joint training with other health-related organizations for both natural disasters and terrorism-related events in 2003 appeared to be relatively common.  As shown in Table 4.2, for natural disasters, 74 (s.e. 7) percent of hospitals reported joint training efforts with emergency medical services (EMS) agencies, 49 (s.e. 7) percent with other hospitals, and 59 (s.e. 7) percent with their local health department.  Among local health departments, 53 (s.e. 6) percent reported training with EMS, 58 (s.e. 6) percent with local hospitals, and 57 (s.e. 6) percent with other health departments.  For hospitals, the proportion involved in joint training efforts for terrorism-related events was quite similar to that for natural disasters.  In contrast, for health departments, we found higher rates of joint training for terrorism-related events than for natural disasters in 2003.  Specifically, 72 (s.e.. 6) percent of health departments reported having trained or educated local health care and/or first-responder personnel on bioterrorism and terrorism response preparedness.  Moreover, 68 (s.e. 5) percent of health departments reported having increased funding for such joint training or education efforts in the two years following the terrorist attacks of September 11, 2001.

In addition to joint training efforts, the 2003 survey also asked health departments and hospitals whether they had participated in either tabletop or field exercises with other local groups. As indicated in Table 4.2, hospital participation in exercises with other local organizations was most common for natural disasters and chemical incidents. Although the pattern is similar for local health departments, we saw statistically significant (a=0.05) lower rates of exercise participation with other local organizations for natural disasters and chemical incidents.  Although not significant, local health department survey respondents reported lower participation rates than hospitals for biological, radiological, and conventional explosive incidents as well.

Membership in an interagency task force is another coordination mechanism that appeared to be quite prevalent in 2003, with 90 (s.e. 3) percent of hospitals and 92 (s.e. 3) percent of local health departments indicating participation. While it is encouraging to see such high participation rates, it is very likely that the effectiveness of work groups or task forces varied substantially with such characteristics as size of the group, quality of the leadership, and level of funding. Unfortunately, the survey data do not provide any insight into the workings of these task forces, which organization served as the main coordinating body, the degree of participation of hospitals and health departments in these task forces, and how effective these task forces or work groups have been in coordinating preparedness activities across multiple organizations.

Table 4.2 Feedback Coordination Mechanisms Used by Hospitals and Local Health Departments in 2003

NOTES:  Standard error (s.e.) of the estimate is shown in parentheses. na=indicates that the question was not asked of that group. Refer to Table 4.1’s footnote regarding comparisons across organizational types.

Changes in Coordination Between 2001 and 2003

The longitudinal nature of the survey data provides a unique opportunity to examine how coordination had changed between the summer 2001 (just prior to the terrorist attacks) and in the two years following the September 11 attacks and the subsequent influx of bioterrorism funding to improve public health and hospital preparedness. Nevertheless, our ability to examine changes over time is somewhat limited by differences in the questionnaires that were fielded in 2001 and 2003. In this section, we provide information on changes over time for a limited set of variables that were consistently measured across the two waves of data. These variables are summarized in Table 4.3.  It is important to note, however, that even though the questions may be worded in the same way, the events of September 11, 2001, clearly changed health professionals’ perceptions of preparedness and increased their awareness of preparedness-related issues. Therefore, any changes observed between 2001 and 2003 likely reflect both actual changes in coordination and changes in respondents’ perceptions and awareness. While this is true to some extent for all questions, it will be particularly true for the subjective questions regarding the quality of organizations’ emergency response plans. 

Focusing first on programming mechanisms of coordination, we found that, because nearly all hospitals had an emergency response plan in place in 2001 (Table 4.3), there was little scope to increase the proportion by 2003. In contrast, among local health departments, the proportion with an emergency response plan in place increased 23 (s.e. 6) percentage points from 2001 to 2003, among the organizations responding in both years.⁶ It is possible that the rate for health departments appears to be low because some of the health departments without an emergency response plan were included as part of the local Office of Emergency Management’s (OEM) plan. We were able to verify this possibility in the 2001 data, in which 81 (s.e. 9) percent of the health departments without a plan reported being part of their city or county’s emergency management plan. The data from 2003 do not allow us to investigate this possibility.

Among those organizations with a written response plan, the level of plan integration with local, state, and federal plans grew substantially between 2001 and 2003 for both hospitals and local health departments. The largest gains were found in hospital integration with state and federal plans, with increases of 45 (s.e. 10) and 42 (s.e. 10) percentage points, respectively, among those responding in both years.  There was also an increase for both organizational types in the subjective rating of plan adequacy. For hospitals, the average rating increased 1.2 (s.e. 0.2) on a 5-point scale, among those responding in both years.  Similarly, for local health departments, the average rating also increased. Although these increases seem rather small, given the influx of resources from the federal level that have been allocated to states to improve public health preparedness efforts, it is likely that this difference reflects the counteractive effects of increases in overall preparation and increases in the respondents’ standards for what is adequate. 

Table 4.3 Comparison of Coordination Mechanisms Used by Local Health Departments and Hospitals in 2001 and 2003^a

NOTES:  Standard error of the estimate is shown in parentheses. Numbers in the table represent percentages unless otherwise indicated.

a Individual cell entries reflect all responses for the given year, regardless of whether individual organizations responded in both years. 2003 values marked with a superscript “i” indicate a significant increase from 2001 to 2003 at the a=0.05 level, based upon a two-sided hypothesis.  Similarly, “ii” indicates a significant increase at the a=0.10 level.  Hypothesis tests for an increase from 2001 to 2003 are based on only those organizations responding in both 2001 and 2003.
b CBRNE = chemical, biological, radiological, nuclear, or high-yield explosives.

Turning to feedback mechanisms for coordination, we found that while there was not a great deal of change in the proportion of health departments and hospitals involved in joint training efforts aimed at natural disasters, that both hospitals and local health departments reported increases in joint training with other local health departments. In contrast, there were substantial increases in participation in joint training efforts for terrorism-related events.⁷ As seen in Table 4.3, hospitals and local health departments indicated across-the-board increases in joint training for terrorism-related incidents with other local health-related entities.  Though the increases are intuitively appealing, it is important to note that the increases seen here may be due in part to wording changes between the 2001 and 2003 questionnaires. In 2001, the survey asked about joint training for incident response to weapons of mass destruction, whereas in 2003 the survey asked about joint training for terrorism-related incident response. If terrorism was interpreted more broadly than use of weapons of mass destruction, then the observed changes may not reflect actual increases in joint training.

The estimates also indicate a substantial increase in participation in interagency task forces and working groups between 2001 and 2003. Among hospitals, participation increased 24 (11) percentage points from 2001 to 2003.⁸ Similarly, among health departments, we see an increase of 25 (s.e. 6) percentage points.

Discussion

The survey results enable us to paint a broad national picture of the types of coordination mechanisms health departments and hospitals have utilized and how that use has changed over time.  In 2001, hospitals appeared to rely more on formal or programming mechanisms to coordinate their preparedness activities than did local health departments.  By 2003, both hospitals and health departments reported utilizing extensively such programming mechanisms as having disaster response plans that addressed communication with other health responders and, to varying degrees, plans that were integrated with other local emergency response plans or with state and federal response plans.  Other programming mechanisms reported by local health departments in 2003 included having a communications system in place to rapidly deliver health alerts to the major health care facilities in their area or county.  Although in 2003 nearly all health departments reported having worked with hospitals to develop a smallpox plan, the depth of coordination appeared to be relatively low, with many health departments reporting only one coordination activity in this area (e.g., helping to vaccinate local hospital staff, having designated a specific health facility to serve as the smallpox treatment center, or working with hospitals to make plans for how patients could be treated at home).    

By 2003, both hospitals and health departments reported extensive use of feedback coordination mechanisms, such as joint training, participation with other organizations in tabletop or field exercises, and participation in interagency task forces to address preparedness for natural disasters and terrorism-related incidents. The survey results suggest that the influx of bioterrorism funding that followed the September 11, 2001, attacks may have helped facilitate public health’s adoption of both formal and informal mechanisms of coordination over time.

The above results also imply what we might expect to find in the in-depth case studies reported in the next chapter.

Limitations of the Survey Analysis

As is true of survey research in general, there are several study limitations that the reader should keep in mind.  The survey focused on the organizational level.  However, we recognize that the survey responses depend on how informed a particular individual is about his or her organization’s experiences with preparedness and with his or her organization’s emergency response and planning activities.  We attempted to control for this dependence by asking the organizational head to designate the person within their organization who would be most informed about public health and emergency preparedness activities to fill out the survey. 

It is also important to keep in mind that the general acute-care hospitals in our sample may not fully represent what is occurring within a particular county at the hospital level.  Within each county, we randomly selected one organization from each of the following respondent groups—general acute-care hospitals and local health departments—to be in our survey.  Depending on the size of the county and the number of general acute-care hospitals within a county, this meant that the chance of a hospital within a particular county being selected into the study varied.  In contrast, most counties either had a single city health department or a county health department, which meant that, at the county level, these organizations had a greater chance of being selected into the sample than did hospitals.  Therefore, it is possible that some of the general acute-care hospitals in our survey may not have been as actively engaged in preparedness (and coordination) as other hospitals within a particular county, which did not make it into our sample.

The 2001 survey was conducted in summer 2001, just prior to the September 11, 2001, terrorist attacks and the 2003 survey was conducted two years following this incident.  One might expect that, in 2003, respondents within local health departments and hospitals may have been more sensitized to the issue of terrorism and public health preparedness than those who responded in 2001.  It is difficult to predict the direction of the bias.  For example, one might argue that heightened sensitivity may lead some individuals in the 2003 survey to understate the level of preparedness of their organization—that is, the more you know, the more likely you are to assess your organization as being less prepared.  Then again, individuals who were less sensitized to this issue in 2001 or less aware of the issue may be more likely to overstate their organization’s level of preparedness.  We suspect that the direction and magnitude of the bias resulting from this sensitivity will differ for objective versus subjective measures.  In this report, we largely focus on objective measures, such as planning, training, and other preparedness activities (e.g., whether an organization has a response plan as opposed to self-assessed ratings of preparedness).

Chapter Five.  Case-Study Results

Introduction

The conceptual framework identified a number of factors that may facilitate or hinder interagency coordination, including the role of leadership and buy-in from agency directors; differences in organizations’ missions and cultures; the characteristics of organizations and the environment (such as number of stakeholders, organizational size, and degree of geographic isolation); and funding.  In this chapter, we report on the experience of health departments and hospitals within the six case-study sites in coordinating on different aspects of preparedness.  We also highlight some promising approaches to coordination that may be generalizable to other localities.

Case-Study Results

As noted in Chapter Two, our selection criteria included:  (1) localities that national public health and hospital leadership identified as having innovative public health or hospital preparedness programs, (2) programs from different types of counties (e.g., large versus small, urban versus rural), and (3) programs from different regions of the country.  We used a multipronged approach to identify candidate sites for the case-study analysis.  We drew on an analysis of the 2003 Survey of Federal Preparedness Programs for Combating Terrorism for local health departments to identify those counties that appeared to be more actively engaged in addressing emergency preparedness in general and for which coordination among the health department and hospitals (as well as with other response organizations) appeared to be taking place. The information from the survey data was combined with results from the previous year’s tasks on exemplary practices and input from the National Association of County and City Health Officials (NACCHO) and the Association of State and Territorial Health Officials (ASTHO) to identify counties and sites that merited closer examination.  Table 5.1 shows the characteristics of the six case-study sites.

Table 5.1 Characteristics of Case Study Sites

NOTE: We interviewed between 4 and 12 individuals per site, including representatives from the health department, local hospitals, hospital associations, interagency working groups, and emergency management.

In selecting sites for the case studies, we sought to identify successful models of coordination between local health departments and hospitals. However, our visits to the selected sites revealed that, in some cases, there was a disconnect between the “paper” version of coordination efforts and the reality.  We found that substantial coordination between local health departments and hospitals was occurring in only three of the six sites. Nonetheless, our visits to the other three sites enabled us to extract valuable information regarding coordination barriers.

The in-depth interviews generated a great deal of information regarding the characteristics of successful coordination efforts, as well as common barriers to coordination. In our synthesis of these data, we found that the majority of issues could be categorized into the four key areas identified in the conceptual framework: leadership, culture and mission, organizational characteristics and environment, and funding.

Leadership

The importance of leadership in promoting coordination is a theme that emerged across all sites.

Leadership at all levels is needed to promote coordination efforts. Strong leadership is needed at all levels--federal, state, and local. At the federal and state levels, leaders need to provide a clear and consistent statement of the missions, goals and of objectives to guide local preparedness activities. While HRSA and CDC do provide guidance, a number of interviewees noted that this guidance was unclear and changed from year to year, making it very difficult to move forward with preparedness activities. In some cases, local health departments were reluctant to begin coordination efforts until they felt that they had a clear set of goals and objectives to guide their activities from the federal government that would not change from year to year. Similarly, in a number of localities, interviewees indicated that no clear targets for preparedness efforts or measurable outcomes had been set by the state, leaving the localities confused about the priority areas they should be focusing on. It is important to note that the lack of clear guidance from federal and state officials stymied coordination efforts at some sites, but not at all sites. In at least one case where there was a state-level lack of leadership, local preparedness overall and, specifically, coordination efforts were thriving.  

Although leadership from the federal and state levels facilitates local coordination, local leadership is key. The organizational location (whether within the health department or the hospital or a healthcare association or interagency group) of the leadership is less important than its quality.  Judging from our site visits, we found that the coordinating body for local public health preparedness need not be the health department. In one of the successful models, a healthcare association (representing all hospitals in the state) took