Enhancing Surveillance to Detect and Characterize Infectious Disease Threats
Infectious disease surveillance is conducted at all levels of government, with most surveillance legally authorized and performed by a heterogeneous set of state and local public health departments that voluntarily collaborate with the federal government (see Chapter 4). In practice, many separate systems are engaged in activities that contribute to public health surveillance at local, state, and national levels.
The multiplicity of surveillance systems (many of them monitoring specific diseases), the unevenness of their capabilities, and both the strengths and limitations of current approaches to surveillance through public health and health care systems have been the subject of many different studies, task forces, commissions, and other efforts both to assess the status of the situation and to propose changes to remedy current problems and bring about improvements (e.g., CDC, 2001; Baker and Koplan, 2002; IOM, 2003; Baker et al., 2005; Lurie et al., 2006).1 These analyses share some broad themes.
The nation is facing an increased threat from infectious disease outbreaks, both intentional from terrorists and natural from emerging and re-emerging pathogens.
The nature of many bioterror threats (e.g., a high degree of infectiousness or severe morbidity, short incubation period, decreasing effectiveness of
treatment as the disease progresses) necessitates a very rapid medical response to prevent most of the casualties.
Local and state public health agencies play a vital role in monitoring disease trends and outbreaks but do not all have sufficient staff, tools, and resources to be as effective as they need to be for rapid detection and response to outbreaks (e.g., adequate numbers of trained personnel, modernized information and communication systems, access to and capacity to fully use epidemiologic data systems and analysis tools, expanded and modernized public laboratories).
The sharing of surveillance information between the health care system and state and local public health agencies needs to be improved to detect health threats and detect them earlier.
Automated systems to improve the sharing of surveillance information between the health care system and state and local public health agencies to detect widespread outbreaks earlier and manage them better also need to be improved.
The integration of human health information with information about infectious agents derived from surveillance of animal disease, water quality, and air quality—or “biosurveillance”—is needed to defend against bioterrorism and natural pandemics.
With its focus on the BioWatch system and a short timeframe in which to conduct its work, this committee acknowledges the large number of completed and ongoing efforts over the past decade, especially since 2001, to improve infectious disease surveillance and detection of disease outbreaks. In this chapter, the committee describes opportunities it has identified for further enhancing the detection of public health threats, especially threats from biological hazards, including bioterrorism. Many of the enhancements derive from the increasing digitalization of health information and resulting opportunities for better information technology systems and tools to aid case recognition, reporting, and analysis of information. A central and desirable aspect of the enhancements—in the context of bioterrorism surveillance—is to improve the timeliness of individual case and outbreak recognition, reporting, and analysis of information.
The opportunities for enhancement of surveillance through the public health and health care systems fall into broad and overlapping categories of improving legally mandated reporting, establishing automated linkages between health care information systems and public health systems, increasing laboratory and diagnostic testing capacity, and promoting information integration and knowledge sharing. These enhancements bring with them the risk of information overload, which must also be addressed through improved information and knowledge management (the field of public health informatics) (Yasnoff et al., 2000). Information tools to manage
such floods of information must be specifically designed for the end user to ensure that salient information is correctly interpreted in a timely manner (Endsley et al., 2003).
Skilled staff and other infrastructure resources are needed to use these systems effectively on a day-to-day basis and to evaluate and respond to these additional or enhanced information streams. Automated connections between health care and public health systems are unlikely to be fully effective in the absence of mutual trust, deserved respect for one another’s expertise, and effective personal communication links between the health care providers and public health officials, particularly at the local level. In the final portion of the chapter, the committee provides recommendations for making the most of the enhancements and providing critical baseline capabilities to benefit from them.
LEGALLY MANDATED REPORTING
As discussed in Chapter 4, each state has laws requiring health care providers, laboratories, and other entities to report certain diseases and other conditions to local or state public health authorities. Typically, state laws also mandate reporting of unusual clusters of disease or outbreaks, even if the disease in question is not on the list of reportable conditions.
The Council of State and Territorial Epidemiologists (CSTE) has proposed the establishment of a common list of nationally notifiable conditions to be reported to all levels of the public health system (CSTE, 2007). CSTE has also proposed establishment of a list of “immediately notifiable conditions” in addition to routinely notifiable conditions (CSTE, 2008). Immediately notifiable conditions would be those that might constitute a “public health emergency of international concern,” as defined in International Health Regulations (WHO, 2008). They include such conditions as smallpox, a novel strain of influenza, wild-type polio, and severe acute respiratory syndrome (SARS). The list also includes diseases caused by pathogens in Category A on the Centers for Disease Control and Prevention’s (CDC’s) list of possible bioterrorism agents and toxins (e.g., anthrax, plague, tularemia). These immediately notifiable diseases have been largely eliminated in the United States (e.g., measles) or globally, or are otherwise designated by the Nationally Notifiable Disease process.
In addition, standardized disease investigation forms with required data elements are needed for all notifiable diseases. CSTE and CDC are currently working together to establish an updated list of Nationally Notifiable Conditions, which will include nationally standardized case definitions,2 forms,
Many case definitions are now available at http://www.cdc.gov/ncphi/disss/nndss/casedef/case_definitions.htm.
and disease-specific codes and data elements, so that information can be readily and quickly shared, compared, and analyzed. This standardized approach to reporting the diseases of greatest concern is important, because a biothreat or naturally emerging infectious disease may strike more than one location, or terrorists may target an airport, state fair, sports arena, or other place with people travelling to and from multiple jurisdictions.
Electronic Laboratory Reporting Systems
Electronic reporting of laboratory results to public health officials can improve the timeliness and completeness of information regarding notifiable diseases and other agents of potential concern (Effler et al., 1999; Panackal et al., 2002; Babin et al., 2007). Such systems can reduce the delays and incompleteness of reporting based on busy laboratorians having to fill out and submit paper forms by mail or fax. While the benefits of more prompt reporting will vary by disease, depending on the urgency of beginning prophylaxis or treatment, electronic reporting can at least shorten one critical step in the transfer of information about the occurrence of notifiable diseases (CDC, 2008b). As of late 2008, about 85 percent of the states have some capacity for electronic laboratory reporting (ELR) (TFAH, 2008).3
In order for ELR systems to be deployed in a jurisdiction, standardized information exchange is needed between laboratories and public health reporting systems (i.e., standard vocabularies and codes and electronic messaging and transmission formats). The main impediment is the time required for both laboratory information technology (IT) staff and public health staff to map out a customized interface from each laboratory information system to the standardized format required for each notifiable disease, especially if a disease is confirmed by more than one test. This interface problem is reduced when regional health information exchanges (HIEs) electronically transmit information, including laboratory reports, between multiple users across the community. These information exchange systems can be politically and technically challenging, but they offer the possibility of creating a limited set of interfaces rather than unique interfaces for every entity and every application. Thus they can speed the process of establishing ELR in a community where it is not yet established. ELR has substantially increased the speed of delivery, the number of case reports, and the quality
of patient-level information (e.g., address, demographics) provided in case reports (Overhage et al., 2008).
Better case ascertainment from ELR also creates higher information flow and workload demands on public health staff. For this to translate into better public health response, sufficient numbers of skilled public health professionals and high-quality information management systems are needed to quickly assess if reports meet disease case criteria and to implement appropriate case and contact investigations.
Notifiable Disease Reporting by Health Care Providers
As discussed in Chapter 4, all states mandate that physicians, veterinarians, laboratories, and other health care providers report certain health conditions—mainly infectious diseases—to their local or state department of public health, or both. Although health care providers are generally aware of the requirement to report, compliance is a problem (Jajosky and Groseclose, 2004). Providers may not be fully aware of the scope of reporting requirements, may assume that others (such as the laboratory or hospital infection control staff) will report, or be too busy with patient care to stop work and report a disease. Some may also doubt that their input is necessary or would be used effectively.
The challenge of complying with reporting requirements is even greater in clinical environments such as busy hospital emergency departments (EDs) and acute care clinics. Because these sites provide readily accessible care 24 hours a day, clinicians working in these settings may be the first to encounter patients with signs or symptoms of illness from a bioterror agent or an emerging infectious disease (IOM, 2007). The crowded conditions that characterize many EDs are also a problem, because they can extend the waiting time to see a physician and promote patient-to-patient spread of virulent infectious diseases, such as SARS (Augustine et al., 2004; Cass, 2005). Clinicians working in these environments are expected to diagnose and treat a wide range of conditions of widely varying severity, from immediately life-threatening to chronic. The frequent interruptions, need for multitasking, and circadian stress of shift work reduce the probability that a busy emergency physician or nurse will quickly report a diagnosis. In addition, the results of microbiologic tests may not be available for hours or days after they are submitted. Reporting rates from EDs could be enhanced if steps were taken to raise awareness among clinicians of the importance of case reports in monitoring public health and identifying disease outbreaks, or to provide additional support from infectious disease physicians or infection control staff who have somewhat more time to review differential diagnostic information.
Recognizing and diagnosing unusual diseases in EDs and acute care clinics is also a challenge. For bioterrorism preparedness, primary care provid-
ers and key specialists (e.g., in emergency medicine, infectious disease, dermatology, radiology, and microbiology) need to acquire and maintain “front of mind” awareness of the clinical manifestations of infection with Category A and B agents, which for many clinicians may be a once-in-a-lifetime diagnosis. Health alert systems, such as Health Alert Networks (HANs) located at CDC and all state and some local health departments, which broadcast email and faxes to health care providers and systems, have been put in place to rapidly notify providers of potential or breaking public health emergencies, but the task of signing up all providers is not complete in most places. In the event of an outbreak, HAN alerts should be used to increase the index of suspicion for providers by providing them with the specific clinical and epidemiologic criteria, public health contact information for reporting cases, and instructions for obtaining confirmatory laboratory testing.
In addition to maintaining provider outreach efforts on the requirements for reporting of notifiable diseases, local and state public health departments should take steps to ease the burden of reporting and ensure that staff who triage calls from providers are responsive and adequately trained. These steps, which have been taken by many but still not all public health systems, include:
24/7 toll-free call lines at the local or state health departments with the capability to rapidly connect providers with an appropriate staff person to triage their calls, including an on-call medical epidemiologist if indicated;
easy-to-use Internet-based reporting systems; and
feedback to clinicians regarding local disease trends and the impact of their reporting.
Enhancing the Infection Control Professional Workforce
Infection control professionals (or infection preventionists) are an important human link connecting the clinical and public health sectors. They often report notifiable conditions to public health, assist clinicians with definitive diagnosis, assist public health authorities with case investigation, and improve the speed and consistency of the application of infection control measures such as isolation. While common in hospitals, they are less often present in ambulatory care settings. Strengthening the system of infection control professionals across the health care system could provide an important tool to improve capacity for surveillance and response to bioterrorism or emerging infectious diseases, while simultaneously reducing health care-associated infections.
Electronic Death Reporting Systems
Making death reporting systems electronic may speed the detection of unusual trends in mortality. Tracking causes of mortality is the oldest form of public health surveillance and has been done for centuries. Ironically, the data systems in place to track mortality in most states use the same technology employed in these earlier centuries—i.e., paper—and are therefore slow. Electronic death reporting could improve timeliness of detection of clusters of deaths, identification of cases of infectious disease that should have been reported to public health by the treating physician, and identification of seasonal trends in infectious causes of death (Fallon and Boone, 2004).
In New Hampshire, death certificates are filed electronically within 24 hours of being signed by a physician. A surveillance coordinator in the New Hampshire Department of Health and Human Services reviews death certificates daily using a query developed to identify more than 50 illnesses potentially related to terrorism (Fallon and Boone, 2004).
The Los Angeles coroner reports on each day’s cases to the Los Angeles County Department of Public Health (LACDPH). Of the approximately 70,800 coroner’s cases reported between 2003 and 2006, the LACDPH investigated 424, and among the investigated cases, 196 (46 percent) had an infectious disease as the cause of death. Of these infectious disease cases, 81 (41 percent) had not been properly reported prior to death, including cases of hantavirus and rabies. There were also 505 cases with causes of death attributed explicitly to reportable communicable diseases, of which more than half (56 percent) had not been previously reported (Peterson and Terashita, 2008).
AUTOMATION OF HEALTH CARE INFORMATION SYSTEMS AND PUBLIC HEALTH LINKAGES
Advances in IT make it possible to collect and analyze public health information more quickly, accurately, and comprehensively than is possible with traditional manual methods. IT can greatly ease the demands of reporting for providers, as discussed above, especially for initiating case reports. However, additional input from clinicians is likely to continue to be necessary to complete case reports in many instances (e.g., to obtain information on patients’ modes of exposure or more nuanced information necessary to verify that patients meet surveillance case definitions). Electronically reported data can be more easily aggregated and analyzed for patterns that indicate a possible outbreak of naturally occurring or bioterrorism-caused disease. Early trend detection by public health can be shared with clinicians using HAN and other notification systems, thus creating a circle of enhanced diagnosis, case reporting, infection control, and treatment.
Clinical Decision Support Tools
Many outbreaks are identified and reported by astute clinicians who encounter cases or clusters of cases with suspicious features, such as monkeypox in 2003 (Reed et al., 2004), anthrax in 2001, West Nile virus in 1999 (Fine and Layton, 2001), and hantavirus in 1993. In fact, the most crucial step in disease detection is the first one—recognizing that an ill patient has a potentially unusual disease or disease manifestation that warrants further investigation and notification of public health officials. Although it is widely assumed that all clinicians working in busy acute care settings intrinsically possess the knowledge, skills, and means to promptly identify, isolate, and report suspicious cases, this is not necessarily so. Most health care providers—particularly those who work in EDs and other acute care settings—constantly juggle competing demands for their time and attention. Yet it is vital that these same clinicians promptly recognize index cases that might signify a bioterrorist attack or an emerging outbreak of a dangerous infectious disease.
Traditionally, health departments and professional societies have relied on continuing education, posters, and pocket cards to educate health care professionals about the signs and symptoms of an illness caused by bioterrorism agents and other dangerous diseases. Although these measures are useful, they are not enough (IOM, 2000). Health care providers need decision support tools in their clinical environments that are designed to help them with their daily tasks (IOM, 2007). Examples include rapid dissemination of paper triage screening forms for a particular disease of concern (e.g., for SARS [Foldy et al., 2004b]), computer-assisted triage systems (e.g., Dong et al., 2007; Bullard, 2008), web-based interfaces or other electronic tools to help clinicians formulate differential diagnoses and access guidance about the clinical management of patients with rare or highly dangerous diseases (e.g., Papier, 2008), and electronic medical records (EMRs) with imbedded decision prompts that remind busy clinicians to consider certain diagnoses and report particular diseases (see Recommendation 10, below). Computer-assisted triage systems, for example, could be enhanced by programming them to automatically alert the triage nurse if an ED patient has history, symptoms, or clinical signs suggestive of exposure to a bioterrorism agent or other public health threat (IOM, 2007). In addition to reducing medical errors and streamlining patient flow, routine use of decision support systems should increase the likelihood that index cases of bioterrorism or emerging infectious disease are recognized in the earliest stages of an outbreak of potential public health significance. Although diagnosis of most diseases caused by bioterrorism agents will require more information (e.g., from a physician’s examination, laboratory tests, or X-rays), the decision support tool may raise the index of suspicion to make appropriate
care more likely. This could enhance prompt reporting, use of appropriate diagnostic tests, institution of appropriate isolation measures, and swift treatment of both ill and exposed individuals.
Technologies exist to assist busy ED clinicians and hospital-based adult internal medicine and pediatric specialists in making clinical decisions concerning prevention and monitoring guidelines, prescribing of drugs, and diagnosis and management (IOM, 2001). Studies show that clinical decision support systems can increase clinician compliance with clinical guidelines and drug selection, screening for interactions, and monitoring for adverse side effects; but their utility in improving diagnosis is still unclear, especially in the ED setting (IOM, 2007). Further development of these systems is warranted, because of their promise in improving diagnosis and treatment, but additional research on their accuracy, effectiveness, and safety is also warranted.
Ideally, these tools should be modifiable and also allow for a multidirectional information flow. It should be possible to modify triage protocols to be specific for outbreak-related cases when warranted. And in addition to facilitating reporting of suspicious cases to the health department, they should allow the health department to alert front-line health care providers of a real or potential outbreak. For example, when a BioWatch Actionable Result (BAR) is declared or aberrancy is detected by syndromic surveillance, planners invariably want timely information from hospital EDs and other acute care providers to assess the validity of the signal.
This process could be dramatically enhanced in a setting of better bilateral electronic communication between clinicians and public health agencies. Unfortunately, the care settings in which this information exchange must occur (hospital EDs, intensive care units, and acute care adult and pediatric wards in hospitals and clinics) often have weak links with health departments, which tend to have more established ties with infectious disease departments. This represents a major opportunity for improvement in our nation’s disease surveillance system. One approach is the use of regional emergency medical Internet (REMI) systems to communicate with and collect information from emergency services providers on a near-real-time basis, on scales ranging from local to multistate alerting and surveillance (Barthell et al., 2003; Foldy et al., 2004c).
Electronic Medical Records
Greater use of information from EMRs and other electronic health information sources, with linkages to public health, could enhance disease surveillance by detecting unusual cases of disease, clusters, or trends (Klompas et al., 2008; Lazarus et al., 2009). While adoption of EMRs is
proceeding slowly (Jha et al., 2009), their widespread adoption would help improve bioterrorism preparedness, but only if the standards for linking EMR and public health systems provide for timely reporting to public health and include the information needed to rapidly detect and respond to potential biothreats.
Electronic health record systems that are linked together by regional health information systems and to local and state public health departments could accomplish a number of essential tasks, including:
increasing the capacity to prompt or automate key steps in notifiable disease reporting for patients meeting certain criteria (e.g., an ICD-9 diagnosis or laboratory report);
increasing the capacity to prompt enhanced surveillance in response to a suspected or confirmed outbreak (e.g., automatic prompts to consider specific diagnostic testing for patients meeting clinical criteria); and
increasing the capacity for public health officials to review more detailed clinical information (e.g., radiologic and laboratory findings) when a suspect disease cluster is detected or conduct more detailed chart reviews electronically.
Regional Health Information Exchanges
In a growing number of communities, information from EMRs, electronic laboratory reporting, hospital and other clinical registration systems, and other electronic health data are securely shared across organizational boundaries in regional health information exchanges (eHealth Initiative, 2008). In 2008, there were 42 operational HIE systems in the United States (eHealth Initiative, 2008).
These systems offer many opportunities to enhance both surveillance and information sharing between clinical providers and public health authorities. In Wisconsin, for example, data from over 10 hospitals and many clinics are made available (with personal identifiers removed) for real-time syndromic surveillance by local and state health departments (Foldy et al., 2008). CDC is preparing to pilot electronic feeds from three other HIEs into the BioSense syndromic surveillance system (Lenert, 2007). In Indianapolis, the Indiana Health Information Exchange has facilitated a great increase in the completeness and timeliness of electronic laboratory reporting to public health (Overhage et al., 2008). Indeed, in 2008, 6 of the 42 operational HIEs indicated that they were providing surveillance information to public health agencies, and 5 were exchanging electronic laboratory results with public health agencies (eHealth Initiative, 2008).
If HIEs are more widely adopted and become a universal medium for sending and receiving information in regional health care markets, they
may replace existing stand-alone public health alerting systems for communicating with clinicians and offer two-way communication between public health and health care partners to improve completion of case reports and similar collaborative communications. In the future, as the interoperability of HIEs with medical practice support systems like EMRs becomes more sophisticated, HIEs may offer the opportunity for public health officials to adjust electronic decision support systems that guide clinician practice (e.g., a rising incidence of pertussis or suspicion of an terrorist release of anthrax spores leading to specific prompts in the decision support tool to raise the provider’s index of suspicion and prompt requests for diagnostic tests for a particular symptom complex) (Hanrahan et al., 2006). While such opportunities are currently speculative, if HIEs do survive and thrive, there are few technical reasons why public health systems and health information exchange could not converge over time, allowing more real-time and less labor-intensive information sharing between health care and public health professionals.
The federal government has already recognized the advantages of standardizing health information exchange at the national level. The second goal of the Federal Health IT Strategic Plan: 2008–2012 reads, “Population Health—supports the use of electronic health information—primarily, but not exclusively, generated as a by-product of health care delivery—for critical national needs relating to public health, biomedical research, quality improvement, and emergency preparedness. Such use would promote early and effective management of infectious disease outbreaks, improved tracking of chronic disease management, the ability to gather data for research purposes, and the evaluation of health care based on value, by way of comparable price and quality information” (ONC, 2008, pp. 2–3). This commitment has been made more concrete by the health information technology programs passed in the 2009 American Recovery and Renewal Act, which created economic incentives for health care providers to use electronic medical records and engage in “meaningful use” of regional health information exchange.
The standardization of health information necessary to facilitate such clinical information exchange will similarly facilitate the horizontal transmission of critical public health information between jurisdictions as well as vertically from local to state to federal levels. As HIEs develop further, continued involvement of public health professionals will be needed to facilitate and evaluate this enhanced information sharing.
Although regional health information exchange systems can reduce the need for many unique and expensive interfaces tying EMR feeds and other data to public health systems, achieving wider adoption must overcome important barriers. Some of the barriers include establishing a basis for financial sustainability, addressing concerns about preserving privacy and
confidentiality of shared patient data, reaching agreements between public and private contributors on governance of the exchange system, and creating incentives for participation.
LABORATORY AND DIAGNOSTIC TESTING CAPACITY
Technology is expected to increase the availability, speed, accuracy, and utility of diagnostic tests, which would contribute to recognition of natural and terrorist disease outbreaks.
Rapid “Point-of-Care” Diagnostic Tests
The development and appropriate use of rapid, multiplexed “point-of-care” diagnostic tests seem likely to strengthen both surveillance and bedside patient care by helping clinicians pinpoint the cause of an infection (e.g., Mahony, 2008). This is still an emerging technology. In 2008, the Food and Drug Administration (FDA) approved the first multiplex system that simultaneously tests for 12 different respiratory viruses (FDA, 2008), as well as at least one other system that tests for a smaller set of respiratory viruses. Many bioterrorism agents produce nonspecific prodromes such as fever, cough, and myalgia that make it difficult to promptly diagnose the source of an infection if a patient presents for care in the first few days of illness before more characteristic signs and symptoms appear. In theory, the availability and use of validated multiplex point-of-care bioassays at frontline acute and primary care clinical settings would allow clinicians to test patients in response to public health alerts for persistent and seasonal pathogens, as well as bioterrorism agents and other pathogens of potentially great public health significance (e.g., avian influenza, SARS). In the event of a BAR or a public health alert, such testing might be used to detect or rule out a potential outbreak.
However, several factors can be expected to influence the implementation of such testing, especially for the purpose of detecting infections related to bioterrorism. Tests to be used in a clinical setting must be approved by the FDA, and the path to FDA approval of such diagnostic tests has not yet been made clear. Studies are needed to assess the costs of widespread use of such testing and how those costs will be met. For testing to be routinely conducted in a clinical setting, it may be necessary for the Medicare program and other insurers to agree to cover these costs to make their adoption economically attractive. As the current generation of multiplex tests for respiratory viruses are adopted, their contribution to surveillance will need to be evaluated. Rapid tests for clinical settings still need to be developed for the bioterrorism agents of greatest concern. If they are developed and approved, they, too, will require piloting and evaluation to establish their usefulness (see Recommendation 10, below).
Characterization of Pathogens
Many efforts are under way to extend capacities to molecularly characterize pathogens and compare findings to identify the emergence or spread of related cases of infectious disease (e.g., PulseNet model). Laboratories that diagnose and report cases now often send specimens to public health laboratories for phenotypic analysis or genetic “fingerprinting” (e.g., pulsed-field gel electrophoresis, insertion segment analysis, and nucleotide sequencing), which can be compared to other isolates around the nation or world. This has allowed cases of illness to be linked to common sources that would otherwise have gone unnoticed, and it has permitted the detection and characterizations of several recent national and international outbreaks of food-related disease (e.g., CDC, 2006, 2009b).
Collection and Testing of Clinical Specimens
The ability and capacity to collect and test clinical specimens as part of public health surveillance systems, either as an ongoing process or in response to possible alerts, needs to be expanded or enhanced. This involves improving the capability of hospital and commercial laboratories to serve as sentinel facilities to conduct preliminary testing of suspect bioterrorism agents and to rapidly recognize and transport suspect specimens to the nearest public health reference laboratory in the Laboratory Reference Network (LRN) for standardized confirmatory testing. It should be noted that attempts to constrain health care costs make it likely that microbiology tests will be used less frequently, especially in situations in which the clinical management of individual patients is unlikely to be dependent on definitive diagnostic testing. Thus, such surveillance may need to be supported financially by methods separate from personal health insurance.
There is a critical need to increase public health reference laboratory capacity in every state for
molecular subtyping by DNA fingerprinting and DNA sequencing and serologic type testing;
participation in networked comparisons of fingerprinting, such as CDC’s PulseNet (which compares pulse-field gel electrophoresis of enteric organisms). This allows detection of the emergence of similar bacterial isolates from patients across state or even national boundaries, which may indicate widely separated cases in a disease cluster resulting from a single source, such as a contaminated food product;
rapid molecular confirmation of Category A and B agents and other known infectious disease agents of public health significance (e.g., SARS and avian influenza [H5N1]);
rapid development and deployment of molecular tests for new pathogens as they emerge (e.g., as was accomplished for novel influenza A [H1N1]); and
electronic linking of public health laboratory information management systems with epidemiology and surveillance systems to enhance rapid reporting and detection of patterns of disease over time and space.
INFORMATION INTEGRATION AND KNOWLEDGE SHARING
In many public health systems, there are multiple surveillance systems and databases, typically created piecemeal over time for specific diseases and conditions or other purposes. Integrating these across local and state levels would permit greater awareness of patterns and trends and increase the likelihood that anomalies that might indicate a disease outbreak are detected and investigated.
Similarly, joining public health information with the information from other sources, such as information from law enforcement on terrorist activities, from surveillance of animal health, and from monitoring air and water quality, would enhance biosurveillance. Issues of privacy, confidentiality, and security must be carefully managed for such “fusion” of health information with other intelligence information (Riegle, 2009).
Integration of Public Health Information
Integration and analysis of public health information within and across public health jurisdictions promises to enhance surveillance by making it possible to associate related phenomena that otherwise would remain isolated. It might provide situational awareness of normal events, such as the beginning of the influenza season, while potentially enabling earlier detection of unexpected disease outbreaks caused by terrorism or that occur naturally. It may also help inform the response to an outbreak. As with other tools discussed here, its best use should be an area of cautious implementation and active evaluation.
Public health information originates from a variety of sources external to public health agencies, and even within a public health agency surveillance information is often compartmentalized by program (e.g., communicable disease, maternal and child health, environmental health). The major external source of public health information, of course, is the health care sector. Generally, health care providers are already required to report suspected cases of disease with major public health consequences, and approaches to enhancing the reporting of cases to public health have been discussed above. But other external sources of information important for biosurveillance do not necessarily report information to public health
authorities. These include law enforcement and intelligence agencies, agriculture and wildlife agencies that monitor animal health, municipal water authorities that monitor water quality, and air pollution agencies.
There is an urgent need for better situational awareness, both to detect anomalies that might signal a disease outbreak, and, in the event of an outbreak, to help responders monitor the spread of the disease and institute mitigating actions. This is specifically mandated in the Pandemic and All-Hazards Preparedness Act of 2006 (P.L. 109-417), which states that the Secretary of Health and Human Services shall “establish a near real-time electronic nationwide public health situational awareness capability.” In response, CDC established a public health information fusion center in 2008, BioPHusion, to “incorporate information from multiple disparate data sources, facilitate the exchange of information across programs, and analyze aggregated interpreted data (information) from existing surveillance systems in order to enhance agency-wide situational awareness” (Rolka et al., 2008, p. 3). Currently, BioPHusion receives information from CDC programs, including BioSense and Epi-X, and from open sources outside CDC. It distributes a daily situational awareness report to the CDC director, division directors, and branch chiefs, and to selected external partners regarding infectious disease outbreaks, toxic spills and other accidental exposures, and natural disasters such as earthquakes and hurricanes. BioPHusion receives information about BioSense anomalies and the BioSense news digest. The program is working out memoranda of agreement to share information with other federal agencies with public health-related data, including the U.S. Department of Agriculture, which conducts veterinary surveillance. The plan for the future is to create online communities of public health practitioners who collect and exchange real-time information for situational awareness at the state and local levels (Rolka et al., 2008).
In 1994, problems with the local water supply in the Milwaukee area resulted in a cryptosporidium outbreak that sickened approximately 400,000 residents. The outbreak was not recognized until shortages of diarrhea medications and enteric culture media were reported (Proctor et al., 1998). The outbreak might have been detected earlier if information held by different agencies and organizations had been shared (Foldy, 2004). In response to that experience, the City of Milwaukee Health Department successfully piloted data integration, including information “dashboards” in which multiple data streams could be compared in the same time sequence. Information was accumulated opportunistically, including ED chief complaints, laboratory reports, hospital ED diversions, ambulance runs, medical examiner reports of unusual or suspicious deaths, poison control and nursing hotline call volumes, and over-the-counter pharmacy sales (Foldy et al., 2004a). In addition to analyzing this information in the public health agency, results were also supplied to the various data providers. Such “situ-
ational awareness” or “fusion” systems are increasingly common in public health, but must be well designed to support decision making rather than drown the user with data of limited value (Endsley et al., 2003).
Integration of Public Health with Other Information Sources
A further step is to integrate public health information with other information that provides contextual information for interpreting public health events. For example, information from intelligence sources on potential or actual threats of bioterrorism could be used to enhance the interpretation of public health surveillance data or prompt active surveillance. A syndromic surveillance alert based on an unexpected increase in influenza-like symptoms might be seen as a stronger signal if intelligence information were available that a terrorist group was planning an attack. Active surveillance for cases would no doubt be instituted, and, if the intelligence could identify the biological agent, the surveillance could be focused on signs and symptoms associated with that agent.
At the national level, the Department of Homeland Security (DHS) has established the National Biosurveillance Integration Center (NBIC) to integrate biosurveillance information with intelligence information about threats to improve situational awareness and early warning of bioterrorism. As noted in Chapter 4, NBIC has been slow to achieve full operational capability and exemplifies the complexity of integrating surveillance data across agencies.
Most states and some localities have established fusion centers, which are in various stages of development. Most of them are headed by law enforcement agencies (GAO, 2007). State and local fusion centers work most closely with the Federal Bureau of Investigation and, to a lesser extent, DHS, which assign personnel to the centers. The original impetus for fusion centers was the threat of terrorism, but most of them also collect, analyze, and disseminate criminal information. State health departments are partners in a number of fusion centers, but these centers are not generally viewed as the locus of situational awareness for public health agencies.
NATIONAL BIOSURVEILLANCE STRATEGY
In October 2007, Homeland Security Presidential Directive 21 (HSPD-21) directed the Secretary of Health and Human Services to establish an “operational national epidemiologic surveillance system for human health” (The White House, 2007). According to HSPD-21, the national surveillance system should be built on existing federal, state, and local surveillance systems; provide incentives for public health agencies to implement local surveillance systems where they do not exist; be built using electronic health information
systems; enable two-way information flow between federal, state, and local government public health authorities and clinical health care providers; and integrate the data into a national biosurveillance common operating picture (Box 5-1).
HSPD-21 also calls for the Secretary of Health and Human Services to establish an Epidemiologic Surveillance Federal Advisory Committee, including representatives of state and local public health and private-sector health care, “to ensure that the federal government is meeting the
Biosurveillance: A Critical Component of Public Health and Medical Preparedness
Homeland Security Presidential Directive 21 Public Health and Medical Preparedness says, in part:
The United States must develop a nationwide, robust, and integrated biosurveillance capability, with connections to international disease surveillance systems, in order to provide early warning and ongoing characterization of disease outbreaks in near real-time. Surveillance must use multiple modalities and an in-depth architecture. We must enhance clinician awareness and participation and strengthen laboratory diagnostic capabilities and capacity in order to recognize potential threats as early as possible. Integration of biosurveillance elements and other data (including human health, animal health, agricultural, meteorological, environmental, intelligence, and other data) will provide a comprehensive picture of the health of communities and the associated threat environment for incorporation into the national “common operating picture.” A central element of biosurveillance must be an epidemiologic surveillance system to monitor human disease activity across populations. That system must be sufficiently enabled to identify specific disease incidence and prevalence in heterogeneous populations and environments and must possess sufficient flexibility to tailor analyses to new syndromes and emerging diseases. State and local government health officials, public and private sector health care institutions, and practicing clinicians must be involved in system design, and the overall system must be constructed with the principal objective of establishing or enhancing the capabilities of State and local government entities.
SOURCE: The White House (2007).
goal of enabling state and local government public health surveillance capabilities.” The National Biosurveillance Advisory Subcommittee (NBAS) was established in May 2008 as a subcommittee to the Advisory Committee to the Director of CDC to meet this mandate. CDC expects NBAS to “review, research, guide, and endorse the National Biosurveillance Strategy for Human Health on an annual basis” and “serve as an innovative engine for advancing nationwide biosurveillance capability” (Sosin, 2009).
CDC was assigned the lead in the interagency effort to develop the operational national surveillance system. CDC’s Biosurveillance Coordination Unit worked with CDC biosurveillance strategy management and advisory teams; a state, local, territorial, and tribal work group; the interagency HSPD-21 Federal Biosurveillance Work Group, and representatives of other organizations to develop a working draft, National Biosurveillance Strategy for Human Health 2008–2013 (The Strategy), which was released for public review and comment in December 2008 (CDC, 2008a).
The Strategy summarizes the well-known shortcomings of current surveillance systems in a short paragraph (CDC, 2008a, p. 10):
Notifiable disease reports are often accurate, but not timely. Syndromic surveillance systems can be timely, but may lack specificity needed to validate and interpret signals and cues. Communication between laboratorians and epidemiologists is sometimes poor and all too frequently still manual and paper-based. Cues to human health threats from animal, plant, and environmental sources are inconsistently identified and shared. In addition, professionals within the public health system may have pieces of information that are not being effectively integrated or shared to achieve accurate situation awareness for themselves and other responders.
The Strategy identifies the following six priority areas “to address critical gaps and opportunities.”
Electronic Health Information Exchange: strengthening and expanding upon multidirectional health information exchanges with health care and public health entities.
Electronic Laboratory Information Exchange: strengthening information exchanges between and among clinical and public health laboratories and between laboratories and public health programs for use in investigations. This is a part of electronic health information exchange but was broken out to underscore the importance of laboratories in biosurveillance.
Unstructured Data: leveraging digital information (e.g., text and image) that is not in a database format for biosurveillance for human health (e.g., mentions of disease names or symptoms in media reports)
Integrated Biosurveillance Information: generating actionable health intelligence by increasing access to information resources and synthesizing multiple streams of information into one coherent picture.
Global Disease Detection and Collaboration: ensuring the United States’ ability to contribute to and participate in global disease detection and response through increased global capacity and coordinated international action.
Biosurveillance Workforce of the Future: addressing the need for a workforce that is available, prepared, and collaborating to adapt to evolving threats and crises.
The six priority areas could be considered to be the Department of Health and Human Service’s (HHS’s) plans to enhance the current surveillance system (i.e., take advantage of new opportunities), but the priorities also include efforts to complete the basic infrastructure (i.e., fill gaps).
The committee finds The Strategy to be very compatible with its views of areas that should and could be enhanced to improve early detection of biological threats. For example, The Strategy acknowledges that biosurveillance includes early detection of environmental threats as well as early suspicion and confirmation of cases of human infection and disease. In other words, environmental monitoring systems such as BioWatch are to be considered an integral part of surveillance of public health threats and should be integrated with surveillance by public health agencies and the health care system. (This topic is addressed more fully in Chapter 6.)
The intent of The Strategy is laudable, but it needs to be accompanied by details of its execution and, importantly, its funding. Furthermore, investment in surveillance systems should be accompanied by investment in the evaluation of their effectiveness. The main barriers to a national system, even a federated system of systems, are organizational and financial rather than technical (not that the technical issues are easy to resolve). A concept of operations document is planned for completion in 2009, and it may provide more detail on federal, state, and local responsibilities (Sosin, 2009).
COSTS OF INFECTIOUS DISEASE SURVEILLANCE AND ENHANCEMENTS
Part of the charge to the committee was to “assess the costs and benefits of an enhanced national surveillance system that relies on U.S. hospitals and the U.S. public health system … and its effectiveness compared to that of the current BioWatch approach.” The committee engaged Industrial Economics, Incorporated (IEc), to collect and analyze information on the costs of the public health system as it currently exists and the costs of an
enhanced public health surveillance system. After its effort, IEc reported on available cost data and concluded that “the available data do not support a comprehensive cost analysis of either current or enhanced public health activities related to biosurveillance and outbreak response” (IEc, 2009; see Appendix C).
IEc identified several significant obstacles to performing a cost analysis of public health programs in general or public health surveillance programs specifically. The primary obstacle is the lack of financial transparency. Information on spending by functional category (e.g., surveillance) is not comparable across jurisdictions. Unlike hospitals and health systems, schools and school districts, and institutions of higher education, public health departments generally do not have (1) uniform classifications for revenues and expenditures, (2) electronic management information systems, (3) nationwide standardized financial analysis and reporting practices, or (4) professional associations of financial analysts and managers (Honoré et al., 2007). This lack of standardization is reinforced by the absence of a universally accepted agreement regarding what practices or activities constitute surveillance (despite general acceptance of the standard definition promulgated by CDC) and by the organizational variability of public health agencies in terms of program mix, division of labor between state and local agencies, and accounting systems (Sensenig, 2007).
In addition, most public health agencies order their budgets and financial reports by organizational unit (e.g., Communicable Disease Bureau, Laboratory Science Bureau, etc.) or object classification (e.g., personnel, supplies, etc.) rather than by function or mission (e.g., monitoring community health status, surveillance and identification of health threats, disease prevention and health promotion, personal health services, regulation of health and safety, etc.). Moreover, CDC (2009a) does not currently collect or track estimates of the costs at the federal, state, and local levels for the CDC disease surveillance programs relevant to biodefense and other public health emergencies.
Some past efforts to assess the costs of surveillance and other public health activities (Eilbert et al., 1996; Barry et al., 1998, 2000) tried using a set of 10 “essential public health services” (Box 5-2) as a basis for apportioning costs. Surveillance and detection of infectious disease threats is part of the second of these services: “Diagnose and investigate health problems and health hazards in the community.” While these studies concluded that public health expenditures can be measured reliably and consistently using the essential services framework, they noted a number of serious limitations in this approach, including differences in state and local public health agency organizational structures, variability in the interpretation of terminology, lack
The 10 Essential Public Health Services
SOURCE: Public Health Functions Steering Committee (1995).
of information on public health expenditures by non-public health agencies, and variability in the quality of local health department data.
In 2004, the Missouri Department of Health and Senior Services (MDHSS) categorized its expenditures using the essential services framework and compared them with performance measures. One of the limitations noted in that study was the difficulty of determining which expenditures should be assigned to each essential function. MDHSS had to use a cross-walk between its accounting categories and the essential services categories, which introduced an element of subjectivity and arbitrariness (Honoré and Schlechte, 2007).
In 2003, the IOM recommended that HHS conduct an annual assessment of the state of the nation’s public health infrastructure and capacity to provide the 10 essential public health services to every community and report on the assessment to Congress. “The assessment should include a thorough evaluation of federal, state, and local funding for the nation’s governmental public health infrastructure” (IOM, 2003, p. 7). This recommendation was not adopted.
CDC and six other public health organizations have been promoting standardization through the National Public Health Performance Standards Program (NPSPHP) since 2002.4 NPSPHP is organized around the 10 essential services of public health. Under essential service #2 (“Diagnose and investigate health problems and health hazards in the community”), NPSPHP (no date) has identified three primary types of activities at the local level:
Identification and Surveillance of Health Threats. This includes surveillance systems to monitor health problems and identify health threats, timely submission of reportable disease information, and adequate resources to support surveillance and investigation activities.
Investigation and Response to Public Health Threats and Emergencies. This includes written protocols for case finding, contact tracing, source identification, and containment; current epidemiological case investigation protocols; a designated emergency response coordinator; capacity of personnel to respond quickly in emergencies or disasters; and periodic evaluations of public health emergency response preparedness and performance.
Laboratory Support for Investigation of Health Threats. This includes ready access to laboratories for routine diagnostic and surveillance needs; ready access to laboratories able to support investigation of public health threats, hazards, and emergencies; licensed and/or credentialed laboratories; and maintenance of guidelines or protocols for handling laboratory samples.
To date, the NPSPHP has been used by public health agencies in more than 35 states for self-assessment. However, the program does not collect or assess data on expenditures in relation to performance, making it impossible to conduct cost-effectiveness analyses. For example, a performance measure might be the number of graduate-level epidemiologists or the development or expansion of an ELR system or the number of Category A agents the state laboratory can test for; but it is not possible to determine the cost-effectiveness of improvements in these measures without cost information.
Given the lack of nationally comparable expenditure data, IEc concentrated on collecting examples of cost data for enhanced surveillance and outbreak response activities or programs under way at the state or local level. The intent was to provide some insight into the potential orders of
The other organizations are the American Public Health Association, Association of State and Territorial Health Officials, National Association of County and City Health Officials, National Association of Local Boards of Health, National Network of Public Health Institutes, and Public Health Foundation. Information about the program is at http://www.cdc.gov/od/ocphp/nphpsp/index.htm.
magnitude or range of costs of developing and maintaining systems that improve biosurveillance and response.
IEc obtained one national-level cost estimate—the cost of implementing a national ELR infrastructure to enable quicker communication of results from public health and private laboratories to the appropriate public health authorities. Local, state, federal, and private-sector experts at a 2007 Association of Public Health Laboratories meeting sponsored by CDC estimated it would cost $25 million to plan the system, $25 million to build state public health laboratory information systems where needed,5 $600 million to implement the system, and $100 million per year to operate and maintain it (APHL, 2007).
IEc was most successful in identifying cost information for the larger IT-related efforts: ELR systems, syndromic surveillance systems, HIEs, and electronic disease surveillance and outbreak management software. However, these data were often incomplete. For example, in some cases the systems are too new to provide operations and maintenance costs. IEc was least successful in obtaining cost estimates for marginal expansion of laboratory testing and diagnostic capabilities; clinical decision support tools and point-of-care diagnostic tools; health alert networks; and fusion centers. In some instances, anticipated costs varied greatly between two different localities’ planned implementation of enhancements (e.g., regional health information exchanges in Tarrant County, Texas, and planned by Pennsylvania and Ohio), illustrating the difficulties in arriving at costs for nationwide implementation and operation. Appendix C provides details regarding the costs for specific examples of enhancements to surveillance in the public health and health care systems.
In this chapter, the committee considered several potential enhancements of public health and health systems for surveillance for the detection of bioterrorism and other significant infectious disease threats. The emergence of the novel influenza A (H1N1) outbreak in April 2009, as this report was being finished, highlighted many of the challenges of infectious disease surveillance as well as opportunities to apply enhanced approaches like those discussed in this chapter. Box 5-3 illustrates the experience of one state (Wisconsin) at the early stages of the outbreak. Other states experiences will vary, and these differences can be explored to learn more about strengths and weakness of a range of disease surveillance resources.
Pandemic Influenza A (H1N1): Implications for Enhanced Surveillance
On April 21, 2009, the Centers for Disease Control and Prevention (CDC, 2009c) confirmed that two children had been infected in late March by a pandemic influenza A (H1N1) strain. Within a week, CDC had disseminated a set of working case definitions and strategies for testing, isolation, and treatment of suspect cases, and it had received approval for use of a new real-time PCR test to confirm pandemic influenza A (H1N1) infection. Many suspected cases that were previously “untypable” could be confirmed as the novel H1N1 virus.
The surveillance challenges facing states and communities have illustrated many of the concerns highlighted in this report. But the rapidity of the response also reflects past funding and innovation directed specifically at the management of a widely anticipated influenza pandemic. It does not necessarily reflect the state of the nation’s capability to mount surveillance and response for other types of biological threats, including bioterrorism. Events in Wisconsin during the first few weeks of the outbreak serve as an early case study.
With the late-April information from CDC, the Wisconsin Division of Public Health and the Wisconsin State Laboratory of Hygiene implemented a new system of surveillance and proactively took steps to manage a possible outbreak of uncertain virulence. An around-the-clock incident command system was established. Instructions regarding the recognition and reporting of suspect cases, the submission of specimens for testing, isolation and other infection control measures, and treatment were disseminated to clinicians, hospitals, local health departments, and infection control professionals using email-lists, Wisconsin’s Health Alert Network, and the Wi-Trac website for health care situation awareness and alerting.
Existing influenza surveillance systems, including sentinel clinic networks for assessing visits for influenza-like illness (ILI) and death record reviews for pneumonia or influenza, shifted from weekly to daily reporting. The Wisconsin Health Information Exchange (WHIE) provided near-real-time information on Milwaukee emergency department visits, hospital admissions, and the proportions of ILI among both. The Wisconsin Electronic Disease Surveillance System (WEDSS) (in use by approximately one-third of local health jurisdictions) incorporated reporting and case management for pandemic H1N1, and state and local public health laboratories implemented electronic laboratory reporting of test results. Remaining local health jurisdictions were provided reading rights on WEDSS to view laboratory results and other epidemiologic information in real time on suspected and confirmed cases in their jurisdictions.
Wisconsin’s first three probable cases (untypable Influenza A) were reported on April 29. Many probable cases were also identified by a Milwaukee-area private research laboratory whose methods were validated by the state
Laboratory of Hygiene, and virtually all probable cases were later confirmed by additional analysis. The number of probable cases increased daily, including many among Milwaukee school children. National guidance then in effect recommended consideration of partial or total school system closure to reduce community transmission of the virus when multiple schools experienced cases—a high-consequence action for schools, families, and employers. Analysis of multiple surveillance systems showed that while emergency department visits for ILI were above normal and rising (with surges noted after major press announcements), no major increase had occurred in hospitalization for respiratory illness or deaths from pneumonia or influenza. This finding was consistent with information from CDC and other national authorities and with public health workers’ interviews of patients with probable or suspect illness. Taken together, this information showed the virulence of the new virus might not warrant such extraordinary social distancing measures as school-system closings, and Milwaukee schools remained open.
By early May, three Wisconsin laboratories (two public and one private) were validated to confirm cases, speeding results and reducing the load on CDC laboratories. The number of laboratory-confirmed cases increased rapidly, making Wisconsin’s case count the highest in the nation. However, Wisconsin also had significantly lower case-hospitalization and case-mortality rates than the national average, suggesting more complete ascertainment of mild illness. ILI rates and other measures of disease incidence remained similar to other states. The state’s surveillance and laboratory capacity helped improve understanding the virulence of the virus. At the beginning of June 2009, however, Wisconsin’s ILI rates began to rise, peaking the week of June 20. Comparison of syndromic, hospitalization, and death surveillance between Wisconsin and other states appeared to confirm that Wisconsin experienced exceptionally high H1N1 activity during this period. Three weeks later, at the conclusion of the public school year, the Wisconsin rates declined markedly.
The emergence of pandemic influenza A (H1N1) in Wisconsin illustrates several conclusions by the committee. Early warning from international and national authorities prompted rapid establishment and refinement of state and local surveillance systems. Clinicians were alerted electronically to initiate case finding and informed about how to recognize cases. LRN laboratories provided standardized and reliable testing capability within days of the first report of the virus. Electronic laboratory reporting into a shared electronic case-management environment facilitated case investigation and management; managing the rush of thousands of cases on paper would have been impossible. Near-real-time monitoring of ILI in primary care, emergency rooms, and hospitals, including information from a regional Health Information Exchange, helped gauge the spread of the outbreak, and more importantly provided valuable, if preliminary, information regarding the virulence of a novel pathogen. Rudimentary systems for comparing Wisconsin’s syndromic information with other areas (e.g., BioSense and ILINet) proved helpful in distinguishing spurious from real differences in disease incidence.
The committee tried to assess the benefits and costs of these enhancements, based on examples around the country. In all but one case, however, it was unable to determine the costs of adopting them nationally or to quantify their benefits with any rigor. Although the committee could not recommend enhancements based on a formal benefit-cost analysis, a consensus exists that certain steps should be taken because of the urgent need to fill gaps and strengthen the infrastructure for surveillance of infectious disease. These steps are the subject of the four recommendations that follow.
Enhancing Methods for Surveillance
Infectious disease surveillance is a key public health practice, and it is the responsibility of state and local government. One federal role—aggregation and analysis of surveillance data—is achieved primarily through mutual agreement and cooperation with the state and local public health agencies and through federal funding incentives. Many novel and promising surveillance techniques and programs have been developed rapidly at the local, state, and federal level in recent years, spurred in part by funding for bioterrorism and public health emergency preparedness since 2001. However, as noted in Chapter 4, there has been limited evaluation of surveillance approaches, and national standards for surveillance data and interoperability between systems are incompletely developed. Surveillance capacities are unevenly distributed among states and localities, and limited year-to-year funding has made long-term planning and recruitment of qualified personnel difficult. This complex series of problems must be addressed by a dedicated, strategic, integrated, and adequately funded program. The CDC definition of surveillance is explicit in noting that surveillance is an activity intended to inform specific prevention or control programs; thus, it is essential that efforts to reform surveillance be led by those responsible for the programs the surveillance systems are intended to serve.
RECOMMENDATION 8: HHS should support the development, testing, and evaluation of improved methods for surveillance for infectious disease outbreaks. HHS, through CDC, should take a stronger leadership role in evaluating and enhancing efforts for automating both traditional provider and laboratory reporting and syndromic and other approaches to surveillance, including the development of standards, coordination of state and local initiatives, and integration of federal programs with state and local activities. HHS should assign this leadership role to those responsible for the prevention and control programs these surveillance systems are intended to serve, and it should rigorously evaluate these surveillance efforts.
HHS should coordinate a strategic, goal-oriented, integrated program of intra- and extramural research and development, pilot-testing, and operational evaluation of improved public health surveillance methods. This program should be implemented in partnership with DHS, the Agency for Healthcare Research and Quality, the National Institutes of Health, and other federal, state, and local agencies that have a role in monitoring threats to human health. Program planning should identify the need for additional evidence regarding effectiveness, identify gaps in the geographic deployment and quality of public health surveillance, identify and evaluate promising methods and technologies, and integrate and harmonize approaches across the many surveillance programs used by CDC and others in the public health community.
The focus of this effort should be to improve
notifiable condition reporting by clinicians and laboratories, including automated, electronic reporting methods;
syndromic and other automated health information monitoring;
public health reference laboratory services;
situational awareness based on integration of multiple surveillance and other information streams, including intelligence on terrorism threats;
efficiency, effectiveness, and agility of information management at the state and local levels;
horizontal and vertical information sharing across jurisdictions;
surveillance support for rapid decision making and response; and
methods to compare the utility and cost-effectiveness of surveillance methods.
Effective methods resulting from this effort should be deployed across state and local jurisdictions through a combination of federal funding and local investments.
Enhancing Mechanisms for Information Sharing
The threat of bioterrorism is just one of several reasons for the United States to improve means for sharing disease surveillance information. Another reason is to enhance detection of naturally occurring diseases, including emerging infectious diseases such as SARS and possibly avian influenza, and the re-emergence of traditional threats such as tuberculosis. At the same time, scientific and technological advances make such information sharing more effective and useful. Some advances make it quicker and easier to collect and transmit data and others make it easier to analyze them for
patterns and anomalies. Others, such as genetic sequencing and polymerase chain reaction (PCR) allow new tests to be developed rapidly. The sharing of phenotypic and genotypic fingerprinting results enables molecular epidemiology—comparing and linking strains to connect widely separated cases—in ways not possible before.
In addition to quicker and more effective outbreak detection, information sharing can be expected to improve situational awareness and response capabilities. It should enable a jurisdiction to better track an outbreak, focus its response activities, and determine the effectiveness of response efforts. Better information sharing should also enable affected jurisdictions to see the bigger picture and coordinate their actions for their mutual benefit.
RECOMMENDATION 9: DHS and HHS should enhance the efforts to develop a mechanism for providing a national situational awareness of biological threats and significant disease outbreaks, to better inform rapid decision making and response through cross-jurisdictional data sharing and analysis of data. To this end, DHS and HHS should facilitate the development of an interoperable, secure, bidirectional, nationwide information-sharing infrastructure and ensure that local and state health officials have ready access to the system.
“National” in this context does not mean federal or centralized, although an important feature of such an information-sharing infrastructure is the ability to analyze data aggregated across multiple jurisdictions. This can sometimes make it possible, for example, to recognize a disease outbreak when the number of cases in any one jurisdiction is too small to generate concern. Or, if the exposure is in a transportation hub such as an airport or train station, it may make it possible to recognize that geographically scattered cases are the result of exposure in a single incident.
A nationwide information-sharing infrastructure that is decentralized and a cooperative effort by multiple jurisdictions will also require a bidirectional information flow. Higher levels of data aggregation and analysis may be necessary to detect an outbreak in its earliest stages, but an effective response depends on well-informed actions by and cooperation among the state and local public health agencies where people are sick. A decentralized cooperative infrastructure should also enhance data sharing because the data providers would find it to be useful in carrying out their missions.
Accordingly, CDC and the states should focus their efforts on developing and deploying methods for intra- and interstate, cross-jurisdictional integration, sharing, analysis, and display of public health surveillance information. This system should function to monitor the presence of naturally occurring or deliberately introduced infectious agents, discover emerging or yet undefined threats to public health, and integrate data from the
BioWatch program and the public health and health care sectors. Priority should be given to results of tests for laboratory-reportable diseases, LRN biothreats, and BioWatch samples.
Improving Technologies for Clinical Case Recognition and Reporting
If environmental surveillance systems such as BioWatch do not detect a bioaerosol attack (or if a biological attack takes another form), early detection may depend upon an astute clinician’s recognition of history, symptoms, or signs that are consistent with exposure to a bioterrorism agent. However, most clinicians will be seeing what amounts to a rare disease for the first time and may not recognize it, especially in the early stages when symptoms are still consistent with other, much more common (and less dangerous) diseases. The availability, and use, of valid and reliable automated assistance may aid in recognizing rarely encountered diseases that pose a large threat to public health.
RECOMMENDATION 10: HHS should promote the development, testing, and evaluation of technologies that strengthen the accuracy, timeliness, consistency, and completeness of clinical diagnosis of infectious diseases of public health importance, and that facilitate timely reporting of these diagnoses to public health authorities.
Promising strategies, described earlier in this chapter, include
computer-assisted emergency department triage;
bedside decision support tools with automated case reporting;
rapid point-of-care laboratory testing; and
multiplex assays that could be used for epidemiologic investigations, surveillance, or clinical applications, especially in the context of suspected outbreaks or concerning signals from syndromic systems or BioWatch.
HHS, under the auspices of the Office of Assistant Secretary for Preparedness and Response, should identify and support the deployment of web-based clinical decision support tools to help triage nurses and acute care clinicians identify, report, and manage suspicious cases of diseases of public health concern. These systems should be clinically useful for more routine diseases and, therefore, be employed on an ongoing basis. They should be bidirectional so they can quickly and reliably communicate public health alerts to front-line providers as well as enable busy clinicians to easily report suspect or confirmed cases to the local health department. The systems should be easily modifiable so they can be updated to reflect evolving knowledge of a bioterrorist attack or infectious disease outbreak
and provide event-specific recommendations to providers on recognizing and managing outbreak-associated cases.
Enhancing Public Health Surveillance Capacity
Before the terrorism events of 2001, the governmental public health infrastructure was unevenly, but generally inadequately, staffed and funded to accomplish its traditional tasks, including surveillance (IOM, 2003). During the 1990s, some states and localities began to use information technologies to develop new techniques to counter biological threats, such as syndromic surveillance, monitoring of health-related behaviors such as over-the-counter pharmacy sales and absenteeism, and electronic laboratory information management and reporting systems. CDC began to encourage and subsidize electronic reporting of surveillance data. Congress began to fund state and local public health preparedness in 1999 through CDC’s Public Health Emergency Preparedness Program (PHEP). The events of 2001, however, made the general weakness in the capacity of the public health system painfully evident to policy makers (IOM, 2003).
At that time, there were plans under the Public Health Improvements Act of 2000 to provide grants to improve basic state and local public health infrastructure and under the Public Health Threats and Emergency Act of 2000 to provide grants for state and local bioterrorism preparedness. After 2001, the two efforts were combined and focused on preparedness for bioterrorism and other health emergencies (IOM, 2003). The Health Resources and Services Administration in HHS began a Hospital Preparedness Program (HPP) in 2002. Federal funding for public health and hospital preparedness rose from less than $50 million in fiscal year (FY) 2001 to more than $1 billion in FY 2002.
The substantial infusion of federal funds for public health and medical preparedness has improved the general level of preparedness, but the categorical nature of the funding and the uncertainty of its continuation has generally discouraged public health agencies from integrating these new capacities with traditional programs. At the same time, the basic public health infrastructure, such as staffing and electronic information management and analysis systems, did not experience the same level of investment in most places. This uneven level of basic organizational capacity reduces the national level of preparedness to detect and, especially, to respond to and manage the consequences of a major health emergency. The need for a minimum standard of local public health services to detect public health emergencies, as demonstrated by the rapid global spread of illnesses like SARS and the novel influenza A (H1N1), is also now codified in the World Health Organization’s International Health Regulations (WHO, 2008). Thus there is an enforceable treaty obligation to improve the capability of local health services across the United States.
Meanwhile, the current economic recession is causing significant cutbacks in state and local government programs, including basic public health programs. PHEP and HPP funding has also declined for several years, even though much of the federal investment in capacity building requires continued financial support to be effective. Even with continuing federal investments, achieving the desired minimum state and local capacities for surveillance for bioterrorism and infectious disease threats will be challenging. The task is substantial and will require effective coordination and collaboration across the diversity of configurations of state and local public health systems and across complexities of state and federal relationships and interagency action at the federal level.
RECOMMENDATION 11: HHS and DHS should give high priority to building and sustaining sufficient public health workforce strength and competencies, along with associated laboratory and information management capacities, needed by all states and communities to detect a bioterrorism attack or other public health emergency. They should pursue a nationally consistent minimum level of disease surveillance and communication sufficient to provide early warning and tracking of bioagent attacks and outbreaks of natural disease. Key state and local capacities should include the following:
Adequate amounts and types of staff expertise, including infectious diseases, veterinary health, laboratory science, environmental health, applied epidemiology and biostatistics, and health informatics;
Adequate public health reference laboratory capacity;
Electronic laboratory reporting systems to ensure timely and complete transmission of notifiable disease reports from commercial and hospital-based laboratories to public health;
Universal access to public health reference laboratory services for detecting and confirming biothreats and other emerging infectious diseases and performing molecular typing to link cases in outbreaks;
Robust surveillance and outbreak management information systems;
Electronic death registration systems;
Health alert networks that connect public health departments with all health care facilities and providers in their jurisdictions; and
Integration of public health needs and systems into emerging health information exchanges.
For maximum benefit this investment should be directed at developing and maintaining staff expertise, informed decision making, and response capabilities that would serve in both natural and bioterrorism-related disease outbreaks.
APHL (Association of Public Health Laboratories). 2007. Framework for electronic laboratory reporting: Recommendations to policymakers. A report of the Electronic Laboratory Reporting Consensus-Building Meeting held October 30, 2007, in Arlington, VA. http://www.aphl.org/policy/Documents/Electronic_Lab_Reporting_2007.pdf (accessed March 26, 2009).
Augustine, J.J., A.L. Kellermann, and J.P. Koplan. 2004. America’s emergency care system and severe acute respiratory syndrome: Are we ready? Annals of Emergency Medicine 43(1):23–26.
Babin, S., S. Magruder, S. Hakre, J. Coverly, and J.S. Lombardo. 2007. Understanding the data: Health indicators in disease surveillance. In: Disease surveillance: A public health informatics approach, edited by J.S. Lombardo and D.L. Buckeridge. Hoboken, NJ: Wiley-Interscience. Pp. 43–90.
Baker, E.L., Jr., and J.P. Koplan. 2002. Strengthening the nation’s public health infrastructure: Historic challenge, unprecedented opportunity. Health Affairs 21(6):15–27. http://content.healthaffairs.org/cgi/reprint/21/6/15 (accessed March 25, 2009).
Baker, E.L., M.A. Potter, D.L. Jones, S.L. Mercer, J.P. Cioffi, L.W. Green, P.K. Halverson, M.Y. Lichtveld, and D.W. Fleming. 2005. The public health infrastructure and our nation’s health. Annual Review of Public Health 26:303–318.
Barry, M.A., L. Centra, E. Pratt, Jr., C.K. Brown, and L. Giordano. 1998. Where do the dollars go? Measuring local public health expenditures. Washington, DC: Public Health Foundation. http://www.phf.org/reports/Expend1/loclexpn.pdf (accessed February 23, 2009).
Barry, M.A., R. Bialek, M.R. Fraser, C.K. Brown, and S. Vogel. 2000. Statewide public health expenditures: A pilot study in Maryland. Washington, DC: Public Health Foundation. http://www.phf.org/reports/Expend1/MDmain.pdf (accessed February 23, 2009).
Barthell, E.N., S.L. Foldy, K.R. Pemble, C.W. Felton, P.J. Greischar, R.G. Pirrallo, and W.J. Bazan. 2003. Assuring community emergency care capacity with collaborative Internet tools: The Milwaukee experience. Journal of Public Health Management and Practice 9(1):35–42.
Bullard, M. 2008. Emergency department potential for biosurveillance and BioWatch. Presentation to the Committee on Effectiveness of National Biosurveillance Systems: BioWatch and the Public Health System, Meeting 3, November 3–5, Washington, DC.
Cass, D. 2005. Once upon a time in the emergency department: A cautionary tale. Annals of Emergency Medicine 46(6):541–543.
CDC (Centers for Disease Control and Prevention). 2001. Public health’s infrastructure: A status report. Atlanta, GA: CDC. http://www.uic.edu/sph/prepare/courses/ph410/resources/phinfrastructure.pdf (accessed February 12, 2009).
CDC. 2006. Surveillance for foodborne-disease outbreaks—United States, 1998–2002. Morbidity and Mortality Weekly Report 55(SS-10):1–34. http://www.cdc.gov/mmwr/PDF/ss/ss5510.pdf (accessed August 7, 2009).
CDC. 2008a. National biosurveillance strategy for human health, 2008–2013. Working Paper Version 1.0. December 15. http://sites.google.com/site/nbshh10/graphics-and-documents/NationalBiosurveillanceStrategyv10_121508-1650.pdf?attredirects=0 (accessed March 3, 2009).
CDC. 2008b. Potential effects of electronic laboratory reporting on improving timeliness of infectious disease notification—Florida, 2002–2006. Morbidity and Mortality Weekly Report 57(49):1325–1328. http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5749a2.htm?s_cid=mm5749a2_e (accessed December 22, 2008).
CDC. 2009a. IOM review. Document submitted to the Committee on Effectiveness of National Biosurveillance Systems: BioWatch and the Public Health System, January 13.
CDC. 2009b. Multistate outbreak of Salmonella infections, associated with peanut butter and peanut butter-containing products—United States, 2008–2009. Morbidity and Mortality Weekly Report 58(4):85–90. http://www.cdc.gov/mmwr/PDF/wk/mm5804.pdf (accessed August 7, 2009).
CDC. 2009c. Swine influenza A (H1N1) infection in two children—Southern California, March–April 2009. Morbidity and Mortality Weekly Report 58(5):400–402. http://www.cdc.gov/mmwr/PDF/wk/mm5815.pdf (accessed August 7, 2009).
CSTE (Council of State and Territorial Epidemiologists). 2007. CSTE official list of Nationally Notifiable Conditions. CSTE Position Statement 07-EC-02. http://www.cste.org/PS/2007ps/2007psfinal/EC/07-EC-02.pdf (accessed February 27, 2009).
CSTE. 2008. Criteria for inclusion of conditions on CSTE Nationally Notifiable Condition list and for categorization as immediately or routinely notifiable. CSTE Position Statement 08-EC-02. http://www.cste.org/PS/2008/2008psfinal/08-EC-02.pdf (accessed February 27, 2009).
Dong S.L., M.J. Bullard, D.P. Meurer, S. Blitz, E. Akhmetshin, A. Ohinmaa, B.R. Holroyd, and B.H. Rowe. 2007. Predictive validity of a computerized emergency triage tool. Academic Emergency Medicine 14(1):16–21.
Effler, P., M. Ching-Lee, A. Bogard, M.C. Ieong, T. Nekomoto, and D. Jernigan. 1999. Statewide system of electronic notifiable disease reporting from clinical laboratories: Comparing automated reporting with conventional methods. Journal of the American Medical Association 282(19):1845–1850. http://jama.ama-assn.org/cgi/reprint/282/19/1845 (accessed March 3, 2009).
eHealth Initiative. 2008. Fifth annual survey of health information exchange at the state and local levels: Overview of 2008 findings. Washington, DC: eHealth initiative. http://www.ehealthinitiative.org/HIESurvey (accessed March 3, 2009).
Eilbert, K.W., M. Barry, R. Bialek, and M. Garufi. 1996. Measuring expenditures for essential public health services. Washington, DC: Public Health Foundation. http://www.phf.org/reports/Expend1/Measuring_Expenditures_for_Essential_Public_Health_Services.pdf (accessed February 23, 2009).
Endsley, M.R., B. Bolté, and D.G. Jones. 2003. Designing for situation analysis: An approach to user-centered design. Boca Raton, FL: Taylor & Francis Group.
Fallon, K.K., and D. Boone. 2004. Death certificate surveillance—New Hampshire. Morbidity and Mortality Weekly Report 53(Suppl.):236. http://www.cdc.gov/mmwr/preview/mmwrhtml/su5301a48.htm (accessed February 28, 2009).
FDA (Food and Drug Administration). 2008. FDA clears first test designed to detect and identify 12 respiratory viruses from single sample. http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/2008/ucm116829.htm (accessed July 8, 2009).
Fine, A., and M. Layton. 2001. Lessons from the West Nile viral encephalitis outbreak in New York City, 1999: Implications for bioterrorism preparedness. Clinical Infectious Diseases 32(2):277–282.
Foldy, S. 2004. Linking better surveillance to better outcomes. Morbidity and Mortality Weekly Report 53(Suppl.):12–17. http://www.cdc.gov/mmwr/preview/mmwrhtml/su5301a4.htm (accessed March 30, 2009).
Foldy, S., P. Biedrzycki, E. Barthell, N. Haney-Healy, B.K. Baker, D.S. Howe, D. Gieryn, and K.R. Pemble. 2004a. Syndromic surveillance using regional emergency medicine Internet. Annals of Emergency Medicine 44(3):242–246.
Foldy, S.L., P.A. Biedrzycki, B.K. Baker, G.R. Swain, D.S. Howe, D. Gieryn, E.N. Barthell, and K.R. Pemble. 2004b. The public health dashboard: A surveillance model for bioterrorism preparedness. Journal of Public Health Management and Practice 10(3):234–240.
Foldy, S., E.N. Barthell, J.C. Silva, P. Biedrzycki, D.S. Howe, M. Erme, B. Keaton, C.L. Hamilton, L.K. Brewer, G. Miller, R. Bernstein, E. Eby, K. Pemble, and C. Fenton. 2004c. SARS surveillance project: Internet-enabled multi-region syndromic surveillance for rapidly emerging disease. Morbidity and Mortality Weekly Report 53(Suppl.):215–220.
Foldy, S., K. Pemble, R.T. Heffernan, E.N. Barthell, and J. Beix. 2008. Whoosh!: Building and controlling Wisconsin HIE’s clinical data stream for local and state health departments. Public Health Information Network annual meeting, Aug. 28. Atlanta, GA: Centers for Disease Control and Prevention. http://cdc.confex.com/recording/cdc/phin2008/pdf/free/4db77adf5df9fff0d3caf5cafe28f496/paper17528_1.pdf (accessed July 30, 2009).
GAO (Government Accountability Office). 2007. Homeland security: Federal efforts are helping to alleviate some challenges encountered by state and local information fusion centers. GAO-08-35. http://www.gao.gov/new.items/d0835.pdf (accessed March 30, 2009).
Hanrahan, L.P., S. Foldy, E.N. Barthell, and S. Wood. 2006. Medical informatics in population health: Building Wisconsin’s strategic framework for health information technology. Wisconsin Medical Journal 105(1):16–22.
Honoré, P.A., and T. Schlechte. 2007. State public health agency expenditures: Categorizing and comparing to performance levels. Journal of Public Health Management and Practice 13(2):156–162. http://journals.lww.com/jphmp/Fulltext/2007/03000/State_Public_Health_Agency_Expenditures_.11.aspx (accessed August 7, 2009).
Honoré, P.A., R.L. Clarke, D.M. Mead, and S.M. Menditto. 2007. Creating financial transparency in public health: Examining best practices of system partners. Journal of Public Health Management and Practice 13(2):121–129. http://journals.lww.com/jphmp/Fulltext/2007/03000/Creating_Financial_Transparency_in_Public_Health_.7.aspx (accessed August 7, 2009).
IEc (Industrial Economics, Incorporated). 2009. Summary of research into the costs of enhanced public health surveillance systems. Commissioned paper. Committee on Effectiveness of National Biosurveillance Systems: BioWatch and the Public Health System, Washington, DC.
IOM (Institute of Medicine). 2000. To err is human: Building a safer health system. Washington, DC: National Academy Press.
IOM. 2001. Crossing the quality chasm: A new health system for the 21st century. Washington, DC: National Academy Press.
IOM. 2003. The future of the public’s health in the 21st century. Washington, DC: The National Academies Press.
IOM. 2007. Hospital-based emergency care at the breaking point. Washington, DC: The National Academies Press.
Jajosky, R.A., and S.L. Groseclose. 2004. Evaluation of reporting timeliness of public health surveillance systems for infectious diseases. BMC Public Health 4:29.
Jha, A.K., C.M. DesRoches, E.G. Campbell, K. Donelan, S.R. Rao, T.G. Ferris, A. Shields, S. Rosenbaum, and D. Blumenthal. 2009. Use of electronic health records in U.S. hospitals. New England Journal of Medicine 360(16):1628–1638.
Klompas, M., R. Lazarus, R. Platt, X. Hou, F. Campion, B. Kruskal, G. Haney, J. Daniel, A. DeMaria, and S.J.N. McNabb. 2008. Automated detection and reporting of notifiable diseases using electronic medical records versus passive surveillance—Massachusetts, June 2006–July 2007. Morbidity and Mortality Weekly Report 57(14):373–376.
Lazarus, R., M. Klompas, F.X. Campion, S. McNabb, X. Hou, J. Daniel, G. Haney, A. DeMaria, L. Lenart, and R. Platt. 2009. Electronic support for public health: Validated case finding and reporting for notifiable diseases using electronic medical data. Journal of the American Medical Informatics Association 16(1):18–24. http://www.jamia.org/cgi/reprint/16/1/18 (accessed February 23, 2009).
Lenert, L. 2007. The evolution of public health surveillance to a GRID model. Presentation to the 2007 International Society for Disease Surveillance Sixth Annual Conference. http://www.syndromic.org/conference/2007/powerpoint/5%20CDC.Linert.ppt (accessed July 27, 2009).
Lurie, N., J. Wasserman, and C.D. Nelson. 2006. Public health preparedness: Evolution or revolution? Health Affairs 25(4):935–945. http://content.healthaffairs.org/cgi/reprint/25/4/935 (accessed March 25, 2009).
Magnuson, J.A. 2008 (unpublished). 2007 National electronic laboratory reporting snapshot survey: Summary of results. Portland, OR: Oregon State Public Health, Acute and Communicable Disease Prevention.
Mahony, J.B. 2008. Detection of respiratory viruses by molecular methods. Clinical Microbiology Reviews 21(4):716–747.
NPHPSP (National Public Health Performance Standards Program). No date. Local Public Health System Performance Assessment – Model Standards. Version 2. http://www.cdc.gov/od/ocphp/nphpsp/documents/FINAL%20Local%20MS.pdf (accessed July 27, 2009).
ONC (Office of the National Coordinator for Health Information Technology). 2008. Federal health IT strategic plan: 2008–2012.Washington, DC: Department of Health and Human Services. http://healthit.hhs.gov/portal/server.pt/gateway/PTARGS_0_10731_848083_0_0_18/HITStrategicPlan508.pdf (accessed July 27, 2009).
Overhage, J.M., S. Grannis, and C.J. McDonald. 2008. A comparison of the completeness and timeliness of automated electronic laboratory reporting and spontaneous reporting of notifiable conditions. American Journal of Public Health 98 (2):344–350.
Panackal, A.A., N.M. M’ikanatha, F.-C. Tsui, J. McMahon, M.M. Wagner, B.W. Dixon, J. Zubieta, M. Phelan, S. Mirza, J. Morgan, D. Jernigan, A.W. Pasculle, J.T. Rankin, Jr., R.A. Hajjeh, and L.H. Harrison. 2002. Automatic electronic laboratory-based reporting of notifiable infectious diseases at a large health system. Emerging Infectious Diseases 8(7):685–691. http://www.cdc.gov/ncidod/EID/vol8no7/pdf/V8n7_final.pdf (accessed March 3, 2009).
Papier, A. 2008. Bioterrorism: The challenge of early diagnosis. Presentation to the Committee on Effectiveness of National Biosurveillance Systems: BioWatch and the Public Health System, Meeting 3, November 3–5, Washington, DC.
Peterson, A., and D. Terashita. 2008. The unusual death surveillance system: Real time monitoring of deaths using coroner’s data, 2003–2006. In: 2007 Special Studies Report, Acute Communicable Disease Control Program, Los Angeles County Department of Public Health. Pp. 67–72. http://www.lapublichealth.org/acd/reports/annual/2007%20Special%20Studies%20Report%20for%20Web%20Posting.pdf (accessed August 7, 2009).
Proctor, M.E., K.A. Blair, and J.P. Davis. 1998. Surveillance data for waterborne illness detection: An assessment following a massive waterborne outbreak of Cryptosporidium infection. Epidemiology and Infection 120(1):43–54.
Public Health Functions Steering Committee. 1995. Public health in America: The essential public health services. Washington, DC: Public Health Functions Steering Committee, U.S. Public Health Service. http://www.health.gov/phfunctions/public.htm (accessed February 27, 2009).
Reed, K.D., J.W. Melski, M.B. Graham, R.L. Regnery, M.K. Sotir, M.V. Wegner, J.J. Kazmierczak, E.J. Stratman, Y. Li, J.A. Fairley, G.R. Swain, V. Olson, E.K. Sargent, S.C. Kehl, M.A. Frace, R. Kline, S.L. Foldy, J.P. Davis, and I.K. Damon. 2004. The initial detection of human monkeypox in the western hemisphere: Association with infected prairie dogs. New England Journal of Medicine 350(4):342–350.
Riegle, R. 2009. The future of fusion centers: Potential promise and dangers. Testimony before the Committee on Homeland Security, Subcommittee on Intelligence, Information Sharing, and Terrorism Risk Assessment, April 1. http://www.dhs.gov/ynews/testimony/testimony_1238597287040.shtm (accessed July 30, 2009).
Rolka, H., J.C. O’Connor, and D. Walker. 2008. Public health information fusion for situation awareness. In: Lecture Notes in Computer Science: Biosurveillance and Biosecurity, edited by D. Zeng, H. Chen, H. Rolka, and W.B. Lober. Raleigh, NC: Springer. Pp. 1–9.
Sensenig, A.L. 2007. Refining estimates of public health spending as measured in national health expenditures accounts: The United States experience. Journal of Public Health Management and Practice 13(2):103–114. http://journals.lww.com/jphmp/Fulltext/2007/03000/Refining_Estimates_of_Public_Health_Spending_as.5.aspx (accessed August 7, 2009).
Sosin, D. 2009. Enhancing nationwide biosurveillance for human health. Presentation at the Integrated Surveillance Seminar, CDC, February 11. http://www2a.cdc.gov/ncphi/disss/seminars/docs/2-11-09DanSosin.ppt (accessed March 3, 2009).
TFAH (Trust for America’s Health). 2008. Ready or not? Protecting the public’s health from diseases, disasters, and bioterrorism. Washington, DC: TFAH. http://healthyamericans.org/assets/files/bioterror-report-2008.pdf (accessed July 17, 2009).
The White House. 2007. Public health and medical preparedness. Homeland Security Presidential Directive (HSPD) 21. Washington, DC. http://www.dhs.gov/xabout/laws/gc_1219263961449.shtm#1 (accessed March 3, 2009).
WHO (World Health Organization). 2008. International health regulations (2005), 2nd ed. Geneva, Switzerland: WHO.
Yasnoff, W.A., P.W. O’Carroll, D. Koo, R.W. Linkins, and E.M. Kilbourne. 2000. Public health informatics: Improving and transforming public health in the information age. Journal of Public Health Management and Practice 6(6):67–75.