Missouri Hospitals Launch Syndromic Surveillance System to Track Early Signs of Bioterrorism or Disease Outbreaks
From 2003 to 2005, the Missouri Department of Health and Senior Services (DHSS) designed and implemented a new "syndromic" surveillance system.
It allows public health officials to use hospital electronic medical records to detect unusual clusters of disease symptoms that might signal an outbreak related to bioterrorism agents, such as anthrax, plague and smallpox, or more common conditions, such as influenza or sexually transmitted diseases.
- In September 2005, 41 of Missouri's 126 hospitals were participating in the new bioterrorism surveillance system, reporting their disease symptom data electronically to state public health officials daily. By June 2006, 82 of the state's hospitals were participating. Psychiatric and rehabilitation hospitals; small, rural hospitals; and those without emergency rooms are exempt from the legal requirement to participate in the system.
The Robert Wood Johnson Foundation (RWJF) supported this project with an unsolicited grant of $343,788. The project also received funding under two programs of the Centers for Disease Control and Prevention (CDC): the Emergency Preparedness & Response, Bioterrorism program and the National Electronic Disease Surveillance System (NEDSS).
A week after the terrorist attacks of September 11th 2001, letters containing anthrax bacteria arrived at the offices of two U.S. senators and members of the news media. These events alerted public health officials to the need for improved surveillance systems that could provide an early warning of a bioterrorism attack.
Such an attack could include the intentional release of viruses, bacteria or other agents, including anthrax, smallpox or plague.
Missouri and other states responded by establishing a "syndromic" surveillance system. Unlike traditional disease surveillance, which relies on time-consuming laboratory tests to confirm a final diagnosis, syndromic surveillance is a computer-based system that typically monitors data on patients' initial symptoms.
Unusual patterns or clusters of common symptoms, such as headache, vomiting or muscle pain, can provide an early warning of a possible bioterrorism attack.
The original Missouri syndromic surveillance system required hospitals to collect data on patients admitted with seven complaints that could potentially be associated with bioterrorism-related diseases:
- Gastrointestinal illness.
- Hemorrhagic disease.
- Respiratory illness.
- Neurologic illness.
- Rash illness.
- Fever illness of unknown origin.
- Chemical exposure of sudden onset.
Hospital personnel manually tabulated the data daily and called or faxed it in to the local or district health agency three times a week. The data were then entered into a state database for analysis.
Many hospitals refused to participate because they said the system was too labor-intensive. In response, public health officials met with hospitals throughout the state and agreed to develop a less cumbersome surveillance system that used existing electronic hospital data.
RWJF seeks to strengthen America's public health system to make it better prepared to promote health and protect all Americans from a wide range of threats from bioterrorism to emerging infectious diseases to health problems such as obesity, tobacco use and asthma.
One of RWJF's strategies in this area is to spur participation among state and local public health agencies in exchanging health information through collaborations that allow the rapid and timely sharing of information between public health and health care systems, between public health systems and local communities, and among public health agencies within and across jurisdictions.
This grant falls under that strategy.
RWJF has made a number of grants specifically addressing the issue of bioterrorism since the fall of 2001:
- Educating the Public About Improving the Public Health System to Deal With Bioterrorism. See Grant Results on ID# 045388.
- Health Experts Analyzing the Effects of Increased Federal Funding on State Bioterrorism Preparedness. See Grant Results on ID# 045458.
- A Survey Assessing Public Attitudes, Knowledge and Misinformation About Bioterrorism. See Grant Results on ID# 044111.
- Three Reports Identifying Weaknesses in the Public Health System. See Grant Results on ID# 044873.
This is the only RWJF funding directly intended to support syndromic surveillance for the purposes of early detection of outbreaks.
From 2003 to 2005, staff of the Center for Health Information Management and Evaluation (CHIME) of the Missouri Department of Health and Senior Services (DHSS) created a new "syndromic" surveillance system that uses information from hospital electronic medical records to detect possible outbreaks of bioterrorism related diseases, like anthrax.
Since bioterrorism-related diseases are rare, CHIME designed the surveillance system also to be useful in tracking and more quickly responding to other, more common conditions, such as influenza and sexually transmitted diseases.
According to former project director, Garland Land, the new system represented an advance over the previous system of manual data collection because it used existing data from the patient's electronic medical record, including date of encounter, date of birth, sex, address and the chief complaint as described by the patient.
Hospitals participating in the system report data to the DHSS via the Internet daily. Data received from hospitals are then compared with a baseline figure for each condition and hospital. If the number of cases exceeds the expected level, state public health officials contact the hospital to obtain additional clinical information on the affected patients.
To implement the surveillance system, the project team:
- Held numerous meetings with representatives of affected departments to gain their participation and ownership for an enhanced surveillance system. Although this old system of manual data collection was resource intensive, staff at the department had expended time and effort to implement it. Therefore, project staff held numerous discussions and meetings with representatives from the Office of Surveillance, the State Epidemiologist's Office, the Missouri Hospital Association, Information Technology and CHIME.
- Formed an advisory committee. To further overcome resistance to the new system and increase the prospects for widespread adoption, CHIME staff formed an advisory committee of representatives from major hospitals, the Missouri Hospital Association, smaller hospitals and from DHSS's Office of Surveillance and its departments of information technology and epidemiology. In three regional meetings, committee representatives completed an implementation plan for the CHIME staff to follow in developing the new bioterrorism surveillance system. (See Challenges for a discussion of resistance to the system.)
- Selected a software conversion system that could translate the natural language in which patients report symptoms into a commonly recognized disease code. Hospital staff in emergency rooms recorded data on patient complaints in natural language or "literal" form a symptom such as a cough or earache, for example. Before submitting these data to state public health officials, they had to be translated by a software conversion system into an associated code known as an ICD-9-CM code (International Classification of Diseases, Ninth Revision, Clinical Modification). (See Challenges for a discussion of the software selection process.)
- Secured amendments to existing state regulations on hospital data reporting requirements. State regulations were revised in 2003 to require hospitals to report data collected on patient symptoms to DHSS daily. By November 2003, hospitals were required to submit a plan for initiating daily electronic data submission. Hospitals were allowed to phase in their participation over several years based upon location, annual number of ER patients, electronic reporting capacity, specialty and special circumstances.
- Many physicians, epidemiologists and hospital personnel were afraid that the new surveillance system would increase their workload without adding to their capacity to detect disease. To win over skeptics, project staff had to demonstrate that the previous surveillance system was not as effective as they believed. Many who resisted the new technology, for example, were unaware that sexually-transmitted diseases and other reportable conditions were underreported under the old system. Using medical records from the emergency departments of seven Missouri hospitals, the project team showed that, under the existing reporting system, only 20 percent of reportable communicable diseases were being detected. (See Lessons Learned for more discussion on ensuring acceptance of new technology.)
- Finding a cost-effective and accurate natural language conversion package was time-consuming. CHIME staff first chose a software conversion system from a private vendor. Rapid price increases, however, put it out of range for many financially stressed hospitals, particularly the large urban hospitals in St. Louis. Staff replaced the system with ESSENCE (Electronic Surveillance for the Early Notification of Community-Based Epidemics), an equally effective but less costly system developed by the Department of Defense.
- The CHIME information specialists who designed the new surveillance system and the clinicians and epidemiologists who were its intended users did not communicate effectively about their different priorities. According to Garland Land, the original project director and the former director of CHIME, information technologists were interested in creating a more efficient electronic system. Many clinical program staff, however, particularly those at lower levels, remained invested in the old syndromic surveillance system, which they had worked hard to develop. (See Lessons Learned for more discussion on leadership issues related to implementing new technology.)
- Although the grantee overcame many of the significant hurdles to growing a hospital emergency room surveillance system during the period of the grant, it is clear that "when the original project director left the Department of Health, this was a blow to the project . [It] needed a committed and creative leadership to continue to overcome the hurdles of working with the epidemiologists and clinical staff to utilize the information that the system was bringing in," comments RWJF Senior Program Officer Pamela G. Russo, M.D.
In addition to RWJF support, the project leveraged additional funding under two programs of the Centers for Disease Control and Prevention (CDC): the Emergency Preparedness & Response, Bioterrorism program and the National Electronic Disease Surveillance System (NEDSS).
- By September 2005, 41 of the state's 126 hospitals were participating in the new bioterrorism surveillance system. These hospitals were electronically submitting data daily on symptom clusters to the DHSS to detect a possible bioterrorism-related disease outbreak. Psychiatric and rehabilitation hospitals, rural hospitals and those without emergency rooms were exempt from the legal requirement to submit their data.
- By June 2006 (eight months after the grant ended), 82 of the state's hospitals were participating.
- Bring the "end users" epidemiologists and clinical staff into a systems change project as collaborators from the beginning. This may help lower resistance and turn them into champions of the system. (Program Officer/Russo)
- To gain wider acceptance for a new technology, emphasize its usefulness beyond its narrow, intended purpose. Some health department staff questioned the value of spending time and resources on a system that is used only for extremely rare events, such as the outbreak of anthrax or other bioterrorism-related diseases. To counter this resistance, Project Director Nancy Hoffman recommended exploring its applications in detecting conditions that health departments are required to report routinely, such as flu, tuberculosis or sexually transmitted diseases. Using the new technology regularly also guarantees that staff will know how to use it when it is needed in the event of a bioterrorism-related outbreak. (Project Director/Hoffman)
- Make certain that the organization's leaders remain committed to implementing a new technology. According to Garland Land, former project director, information technologists on the staff of CHIME spent "time and energy setting up" the bioterrorism surveillance system, but then were "too exhausted to use it." Epidemiologists and clinical program staff at DHSS were less committed to implementing the new system. When Land and many CHIME staff left DHSS, the new system lost its champions. To maintain momentum and continuity despite staffing changes, energetic committed leaders are needed to ensure that a system is implemented over the long term. (Project Director/Land)
AFTER THE GRANT
The Missouri DHSS will continue implementing the new system with bioterrorism funding from the CDC.
According to Land, it is too early to determine whether the new surveillance system is effective in detecting symptom clusters indicative of a bioterrorism threat, such as anthrax, or a communicable disease, such as influenza. However, the project did demonstrate that state public health officials could obtain bioterrorism related data from a large group of hospitals "a success in itself," Land said.
GRANT DETAILS & CONTACT INFORMATION
Developing an Active Hospital-Based Bioterrorism Surveillance System
State of Missouri Department of Health and Senior Services (Jefferson City, MO)
Dates: March 2003 to September 2005
Report prepared by: Jayme Hannay
Reviewed by: Richard Camer
Reviewed by: Molly McKaughan
Program Officer: Pamela G. Russo