Essay Instructions: TO DO:
Surveillance and Data Sources: Activities
1. Evaluation of Surveillance Systems
In the Evaluation of Surveillance Systems Worksheet, you will list two quantitative methods for evaluation of a surveillance system and two qualitative methods for evaluation of a surveillance system. One Page document.
2. Other Surveillance Systems
Identify an existing surveillance system that was not described in the lesson. Then create a one-page document listing its name, data sources, two of its strengths, and one of its limitations.
Background Information:
_______________________________________
Surveillance and Data Sources
________________________________________
Page 1 of 8
Surveillance and Data Sources
Overview
This lesson introduces and develops the concept of public health surveillance — the process by which the health conditions and risks in a specific population are monitored. Surveillance requires information systems and personnel to collect and analyze accurate data in an ongoing and systematic manner. However, surveillance is not only working with data; it is also the interpretation and dissemination of the information from the data back to those who may need such information to promote health. In considering outbreak investigation previously, we saw that outbreaks are detected through surveillance systems. Thus, our ability to detect and intervene in outbreaks, or to improve public health in many ways, is very dependent upon the completeness and quality of our surveillance systems.
Objectives
By the end of this lesson, you should be able to:
• List two quantitative methods for evaluation of a surveillance system and two qualitative methods for evaluation of a surveillance system
• Describes a method of surveillance that is not discussed in this lesson
________________________________________
Page 2 of 8
Surveillance and Data Sources: Road Map: Road Map
Note: The purpose of the Lesson Road Map is to give you an idea of what will be expected of you for this lesson. You will be directed to specific tasks as you proceed through the lesson. Each activity in the To Do: section will be identified as individual (I), team (T), graded (G).
To Read
• Centers for Disease Control and Prevention (CDC). (2006). Overview of Influenza Surveillance in the United States. Retrieved January 22, 2007, from http://www.cdc.gov/flu/weekly/pdf/flu-surveillance-overview.pdf.
• Centers for Disease Control and Prevention (CDC). (n.d.). An Overview of the NEDSS Initiative. Retrieved January 22, 2007, from http://www.cdc.gov/nedss/About/overview.html.
• Brackbill, R., Thorpe, L., DiGrande, L., Perrin, M., Sapp, J., Wu, D., et al. (2006). Surveillance for World Trade Center Disaster Health Effects Among Survivors of Collapsed and Damaged Buildings. Morbidity and Mortality Weekly Report (MMWR), Vol. 55, No. SS-2. Retrieved January 22, 2007, from http://www.cdc.gov/mmwr/preview/mmwrhtml/ss5502a1.htm.
To Do
In this lesson you will complete the following activities:
1. Evaluation of Surveillance Systems
2. Other Surveillance Systems
Topics
• Purposes of Public Health Surveillance
• Sources of Surveillance Information
• Coverage of the Surveillance System
• Evaluation of a Surveillance System
________________________________________
Page 3 of 8
Purposes of Public Health Surveillance
Public health surveillance ( ) is the ongoing, systematic collection, analysis, and interpretation of health-related data essential to the planning, implementation, and evaluation of public health practice, closely integrated with the timely dissemination of these data to those responsible for prevention and control. Note that surveillance has two primary components – data and practice. Collection of data without its use in action is meaningless. Conversely, practice that is not founded upon data may be inefficient and ineffective.
Purposes
Surveillance has multiple purposes which include:
• Assessing public health status
• Defining priorities
• Evaluating program and intervention efforts
• Stimulating research
Assessing Public Health Status
First, surveillance is the process by which we assess the health of the public. Without surveillance information, for example, we would not know with confidence the number of influenza cases in a particular geographic area, and we would not be able to tell whether that number is going up or down. Without that information, public health would be unable to know whether intervention to reduce the number of cases is necessary, and if so, where to intervene.
Defining Priorities
Surveillance is necessary to define public health priorities. In most situations, public health and medical authorities do not have the resources necessary to address all the public health issues that they might like to address. For example, authorities may have chosen between having their staff initiate a statewide influenza campaign or a statewide program to reduce exposure to lead. In such a situation, surveillance information will help the authorities determine which problem affects more people and is causing more ill-health. These pieces of information are necessary for authorities to effectively deploy their staff and other resources. Also, surveillance information may detect the beginning of an outbreak when it can best be controlled with well-placed resources.
Evaluating Program and Intervention Efforts
Surveillance is also used to evaluate programs and interventions. If authorities had deployed their staff and resources to address an outbreak of influenza, surveillance information could be used to determine whether the number of new cases was decreasing. If they do go down, then authorities would have additional information that their actions were effective and protected the health of the public.
Stimulating Research
Finally, surveillance is used to stimulate research. Careful analysis and interpretation of surveillance information may demonstrate to authorities that only persons who live in a particular geographic location are being affected by a disease or health condition. This finding would stimulate research to determine that factors that cause only people in that area to be affected, or conversely, what factors are keeping people in other areas healthy.
Recap
Without surveillance information, we would be uninformed about what is actually happening to the health of the public and whether or not important risks to our health are being controlled. Of particular note, surveillance information is critical to public health evaluation of disasters and bioterrorism. In the event of a disaster or bioterrorism, surveillance will:
• Inform us of the magnitude of the problem
• Help us to select the appropriate courses of action
• Identify whether the threat or problem is increasing, decreasing, or remaining the same
• Help us identify and quantify risks from an event or exposure.
Thus, the maintenance and analysis of data from surveillance systems is a critical priority for public health preparedness.
________________________________________
Page 4 of 8
Sources of Surveillance Information
Having considering why surveillance data are so important, we need to understand how they are collected. Fortunately, surveillance data can be collected through different sources and by different methods. For some diseases and conditions, relevant public health organizations maintain their own surveillance systems. For other diseases and conditions, surveillance data are gathered by other systems.
National Notifiable Disease Surveillance System
The National Notifiable Disease Surveillance System (NNDSS) is the primary mechanism by which infectious and acute conditions are monitored in the United States. NNDSS is maintained and supported by the Centers for Disease Control and Prevention (CDC) but relies upon reports from the states that are mandated by state laws or regulations. Mandated reports may originally come from physicians' offices, hospitals, and laboratories. Typically, the information is first reported to the county health department, which then relays information to the state health department, and finally to the national level. Note, that the state, not the national, level, is the mandating body in the United States for surveillance purposes. The NNDSS contains many diseases and health conditions, ranging from acute infectious to chronic to environmental to occupational.
Real World Case: NNDSS
The current NNDSS is located at http://www.cdc.gov/epo/dphsi/nndsshis.htm. Case definitions can often be found at this location also.
The National Electronic Disease Surveillance System (NEDSS) is an initiative in the United States that promotes the use of data and information system standards to advance the development of surveillance systems at federal, state and local levels with primary application to notifiable diseases. A goal of NEDSS is the ongoing, automatic capture and analysis of data that are already available electronically. NEDSS system architecture is designed to integrate and replace several current CDC surveillance systems, including the National Electronic Telecommunications System for Surveillance, the HIV/AIDS reporting system, and the vaccine preventable diseases and systems for tuberculosis (TB) and infectious diseases.
Real World Case: NEDSS
You can learn more about the NEDSS at http://www.cdc.gov/nedss/About/overview.html (which is also included in this lesson's readings).
Laboratory surveillance
Laboratories may also be an excellent source of surveillance data because they may provide a confirmatory assessment of the health status of individuals. Laboratory samples might be blood, serum, tissue (e.g., fat cells), or fecal specimens. Laboratory tests are less likely to be influenced by transitory health status that some patients might experience. Also, laboratory data are frequently computerized and can be made anonymous to protect confidentiality of patients. (Both of these features allow for quick analysis.) However, future contact of patients based upon laboratory data may be difficult unless that possibility has been assured at the time of the original data collection.
Vital records
Vital records ( ) are records of births and deaths. As a surveillance system, vital records are one of our oldest and most commonly used surveillance systems. Being well-developed in most countries, the vital record surveillance system serves as a foundation and backbone of surveillance throughout the world. The data contained in vital records provides necessary information to examine differences between residents of different countries and regions of the world, or of a single country. Also, vital records do not rely upon the recognition by states or countries that a disease or condition be notifiable, as is necessary for a disease or condition to listed in a notifiable disease surveillance system.
One of the drawbacks to vital record surveillance is the fact that the birth or death of individual is a significant event. Also, deaths may result from an exposure that occurred many years previously, or from an exposure that occurred in a geographic location other than one of the decedent, or from continuous exposure over a long time period.
Surveys
Surveys are frequently used to collect surveillance data on health and health risks of people. Survey data may be collected via telephone and personal interviews, or through the Internet. Often these surveys may be called questionnaire surveys. Occasionally, surveys will include a clinical examination. This makes them even more useful for epidemiology and public health evaluation because clinical measures can be used to verify self-reported conditions. Also, surveys may be mobilized fairly quickly such that emergent conditions or risk can be assessed. A disadvantage to surveys is that information on the exposure and the health condition are obtained simultaneously, which may reduce the assurance that the exposure preceded the onset of the health condition or disease.
Administrative Data
Finally, surveillance data may be collected through other systems that are not specifically maintained for surveillance purposes. A primary example of this type of data set would be an administrative data set. Administrative data sets are those data sets that are maintained for purposes other than surveillance, but contain data that can be very useful for surveillance purposes. A good example of an administrative data set is a hospital discharge data set, which is maintained for billing purposes of the hospital. Within this billing information is information on diagnoses and sometimes tests or test results. This information can be very helpful for understanding the health and diseases of people that have been hospitalized.
Multiple Sources
Frequently, a surveillance system uses multiple methods and data sources. A good example of such a surveillance system in the United States is the influenza system. It is composed of seven components including laboratory, sentinel physician, vital records, and other specialized surveillance pieces.
Real World Case: Influenza Surveillance
You can learn more about the seven components of influenza surveillance in the U.S. at http://www.cdc.gov/flu/weekly/pdf/flu-surveillance-overview.pdf (which is also included in this lesson's readings).
________________________________________
Page 5 of 8
Coverage of the Surveillance System
In addition to the method of data collection, another facet of surveillance revolves around the fact that surveillance systems data can be collected from different groups of people. In general there are three different populations on which a surveillance system operates, each giving rise to its own category:
• Registries
• Sentinal Surveillance
• Syndromic Surveillance
Registries
First, we may attempt to collect disease and health data on all affected people in a specific geographic area. These sorts of surveillance systems are often called registries. For example, most states in the United States operate a statewide cancer registry which seeks to collect and analyze information for persons in their state. For most part, statewide cancer registries are very complete because they typically collect data for about 95 percent of all cancer cases. For some cancers, like lung cancer, the completeness is closer to 100 percent.
Sentinal Surveillance
Second, data may be collected through specific samples of people or health care clinics. For example, surveillance information on influenza in the United States is collected through a very carefully designated network of physician offices in selected cities in the United States. This network has a specifically designed method to accurately report influenza cases to the CDC in a timely and efficient manner. This reporting thus allows influenza activity to be monitored around the country so that strategic methods can be emploed if outbreaks should be detected. Often, this type of surveillance, which detects cases from specific areas or by specific methods, is called sentinel surveillance because we are relying upon isolated systems to detect important events which provide us with an early warning.
Syndromic Surveillance
Recently, information analysts and statisticians have become very interested in the potential of using health-related data that precede diagnosis and signal a sufficient probability of a case or an outbreak to warrant further public health response. This has been called syndromic surveillance. Though historically syndromic surveillance has been utilized to target investigation of potential cases, its utility for detecting outbreaks associated with bioterrorism is increasingly being explored by public health officials. Often, investigators have thought that hospital administrative data may be very useful for syndromic surveillance. While most surveillance occurs after a diagnosis has been determined, syndromic surveillance focuses upon analysis of the signs and symptoms, not diagnoses, of individuals. Thus, syndromic surveillance could be very useful in an early alert system because it does not need to wait for full confirmation of the disease. For example, an alert of an increase in the number of persons with a fever and respiratory congestion may be an early sign of increasing influenza activity, thus alerting public health officials of possible impending public health problem related to influenza.
Real World Case: Syndromic Surveillance for Bioterrorism-associated Agents
Detailed definitions of syndrome definitions associated with bioterrorism-associated agents can be found at http://www.bt.cdc.gov/surveillance/syndromedef/index.asp.
________________________________________
Page 6 of 8
Evaluation of Surveillance Systems
Surveillance data are especially useful if they are of high quality and collected in a timely manner because then they accurately represent the real world situation. Even if they are not of highest quality or collected and analyzed immediately, surveillance data may be useful, particularly if their quality and timeliness are well understood. For example, a well-designed sentinel surveillance system that is timely and truly representative of a larger population can be very valuable in identifying emergent trends or conditions despite its not collecting data from the entire population. Therefore, regular evaluation of surveillance systems is an important activity so that we can approach our data with a known level of confidence.
Qualitative Attributes of a Surveillance System
Several features of a surveillance system can make it very useful for the prevention and control of diseases and other adverse health outcomes. These features include the system's:
• Flexibility – This refers to the system's ability to adapt to a change in the health outcome or risk factor.
• Acceptability – This refers to whether or to what degree health care providers, hospitals, and laboratories perceive the surveillance system as acceptable, easy to use, relevant, or minimally time-consuming.
• Simplicity – This is related to acceptability and refers to a surveillance system's ease of use. Complex systems are less likely to be used, or the data in them is less likely to be accurate or timely.
Quantitative Attributes of a Surveillance System
The attributes of a surveillance system may be assessed by two key quantitative measures: sensitivity and specificity. The discussion of those two measures that follows is based on the following sample 2x2 table.
Sample 2x2 Table
Sensitivity
The primary quantitative measure is sensitivity, which is proportion of all true cases of the disease that are detected by the surveillance system. In the 2x2 table above, sensitivity is calculated as follows:
• Sensitivity = A/A+C
Sensitivity tells us the degree to which we can expect a surveillance system to detect all cases of the disease. In other words, sensitivity tells us how good the surveillance system is at detecting cases. Sensitivity is usually expressed as a percentage between 0 and 100.
Specificity
Specificity, another quantitative attribute, is the proportion of all persons who do not have disease that are detected by the surveillance system as being negative. Again, in the 2x2 table above, specificity is calculated as follows:
• Specificity = D/B+D
Like sensitivity, specificity is reported as a percentage between 0 and 100. It is often difficult to calculate specificity because its calculation requires that all negative persons receive the test or screening in a surveillance system. For diseases that are rare (i.e., most diseases) testing all persons, especially because so many will be negative, may be prohibitively expensive or problematic in other ways.
Other Evaluation Considerations
The ability of a surveillance system to identify diseased individuals may be variable among different populations. For example, a test procedure that requires extensive medical testing is more likely to be used only among those persons who are in the health care system (rather than among those who are not in the health care system). In situations or geographic locations where the use of tests is variable, we might believe that the representativeness of the surveillance system is low. Thus, its findings may represent one group better than another. Representativeness of surveillance systems may be quantified through detailed studies, but is often referred to in a qualitative manner.
A final important attribute of a surveillance system is its timeliness. In other words, how long does it take for the result of a test or analysis of surveillance data take to reach to the appropriate authorities? For some diseases, especially chronic diseases like cancer or heart disease, a timeliness that is measured in months or years may be sufficient. However, for diseases or health conditions that are acute or contagious, test results and surveillance data should be reported and analyzed within hours or days.
________________________________________
Page 7 of 8
Lesson Summary
This lesson introduced you to a very important component of public health — surveillance systems. Surveillance systems are the foundation from which public health investigations and control measures follow. Without accurate and timely data from these surveillance systems, outbreaks would go undetected, risks would not be identified, and preventive measures would not be evaluated. Surveillance data can be collected by various data systems — not all are specifically designed solely for surveillance purposes. Finally, evaluation of surveillance systems is important so that the credibility of public health data and information is known.
________________________________________
Page 8 of 8
Works Cited
N/A
