The collection of relevant data pertaining to disease occurrence, when collated, analyzed, and reported, leads to a public health action. This is known as Public Health Surveillance. Without a resultant action, the information gathered is archival and not satisfying a public health function. Disease control and prevention activities cannot be effectively and efficiently accomplished without analyzing occurrence data, which helps direct the development of these activities. The initial data may not be adequate for accomplishing the subsequent public health action and thus additional data may need to be collected or a research project initiated. All of our efforts in disease control and prevention work are predicated by having meaningful disease notification.
Disease reporting was initially developed more than 150 years ago in Great Britain when occurrence data related to epidemic diseases such as cholera and plague were collected. It was the accumulation of data concerning the occurrence of cholera deaths in London in the 1840s that led to epidemiological studies by John Snow that portrayed the natural history of cholera in London from which control and prevention measures were offered and eventually accepted. Subsequently, more specific mechanisms of obtaining disease occurrence data were developed; and in the United States in the early 1900s, laws were passed that initiated the development of the disease reporting system. Through the years, this system has become more sophisticated both in the United States and throughout the world. No part of the world is truly isolated, and thus it is recognized that worldwide disease reporting is important if diseases are to be controlled and prevented. Each country will have its own arrangements for developing surveillance data according to its own needs and resources. The World Health Organization (WHO) maintains worldwide coverage of selected diseases and solicits the reporting of other diseases on a voluntary basis. By international law, plague, cholera, and yellow fever are reportable to WHO. Data on these diseases are summarized in a weekly epidemiologic report prepared and distributed by the WHO.
There are multiple uses of disease notification data. A major use is that of providing information concerning the natural history of disease. For example, data for malaria in the United States from 1920 to 1999 depicts a gradual decrease in incidence, apparently due to control measures, but with a number of sporadic increases, reflecting, for example, returning members of the armed forces or immigrants from malarious areas of the world. The system also may be used to portray changes in the occurrence of a disease. Another use is to monitor the effectiveness of control and prevention measures, such as the decreasing incidence of measles reflecting the nation's vaccination program. Another use of the data is to assist in the detection of a national epidemic that may not be apparent when looking at data from a single state. When data are accumulated from several states, however, the occurrence of an epidemic may become apparent. This is exemplified by the data for the epidemic of salmonellosis in 1984 in the Midwest caused by contaminated milk. The Centers for Disease Control and Prevention (CDC) has developed guidelines, including threshold values, which assist in identifying the occurrence of a disease epidemic. Disease notification data may also be useful in suggesting the need for research, including field investigations in order to explain unusual features of disease occurrence.
Within the Untied States, each state has its own laws concerning what diseases are reportable and how they should be reported to the state health department. The number of diseases reportable within the states varies from thirty-five to 120. These state laws include penalties for not reporting reportable diseases; however, there is no information available concerning whether any health care practitioners have been charged with not reporting a reportable disease.
The CDC is the agency within the United States Public Health Service that is charged with the responsibility for maintaining the notifiable disease system for the United States. The diseases that are reportable from states to CDC are decided upon by agreement between the states and CDC, though this is a voluntary reporting system. The diseases to be reported to CDC are discussed at an annual meeting. As of January 2000, sixty diseases were reportable. In some states, the laws do not include all nationally notifiable diseases. The National Notifiable Diseases Surveillance System (NNDSS) asks for the number of cases of each disease that have been reported within the state on a weekly basis. However, there are supplemental CDC reporting systems for many of these diseases for which the states are asked to fill out a more specific reporting from that calls particularly for data that allows for the determination of potential risk factors. Routine reporting through the NNDSS from the states to CDC is accomplished by computer, whereas the supplemental systems have specific forms that are filled out and forwarded to CDC.
Public health surveillance may either be passive or active. Passive surveillance is the form of surveillance that is most commonly utilized throughout the world and consists of the routine reporting of a disease being initiated by the health care provider. Occasionally a specific request for reporting a disease is made by a public health agency, initiated because an acute outbreak of disease requires rapid accumulation of information in order to consider effective control and prevention measures. Usually such reporting is maintained only for a limited period of time. Also, a group of health care practitioners may be specifically requested to report a disease or diseases. This group may be proportionately representative of the entire group of health care providers, or they may be a selected sample in order to get more quantitative reporting. If the reporters are randomly selected on the basis of population density, the resulting data may be extrapolated to the entire population. Disease reporting may also be directed toward a specific population among whom a particular disease has a higher incidence than in the general population.
Reporting is usually the responsibility of the health care provider who makes the diagnosis of a patient's illness. However, a surrogate may be used; for example, in a hospital a ward clerk may be designated to initiate the report. Reporting may also be the responsibility of the laboratory in which the etiologic agent is identified, which allows more specific reporting of certain diseases such as gastrointestinal diseases.
The most common reporting system is morbidity reporting, where each case of a reportable disease is reported. Mortality reporting may also be used, such as with influenza. The weekly occurrence of deaths due to "Influenza and Pneumonia" are used to portray the occurrence of influenza. Reporting may also be accomplished by noting absenteeism in schools or large industrial establishments, or by the utilization of pharmaceuticals purchased in pharmacies. Reporting within states is usually accomplished on a weekly basis using reporting forms, postcards, fax machines, or computers.
The sensitivity of reporting varies, but for the common diseases it is low, maybe as low as 5 percent. However, if the methods of reporting, case definitions, and the reporters do not change, then the data will still adequately represent the trend of the disease occurrence for control and prevention purposes. Sensitivity can be increased by offering a reward (such as financial) for reporting a confirmed case of a disease. Such an award system was very important in the worldwide smallpox eradication program in the 1960s and 1970s.
At regular intervals the accumulated data concerning notifiable diseases should be analyzed by time, place, and person to the level permitted by the details reported. Once analyzed, a report should be prepared providing quantitative information concerning the occurrence of disease, as well as information concerning recommended control and prevention measures. Field investigations can also be reported. CDC prepares the Morbidity and Mortality Weekly Report (MMWR), which is available
The reporting of data analyses back to the health care providers is important so they are encouraged to continue reporting. Not only is this important for continuity, but is also allows them to be updated on disease occurrence, on current method and prevention, and on the success of implementation of these measures.
Disease reporting systems should be evaluated at regular intervals to assure that they are functioning as planned. Important attributes to consider in evaluating such effectiveness include sensitivity, specificity, predictive value positive, representativeness, flexibility, simplicity, acceptability, and timeliness. Sensitivity refers to whether all true cases of the particular disease have been reported. Specificity refers to noncases not being reported as cases. Predictive value positive is the proportion of reported cases that truly are cases. A flexible system can adapt to changes in disease occurrence, or in changes in the reporting system, and still continue to function. Simplicity indicates a minimum of effort on the part of those involved in reporting or maintaining the system, and acceptability that reporting is acceptable to the health care providers who are asked to report. Timeliness refers to the fact that the data reported can be used in a timely manner for the control and prevention of that disease. Representativeness means that the cases reported does represent the occurrence of the disease, as it is occurring at the time of reporting. Another important consideration is cost-effectiveness, that is, whether the disease reporting for a particular disease saves money compared to not maintaining reporting for that disease.
The practice of disease notification is becoming more sophisticated not only in the United States but throughout the world. The use of computers is making reporting quicker and more accurate, and it supports more sophisticated analysis of the data. Improving laboratory support by means of more rapid and detailed identification of etiologic agents has also increased the effectiveness disease reporting.
In the future, health care practitioners, after seeing a patient, will be able to record the history, physical examination, and laboratory findings on a computer that will be programmed to automatically report the reportable data to the local health department, form which it will automatically be reported to the state health department. This will release the health care provider from having to make a special effort to file a report and improve the timing, accuracy, and sensitivity of reporting.
PHILIP S. BRACHMAN
Brachman, P. S. (1998). "Surveillance." In Bacterial Infections of Humans—Epidemiology and Control, 3rd edition, eds. A. S. Evans and P. S. Brachman. New York: Plenum Publishing Corporation.
Langmuir, A. D. (1963). "The Surveillance of Communicable Diseases of National Importance." New England Journal of Medicine 268:182–192.
Thacker, S. B., and Berkelman, R. C. (1988). "Public Health Surveillance in the United States." Epidemiologic Reviews 10:164–190.
Thacker, S. B., and Stroup, D. F. (1998). "Public Health Surveillance." Applied Epidemiology: Theory to Practice, eds. R. C. Brownson and D. B. Petitti. Oxford: Oxford University Press.