Skip directly to search Skip directly to A to Z list Skip directly to navigation Skip directly to page options Skip directly to site content

Lesson 5: Public Health Surveillance

Appendix A. Characteristics of Well-Conducted Surveillance

Acceptability reflects the willingness of individual persons and organizations to participate in surveillance. Acceptability is influenced substantially by the time and effort required to complete and submit reports or perform other surveillance tasks.

Flexibility refers to the ability of the method used for surveillance to accommodate changes in operating conditions or information needs with little additional cost in time, personnel, or funds. Flexibility might include the ability of an information system, whose data are used for surveillance of a particular health condition, to be used for surveillance of a new health problem.

Predictive Value Positive is the proportion of reported or identified cases that truly are cases, or the proportion of reported or identified epidemics that were actual epidemics. Conducting surveillance that has poor predictive value positive is wasteful, because the unsubstantiated or false-positive reports result in unnecessary investigations, wasteful allocation of resources, and especially for false reports of epidemics, unwarranted public anxiety (see Figure 5.10 for how to calculate predictive value positive.)

Quality reflects the completeness and validity of the data used for surveillance. One simple measure is the percentage of unknown or blank values for a particular variable (e.g., age) in the data used for surveillance.

Representativeness is the extent to which the findings of surveillance accurately portray the incidence of a health event among a population by person, place, or time. Representativeness is influenced by the acceptability and sensitivity (see the following) of the method used to obtain data for surveillance. Too often, epidemiologists who calculate incidence rates from surveillance data incorrectly assume that those data are representative of the population.

Sensitivity is the ability of surveillance to detect the health problem that it is intended to detect. (see Figure 5.10 for how to calculate sensitivity.) Surveillance for the majority of health problems might detect a relatively limited proportion of those that actually occur. The critical question is whether surveillance is sufficiently sensitive to be useful in preventing or controlling the health problem.

Simplicity refers to the ease of operation of surveillance as a whole and of each of its components (e.g., how easily case definitions can be applied or how easily data for surveillance can be obtained). The method for conducting surveillance typically should be as simple as possible while still meeting its objectives.

Stability refers to the reliability of the methods for obtaining and managing surveillance data and to the availability of those data. This characteristic is usually related to the reliability of computer systems that support surveillance but might also reflect the availability of resources and personnel for conducting surveillance.

Timeliness refers to the availability of data rapidly enough for public health authorities to take appropriate action. Any unnecessary delay in the collection, management, analysis, nterpretation, or dissemination of data for surveillance might affect a public health agency's ability to initiate prompt intervention or provide timely feedback.

Validity refers to whether surveillance data are measuring what they are intended to measure. As such, validity is related to sensitivity and predictive value positive: Is surveillance detecting the outbreaks it should? Is it detecting any nonoutbreaks?

Figure 5.10 Calculation of Predictive Value Positive, Sensitivity, and Specificity for Surveillance

True case or outbreak
Detected by surveillance?YesTrue positive
False positive
Total detected by surveillance (A + B)
NoFalse negative
True negative
Total missed by surveillance(C + D)
TotalTotal true cases or outbreaks (A + C)Total noncases or non-outbreaks (B + D)Total (A + B + C + D)

Predictive value positive = A ⁄ (A+B)
Sensitivity = A ⁄ (A+C)
Specificity = D ⁄ (B+D)

Adapted from: Centers for Disease Control and Prevention. Updated guidelines for evaluating public health surveillance systems: recommendations from the guidelines working group. MMWR 2001;50(No. RR-13): p. 18.
Protocol for the evaluation of epidemiological surveillance systems [monograph on the Internet]. Geneva: World Health Organization [updated 1997; cited 2006 Jan 20]. Available from:

Table 5.7 Relative Importance of Selected Surveillance Characteristics By Use of Surveillance Findings

Use of surveillance
CharacteristicManaging individual cases of diseaseDetecting outbreaks of diseasePlanning and evaluating health programs
Predictive value positive***********

The number of asterisks reflects the relative importance of each characteristic with more asterisks signifying greater importance.

Adapted from: Sosin DM, Hopkins RS. Monitoring disease and risk factors: surveillance. In: Pencheon D, Melzer D, Gray M, Guest C (editors). Oxford Handbook of Public Health, 2nd ed. Oxford: Oxford University Press; 2006 (in Press).