Skip to main content
hpa logo
Topics A-Z:
Search the site:
Home › Products & Services › Infectious Diseases › Services and Activities › Surveillance ›Sources of surveillance data
Printer friendly page (opens in new window)

Sources of surveillance data

Where do the data come from?

The surveillance of infectious disease involves the collection of data from many different sources.

The main sources are:

  • Notifiable diseases - by law, some diseases are "notifiable", which means that a doctor should report them on the basis of the symptoms alone, rather than waiting for laboratory confirmation.
  • Laboratory reports - if a doctor sends a sample from a patient for testing, the laboratory can confirm what germ (if any) is making the person sick.
  • Clinician reporting - for some diseases of particular public health importance, doctors are asked to report key items of information to ensure that appropriate public health action can be taken.

For a full list of data sources available to CfI, see the diagram below.

Each source of information provides a different perspective on the frequency and distribution (the epidemiology) of disease. Combining the information gathered from these different sources often gives a more complete and accurate picture of disease than is obtained by looking at the data from one source only.

Why do we sometimes publish different numbers?

Combining different data sets is, however, a complicated process, since the different data sets usually overlap and therefore matching reports of infection from different sources must be identified and merged if cases are not to be double counted and a falsely high estimate of disease frequency generated. Because of the effort required to match and merge data sets this process is only undertaken for the diseases for which the most accurate picture of disease frequency and distribution is required.

For other infections the data from different sources are simply presented as separate events of disease. These separate counts can sometimes be quite different for the same disease. This is simply a consequence of the different ways in which the data are generated.

For example, the characteristic nature of the cough in whooping cough will lead many doctors to make the diagnosis without seeking laboratory confirmation. Furthermore, the bacterium that causes whooping cough, Bordetella pertussis can be difficult to grow in the laboratory, and so sometimes if the doctor does take a specimen the organism might not be detected even though the patient genuinely has whooping cough. The result is that there are many more notifications of whooping cough, which can be made without any laboratory confirmation, than there are laboratory reports.

Nonetheless, laboratory reports of Bordetella pertussis remain an important source of data as they provide information on which strains of the bacterium are circulating in the population. In addition, the fact that laboratory reports of Bordetella pertussis show the same seasonal pattern as notifications of whooping cough provides reassurance that the notified cases are indeed whooping cough.

Such reassurance is important since some infections are now so rare that more often than not cases that are notified on the basis of clinical suspicion (i.e. on the basis of symptoms and signs, but without any laboratory test confirmation) turn out not to be the suspected infection. This is now the case for notification of measles, with less than 5% of suspected cases turning out to have actual measles infection.

Sources of surveillance data to CfI

Click to view large version (opens in a new window)

Sources of surveillance data to CfI