Skip Ribbon Commands
Skip to main content

Analysis by place characteristics

Last modified at 9/4/2015 7:31 AM by Vladimir Prikazsky

Plotting surveillance data by geographical location is an important step of the descriptive analysis. When done properly, this method enables the visualization of the geographical extension of a public health event, the identification of disease clusters, and the targeting of affected populations.

Data can be expressed in absolute numbers or rates (incidence, attack rates). Using rates allows for the comparison of distributions across geographic boundaries (based on places of disease notification). However, standardization of rates by age may sometimes be necessary in order to compare areas with different population structures (i.e. younger urban vs. older rural population).

The level of geographical analysis depends on the public health event under consideration. A cluster of a rare disease can only be described at the level of the affected community, while the national picture is of interest for diseases more commonly distributed.

Figure 1: Distribution of viral hepatitis in Lebanon, by district, as of week 2003-15.


Because they are available and denominators are known, administrative boundaries are often used for analyzing surveillance data. In addition, they reflect the structure through which the response to a public health crisis will be implemented. It is important, however, to remember that in some instances they may distort the true distribution of a disease in a population (e.g. when a city is split over several administrative boundaries, diluting any increase in reporting among several places).

Mapping data issued from sentinel surveillance system requires specific techniques in order to extrapolate the information from sentinel sites to the entire country[1].

See additional information on mapping surveillance data in chapters Choosing an appropriate type of map and Which indicator to map

Spatial distribution of diseases

The analysis of the spatial distribution of communicable disease is an important step in the routine analysis of surveillance data. In relation to early warning functions, it explores whether the spatial distribution of the disease follows some geographical pattern that may contributes to identify areas at increased risk of transmission. During outbreak investigations, mapping cases helps generate hypotheses about the mode of transmission.

The spatial distribution of communicable diseases relates to its mode of transmission:

  • Communicable diseases related to a point common source are likely to show spatial clustering. For example, a cluster of infected people are likely to have been exposed to a common source such as a water supply system contaminated by enteric germs or a cooling tower contaminated with legionella.
  • Other communicable diseases transmitted from person to person can also show spatial clustering when people are infected by being exposed to existing cases. Viral hepatitis and meningitis are examples of such diseases.

However, some communicable diseases are unlikely to show meaningful spatial patterns. This is the case for example for:

  • Communicable diseases associated with travel, especially when the incubation period is long enough to allow the traveler to leave the area of exposure before becoming ill, e.g. viral hepatitis or travel associated Legionnaire disease.
  • Communicable diseases related to a widespread common source, such as a food product having a broad distribution in a country.

Steps in mapping data

A map is an analog way to represent the spatial distribution of a disease. Making a meaningful map requires the following steps:


1. Carrat F., Valleron A-J.: Epidemiologic Mapping Using the "Krigging" Method. Application to an Influenza-Like Illness epidemic in France. Am. J. Epidemiol., 1992; 135:1293-1300