Difference between revisions of "Category:Analysis by place characteristics"

From
Jump to: navigation, search
m
m (FEM Editor 2007)
 
(23 intermediate revisions by 2 users not shown)
Line 1: Line 1:
 +
==Visualizing geography==
 
Plotting surveillance data by geographical location is an important step of 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.
 
Plotting surveillance data by geographical location is an important step of 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 comparing distributions across geographic boundaries (based on places of disease notification). However, standardizing rates by age may sometimes be necessary to compare areas with different population structures (i.e., younger urban vs. older rural population).
+
Data can be expressed in absolute numbers or rates (incidence, attack rates). Using [[Risks and Rates|rates]] allows for comparing distributions across geographic boundaries (based on places of disease notification). However, [[standardization of rates|standardizing rates]] by age may sometimes be necessary 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 rare diseases can only be described at the level of the affected community, while the national picture is of interest for diseases more commonly distributed.
 
The level of geographical analysis depends on the public health event under consideration. A cluster of rare diseases 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.  
 
Figure 1: Distribution of viral hepatitis in Lebanon, by district, as of week 2003-15.  
[[File:Place.png|frame|Distribution of viral hepatitis in Lebanon, by district, as of week 2003-15.]]
+
[[File:Place.png|none|600px|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).
 
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 systems requires specific techniques to extrapolate the information from sentinel sites to the entire country[1].
+
Mapping data issued from sentinel surveillance systems requires specific techniques to extrapolate the information from sentinel sites to the entire country <ref>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</ref>.
  
See additional information on mapping surveillance data in chapters Choosing an appropriate type of map and Which indicator to map?
+
See additional information on mapping surveillance data in chapters [[Choosing an appropriate type of map]] and [[Which indicator to map?|Which indicator to map]]
  
Spatial distribution of diseases
+
==Spatial distribution of diseases==
 
Analyzing the spatial distribution of communicable diseases is an important step in the routine analysis of surveillance data. About early warning functions, it explores whether the spatial distribution of the disease follows some geographical pattern that may contribute to identifying areas at increased risk of transmission. During outbreak investigations, mapping cases help generate hypotheses about the transmission mode.
 
Analyzing the spatial distribution of communicable diseases is an important step in the routine analysis of surveillance data. About early warning functions, it explores whether the spatial distribution of the disease follows some geographical pattern that may contribute to identifying areas at increased risk of transmission. During outbreak investigations, mapping cases help generate hypotheses about the transmission mode.
  
 
The spatial distribution of communicable diseases relates to their mode of transmission:
 
The spatial distribution of communicable diseases relates to their mode of transmission:
  
Communicable diseases related to a point common source are likely to show spatial clustering. For example, a cluster of infected people is likely exposed to a common source, such as a water supply system contaminated by enteric germs or a cooling tower contaminated with legionella.
+
* Communicable diseases related to a point common source are likely to show spatial clustering. For example, a cluster of infected people is likely 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 exposure to existing cases. Viral hepatitis and meningitis are examples of such diseases.
+
* Other communicable diseases transmitted from person to person can also show spatial clustering when people are infected by exposure 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:
 
However, some communicable diseases are unlikely to show meaningful spatial patterns. This is the case, for example:
  
Communicable diseases that are 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 that are 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.
+
* Communicable diseases related to a widespread common source, such as a food product having a broad distribution in a country.
Steps in mapping data
+
==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:
 
A map is an analog way to represent the spatial distribution of a disease. Making a meaningful map requires the following steps:
  
Defining the appropriate indicator to represent the magnitude of the disease
+
* Defining the [[Which indicator to map?|appropriate indicator]] to represent the magnitude of the disease
Selecting the appropriate variable to display the spatial disease pattern
+
* Selecting the appropriate variable to display the spatial disease pattern
Choosing the appropriate type of map
+
* [[Choosing an appropriate type of map]]
References
+
 
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
+
==References==
 +
<references/>
 +
 
 +
<div style="display: inline-block; width: 25%; vertical-align: top; border: 1px solid #000; background-color: #d7effc; padding: 10px; margin: 5px;">
 +
'''FEM PAGE CONTRIBUTORS 2007'''
 +
;FEM Editor
 +
:Denis Coulombier
 +
;Original FEM Authors
 +
:Christophe Paquet
 +
:Arnold Tarantola
 +
:Philippe Quenel
 +
:Nada Ghosn
 +
;FEM Contributors -2017
 +
:Lisa Lazareck
 +
:Farshad Pourmalek
 +
:Denis Coulombier
 +
:Arnold Bosman
 +
:Vladimir Prikazsky
 +
</div>
 +
[[Category:Descriptive data analysis]]

Latest revision as of 08:58, 23 April 2023

Visualizing geography

Plotting surveillance data by geographical location is an important step of 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 comparing distributions across geographic boundaries (based on places of disease notification). However, standardizing rates by age may sometimes be necessary 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 rare diseases 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.

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 systems requires specific techniques 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

Analyzing the spatial distribution of communicable diseases is an important step in the routine analysis of surveillance data. About early warning functions, it explores whether the spatial distribution of the disease follows some geographical pattern that may contribute to identifying areas at increased risk of transmission. During outbreak investigations, mapping cases help generate hypotheses about the transmission mode.

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

  • Communicable diseases related to a point common source are likely to show spatial clustering. For example, a cluster of infected people is likely 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 exposure 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:

  • Communicable diseases that are 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:

References

  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

FEM PAGE CONTRIBUTORS 2007

FEM Editor
Denis Coulombier
Original FEM Authors
Christophe Paquet
Arnold Tarantola
Philippe Quenel
Nada Ghosn
FEM Contributors -2017
Lisa Lazareck
Farshad Pourmalek
Denis Coulombier
Arnold Bosman
Vladimir Prikazsky

Pages in category "Analysis by place characteristics"

The following 2 pages are in this category, out of 2 total.