Methodological limitations
Aggregation of data
The incidence of some cancer types is low at the 3- and 4-digit postcode levels. For this reason, the choice was made to combine data over the period 2011-2022. Using data over a longer time period allowed for stable and precise estimates.
Choice of geographic areas
The atlas displays SIRs per 3-digit and 4-digit postcode areas. It does not show spatial variation within an area. Using a different aggregation level might result in a different geographical pattern (this is known as the 'modifiable areal unit problem').
Differences in exposure
A diagnosis was assigned to an area if a patient lived in that area at the time of diagnosis. It is unknown for how long the patient had been living in that area. It is possible that an individual was exposed to a cause of cancer outside of the area where they were living at time of diagnosis. For many types of cancer the period between exposure and development of cancer is 20 years or more . An increased risk of cancer in a specific area therefore does not necessarily mean that a potential causative factor is more present in that area.
No information at individual level
The estimates in the atlas provide information about all residents of an area combined and are not applicable to individuals living in that area. Individuals may have a higher or lower risk than the average estimated risk for that area.