 |
| Block 1: Epidemiology |
| EPI3-10: ANOTHER RELATIVE MEASURE OF EFFECT: THE ATTRIBUTABLE PROPORTION |
|
| Block 1: Epidemiology |
| EPI3-10: ANOTHER RELATIVE MEASURE OF EFFECT: THE ATTRIBUTABLE PROPORTION |
 |
 |
 |
SESSION OBJECTIVE |
|
At the end of this session you should be able to understand the rather more difficult concept of the Attributable Proportion. This is also a relative measure of effect, and expresses that proportion of the disease that is due to the effect of a particular exposure. |
If an association is considered to be causal, the attributable proportion (AP), can be meaningfully assessed (other terms in use are attributable risk percent, attributable fraction, and etiologic fraction).
The attributable proportion is that proportion of the disease occurrence in the overall population that would be eliminated should the exposed group have its incidence reduced to the level of the unexposed group.
It is relative in that the excess relative risk (RR - 1) is expressed as a proportion of the RR. In other words this measures the fraction or proportion of the relative risk that is excess (or above the baseline risk in the unexposed to which the exposed group is being compared and which is expressed as 1). A RR of 1 means that the exposed have the same risk as the unexposed. Anything more than 1 is then the excess relative risk. The excess relative risk as a proportion of the full RR gives an idea of the specific contribution of the exposure associated with the elevation of risk (RR) in the exposed.
By multiplying this with the fraction f of the exposed in the overall population an estimate of the proportion of the disease occurrence in that overall population can be obtained.
If the RR in smokers for lung cancer compared with nonsmokers is 12, and 60% of the population smokes, then the AP = 55% and the implication is that doing away with smoking will eliminate 55% of all lung cancer in that population.
If in the same population as in Example 1, the RR in smokers for ischaemic heart disease (heart attack) is 1.2 compared with non-smokers, then the AP = 10% and doing away with smoking will only prevent 10% of heart attacks. Again given how many heart attacks there are in the population, eliminating 10% would make a material difference to the health services.
 |
Interactive ExamplesThe Examples listed below have been included in order to illustrate the concepts discussed in this Section. |
|
| Question 1 | Question 6 | Question 11 | Question 16 |
| Question 2 | Question 7 | Question 12 | |
| Question 3 | Question 8 | Question 13 | |
| Question 4 | Question 9 | Question 14 | |
| Question 5 | Question 10 | Question 15 |
