Notes

[pic 1]

Risk ratio= incidence in the exposed/incidence in the unexposed Purpose: the RR measures the strength of the association between the exposure & disease. How many times greater is the risk of disease in the exposed people compared to unexposed people?

Risk Difference (attributable risk): incidence in the exposed -incidence in the unexposed. Using example: RD = Ie-Iu = 0.04 -0.01 = 0.03 (or 30/1000) Purpose: the RD measures the amount of absolute risk (that is, incidence rate) that can be attributed to an exposure. RD answers the questions: How much of the incidence rate in exposed people is due to the exposure? 30 / 1000 If we could eliminate this exposure, what incidence rate of the disease would be eliminated in exposed people? 30 / 1000

Etiologic fraction (percent): (attributable fraction %): (Incidence in the exposed -Incidence in the unexposed)/ Incidence in the exposed x 100. EF% = Ie-Iu/ Ie= ( 0.04 -0.01 / 0.04 ) x 100 = (0.03 / 0.04) x 100 = 75% EF% Purpose: measures the percent of rate of disease in the exposed due to the exposure. answers the questions: 1. What percent of the risk of disease in the exposed is attributable the exposure? 75% 2. How important are other causes of this disease? 25% 3. How much of the disease rate in the exposed group would be prevented if the exposure were eliminated? 75%

Population excess risk: (population attributable risk): PER = Incidence in whole population -Incidence in the unexposed group PER = IP-Iu= 0.02 -0.01 = .01 (or 10/1000) Purpose: measures the incidence rate in the whole population due to the exposure. PER answers the questions: 1. What amount of incidence rate in the whole population is due to the exposure? 10/1000 2. How much could we lower the incidence rate of this disease in the whole population if we eliminated this exposure? 10/1000 (Iu)Population etiologic fraction% : (population attributable fraction): (Incidence in whole population -Incidence in the unexposed group)/ Incidence in whole population X100. PEF% = ( IP-Iu/ IP )x 100 = ( 0.02 -0.01 /0.02) x100 = 50%. Purpose: measures the percent of disease rate in the whole population due to this exposure. PEF answers the questions: 1. What percentage of this disease is due to this exposure in this whole population? 50% 2. What percent of disease in this whole population would be prevented if we could eliminate this exposure? 50%DOSE RESPONSE: a change in the amount of the exposure is associated with a change in the outcome. Significance of an observed dose-response relationship = Stronger argument for causation.