Mehdi Moradinazar Dept of Epidemiology Kermanshah University of

Odds vs Probability Odds = number of favorable outcomes number of unfavorable outcomes Probability

Measures in Case-Control Studies Cases control Exposed A B Not Exposed C D Case-control

Measures in Cohort Studies Disease + Disease - Exposure + A B Exposure -

(Case-control) + (Cases) (Controls) (Exposed) (Cohort) (Not Exposed) A B C D Case-control Analysis:

�� Disease + Disease - Total Exposure + a b a+b Exposure - c

Protective Factor 0 OR RR No association 1 Risk Factor OR RR No association

The odds ratio is a good estimate of the relative risk when a disease

The odds ratio is not a good estimate of the relative risk when a

: ﺩﺭﺻﺪ ﺧﻄﺮ ﻗﺎﺑﻞ ﺍﻧﺘﺴﺎﺏ Attributable Risk Percent Re - R u ARP =

: ﺩﺭﺻﺪ ﺧﻄﺮ ﻗﺎﺑﻞ ﺍﻧﺘﺴﺎﺏ ﺩﺭ ﺟﺎﻣﻌﻪ Population Attributable Risk Percent Rpop. - Ru

Population Attributable Risk (PAR) q Provides information about the excess risk or rate of

Population Attributable Risk Percent (PAR%) Population attributable risk percent provides information about the excess

Summary of Attributable Risk Calculations M. Moradinazar

Slides: 31

Download presentation

ﺑﺮﺍﻭﺭﺩ ﺧﻄﺮ Mehdi Moradinazar Dept. of Epidemiology, Kermanshah University of Medical Sciences, Iran m. moradinazar@gmail. com m. moradinazar@kums. ac. ir M. Moradinazar

M. Moradinazar

Odds vs Probability Odds = number of favorable outcomes number of unfavorable outcomes Probability = number of favorable outcomes total number of outcomes Odds = 1/5 Probability =1/6 M. Moradinazar

Measures in Case-Control Studies Cases control Exposed A B Not Exposed C D Case-control Analysis: OR = A/C = AD B/D BC

Measures in Cohort Studies Disease + Disease - Exposure + A B Exposure - C D Cohort Analysis: OR = A/B = AD C/D BC M. Moradinazar

(Case-control) + (Cases) (Controls) (Exposed) (Cohort) (Not Exposed) A B C D Case-control Analysis: Cohort Analysis: OR = A/C = AD B/D BC OR = A/B = AD C/D BC M. Moradinazar

M. Moradinazar

�� Disease + Disease - Total Exposure + a b a+b Exposure - c d c+d a+c b+d N Total a /(a+b) Risk Ratio = c / (c+d)

Protective Factor 0 OR RR No association 1 Risk Factor OR RR No association M. Moradinazar ∞

The odds ratio is a good estimate of the relative risk when a disease is infrequent. M. Moradinazar

The odds ratio is not a good estimate of the relative risk when a disease is not infrequent. M. Moradinazar

Attributable Risk M. Moradinazar

M. Moradinazar

: ﺩﺭﺻﺪ ﺧﻄﺮ ﻗﺎﺑﻞ ﺍﻧﺘﺴﺎﺏ Attributable Risk Percent Re - R u ARP = * 100 Re M. Moradinazar

: ﺩﺭﺻﺪ ﺧﻄﺮ ﻗﺎﺑﻞ ﺍﻧﺘﺴﺎﺏ ﺩﺭ ﺟﺎﻣﻌﻪ Population Attributable Risk Percent Rpop. - Ru PAR = * 100 Rpop. M. Moradinazar

Population Attributable Risk (PAR) q Provides information about the excess risk or rate of disease in the entire population rather than those just exposed. q Just like AR, this is excess risk/rate of disease, except we’re broadening to the total population compared to the unexposed. q If we believe our association to be causal, we can use PAR to estimate the impact of an exposure on population health. Because of the nature of PAR, it will never be larger than AR in a given population. M. Moradinazar

Population Attributable Risk Percent (PAR%) Population attributable risk percent provides information about the excess incidence in the total population as a percentage of incidence in the total population. M. Moradinazar

M. Moradinazar

Summary of Attributable Risk Calculations M. Moradinazar

M. Moradinazar

Related presentations