7. Measures of association new 27.10.2024.pptx

Al-Quds University Faculty Of Medicine Introduction to Public Health (6106223) Measures of Association Lecture 7 2023-2024  Measures of association The central objective of epidemiology is to study the cause of disease---- expo...

Al-Quds University Faculty Of Medicine Introduction to Public Health (6106223) Measures of Association Lecture 7 2023-2024  Measures of association The central objective of epidemiology is to study the cause of disease---- exposure disease What do you compare??? Risk in exposed with the risk of nonexposed----- in a single summary parameter!!! OR (Odds Ratio ) RR ( relative risk ) RD (risk difference ) AR (Attributable risk )  Association Statistical dependence between two or more events, characteristics, or variables An association is present when the probability of occurrence of an event or characteristic, or the quantity of a variable, depends upon the occurrence of one or more other characteristics, or the quantity of one or more variables What is RISK? 1. Risk Difference  What is RISK? probability of developing a condition over a defined period of time # of people developing disease in time period # of disease-free people at start of period # of people developing TB in 5 years # of people free of TB at start of 5 yr period  year follow-up for the-1 development of CHD CHD Total - + 3000 2916 84 + Smoking 5000 4913 87 - Incidence in exposed+ = 84/3000 = 28.0/1000 Incidence in exposed- = 87/5000 = 17.4/1000  CHD Total - + 3000 2916 84 + Smoking 5000 4913 87 - Risk Difference/Attributable Risk Incidence in exposed+ = 84/3000 = 28.0/1000 Incidence in exposed- = 87/5000 = 17.4/1000 RD = 28.0-17.4 = 10.6/1000 Among every 1000 smokers, 10.6 more cases of CHD develop than among 1000 non-smokers  Risk Difference RD > 0 = positive association RD = 0 = no association RD < 0 = negative association 2. Relative risk  Relative Risk Probability of developing disease if risk factor is present Probability of developing disease if risk factor is absent Incidence of stroke in hypertensives Incidence of stroke in normotensives  Relative risk 2x2 table Disease Yes No Yes a b exposure c d No Risk in exposed / risk in non exposed = a / (a+b) c / (c+d)  CHD Total - + 3000 2916 84 + Smoking 5000 4913 87 - Relative Risk or Risk Ratio Incidence in exposed+ = 84/3000 = 28.0/1000 Incidence in exposed- = 87/5000 = 17.4/1000 RR = 28.0÷17.4 = 1.6 Risk for CHD among smokers is 1.6 times higher than among non-smokers  In the conclusion Statement in conclusions should be that smokers have 1.6 more chance to develop CHD than none smokers  Relative Risk RR > 1 = positive association RR = 1 = no association RR < 1 = negative association RR tells us about the strength of association between exposure and outcome but not about the magnitude of absolute risk (incidence)  Example on RR 100 smokers (Exposed) and 100 non-smokers (none exposed) were followed up for 10 years. 80/100 of smokers developed lung cancer while 10/100 have developed lung cancer in nonsmokers (None exposed ) what is the relative risk??? 22 Odds Ratio.3  Case-control study-OR Exposure Disease  Investigator ? ? In a case-control study, participants are selected on the basis of disease status- not possible to calculate risk of disease- but RR can be estimated by calculating the ratio of the odds of exposure among cases to that among controls. Odds of exposure among cases OR = OR~RR Odds of exposure among controls Odds Ratio (or Relative Odds) Odds: The ratio of the probability of occurrence of an event to that of nonoccurrence (Last, DoE) The ratio of the probability that something will occur, to the probability that it will not occur (Abramson, MSoD)  What is odd ratio An odds ratio (OR) is a measure of association between an exposure and an outcome. The OR represents the odds that an outcome will occur given a particular exposure, compared to the odds of the outcome occurring in the absence of that exposure. Odds ratio are most commonly used in case- control studies, however they can also be used in cross-sectional and cohort study designs as well (with some modifications and/or assumptions). 26  Odds ratios are used to compare the relative odds of the occurrence of the outcome of interest (e.g. disease or disorder), given exposure to the variable of interest (e.g. health characteristic, aspect of medical history). The odds ratio can also be used to determine whether a particular exposure is a risk factor for a particular outcome, and to compare the magnitude of various risk factors for that outcome. 27  Odds Ratio OR > 1 = positive association OR = 1 = no association OR < 1 = negative association 30 31 32  Example of OR Interpretation Interestingly, the odds of persistent suicidal behavior in this group given presence of borderline personality disorder at baseline was twice that of depression (OR 3.8, 33  “Odds is understood intuitively‫ حدسي‬only by statisticians and professional gamblers”  Odds Ratio (OR) outcome a b Positive exposure c d  Odds Ratio (OR) outcome a b Negative exposure c d  Odds Ratio (OR) outcome a b exposur e c d Prospective study (Disease OR) Ratio of odds of outcome in exposed (a/b) to the odds of outcome in the unexposed (c/d): a/b ad = c/d bc  CHD Total - + 3000 2916 84 + Smoking 5000 4913 87 - Odds Ratio odds in exposed+ = 84/2916 = 28.8/1000 odds in exposed- = 87/4913 =17.7/1000 OR = 28.8÷17.7 = 1.63 ODDS for CHD among smokers is 1.63 times higher than among non-smokers  …OR  RR when 1. Outcome is rare because then: b  (a + b) and d  (c + d) 2. Cases are representative of all cases in the population re. exposure history 3. Controls representative of all non-cases in the population re. exposure history outcome a b a+b exposur sample community OR in a retrospective study outcome a b exposure c d  Odds Ratio (OR) outcome a b exposure c d  outcome a b exposur e c d Retrospective study (Exposure OR): Ratio of odds of exposure in cases (a/c) to the odds of exposure in controls (b/d): a/c ad = b/d bc  RR & OR OR is a useful measure of association in and of itself, in both case-control and prospective studies In a cohort study, RR can be calculated directly In a case-control study, RR cannot be calculated directly, so OR is used as an estimate of RR when the risk of disease is low  Advantages of Odds Ratio Suitable for case-control studies Identical in both directions Basis for logistic regression analyses  In the conclusion part We say that the city dwelling has 1.68 the odds of developing lung cancer as compered to regional dwelling Exercise  Suicide attempted History of No Yes sexual abuse 9 14 Yes 149 49 No Deykin & Buka (1994) AJPH 84:634-9  Suicide attempted History of Total No Yes sexual abuse 23 9 14 Yes 198 149 49 No Rexp+ = 14/23 = 60.9% Rexp- = 49/198 = 24.7% RR = 60.9/24.7 = 2.46  Suicide attempted History of Total No Yes sexual abuse 23 9 14 Yes 198 149 49 No 14/23 RR = = 2.46 49/198 Suicide attempted History of No Yes sexual abuse 9 14 Yes 149 49 No 14/9 OR = = 4.73 49/149  Question: Why do the OR and RR differ? Answer: high incidence (61% in exposed, 25% in unexposed) therefore b  a + b and d  c + d Attributable risk.4  AR Definition Attributable risk is a calculation used to determine how much of disease can be contributed to a certain exposure or risk factor. It is calculated by taking the difference between incidence of disease due to a risk factor and incidence of disease in the face of no risk factors.  attributable risk or excess risk is a term synonymous to risk difference, that has also been used to denote attributable fraction among the exposed and attributable fraction for the population 58 Population attributable risk.4 (PAR) Population-attributable risk (PAR) is different from AR. – PAR estimates the proportion of disease in the study population that is attributable to the exposure. – In order to calculate PAR, the incidence of exposure in the study population must be known or estimated. – PAR is usually expressed as a percentage  Example on PAR 100 smokers (Exposed) and 100 non-smokers (none exposed) were followed up for 10 years. 80/100 of smokers developed lung cancer while 10/100 have developed lung cancer in nonsmokers (None exposed ) what is the relative risk???