1 / 30

Strenght of association

Strenght of association. Absolute, Relative and Attributable Risks. Hypothesis testing in medicine. Outcomes or differences that we are interested in: Differences in means or proportions Odds ratio (OR ) – association of two variables Relative Risk (RR ) – association of two variables

alexa-brooks
Download Presentation

Strenght of association

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Strenghtofassociation Absolute, Relative and Attributable Risks

  2. Hypothesistestingin medicine Outcomes or differences that we are interested in: • Differences in means or proportions • Odds ratio (OR) – association of two variables • Relative Risk (RR) – association of two variables • Correlation coefficient – association of two variables

  3. Hypothesistestingin medicine Outcomes or differences that we are interested in: • Differences in means or proportions • Odds ratio (OR) – association of two variables • Relative Risk (RR) – association of two variables • Correlation coefficient – association of two variables

  4. Associationoftwovariables • Quantitative • Correlation coefficient • Qualitative • Chisquare (χ2) test, McNemartest • Oddsratio, relativerisk

  5. Associationoftwoquantitativevariables:Patterns of Scatter Diagrams… Linearity and Direction are two concepts we are interested in Positive Linear Relationship Negative Linear Relationship Weak or Non-Linear Relationship

  6. Associationoftwoquantitativevariables: Correlationcoefficient Correlationcoefficient is themeasureofdirectionandstrengthofassociations!

  7. Associationoftwoquantitativevariables: Correlationcoefficient

  8. Example 1 • A nurse wanted to be able to predict the laboratory HbA1c result (ameasure of blood glucose control) from the fasting blood glucoses whichshe measured in her clinic. On 12 consecutive diabetic patients she notedthe fasting glucose and simultaneously drew blood for HbA1c.

  9. Example 2 • An occupational therapist developed a scale for measuring physicalactivity and wondered how much it correlated to Body Mass Index (BMI)in 12 of her adult patients.

  10. Associationoftwoqualitativevariables: • Cross Tableis usedto calculateassociationof twoqualitative variables • If first variable has r categories, second variable c categories, then we have an r×c cross table.

  11. Associationoftwoqualitativevariables: 1. Build a Cross Table Cross Table – associationsof YPEL5 genotypes withdisease X

  12. Associationofthediseasewiththeriskfactor: 1. Build a Cross Table Input for calculationofrisk, relativerisk (RR), attributable risk, oddsratio (OR)

  13. Associationoftwoqualitativevariables: Risk Risk is the probability that an eventwillhappen. Riskofgeting a diseaseintheexposure group: a/(a+b) Riskofgeting a diseaseinthe no-exposure group: c/(c+d) People at risk

  14. Risk • If one in every 100 patients suffers a side-effect froma treatment, the risk is • 1⁄100 = 0.01

  15. Riskratio (RR) • Calculatedby dividing therisk in the treated or exposed group by the risk in thecontrol or unexposed group. • RR=1 - no difference in riskbetween the groups • RR>1 - the rate oftheevent is increased compared to • controls. • RR<1 - the rate of theevent is reducedcompared to • controls.

  16. Riskratio (RR) – Watchout for… Always check for 95% CI of RR!!! If 95% CI for a risk ratio does not include 1(no difference in risk), it is statistically significant. • RR=1 - no difference in riskbetween the groups • RR>1 - the rate oftheevent is increased compared to • controls. • RR<1 - the rate of theevent is reducedcompared to • controls.

  17. Example A cohort of 1000 regular football players and 1000 non-footballers werefollowed to see if playing football was significant in the injuries that theyreceived.After 1 year of follow-up there had been 12 broken legs in the footballplayers and only four in the non-footballers. The risk of a footballer breaking a leg was therefore 12/1000 or 0.012. Therisk of a non-footballer breaking a leg was 4/1000 or 0.004. The risk ratio of breaking a leg was therefore 0.012/0.004 which equals 3 The 95% CI was calculated to be 0.97 to 9.41. As the CI includes thevalue 1 we cannot exclude the possibility that there was no difference inthe risk of footballers and non-footballers breaking a leg. However, giventhese results further investigation would clearly be warranted.

  18. Relativerisk (RR) • Usedin “cohort studies” • Prospectivestudies that follow a group (cohort) over a period oftime and investigate the effect of a treatment or riskfactor.

  19. Oddsratio (OR) • Used by epidemiologists in studies looking for factorswhich do harm • Itis a way of comparing patientswho already have a certain condition (cases) withpatients who do not (controls) – a “case–controlstudy”. • For rare events its value approximates that of therelative risk (RR)

  20. Odds • Calculatedby dividing thenumber of times an event happens by the number oftimes it does not happen. • Oddsofcasesbeingexposed: a/c • Oddsofcontrols beingexposed: b/d

  21. Odds • One boy is born for every two births, so the odds ofgiving birth to a boy are 1:1 (or 50:50) = 1⁄1 = 1 • If one in every 100 patients suffers a side-effect froma treatment, the odds are1:99 = 1⁄99 = 0.0101

  22. Oddsratio (OR) • Calculatedby dividing the odds of having been exposed to a risk factor bythe odds in the control group. • OR=1 - no difference in riskbetween the groups (odds are same) • OR>1 - the rate oftheevent is increased inpatientswho have • been exposedto the risk factor. • OR<1 - the rate of theevent is reduced

  23. Oddsratio (OR) – Watchout for… Always check for 95% CI of OR!!! If 95% CI for a odds ratio does not include 1(no difference in odds), it is statistically significant. • OR=1 - no difference in riskbetween the groups (odds are same) • OR>1 - the rate oftheevent is increased inpatientswho have • been exposedto the risk factor. • OR<1 - the rate of theevent is reduced

  24. OR - Levin’s Case-Control Studies of Lung Cancer OR =ad/bc=2.29 Odds for cases (patientswithcancer) beingsmokers are 2.29 timesgreaterthan for controls

  25. RISK REDUCTION ANDNUMBERS NEEDED TOTREAT • Helpfulin trying to work out how worthwhile atreatment is in clinical practice.

  26. Absoluteriskreduction (ARR) andNumberneeded to treat (NNT) • ARR is the difference between the event rate in theintervention group and that in the control group. • NNT is the number of patients who need to betreated for one to get benefit. • RRR is the proportion by which the interventionreducestheevent rate.

  27. ARR = improvement rate in the intervention groupimprovement rate inthe control group = 80% – 60% = 20%

  28. NNT = 1/ARR=1/20%=1/0.2=5 Fivewomen have to be treated for one to get benefit.

  29. The incidence of candidiasis was reduced from 40% with placebo to 20%with treatment , i.e. by half. RRR=20% / 40%=50%

More Related