PH6415 Review Questions. 2 Question 1 A journal article reports a 95%CI for the relative risk (RR) of an event (treatment versus control as (0.55, 0.97).

Slides:



Advertisements
Similar presentations
Logistic Regression I Outline Introduction to maximum likelihood estimation (MLE) Introduction to Generalized Linear Models The simplest logistic regression.
Advertisements

Logistic Regression.
Simple Logistic Regression
Introduction to Survival Analysis October 19, 2004 Brian F. Gage, MD, MSc with thanks to Bing Ho, MD, MPH Division of General Medical Sciences.
HSRP 734: Advanced Statistical Methods July 24, 2008.
SC968: Panel Data Methods for Sociologists
Multiple Logistic Regression RSQUARE, LACKFIT, SELECTION, and interactions.
April 25 Exam April 27 (bring calculator with exp) Cox-Regression
Logistic Regression Multivariate Analysis. What is a log and an exponent? Log is the power to which a base of 10 must be raised to produce a given number.
1 Statistics 262: Intermediate Biostatistics Kaplan-Meier methods and Parametric Regression methods.
Some Terms Y =  o +  1 X Regression of Y on X Regress Y on X X called independent variable or predictor variable or covariate or factor Which factors.
the Cox proportional hazards model (Cox Regression Model)
Chapter 11 Survival Analysis Part 3. 2 Considering Interactions Adapted from "Anderson" leukemia data as presented in Survival Analysis: A Self-Learning.
Survival analysis1 Every achievement originates from the seed of determination.
Proportional Hazard Regression Cox Proportional Hazards Modeling (PROC PHREG)
Chapter 11 Survival Analysis Part 2. 2 Survival Analysis and Regression Combine lots of information Combine lots of information Look at several variables.
EPI 809/Spring Multiple Logistic Regression.
Nemours Biomedical Research Statistics April 23, 2009 Tim Bunnell, Ph.D. & Jobayer Hossain, Ph.D. Nemours Bioinformatics Core Facility.
Notes on Logistic Regression STAT 4330/8330. Introduction Previously, you learned about odds ratios (OR’s). We now transition and begin discussion of.
Adjusting for extraneous factors Topics for today More on logistic regression analysis for binary data and how it relates to the Wolf and Mantel- Haenszel.
Introduction to Survival Analysis PROC LIFETEST and Survival Curves.
Logistic regression for binary response variables.
Survival analysis. First example of the day Small cell lungcanser Meadian survival time: 8-10 months 2-year survival is 10% New treatment showed median.
Assessing Survival: Cox Proportional Hazards Model Peter T. Donnan Professor of Epidemiology and Biostatistics Statistics for Health Research.
Survival Analysis A Brief Introduction Survival Function, Hazard Function In many medical studies, the primary endpoint is time until an event.
Analysis of Complex Survey Data
STAT E-150 Statistical Methods
Survival analysis Brian Healy, PhD. Previous classes Regression Regression –Linear regression –Multiple regression –Logistic regression.
1 Kaplan-Meier methods and Parametric Regression methods Kristin Sainani Ph.D. Stanford University Department of Health.
Logistic Regression II Simple 2x2 Table (courtesy Hosmer and Lemeshow) Exposure=1Exposure=0 Disease = 1 Disease = 0.
Logistic Regression III: Advanced topics Conditional Logistic Regression for Matched Data Conditional Logistic Regression for Matched Data.
Essentials of survival analysis How to practice evidence based oncology European School of Oncology July 2004 Antwerp, Belgium Dr. Iztok Hozo Professor.
1 Survival Analysis Biomedical Applications Halifax SAS User Group April 29/2011.
Survival Data John Kornak March 29, 2011
G Lecture 121 Analysis of Time to Event Survival Analysis Language Example of time to high anxiety Discrete survival analysis through logistic regression.
Dr Laura Bonnett Department of Biostatistics. UNDERSTANDING SURVIVAL ANALYSIS.
Biostatistics Case Studies 2005 Peter D. Christenson Biostatistician Session 4: Taking Risks and Playing the Odds: OR vs.
April 11 Logistic Regression –Modeling interactions –Analysis of case-control studies –Data presentation.
Assessing Survival: Cox Proportional Hazards Model
Analyses of Covariance Comparing k means adjusting for 1 or more other variables (covariates) Ho: u 1 = u 2 = u 3 (Adjusting for X) Combines ANOVA and.
Excepted from HSRP 734: Advanced Statistical Methods June 5, 2008.
Time-dependent covariates and further remarks on likelihood construction Presenter Li,Yin Nov. 24.
April 6 Logistic Regression –Estimating probability based on logistic model –Testing differences among multiple groups –Assumptions for model.
2 December 2004PubH8420: Parametric Regression Models Slide 1 Applications - SAS Parametric Regression in SAS –PROC LIFEREG –PROC GENMOD –PROC LOGISTIC.
Applied Epidemiologic Analysis - P8400 Fall 2002
AN INTRODUCTION TO LOGISTIC REGRESSION ENI SUMARMININGSIH, SSI, MM PROGRAM STUDI STATISTIKA JURUSAN MATEMATIKA UNIVERSITAS BRAWIJAYA.
1 היחידה לייעוץ סטטיסטי אוניברסיטת חיפה פרופ’ בנימין רייזר פרופ’ דוד פרג’י גב’ אפרת ישכיל.
LOGISTIC REGRESSION A statistical procedure to relate the probability of an event to explanatory variables Used in epidemiology to describe and evaluate.
HSRP 734: Advanced Statistical Methods July 17, 2008.
Introduction to Survival Analysis Utah State University January 28, 2008 Bill Welbourn.
April 4 Logistic Regression –Lee Chapter 9 –Cody and Smith 9:F.
Applied Epidemiologic Analysis - P8400 Fall 2002 Lab 9 Survival Analysis Henian Chen, M.D., Ph.D.
Lecture 12: Cox Proportional Hazards Model
1 Multivariable Modeling. 2 nAdjustment by statistical model for the relationships of predictors to the outcome. nRepresents the frequency or magnitude.
Multiple Logistic Regression STAT E-150 Statistical Methods.
We’ll now look at the relationship between a survival variable Y and an explanatory variable X; e.g., Y could be remission time in a leukemia study and.
Applied Epidemiologic Analysis - P8400 Fall 2002 Labs 6 & 7 Case-Control Analysis ----Logistic Regression Henian Chen, M.D., Ph.D.
Logistic regression (when you have a binary response variable)
Probability and odds Suppose we a frequency distribution for the variable “TB status” The probability of an individual having TB is frequencyRelative.
Applied Epidemiologic Analysis - P8400 Fall 2002 Labs 6 & 7 Case-Control Analysis ----Logistic Regression Henian Chen, M.D., Ph.D.
1 “The Effects of Sociodemographic Factors on the Hazard of Dying Among Aged Chinese Males and Females” Dudley L. Poston, Jr. and Hosik Min Department.
Analysis of matched data Analysis of matched data.
Additional Regression techniques Scott Harris October 2009.
Marshall University School of Medicine Department of Biochemistry and Microbiology BMS 617 Lecture 13: Multiple, Logistic and Proportional Hazards Regression.
April 18 Intro to survival analysis Le 11.1 – 11.2
Multiple logistic regression
Categorical Data Analysis Review for Final
Logistic Regression.
Introduction to Logistic Regression
Presentation transcript:

PH6415 Review Questions

2 Question 1 A journal article reports a 95%CI for the relative risk (RR) of an event (treatment versus control as (0.55, 0.97). What can be said of the p-value associated with testing Ho: RR=1 vs. Ha: RR not equal 1? A journal article reports a 95%CI for the relative risk (RR) of an event (treatment versus control as (0.55, 0.97). What can be said of the p-value associated with testing Ho: RR=1 vs. Ha: RR not equal 1? The p-value is < The p-value is < The p-value is < The p-value is < The p-value is > 0.05 The p-value is > 0.05 No statement can be said about the p-value. No statement can be said about the p-value.

3 Question 2 If S (t) is the survival function and t is in years what is the meaning of S(3). If S (t) is the survival function and t is in years what is the meaning of S(3). The probability of dying at year 3. The probability of dying at year 3. The probability of surviving to year 3. The probability of surviving to year 3. The probability of dying by year 3 The probability of dying by year 3 The hazard of dying at year 3. The hazard of dying at year 3.

4 Question 3 In logistic regression with a continuous variable age what is the meaning of  1 ? In logistic regression with a continuous variable age what is the meaning of  1 ? The difference in log odds between two persons 1 year apart in age The difference in log odds between two persons 1 year apart in age The relative odds between two persons 1 year apart in age The relative odds between two persons 1 year apart in age The difference in probabilities between two persons 1 year apart in age The difference in probabilities between two persons 1 year apart in age

5 Question 4 If the probability of developing diabetes is 0.20 among Hispanics and 0.15 among whites, what is the relative odds (Hispanics v white) of developing diabetes. If the probability of developing diabetes is 0.20 among Hispanics and 0.15 among whites, what is the relative odds (Hispanics v white) of developing diabetes

6 Question 5 Suppose the logistic regression model: log odds =  0 +  1 X 1 +  2 X 2 +  3 X 1 *X 2 where X 1 is an indicator for treatment and X 2 is an indicator for male gender. The relative odds (treatment versus no treatment) for women is: Suppose the logistic regression model: log odds =  0 +  1 X 1 +  2 X 2 +  3 X 1 *X 2 where X 1 is an indicator for treatment and X 2 is an indicator for male gender. The relative odds (treatment versus no treatment) for women is: exp(  1 ) exp(  1 ) exp(  2 ) exp(  2 ) exp(  1 +  3 ) exp(  1 +  3 ) Exp(  1 -  3 ) Exp(  1 -  3 )

7 Question 6 The probability and odds of an event will be nearly equal if: The probability and odds of an event will be nearly equal if: The probability of the event is small The probability of the event is small The probability of the event is large The probability of the event is large The probability of the event is 0.50 The probability of the event is 0.50

8

9 Cox Proportional Hazards Regression in SAS A Review Goto: C Read uis.readme file on website. Read uis.readme file on website. Input data from uis (SAS data set) Input data from uis (SAS data set) Use Proc Lifetest to plot the Kaplan-Meier Curve for each categorical predictor separately. Look to see if the survival curves are approximately parallel and if there appears to be a difference in survival. Use Proc Lifetest to plot the Kaplan-Meier Curve for each categorical predictor separately. Look to see if the survival curves are approximately parallel and if there appears to be a difference in survival. Use Proc PHREG to with model containing age, number of previous drug treatments, treatment and site. Use Proc PHREG to with model containing age, number of previous drug treatments, treatment and site.

10 Cox Proportional Hazards Regression in SAS A Review C Consider the interaction between age and site. Is this interaction significant? Consider the interaction between age and site. Is this interaction significant? Consider the final model of age, number of previous drug treatments, site and age_site. Consider the final model of age, number of previous drug treatments, site and age_site.

11 Questions About the Survival Curves What does the log-rank test of equality across strata indicate for the survival curves of the short and long treatment programs? What does the log-rank test of equality across strata indicate for the survival curves of the short and long treatment programs? What does the log-rank test of equality across strata indicate for the survival curves of the two different sites? Why might the p-value for the log-rank test be inflated? What does the log-rank test of equality across strata indicate for the survival curves of the two different sites? Why might the p-value for the log-rank test be inflated? What does the log-rank test of equality across strata indicate for the three combinations of heroine and cocaine use? Do the curves overlap? What does the log-rank test of equality across strata indicate for the three combinations of heroine and cocaine use? Do the curves overlap?

12 Questions about Survival Data What is the median time to relapse for those at site A? What is the median time to relapse for those at site B? What is the median time to relapse for those at site A? What is the median time to relapse for those at site B? How many people relapsed at site A? What percent of site A relapsed? How many people relapse at site B? What percent of site B relapsed? How many people relapsed at site A? What percent of site A relapsed? How many people relapse at site B? What percent of site B relapsed? When did the first person relapse at site A? When did the first person relapse at site B? When did the first person relapse at site A? When did the first person relapse at site B?

13 Questions about Censoring What percent of people where censored in the long treatment program compared to the short treatment? What percent of people where censored in the long treatment program compared to the short treatment? For both treatment groups, does that censoring appear to be patients who do not relapse or patients who are loss to follow-up? For both treatment groups, does that censoring appear to be patients who do not relapse or patients who are loss to follow-up?

14 Questions about parameters in Cox Proportional Hazards What is the relative risk of relapse for a one unit increase in previous drug treatments if all other variables are held constant? This represents a ________ percent increase in rate of relapse. What is the relative risk of relapse for a one unit increase in previous drug treatments if all other variables are held constant? This represents a ________ percent increase in rate of relapse. If treatment length is altered from short(trt =0) to long (trt=1), while holding all other variables constant, the rate of relapse decreases by ______ percent. (RR of trt=1/trt=0). If treatment length is altered from short(trt =0) to long (trt=1), while holding all other variables constant, the rate of relapse decreases by ______ percent. (RR of trt=1/trt=0).

15 Considering Interactions in Cox Proportional Hazards What is the relative risk of relapse for a person who is 30 compared to 25 if they are at site A (site=0) with all other variables held constant? This translates to a _____ percent decrease in rate of relapse. What is the relative risk of relapse for a person who is 30 compared to 25 if they are at site A (site=0) with all other variables held constant? This translates to a _____ percent decrease in rate of relapse. What is the relative risk of relapse for a person who is 30 compared to 25 if they are at site B (site=1) with all other variables held constant? This translates to a _____ percent decrease in relapse. What is the relative risk of relapse for a person who is 30 compared to 25 if they are at site B (site=1) with all other variables held constant? This translates to a _____ percent decrease in relapse. Is this difference in rate of relapse for a five year increase in age between the two sites significant? Is this difference in rate of relapse for a five year increase in age between the two sites significant?

16 Logistic Regression Review Can age, educational level and gender (female=1) predict the odds that someone votes for a particular candidate? Let  = proportion of voters who vote for candidate “Superman”. Can age, educational level and gender (female=1) predict the odds that someone votes for a particular candidate? Let  = proportion of voters who vote for candidate “Superman”. Model: Model:

17 Logistic Regression Review The following is a sample of logistic output: The following is a sample of logistic output: df bSE(b) X2X2X2X2P-value Intercept Age Education Gender

18 Questions for Logistic What is the equation of the estimated Log(odds)? What is the equation of the estimated Log(odds)? What do we predict the odds to be for a 35 year-old male with 16 years of school? What do we predict the odds to be for a 35 year-old male with 16 years of school? What is the probability a 35 year-old male with 16 years of school will vote for “Superman”? What is the probability a 35 year-old male with 16 years of school will vote for “Superman”? What is the odds a woman will vote for “Superman” compared to a man (all other covariates held fixed)? What is the odds a woman will vote for “Superman” compared to a man (all other covariates held fixed)?

19 TOMHS Example Question: Does the effect of active blood pressure treatment on CVD differ for young versus older persons? Question: Does the effect of active blood pressure treatment on CVD differ for young versus older persons? Looking at an interaction effect (effect modification) Looking at an interaction effect (effect modification) Compare Compare Odds CVD (treatment/placebo) in younger patients Odds CVD (treatment/placebo) in younger patients Odds CVD (treatment/placebo) in older patients Odds CVD (treatment/placebo) in older patients

20 Logistic Model For Interaction X1 = 1 for active treatment and 0 for placebo X2 = 1 for age ≥ 55 and 0 for age < 55 X3 = X1 * X2 So, X3 = 1 for active treatment and age > 55 X3 = all other combinations. X3 = all other combinations.

21 Logistic Model For Interaction X1 = 1 for active treatment and 0 for placebo X2 = 1 for age ≥ 55 and 0 for age < 55 X3 = X1 * X2 Log Odds (placebo, young) =  0 Log Odds (active, young) =  0 +  1 Log Odds (placebo, old) =  0 +  2 Log Odds (active, old)=  0 +  1 +  2 +  3 Dif =  1 ; exp(  1 ) is odds (A v P) for young Dif =  1 +   ; exp(  1 +  3 ) is odds (A v P) for old

22 Log Odds (placebo, young) =  0 Log Odds (active, young) =  0 +  1 Log Odds (placebo, old) =  0 +  2 Log Odds (active, old)=  0 +  1 +  2 +  3 exp(  1 ) is odds (A v P) for young exp(  1 +  3 ) is odds (A v P) for old What does  3 Mean? = Odds (A v P) for Oldexp(  1 +  3 ) Odds (A v P) for Youngexp (  1 ) exp (  3 ) = A ratio of ratios!!

23 Interaction Hypothesis Q: Does the effect of active treatment on CVD differ for young versus older persons? Ho:  3 = 0 Ha:  3 ≠ 0 Test in SAS just like any other coefficient

The LOGISTIC Procedure Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq Intercept <.0001 active old active_old Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits active old active_old b1 b2 b3 Odds CVD (A v P) for younger patients = exp(b1) = Odds CVD (A v P) for older patients = exp(b1 + b3) = exp(-0.11) = = 0.90/.415 Ratio of Odds Ratios

25 In patients < age 55 the CVD risk was 58% lower in the active treatment (OR: 0.42) – Exp(b1) For patients over 55 years of age the CVD risk was only 10% lower (OR:.90). - Exp(b1+b3) The test for interaction between treatment and age approached significance (p=.07). Description of Findings