Oyindamola B. Yusuf (Ph.D) Department of Epidemiology and Medical Statistics, College of Medicine, University of Ibadan, Ibadan, Nigeria 1.

Slides:



Advertisements
Similar presentations
More than two groups: ANOVA and Chi-square. First, recent news… RESEARCHERS FOUND A NINE- FOLD INCREASE IN THE RISK OF DEVELOPING PARKINSON'S IN INDIVIDUALS.
Advertisements

Two-sample tests. Binary or categorical outcomes (proportions) Outcome Variable Are the observations correlated?Alternative to the chi- square test if.
Logistic Regression.
Departments of Medicine and Biostatistics
Learning Objectives Copyright © 2004 John Wiley & Sons, Inc. Bivariate Correlation and Regression CHAPTER Thirteen.
Penalized Maximum Likelihood Logistic Regression
Overview of Logistics Regression and its SAS implementation
Log-linear Analysis - Analysing Categorical Data
© 2009, KAISER PERMANENTE CENTER FOR HEALTH RESEARCH Getting to Know Your Data Basic Data Cleaning Principles.
Programme in Statistics (Courses and Contents). Elementary Probability and Statistics (I) 3(2+1)Stat. 101 College of Science, Computer Science, Education.
Introduction to Logistic Regression. Simple linear regression Table 1 Age and systolic blood pressure (SBP) among 33 adult women.
Modeling Gene Interactions in Disease CS 686 Bioinformatics.
Validation of predictive regression models Ewout W. Steyerberg, PhD Clinical epidemiologist Frank E. Harrell, PhD Biostatistician.
Marshall University School of Medicine Department of Biochemistry and Microbiology BMS 617 Lecture 12: Multiple and Logistic Regression Marshall University.
David Yens, Ph.D. NYCOM PASW-SPSS STATISTICS David P. Yens, Ph.D. New York College of Osteopathic Medicine, NYIT l PRESENTATION.
Categorical Data Analysis School of Nursing “Categorical Data Analysis 2x2 Chi-Square Tests and Beyond (Multiple Categorical Variable Models)” Melinda.
1 Journal Club Alcohol, Other Drugs, and Health: Current Evidence January–February 2014.
Learning Objective Chapter 14 Correlation and Regression Analysis CHAPTER fourteen Correlation and Regression Analysis Copyright © 2000 by John Wiley &
POTH 612A Quantitative Analysis Dr. Nancy Mayo. © Nancy E. Mayo A Framework for Asking Questions Population Exposure (Level 1) Comparison Level 2 OutcomeTimePECOT.
How to use SPSS ? Aug, 17, 2011 Hirohide Yokokawa, M.D., Ph.D. Department of General Medicine, Juntendo University School of Medicine.
Week 6: Model selection Overview Questions from last week Model selection in multivariable analysis -bivariate significance -interaction and confounding.
LOGISTIC REGRESSION A statistical procedure to relate the probability of an event to explanatory variables Used in epidemiology to describe and evaluate.
APPLIED DATA ANALYSIS IN CRIMINAL JUSTICE CJ 525 MONMOUTH UNIVERSITY Juan P. Rodriguez.
April 4 Logistic Regression –Lee Chapter 9 –Cody and Smith 9:F.
Week 5: Logistic regression analysis Overview Questions from last week What is logistic regression analysis? The mathematical model Interpreting the β.
Assessing Binary Outcomes: Logistic Regression Peter T. Donnan Professor of Epidemiology and Biostatistics Statistics for Health Research.
Chapter Thirteen Copyright © 2006 John Wiley & Sons, Inc. Bivariate Correlation and Regression.
A generalized bivariate Bernoulli model with covariate dependence Fan Zhang.
Linear Methods for Classification Based on Chapter 4 of Hastie, Tibshirani, and Friedman David Madigan.
Organization of statistical research. The role of Biostatisticians Biostatisticians play essential roles in designing studies, analyzing data and.
1 Chapter 16 logistic Regression Analysis. 2 Content Logistic regression Conditional logistic regression Application.
Applied Epidemiologic Analysis - P8400 Fall 2002 Labs 6 & 7 Case-Control Analysis ----Logistic Regression Henian Chen, M.D., Ph.D.
Qualitative and Limited Dependent Variable Models ECON 6002 Econometrics Memorial University of Newfoundland Adapted from Vera Tabakova’s notes.
BIOSTATISTICS Lecture 2. The role of Biostatisticians Biostatisticians play essential roles in designing studies, analyzing data and creating methods.
Logistic Regression Saed Sayad 1www.ismartsoft.com.
ALISON BOWLING MAXIMUM LIKELIHOOD. GENERAL LINEAR MODEL.
Exact Logistic Regression
Does your logistic regression model suck?. PERFECTION!
Analysis of matched data Analysis of matched data.
1 G Lect 10M Contrasting coefficients: a review ANOVA and Regression software Interactions of categorical predictors Type I, II, and III sums of.
Marshall University School of Medicine Department of Biochemistry and Microbiology BMS 617 Lecture 13: Multiple, Logistic and Proportional Hazards Regression.
LOGISTIC REGRESSION. Purpose  Logistical regression is regularly used when there are only two categories of the dependent variable and there is a mixture.
Bootstrap and Model Validation
Logistic Regression When and why do we use logistic regression?
EHS Lecture 14: Linear and logistic regression, task-based assessment
Sutapa Agrawal1, Shah Ebrahim1,2
Logistic Regression APKC – STATS AFAC (2016).
Notes on Logistic Regression
Chapter 13 Nonlinear and Multiple Regression
Biosafety Implications of GeneXpert MTB/RIF assay: Experience of the National Tuberculosis Reference Laboratory , Lagos Nwokoye N, Onubogu CC, Nwadike.
Vitamin D insufficiency, preterm delivery and preeclampsia in women with type 1 diabetes – an observational study MARIANNE VESTGAARD1,2,3 , ANNA L. SECHER1,2.
The 1st National Health Sciences Students Congress
Table 1. Structural Form Parameter Estimates of inflation rate
METHODOLOGY OF RESEARCH IN SPORT Goran Sporis, PhD.
Low Level of Pre-Vaccination Measles Antibody among Infants Receiving Measles Immunization In Ilorin,
CJT 765: Structural Equation Modeling
Jeffrey E. Korte, PhD BMTRY 747: Foundations of Epidemiology II
The Influence of Medications on Falls in Community Dwelling Elderly
Logistic Regression.
Lexico-grammar: From simple counts to complex models
Variables Coefficient Prob.
Intermediate methods in observational epidemiology 2008
Logistic Regression.
Research Techniques Made Simple: Interpreting Measures of Association in Clinical Research Michelle Roberts PhD,1,2 Sepideh Ashrafzadeh,1,2 Maryam Asgari.
Nazmus Saquib, PhD Head of Research Sulaiman AlRajhi Colleges
© The Author(s) Published by Science and Education Publishing.
Std. Error of the Estimate
© The Author(s) Published by Science and Education Publishing.
© The Author(s) Published by Science and Education Publishing.
© The Author(s) Published by Science and Education Publishing.
Presentation transcript:

Oyindamola B. Yusuf (Ph.D) Department of Epidemiology and Medical Statistics, College of Medicine, University of Ibadan, Ibadan, Nigeria 1

 Logistic regression is a mathematical method for investigating the association of a binary dependent variable with one or more independent variables that may be binary, categorical or continuous  Widely used in epidemiological/public health research  Understanding of the underlying principles is important  Techniques fail because the algorithm for maximum likelihood does not converge. 2

 A complete separation occurs when the two categories of the outcome variable can be separated perfectly by values of one of the independent variables  Can also occur if there is a coding error or an alternate outcome variable is mistakenly included as a predictor  Allison (2004) described why numerical algorithm fail to converge 3

 Occurs whenever there is a zero in any cell of a 2x2 table  If there is a zero in the 2x2 table formed by that variable and the dependent variable, the ML estimate for the regression coefficient will not exist. 4

 Delete the variables  Collapse the number of variable categories  Do nothing and report likelihood ratio chi- squares  Make exact inference  Use Bayesian techniques 5

 To evaluate convergence issues in published articles that employed logistic regression analysis. 6

 We reviewed published articles between 2004 and 2013 in an African Journal. 7

 A total of 581 articles were published  Forty (6.9%) used binary logistic regression.  Twenty-four (60.0%) stated the use of logistic regression in the methodology,  Three (12.5%) described the procedures  None of the articles assessed model fit  Majority presented insufficient details of the procedures  Convergence occurred in 6(15.0%) of the articles. 8

9

10 Extracted from Desalu et al *complete separation Multivariate logistic regression of ever smoked tobacco and determinant of tobacco smoking in HIV patients

11 Extracted from Kehinde et al 2010 *Quasi complete separation

12 Extracted from Kehinde et al 2010 *Quasi complete separation

 Logistic regression was poorly implemented in studies published between 2004 and 2013  The procedure may not be well understood by researchers  Researchers may be unaware of the problem of convergence or how to deal with it  Researchers need to report the type of logistic regression, how variables were entered into the model, how model fit was assessed, and how the problem of convergence was resolved when it occurred 13

 Allison Paul, (2004) Convergence Problems in Logistic Regression In Altman Micah, Gill Jeff and McDonald Michael. Numerical Issues in Statistical Computing for the social scientist. A Wiley- Inter science Publication. John Wiley & Sons, INC  Desalu O.O, Oluboyo P.O, Olokoba A.B. et al. (2009). Prevalence and determinants of tobacco smoking among HIV patients in North Eastern, Nigeria. Afr. J. Med. Med. Sci. 38:  Kehinde A.O, Baba A, Bakare R.A. et al. (2010). Risk factors for pulmonary tuberculosis among health-workers in Ibadan, Nigeria. Afr. J. Med. Med. Sci. 39: