Measures of Associa-on
& Measures of Eect
POPM*3240 Epidemiology
Dr. A. Jones-Bi>on
Outline
l
1. Measures of
Associa-on
l
l
l
RelaAve Risk (RR)
Odds RaAo (OR)
Incidence Rate RaAo (IR)
l
l
2. Measures of Eect
2a. Measures of Eect in
Exposed Gro
CROSS-SECTIONAL STUDIES
POPM*3240
Dr. A. Jones-Bi6on
Outline
Review: descrip>ve vs. analy>cal studies
Cross-sec>onal study design
Advantages
Disadvantages
Biases
Example: stress, substance abuse & behavioural
addic>ons in university studen
Introduc)on to Epidemiology
& Historical Approaches
POPM*3240
Session 2
Dr. A. Jones-BiGon
[Related reading: Chapter 1, Gordis text]
1
Outline
p Dening epidemiology
p Epidemiologists of note:
Edward Jenner
n James Lind
n Percival PoG
n Joh
INTERPRETING
MULTIVARIABLE MODELS
POPM*3240 Dr. A. Jones-Bi7on
Outline
Review:
Confounding
Methods to control for confounding
Dependent / outcome variables
Independent / explanatory variables
Linear regression
LogisNc regression
PracNce
1
STATISTICAL ANALYSES OF
EPIDEMIOLOGICAL DATA:
A REVIEW
POPM*3240
Session 5
Dr. A. Jones-BiEon
2
Outline
Use of staLsLcs in Epi
Means
Central Limit Theorem
t-tests
ProporLons and RaLos
Chi-square tests
p-values
Null Hypothesis
Con
INTRODUCTION TO
INFECTIOUS DISEASE MODELING
Kelsey Spence, PhD Student
March 18th, 2016
With acknowledgements to Dr. Amy Greer for use of some slides
Outline
Infectious diseases are different!
o Reproductive numbers
o Adding time to an epidemic: serial in
Screening & Diagnos.c Tests -
Part II
POPM*3240 Epidemiology
Dr. A. Jones-Bi>on
(with acknowledgements to Dr. Sandi Lefebvre for use of slides)
Recall: Deni.on of a test
lA [email protected][email protected], [email protected] or
[email protected]
lTests are not just conned
Measuring Disease
Frequency
POPM*3240
Dr. A. Jones-Bi6on
With acknowledgments to Dr. Nathan Lachowsky for use of some slides
References:
l
Epidemiology, 5th ediIon, l Veterinary Epidemiology, by
MarIn, Meek & Willeberg
by Gordis (course text)
l
Founda'ons in Health and
Disease Transmission
POPM*3240
Sessions 3 & 4
Dr. A. Jones-BiCon
Related reading: Chapter 2, Gordis text
Outline
n
n
n
n
n
n
n
Health and Disease
Epidemiological Triad
Infec'ous disease transmission
Development of dise
Case-Control Studies
Epidemiology POPM*3240
Dr. A. Jones-Bi>on
Reference:
Gordis, Chapter 10 (up to p. 203)
& Chapter 13 (p. 239-242)
Recall:
ObservaQonal studies
Natural experiments where the invesQgator does
not manipulate treatment or expo
Critical Appraisal
POPM*3240
Dr. A. Jones-Bitton
Outline
Definitions
Critical appraisal
Evidence based medicine
10 questions to ask in a critical appraisal
Working through an example
Critical Appraisal
Systematic process used to identify
the strengt
Screening & Diagnos.c Tests
Part I
POPM*3240 Epidemiology
Dr. A. Jones-Bi>on
Outline
lScreening vs. DiagnosCc tests
l2 x 2 tables
lTrue and Apparent Prevalence
lSensiCvity and Specicity
lFalse NegaCves and False PosiCves
lPosiCve and NegaCve P
Bias & Causa)on
POPM*3240 Epidemiology
Dr. A. Jones-Bi>on
Addi)onal References:
NOTE: this lecture presents some of the more
common biases not an exhausJve list!
Good discussion and review quesJons in Gordis
text!
This may be (probably is?) a t
POPM* 3240 Midterm I Notes
Session 6: Sampling I
SIGNIFICANCE OF SAMPLING
Census: the whole population is surveyed
- E.g. Statistics Canada Population Census
Sample: a portion of the population is surveyed
- E.g. 10% of Ontario residents are randomly
POPM*3240 Final Exam Notes
Critical Appraisal
DEFINING CRITICAL APPRAISAL
Systematic process used to identify the strengths and weaknesses of a research article in order to
assess the usefulness and validity of research findings
Application of ru
POPM*3240 Midterm II Notes
Measures of Association
RELATIVE RISK OR RISK RATIO (RR)
Calculating RR
Relative Risk = _Risk of disease in exposed group_
Risk of disease in non-exposed group
Risk in Exposure + = a/(a+b)
Risk in Exposure - = c/(c+d)
RR= [a
NAME: _
Student ID #:_
University of Guelph
Department of Population Medicine
Midterm Examination I Epidemiology (POPM*3240)
Course Coordinator: Dr. Andria Jones-Bitton
February 10, 2016
This exam is out of 30 marks and is worth 20% of the total course gr
NAME: _
Student ID #:_
University of Guelph
Department of Population Medicine
Midterm Examination I Epidemiology (POPM*3240)
Course Coordinator: Dr. Andria Jones-Bitton
October 6, 2014
This exam is out of 31 marks and is worth 20% of the total course grad
Name: _
Student ID#: _
University of Guelph
Department of Population Medicine
Midterm I Examination Epidemiology (POPM*3240)
Course Coordinators: Dr. Andria Jones-Bitton, Shiona Glass-Kaastra
October 5, 2012
This exam is out of 25 marks and is worth 25% o
Last Name: _ First Name: _ Student ID#: _
University of Guelph
Department of Population Medicine
Midterm #2 Examination
POPM*3240
November 1st, 2013
This is exam is out of 25 marks (10 multiple choice, 15 short answer) and is worth
20% of the total course
Sample Size Practice Questions POPM*3240
Formulas will be provided on the midterm/final exams; however, you must know which one
you need to use. No sample size calculations using the Excel spreadsheet will be given on the
midterm/final exams.
Question 1
Y
Calculating Kappa
ADDITIONAL PRACTICE QUESTIONS & ANSWERS
POPM*3240
It is always important to interpret the kappa statistic based on the following
guidelines:
< 0.2
slight agreement
0.2 - 0.4
fair agreemen
Test Properties
ADDITIONAL PRACTICE QUESTIONS & ANSWERS
POPM*3240
These practice sets relate to material introduced in the Screening & Diagnostic
lectures and include: sensitivity, specificity, positive predictive valu
Calculating Sample Sizes to Compare Two Groups - Proportions
What is Z-alpha?
What is Z-beta?
1.96
-0.84
What is p1? (enter as 0.XX) =
What is p2? (enter as 0.XX) =
p=
Sample size per group is approx:
0.6
0.8
0.7
q1 = 0.4
q2 = 0.2
q= 0.3
For cohort studie
POPM*3240 Sampling Practice Assignment #2
1. An investigator wants to determine the average level of cholesterol in men age 45-50 in Canada (with 95%
confidence). She recruits the provinces of Ontario, Quebec, Alberta, and British Columbia into her st
Helpful Notes for Calculating Sample Sizes
The Z 2pq/L2 and Z 2 2/L2 sample size formulae are just used for estimating a
proportion or mean in one population, respectively. If a study (e.g. a crosssectional/observational study) was just looking to est
Name: _
Student ID#: _
University of Guelph
Department of Population Medicine
Midterm II Examination Epidemiology (POPM*3240)
Course Coordinators: Dr. Andria Jones-Bitton, Shiona Glass-Kaastra
November 2, 2012
This exam is out of 25 marks and is worth 25%
Name:_Student ID#:_
University of Guelph
Department of Population Medicine
Midterm Examination POPM*3240
October 12, 2011
This exam is out of 30 marks and is worth 25% of the total course grade
Instructions:
1. Print your name and student identification n
Unit 6: Observational Study Design
Descriptive vs. Analytical Studies
Descriptive Studies: Collecting information (frequency/ distribution) about disease without
establishing an association between cause and effect. Good for generating hypotheses
Case re
Unit 4- Measures of Association and Disease Frequency
Ratios: Dimensionless, unitless comparison of two counts (1:2)
Proportions: Ratio where number is included in the denominator (1/100, 10%, 0.11)
Risks: proportions that express the probability that som