BIST 511: Statistical Inference
Fall Semester, 2016
Homework 5 (Due 12/07/16)
1. Suppose that n randomly selected adult male hypertensive patients are administered a new
blood pressure lowering drug during a clinical trial designed to assess the efficacy

Methods for Summarizing the Evidence
Meta-Analyses and Pooled Analyses
Dr. Jeanine Genkinger and Dr. Christopher Loffredo
Why summarize the evidence?
Evaluate the evidence to date
PubMed search: fruit AND vegetable AND breast
cancer 146 articles
Identify

Downloaded from http:/oem.bmj.com/ on October 28, 2015 - Published by group.bmj.com
CONFOUNDING AND CONFOUNDERS
R McNamee
227
*
Occup Environ Med 2003;60:227234
C
onfounding should always be addressed in studies concerned with causality. When present,
it

BIST 541
Screening and Diagnostic Tests
Sensitivity, Specificity, and
Predictive Value
Screening and Diagnostic Tests
Diagnostic and screening tests are used to detect
the presence or severity of disease in individuals.
Clinicians rely on these tests to

Methods for Summarizing the Evidence:
Meta-Analyses and Pooled Analyses
Based on a lecture by Drs. Jeanine Genkinger and Chris Loffredo
Why summarize the evidence?
Evaluate the evidence to date
PubMed search: fruit AND vegetable AND breast
cancer 146 arti

American Journal of Epidemiology
Copyright 2003 by the Johns Hopkins Bioomberg School of Public Health
All rights reserved
Vol. 158, No.5
Printed in USA.
DOI: 10.1 093/aje/kwgi 75
Obesity and the Risk of Newly Diagnosed Asthma in School-age Children
Fra

American Journal of Epidemiology
Copyright 2003 by the Johns Hopkins Bloomberg School of Public Health
All rights reserved
Vol. 158, No.5
Printed in USA.
DOI: 10.1093/aje/kwg174
Prior History of Allergies and Pancreatic Cancer in the San Francisco Bay A

Principles of Epidemiology:
BIST 541
Fall 2016
Faculty:
Anca D. Dragomir, Ph.D. (Course Director)
Invited Lecturers:
Christopher A. Loffredo, Ph.D.
Chiranjeev Dash, Ph.D.
Introduction to the Course
What is epidemiology?
Relation epidemiology - public hea

Principles of Epidemiology
BIST 541
Disease Associations and Rates
by Time, Place and Person
Descriptive Epidemiology
DISTRIBUTIONS OF DISEASE
OVER TIME
Trends in Disease Rates over Time
Mortality statistics (number of deaths in a defined
period of time)

Principles of Epidemiology
BIST 541
Clinical Trials and
Randomized Controlled Trials
Experimental Studies in
Epidemiology
Called:
Intervention Studies or
Randomized Controlled Trials
Conducted much like the controlled experiments
scientific research
in
Th

650
J Epidemiol Community Health 2000;54:650651
Debate
CoVee, pancreatic cancer and the question of causation
The article by Miquel Porta and his colleagues1 indicating
that pancreatic cancer cases without activating mutations
in the K-ras gene drank sign

BIST 541
Chi-Square Tests
Contingency Tables for Analysis
of Categorical Data
When working with categorical variables
(such as gender) or continuous data grouped
into categories (such as BMI categorized as
underweight, normal, overweight, obese),
we firs

652
J Epidemiol Community Health 2000;54:652653
Exposures, mutations and the history of causality
The article by Trichopoulos and colleagues (in this issue)
raises some interesting problems concerning the complex
relations between exposures, cancer aetiol

GEORGETOWN UNIVERSITY
Department of Biostatistics, Bioinformatics, and Biomathematics
BIST 511: Statistical Inference
Final Exam - Fall Semester, 2013
December 4, 2013
DUE DATE: MIDNIGHT 12/11/2013
Attempt ALL questions in Part A and Part B
ar stu
ed d
vi

BIST 511: Statistical Inference
Fall Semester, 2016
Homework 3 (Due 11/16/16)
1. A certain company manufactures stitches for coronary bypass graft surgeries. The distribution
of the length X (in feet) of defect-free stitches manufactured by this company i

BIST 511: Statistical Inference
Fall Semester, 2016
Homework 2 (Due 11/09/16)
1. Let X 1 and X 2 be two independent Bernoulli trials, with success probability p, and let
T X 1 X 2 . Using the conditional distribution of X 1 , X 2 given T , show that T is

BIST 511: Statistical Inference
Fall Semester, 2016
Homework 4 (Due 11/29/2016)
1. Suppose that X 1 , X m form a random sample from a parent density
f ( x | ) (1 )( x ) 2 , x 1, 1
(a) Develop an explicit expression for the form of the MP rejection region

Case studies of bias in real life epidemiologic studies
Bias File 2. Should we stop drinking coffee? The story of coffee and pancreatic cancer
Compiled by
Madhukar Pai, MD, PhD
Jay S Kaufman, PhD
Department of Epidemiology, Biostatistics & Occupational He

1997
Salt and Gastric Adenocarcinoma: A Population-Based
Cohort Study in Norway
Krister Sjodahl,1 Chongqi Jia,1,2 Lars Vatten,3 Tom Nilsen,4 Kristian Hveem,3
and Jesper Lagergren1
Unit of Esophageal and Gastric Research, Department of Molecular Medicine a

obesity reviews
doi: 10.1111/j.1467-789X.2011.00952.x
Obesity Comorbidity Diagnostic
obr_952
275.286
Waist-to-height ratio is a better screening tool
than waist circumference and BMI for adult
cardiometabolic risk factors: systematic review
and meta-analy

Introduction to
Propensity Score Methods
BIST 541
Principles of Epidemiology
Fall 2016
Dealing with Confounding
Confounders are variables associated with both the exposure of
interest and the disease under investigation.
The confounders may introduce bi

When to be Concerned
About Confounding Bias
When there are multiple causal
pathways to a disease, i.e. any one of
several risk factors may cause disease
Since most diseases and adverse health
effects can be caused by more than one
exposure or risk facto

Multivariate Behavioral Research, 46:119151, 2011
Copyright Taylor & Francis Group, LLC
ISSN: 0027-3171 print/1532-7906 online
DOI: 10.1080/00273171.2011.540480
A Tutorial and Case Study in Propensity
Score Analysis: An Application to
Estimating the Effec

STATA14
Why use STATA?
Very easy to use
Works on all computers (including Macs!)
Free to download
http:/georgetown.onthehub.com
STATA is very simple to use. The dataset that you import into STATA (the
variables are shown under Variables on the right)

BIST 541
Measures of Disease
Occurrence and Risk
Measures of Disease Frequency in Populations
Occurrence of disease in a single group or population:
Prevalence
Incidence
Attack Rate
Case Fatality
Survival Rate
Compare occurrence of disease in two groups

Principles of Epidemiology
Design and Analysis
Of Clinical Trials
Dr. Christopher Loffredo
Objectives
To understand why randomized controlled
trials (also called clinical trials) are
considered the gold standard of medical
evidence.
To understand the desi

BIST 541
Disease Associations by
Time, Place and Person
Epidemiology
The process by which health problems are:
Detected
Investigated
Analyzed
Scientific basis of Public Health
Epidemiology
The study of the distribution in populations:
of disease
o

Principles of Epidemiology
BIST 541
Clinical Case Series
Cross-Sectional Studies
Case-Control Studies
Types of Epidemiological
Studies
Clinical case series
Cross-sectional study
Case-control study
Cohort study
Randomized clinical
trial
Increasing
value of

Principles of Epidemiology
Design and Analysis
Of Clinical Trials
Dr. Christopher Loffredo
Objectives
To understand why randomized controlled
trials (also called clinical trials) are
considered the gold standard of medical
evidence.
To understand the desi