EPI 512 home Introduction     Target audience     Faculty     Objectives     Instructional philosophy     Course format     Lectures     Problem sets     Learning resources     Textbooks     Supplemental texts     Personal assistance     Study groups     Supercourse EPI 512 on-line     Course website     Dropbox for assignments     Gradebook     Electronic mailing list    Grading Other information     Feedback to instructors     Approved guests     Plagiarism     Students with disabilities     Emergency evacuation Schedule overview

Epidemiology 512 Epidemiologic Methods I Autumn, 2011 Course Syllabus

Introduction

EPI 512 is the first course in a two-course sequence on epidemiologic methods. Everyone who takes EPI 512 should plan to take EPI 513, which is offered in Winter Quarter, as neither course can stand alone as an introduction to the field.

Target audience

The EPI 512-513 sequence is designed mainly for graduate students majoring in Epidemiology, for whom these courses are required. It is assumed that EPI 512-513 graduates will actually be conducting research using epidemiologic study designs in the future. These courses are also open to graduate students from other departments who need an in-depth introduction to epidemiologic methods in order to apply them as research tools in related fields.

Non-Epidemiology majors who need a single introductory course in epidemiology should take EPI 511 instead. Note that EPI 512 alone does not satisfy the distribution requirements of the MPH or MS programs for an introductory course in epidemiology; it must be followed with EPI 513 to meet the requirement.

Faculty

Thomas Koepsell
Professor
Office: F-261F Health Sciences
Telephone: 206-543-8830
E-mail: koepsell@uw.edu

Victoria Holt
Professor
Office: F-346A Health Sciences
Telephone: 206-667-4776
E-mail: vholt@uw.edu

Noel Weiss
Professor
Office: F-262D Health Sciences
Telephone: 206-685-1788
E-mail: nweiss@uw.edu

Kristjana Asbjornsdottir
Teaching Assistant
E-mail: kasbjorn@uw.edu

Alison Rustagi
Teaching Assistant
E-mail: asilvis@uw.edu

Jeff Stanaway
Teaching Assistant
E-mail: stanaway@uw.edu

Tom Koepsell will handle most administrative matters concerning EPI 512, and Victoria Holt will do so for EPI 513.

Course objectives

The primary objective of EPI 512-513 is to help you learn how to conduct good epidemiologic research. Secondary objectives are to help you understand and evaluate research reported by others, and to enable you to apply epidemiologic principles in other health-related areas, including clinical medicine, public health practice, and health policy.

Specific learning objectives for EPI 512-513 correspond to those for the MPH, MS, and PhD programs in Epidemiology. Upon successfully completing these two courses, you should be able to:

• Define and apply measures of disease frequency as they reflect the health status of a population
  
  
• Apply widely accepted criteria for causal inference.
  
  
• Define and calculate measures of association between a given risk factor and disease.
  
  
• Describe and interpret variations in disease frequency according to characteristics of person, place, and time.
  
  
• Describe the strengths and weaknesses of alternative epidemiologic study designs for determining whether a given factor is a determinant of disease risk.
  
  
• Use epidemiologic methods to evaluate public health programs and policies.
  
  
• Describe major sources of bias in epidemiologic research (including selection bias, measurement error, and confounding) and how such biases can be evaluated and reduced.
  
  
• Explain and use the basic terms and methods used in outbreak investigation, infectious disease epidemiology, chronic disease epidemiology, disease prevention research, and evaluation of screening tests.
  
  
• Evaluate effect modification.
  
  
• Critically review scientific literature and synthesize findings across studies.
  
  
• Correctly interpret epidemiologic research and place the findings into proper context in relation to other epidemiologic studies, biological or social processes, and public health implications.
  
  
• Design a randomized trial, cohort study, or case-control study to evaluate whether a certain exposure is causally associated with a certain health outcome.

Instructional philosophy

Although we will be discussing many examples dealing with a variety of diseases, EPI 512 is mainly a course about principles and methods.  Examples used in class are generally chosen to illustrate an underlying concept, and any transfer of factual knowledge about the particular disease in question is coincidental. In fact, some studies used as examples will have been chosen specifically because they were flawed and their findings not to be believed.

Many students find the basic principles and tools of epidemiology to be fairly straightforward and easy to grasp in the abstract.  Nonetheless, recognizing when and how epidemiologic concepts and techniques apply to a certain real-world situation can be surprisingly challenging.  Hence, considerable time during the course will be spent on problem sets, which are intended to help you develop skills at linking theory to practice.  Some problems concern real or hypothetical situations in which topics covered earlier must be applied to solve a study design or data interpretation problem. Others involve working with data. Still others focus on a published paper and raise questions about how the study was designed and conducted.

Some common stumbling blocks with applied problems in epidemiologic methods, including those on problem sets and exams, include:

• Being distracted by another disciplinary perspective.  Past training and experience in clinical medicine, statistics, social science, etc., can sometimes channel one's thinking about a problem away from viewing it as an example of applied epidemiology.  If you find yourself answering a question in a way that draws on your own special knowledge about the topic but that makes little or no use of ideas recently covered in class or readings, you may have missed something.  In the short run, you may have to remind yourself to try to think like an epidemiologist.  In the long run, you will benefit by combining your previous training with new expertise in epidemiologic methods. 
  
  
• New and confusing terminology.  Like any other branch of science, epidemiology has its own terminology or "jargon" that takes some getting used to.  Some words may be new and unfamiliar (e.g., "spurious," or "equipoise").  Sometimes words or phrases from everyday speech have a narrower, more specific meaning in epidemiology (e.g., being "at risk" for disease).  Sometimes two or more words that seem synonymous are used by epidemiologists to refer to different things (e.g., "mortality" and "case fatality").  Sometimes the same thing goes by two or more different names (e.g., "relative risk" and "rate ratio").  And sometimes the same term can refer to different things, depending largely on where an epidemiologist received his/her training (e.g., "attributable risk").  Unfortunately, these are facts of life.  Because confusion about terminology can lead to misunderstanding and inefficiency in communication, care needs to be taken in how key words are used.  Problem sets provide a context in which to learn the language of epidemiology.
  
  
• Lack of a single, obvious, "correct" answer.  Sometimes a decision concerning the design of a study or interpretation of a set of data will be found to depend on a judgment--e.g., whether a certain potential bias would be serious enough to worry about, or whether two numbers are meaningfully different from each other--and differences of opinion are legitimate. In other instances, a study question may be of interest because its potential ambiguity focuses attention on the critical underlying assumptions involved in possible competing answers.
  
  
• Plugging numbers into a formula without really understanding it.  An appealing feature of epidemiologic methods is that the pieces tend to fit together pretty nicely.  Most of the key formulas follow from a few basic ideas and assumptions.  When you first encounter a new formula, it is a good idea to spend some time and effort to understand where it comes from and what main assumptions about the context are involved.  Doing so will lessen your need to depend on rote memorization or on a written list of formulas that is easily lost or forgotten.  It will also reduce the chance that a formula is mistakenly applied to a situation that it does not fit.
  
  
• Inventing a new method based on intuition, while ignoring a standard method.  There is certainly room for creativity in epidemiology, but it is not usually necessary or efficient to invent a new method to solve a problem for which there are already good, well-known solutions.  When working on a problem set, take advantage of recent lectures, notes, and readings on the topic.  If you do feel the need to develop another approach, if is a good idea to show that you are aware of existing methods and can demonstrate that your method is better.
  

As will soon become obvious, we as your instructors are fallible. In fact, our experience is that some of the deepest subtleties of epidemiologic methods tend to surface when one is standing in front of a large group of people. Sometimes we may offer an answer to a question based on intuition, only to decide on further reflection that it was the wrong answer. At other times we may need to admit bafflement and defer the question pending some further study. At still other times, you or a clear-thinking classmate may be able to provide the answer, or at least prevent dissemination of misinformation. In any event, we view interaction in class as instructionally useful and encourage you to raise questions.

Course format

Class sessions in EPI 512 include a mixture of lecture and discussion of a problem set that was distributed previously.

Lectures

Lectures on new material will take place in our main classroom, T-625.  For most lectures, a handout based on the lecture slides will be available in electronic form as a PDF file for downloading from the course website at least a day before class.  Paper copies of these handouts will not be routinely distributed in class (except for the first day).  Instead, you should download the handout for each upcoming lecture yourself and bring it to class with you.

Probably the easiest way to take notes during lecture is to bring a printed copy of the handout to class and write your notes on it.  If you prefer to go paperless and have a laptop computer that you can bring to class, you can add your own notes electronically to a PDF file by using the Comment feature in recent versions of Adobe Reader.

Most lectures in EPI 512 will be audio recorded, and the recordings will be available on the course website as MP3 files.  MP3 files can be played back on most personal computers and on various portable audio devices (e.g., iPods and some cell phones).  The audio recordings are intended for occasional use when you must miss class due to illness or out-of-town travel, or if you want to review a point covered in lecture.  The recordings are not intended as a regular substitute for attending class because there is no opportunity to ask questions, recordings and handouts do not capture all of the visual features of the slides, there is always a risk of technical problems with the recordings, and you miss the collegial experience of learning side by side  and interacting with fellow students. In general, you will need to have a copy of the lecture handout in front of you to make sense of an audio recording of the lecture.

Handouts for sessions beyond the next one are not usually available in advance because we may still be working on them and like to reserve the option to make changes based on what has happened in class before they are distributed.

As a courtesy to your fellow students, please turn off cell phones and audible pagers during class.

Problem sets

Written answers are to be submitted on-line for some problem sets but not for others, as indicated on the course schedule and at the top of each set of questions.  Regardless of whether written answers are required, you should work through each problem set before it is scheduled for discussion in class and be prepared to contribute to the discussion.

On several Tuesday mornings, the class will divide up for the first hour into four smaller groups to discuss a problem set.  Consult the small-group roster on the course website to see which group you are in.  The Daily Details page on the website shows room assignments for each day.  The faculty and teaching assistants will rotate as discussion leaders among all four groups.  You may be asked to sign in on a checklist each time your small group meets so that the discussion leader knows who is present.  (Signing in is for informational purposes only and is not used in grading.)  On most Thursday mornings and occasional Tuesdays, problem set discussions will take place as one large group in our main classroom. 

Discussion of problem sets in class is intended to be interactive, involving everyone in thinking through the methodological issues raised by a problem. To that end, we often call on individual students at random to initiate discussion on a problem by proposing their own answer. Calling on students at random promotes broader participation in the discussion, which otherwise tends to be dominated by a few vocal volunteers.  It also provides an incentive for everyone to be prepared for class, even if no written answers are required that day. At random is literally true: a computer program is used to generate a random list of students' names, giving everyone an equal chance of being called upon. The random selection is carried out with replacement, so your probability of being called on for a given question is unaffected by whether you were called on for another question recently.

There is no penalty for giving a "wrong" answer in class. (Some of the problems are intended to be a bit tricky.) You may "pass" if you are called upon for a specific problem but do not want to offer your answer to that problem.  If you are painfully shy, you can opt out from being called on by requesting that your name not be included on the list from which random names are chosen. Contact Tom Koepsell to do so.

If you prefer to be called by a different name than the official one by which you are known to the Registrar's Office (e.g., you prefer to be called Rob rather than Robert), just let us know.

Past EPI 512 students have occasionally complained when some of their classmates seem to ask more than their share of questions during class and dominate the discussion.  If you have already asked a question or two in class on a certain day, please consider holding back and giving others a chance to do so.  From time to time, the small-group discussion leader (often a TA) may deliberately skip over a hand that has been raised before, in order to broaden participation and manage the discussion.  We are always glad to talk with you individually after class or during office hours if you have more questions or comments after discussion of a problem set.

From time to time, optional additional problems (with answers) will be posted on the course website.  These optional problems provide opportunities for extra practice, especially on topics that may be unfamiliar or difficult.  Some of these optional problems are a bit more challenging than those in the required problem sets.

Learning resources

Textbooks

The required textbook is:

Koepsell TD, Weiss NS. Epidemiologic Methods: Studying the Occurrence of Illness. New York: Oxford University Press, 2003.

This book was written by Tom and Noel specifically for use in EPI 512-513 and in similar courses at other universities. We hope you like it.  Work is currently in progress on a second edition, and suggestions for improvement to the book are always welcome.

A few errors escaped detection during proofreading of the current edition and made their way into print.  A list of known errors appears on a Textbook Errata page that is accessible from the course website home page.  Several of them were pointed out by sharp-eyed EPI 512-513 students, to whom we are grateful.  If you come across a new one, please let us know so that it can be corrected in the next edition of the book and, meanwhile, made known to other students.

Optional readings will also be drawn from:

Gordis L. Epidemiology (4th edition). Philadelphia: Saunders Elsevier, 2009.

Before our book was published, an earlier edition of the Gordis book was used as the primary text for EPI 512-513. It is rich with examples and graphical aids and is written with a touch of humor. We expect that it will still be popular with many students. However, it covers many topics more lightly than we will in EPI 512-513.

Both books can be obtained at the South Campus Center branch of University Bookstore or from other booksellers.

Supplemental texts

Several other standard textbooks in epidemiology have been placed on reserve in the Health Sciences Library. They are:

1. Friedman GD. Primer of Epidemiology (5th edition). New York: McGraw-Hill, 2004.
   
   
2. Hennekens CH, Buring JE. Epidemiology in Medicine. Boston: Little, Brown, 1987.
   
   
3. Kelsey JL, Whittemore A, Evans A, Thompson WD. Methods in Observational Epidemiology (2nd edition). New York: Oxford, 1996.
   
   
4. MacMahon B, Trichopoulos D. Epidemiology: Principles and Methods (2nd edition). Boston: Little, Brown, 1996.
   
   
5. Rothman KJ, Epidemiology: An Introduction. New York: Oxford University Press, 2002.
   
   
6. Rothman KJ, Greenland S, Lash TL. Modern Epidemiology (3rd edition). Philadelphia: Lippincott Williams & Wilkins, 2008.
   
   
7. Haynes RB, Sackett DL, Guyatt GH, Tugwell P. Clinical Epidemiology: How to Do Clinical Practice Research (3rd edition). Philadelphia: Lippincott Williams & Wilkins, 2006.
   
   
8. Szklo M, Nieto FJ.  Epidemiology: Beyond the Basics (2nd edition).  Sudbury, MA: Jones and Bartlett, 2007.
   
   
9. Weiss NS. Clinical Epidemiology: The Study of the Outcome of Illness (3rd edition). New York: Oxford University Press, 2005.
   

To view a big table that summarizes our views about which of these books offer the fullest coverage of various topics in EPI 512-513, click here.

Personal assistance

Kristjana, Alison, and Jeff will each have regular office hours when students may stop in for help on any of the course material covered to date.  The current TA office-hours schedule can be accessed from the EPI 512 website home page.

You may also send course-related e-mail questions to epi512@u.washington.edu, which will be regularly monitored by the TAs.  Please use it as the primary address for most questions, including any new ones.  If you want to follow up on an issue that you discussed earlier with a particular TA, feel free to use his/her personal e-mail address, which can be found on the website home page and earlier in this syllabus.

Tom Koepsell is usually in his Department of Epidemiology office all day Tuesdays and Thursdays and less regularly on other days. His weekly schedule is posted on the door of his Epidemiology office. Victoria Holt divides her time between offices at UW and the Fred Hutchinson Cancer Research Center.  She can be reached by telephone, e-mail, or after class. Noel Weiss is usually in his Epidemiology office on most weekdays.  Students are welcome to drop by to see any of the instructors.  E-mail also works well for setting up appointments or for questions that are not too lengthy or difficult.

Study groups

Past students in EPI 512-513 have often found it useful and stimulating to form a study group that works together on problem sets and reviews together for exams. Doing so is perfectly acceptable and, in fact, encouraged. However, once the discussion in a study group is over, each individual is expected to compose and turn in his or her own answers to the problem sets. Other students prefer to work independently, which is also fine. Formation of study groups is entirely up to you and usually occurs informally.

Supercourse

Dr. Ronald LaPorte and colleagues at the University of Pittsburgh manage the Supercourse project, an on-line collection of over 4,000 self-study "lectures" on various topics in epidemiology, including epidemiologic methods. Lectures have been contributed by faculty from all over the world. Past students have found them to be variable as to level of difficulty and quality, but you may wish to explore what is available by visiting the Supercourse website.

EPI 512 on-line

To help manage information flow for our large class, we will rely heavily on the World Wide Web.  EPI 512 has its own website from which most teaching materials can be downloaded.  We will also use several UW Catalyst Web Tools, including: (1) an on-line dropbox to which you can submit homework assignments electronically from anywhere on the Web, and (2) an on-line gradebook that lets you keep track of your scores on homework and exams throughout the course.  If you are new to Catalyst Web Tools, you may want to read about them here.  Besides the basic functions described below, they have other optional features that you may find handy, such as notifying you automatically by e-mail when something new has been posted.

Course website

The course website will be an important resource throughout the course. Handouts from lectures, links to audio recordings of lectures, problem sets, answers to problem sets, answers to exams, and other teaching materials will be added as we go along. Announcements and schedule updates will also be posted on the website.  The URL is:

http://courses.washington.edu/epi512

Parts of the website are publicly accessible, including the home page, syllabus, and TA office-hours schedule.  However, access to other parts is restricted because those areas are intended only for use by students in EPI 512, not for everyone on the Internet.  When you first enter a private area during a session, you will be asked for your UW NetID and password--the same ones that you use to access your UW e-mail account.  Students who are officially registered for EPI 512 are automatically authorized to use the private areas of the website.  The website authorization list is updated automatically every night based on data from the UW Registrar's Office.  If you recently added the course, you should be able to access private areas of the website within a day or so.  Approved guests may also be granted such access.

The website is organized fairly simply, and most links are self-explanatory.  Go ahead and explore.  Probably the most common task will be to get information related to a specific class session, such as room assignments, lecture notes, problem set questions and answers, reading assignments, audio recordings, and other items.  The quickest way is to go first to the Schedule Overview page, then click on the date in question.  That will take you straight to the proper section of the Daily Details and Downloads page, where you should find what you need. (You can also go directly from the course website home page to the Daily Details and Downloads page, but it is rather long and may require quite a bit of scrolling to get to the date you want.)  Once a downloadable item has been posted, its link will become active; until then, the name of the item will just appear in plain text as a placeholder.

A tip: if you want to go back to the home page from a web page in the private area of the EPI 512 website, it is usually easiest to click on the EPI 512 home link in the upper left corner of the page you are on.  (Clicking on your web browser's Back button may bring up an unwanted dialog box instead.)

Teaching materials posted on the website are not in the public domain and are protected by copyright law. You are asked not to share your password with anyone and not to redistribute materials retrieved from the website.

Most downloadable documents are in Portable Document Format (PDF), which can be viewed and/or printed with Adobe Reader on almost any personal computer. Adobe Reader is widely available in on-campus computing labs and can be obtained for free over the Internet from the Adobe website.

The course website has limited storage capacity and is not a permanent archive.  Expect that materials on it will disappear soon after the EPI 512-513 sequence is over for the year.

On-line dropbox

Gradebook

You can view your scores on problem sets and exams at any time in the on-line course gradebook.  A link to it can be found on the course website home page.  You can also access it via MyUW by following the link to Catalyst Web Tools.

Electronic mailing list

Registered EPI 512 students for whom an e-mail address is on file in the UW Registrar's office are automatically included in an electronic mailing list for the course. Announcements about the course may be sent out occasionally to members of this list, as well as being posted on the course home page. The list is updated automatically by the Registrar's Office when students officially add or drop the course.  Unfortunately, there is no easy way to add unregistered attendees to the list.

Grading

Grades in EPI 512 are based on the following factors:

Problem sets	40%
Mid-term exam	20%
Final exam	40%

• Problem sets. Written answers will be required for several of the problem sets, as indicated on the Daily Details page of the course website.  When written answers are to be turned in for a problem set, each student should prepare his or her own set of written answers, even if he/she worked on them as a member of a study group.  Submitted answers for a problem set that are not the student's own work (e.g., identical answers from members of a study group, or answers drawn from official answers from previous years) may result in no credit for that problem set.  Written answers should be submitted on-line before class begins on the due date.
  
  The three Teaching Assistants will review and grade the problem sets.  All answers to a given problem set will be read by the same TA.  Questions about how a particular answer was graded may be directed to epi512@u.washington.edu, where they will be routed to the appropriate TA.
  
  In general, answers that are submitted on time, represent the student's own work, and reflect a good-faith effort to grapple with the issues will receive full credit, even if some of the answers are incorrect.  This policy is an attempt to reward honest effort and to discourage fixation on the number of points earned.  However, a thoughtful response that is given full credit may not necessarily be correct.  You should always try to reconcile what you wrote with what is discussed in class, with the "official" answers posted on the course website, and with any written feedback on your own answers.
  
  Exceptionally well-crafted (and correct) answers to graded homework may earn a "check-plus." Answers that are turned in late, are judged to reflect little effort, or are very far from correct may be given a "check-minus."  At the end of the quarter, students with several check-pluses may qualify for a small boost in course grade, and those with several check-minuses may lose a little, at the instructors' discretion. However, it is possible to get a grade of 4.0 even with no check-pluses.
  
  Late homework will only be accepted for possible full credit if prior arrangement has been made with the TAs.  If you will be unable to turn in answers to a problem set on time, you should make arrangements with a TA about how to turn them in without incurring a late penalty.  One "free" turned-in-late homework will be accepted without prior arrangement during the quarter; however, in any case, the tardy homework is due no later than one week past the original due date.  Regardless of the reason, homework answers that are turned in after the official answers have been posted are not eligible for a check-plus.
  
  
• Exams. Each exam will consist of a series of questions resembling those in the problem sets. Some will require calculation, some a short answer, and some may be based on a published article. You may use any written notes and books you wish during these exams. It will be handy to have a calculator available.  Personal computers and portable communication devices may not be used during the final exam.
  
  
  • The mid-term exam will be a take-home exam, given out on Thursday, 11/3. Written answers will be due on-line before class begins on Tuesday, 11/8.
    
    
  • The final exam will take place on Monday, December 12, from 10:30 AM - 12:20 PM in our regular lecture room, T-625.

Be forewarned that many students perceive the final exam to be harder than the problem sets, the mid-term exam, or the optional practice problems. This is probably because both exams must be done individually, the final exam has a more rigid time deadline, and the point of each question is not an automatic consequence of recent lecture material.  Also, in contrast to the problem sets, answers on the exams receive full credit only if they respond to the point of the question and are correct.

Any registered student who completes satisfactorily all of the required homework assignments and takes both exams is assured a passing grade of at least 2.7. Historically, the average grade in EPI 512 has been about 3.5.

Other information

Feedback to instructors

We welcome any comments, suggestions, criticisms or compliments you care to offer about the course as we go along. You can catch us during the break, after class, send us regular e-mail, or drop us a note in campus mail. You may also send comments anonymously by following the Send feedback to instructors link on the EPI 512 website home page.

Students with concerns about an instructor or teaching assistant should discuss these concerns with the TA and/or the course instructor. If the student is not satisfied with the response, s/he may contact the department chair at 206-543-1065. If concerns are not satisfactorily resolved, s/he may also contact the Graduate School at G1 Communications Building by phone at (206) 543-5900 or by e-mail at elf@u.washington.edu

Course evaluation forms will also be distributed on the last day of class.

Approved guests

Plagiarism

EPI 512-513 students are expected to follow school and university policies against plagiarism, as described here.  Under these policies, forms of plagiarism include, among other practices: (1) using another person's phrases, sentences, outline, or the structure behind them without proper attribution; or (2) citing a source but reproducing the exact words without quotation marks.  Specific examples include copying part or all of another student's answers, or taking answers from course materials used in a previous year.  Penalties for plagiarism range up to suspension or dismissal from the university.  In EPI 512, no credit is given for problem-set or exam answers tainted by plagiarism, and the points lost cannot be recovered.  Historically, plagiarism is the most common reason for not passing the course.

Students with disabilities

If you would like to request academic accommodations due to a disability, please contact Disabled Student Services, 448 Schmitz, 543-8924 (V/TDD). If you have a letter from Disabled Student Services indicating you have a disability that requires academic accommodations, please present the letter to Tom Koepsell so that you and he can discuss the accommodations you might need for class.

Emergency evacuation

Should we need to evacuate the T-625 classroom:

• Exit via either door at the back of the room.
• Turn left and find the stairway that leads down.  Go down two floors to the 4th floor,
  
• When you exit the staircase on the 4th floor, turn right.  In a few steps you come to a corridor that leads outside.  Follow it, and cross the skybridge over Pacific Street.
• Assemble at the north end of the skybridge.

Schedule overview

Last updated: 9/27/11