(modification of) PRACTICUM - 2 0 0 2 (original plan and text by J-P Collet)
Objectives of practicum
To allow students to apply the epidemiological/biostatistical skills learned in the first and second semester to real research questions, through the use of published studies.
Course organization
To facilitate interaction and discussions the class has been divided into three (fixed-membership) teams: 1, 2, and 3. Each week, each team will address a different part of an article: a title, an abstract, or a methods section (see the table below).
| Date: | 09.17 |
09.24 |
10.01 |
10.15 |
10.22 |
10.29 |
11.05 |
11.12 |
11.19 |
| Team 1 | |||||||||
| Team 2 | |||||||||
| Team 3 | M1
M2 |
The articles were chosen to give students an opportunity to become more familiar with various epidemiologic designs and types of data analysis.
Tasks
Dealing with just the Title will probably be the most challenging and rewarding. For the week the team is assigned to this task, the team must propose and justify: the object of study(*), a design (and offer alternatives, if any exist), a sampling strategy, exposure and outcome measurements, analysis, and potential sources of bias to avoid.
(*) See 'Object Design' in Miettinen and Karp's 2012 book, Epidemiologic Research: An Introduction, available as an online eBook from the McGill Library, or as a single pdf file (chapters consolidated by JH) at the bottom of this webpage
Dealing with the Abstract will also require pro-activity. Using the information supplied in the study summary, the team in question must first summarize the study, then comment on: potential sources of bias arising from the choice of subjects, the appropriateness of the measurements chosen, and what analysis was used. In view of the limited descriptions in the abstract, they are also asked to list the main questions/concerns they have that they will want to see addressed in greater detail in the methods section.
The team assigned the Methods that week have a less challenging task, and may be forgiven for being less motivated. First, the print is often smaller in the Methods. Second, each article was published in a prestigious peer-reviewed journal, so who are we to criticize? Third, they have to go last and everyone wants to talk about the next article. Their job, however is to answer the questions raised by the team dealing with the Abstract, and (independently) comment on the ways the authors actually tried to avoid the pitfalls mentioned by the previous reams, and whether they found the approaches satisfactory.
We have chosen articles which are not too complex from a biological/medical point of view (we hope), but which still reflect important public health problems.
The overall goal of this course-cum-journal club is not to introduce a whole array of new methodological, statistical, or substantive material, but to give you the chance to pretend you are real epidemiologists without the hassle of getting grants and writing up the articles. By the end of the course, you should feel more comfortable sitting down with an epidemiologic article and appraising its validity in a logical and organized way.
updated 2012.09.30