M3A2 involves determining which study type an investigator is using to determine whether smoking is a risk factor for pancreatic cancer. The online text contains a discussion of different study types; all you have to do is decide which one is appropriate. Caution: Identifying the correct study type is worth quite a few points and if you choose the wrong study type you will not be discussing the strengths and weaknesses of the correct study type, so you lose additional points. The following discussion is designed to help you with the choice if you are having trouble.
Case reports/case series. These studies involve taking a medical case or series of cases and describing them in detail. This is usually published in a peer-reviewed journal.
Cross sectional study type. A primary feature of this study is that you start by selecting a sample of subjects, and then you determine how many in the sample have a risk factor and how many have the disease of interest. In other words, you do not identify those with the disease of interest until after you select your sample. Cross sectional studies are typically very large and costly, and take time to complete.
Cohort study type. A cohort is a population group or subset of a population group. A prospective cohort study follows the cohort over a period of time. A retrospective cohort study examines the cohort over a period of time in the past. Cohorts are chosen based on some common characteristic. For example you may look at individuals who work in the fiberglass manufacturing industry (an occupational cohort), everyone born in the U.S.A. in 1960 (a birth cohort), everyone born in the state of Washington since 1950, etc. The number of potential cohorts is endless. Once you choose a cohort you examine (either retrospectively or prospectively) individuals in the cohort to determine if there are associations between exposure variables and some outcome. For example you could look at the fiberglass manufacturing cohort to determine the rate of nasal cancer and exposure to formaldehyde during the manufacturing process. Once again, it is important to note that cohorts are not chosen based on whether individuals have a disease or not. You choose the cohort, then evaluate whether individuals have an exposure or not, and whether they have a disease or not. Cohorts are typically very large (at least 1,000’s).
Case-control study type. To perform this type of study participants are chosen based on whether they have a disease (cases) or not (controls). After selecting cases and controls, participants are evaluated for specific exposures and biostatistics is used to determine if an association exists between an exposure and the disease. Remember, for this study type samples are chosen on the basis of the presence or absence of a condition. It is typical to choose at least 2 controls for every case (and sometimes 3 or 4). An attempt is made to match controls to cases on some important variables (age, sex, race, ethnicity, SES, etc.). Case-control studies are smaller (100’s) than cohorts.
Randomized clinical trials. Randomized clinical trials are where individuals are selected to test drugs or therapies.