Epidemiology: Observational Studies

CANNOT GET RATE INFORMATION FROM CASE CONTROL STUDIES

ALSO CANNOT GET RISK INFORMATION FROM CASE CONTROL STUDIES. INSTEAD WE MEASURE ODDS

Features of Descriptive studies

Describe the patterns of disease occurrence by time, place, and person represent the initial step in many epidemiological investigations

What are the types of analytical studies?

Either observational or experimental observational: Cohort, Case-Control experimental: RCT (randomized control study)

Bias is only present in observational studies T/F?

F! Bias is a major issue in virtually any type of epidemiological design

The focus of epidemiological studies is primarily on infectious diseases T/F?

F. Original focus was on infectious diseases. Currently, the domain of epidemiology is much broader. Acute and chronic diseases Health events Conditions Behaviors

It is impossible to test for harm in RCT (randomized control trials). T/F

False. yes, we cannot ASSIGN exposure if we think that exposure is harmful, however we can observe normal events and see if there are any harmful effects. Example: risk of GI bleeding in patients who take NSAIDS. we are not purposely giving patients NSAIDS to check for harm. We are giving patients NSAIDS as a treatment to another disease but we are able to assess harm (GI bleeds)

Observational studies

Involve NO HUMAN INTERACTION in assigning study groups. Investigators simply observe the relationship between exposure and outcome

What does association measure?

We want to know if a certain exposure increases the risk of a particular defined outcome. Is the rate of illness higher in those exposed?

What is a cohort?

a cohort is a group of subjects who have something in common (an exposure) and are followed over time to evaluate outcomes Note that this exposure was not assigned by the investigator. For example, take 100 people, group them based on who was exposed to radiation in chernobyl. Follow them over time. See how many of group A (had exposure in chernobyl) developed disease. See how many in group B (did not have exposure in chernobyl) developed disease

which is less expensive? a prospective cohort or a retrospective cohort?

a retrospective cohort

What is bias?

any SYSTEMATIC error in the design conduct or analysis of a study that results in a mistaken estimate of an exposure's effect on the risk of the disease

cohort vs case control

cohorts are used for RARE EXPOSURES case controls are used for RARE OUTCOMES cohorts are NOT subject to SELECTION BIAS case controls ARE subject to SELECTION BIAS

Case Report and Case series

consist of collections of reports on the treatment of individual patients or a report on a single patient. Because they are reports of cases and use no control groups with which to compare outcomes, THEY HAVE NO STATISTICAL VALIDITY. USEFUL IN GENERATING A HYPOTHESIS

Retrospective Cohort vs Case Control

The difference between the two is : in case control investigator splits individuals by disease status whereas in retrospective cohort, investigator splits study individuals by their exposure status. First one can provide odds ratio, which is the odds of disease among exposed individuals divided by the odds of disease among unexposed. Whereas the later provides relative risk, which is a degree of risk/probability of disease occurrence due to exposure compared to non-exposure.

Purpose of Analytical Studies

Used to TEST A HYPOTHESIS concerning the relationship between a risk factor (exposure) and an outcome, to measure the magnitude of the association and its statistical significance

case control study

persons with a certain condition (cases) and persons without the condition (controls) are studied by looking back in time to determine the frequency of the risk factor and possible causes of the disease.

prospective cohort vs retrospective cohort

prospective cohorts are more common

systematic error vs random error

remember, BIAS IS A SYSTEMATIC ERROR!

Name two commonly used Descriptive studies

Case reports and Case series

Selection of Exposed

Pick an exposure that has a HIGH ATTRIBUTABLE RISK ASSOCIATED WITH EXPOSURE

Completeness of a study

Should be 80% or more (<20% Loss of Follow up)

Case reports and Case series are useful in generating hypotheses T/F?

T These are purely descriptive studies and there is no comparison group so you can't draw too many conclusions about the disease or the disease process.

Cohorts can be useful in testing multiple outcomes to an exposure T/F?

T!

Cohorts tend to be large, and are expensive and time consuming T/F

T!

Presence of a strong risk factor does not guarantee disease development Absence of a strong risk factor does not guarantee freedom from disease

T!

The analysis of a case control study is always retrospective. T/F?

T! persons with a certain condition (cases) and persons without the condition (controls) are studied BY LOOKING BACK IN TIME to determine the frequency of the risk factor and possible causes of the disease.

Observational studies are subject to many potential biases T/F?

T! but by careful design and analysis, many of these biases may be avoided

The unit of observation in epidemiology is groups rather than individuals T/F?

T.

https://www.youtube.com/watch?v=99FyAqwrIdI

https://www.youtube.com/watch?v=99FyAqwrIdI