The flashcards below were created by user
on FreezingBlue Flashcards.
Analytical analysis focuses on which of the following: who, what, when, where, how, why
how and why
True or False: In experimental studies, a portion of participants are assigned the intervention
True or False: Case-control studies and cohort studies are examples of experimental studies
FALSE. Case-control studies and cohort studies are examples of observational studies.
In a case-control study the outcome is always identified before what?
In a case-control study the outcome is always identified prior to the exposure.
Provide the first step in conducting a case-control study
The first step in a case-control study is establishing the diagnostic criteria and the definition of the disease
Where do cases (persons experiencing a health-related state or event) come from in a case-control study?
- 1. records from public health clinics
- 2. physician offices
- 3. hospitals
- 4. health maintenance organizations
- 5. industrial and government sources
How do you select a control group in a case-control study?
A control subject should look like the case subjects minus having the disease under study.
What is the epidemiological assumption?
The epidemiological assumption assumes that the controls are representative of the general population in terms of probability and exposure and that controls have the same possibility of being selected or exposed as the cases.
Controls can be selected from what three areas?
The general public, hospitals, friends/family
Give the disadvantage controls from hospitals have.
Hospitals differ from healthy people such that they do not accurately represent the exposure distribution in the population where cases were obtained.
Give examples of disadvantages the controls from general population have.
More costly and timely in comparison to hospital controls
May be difficult to contact healthy persons with work and leisure activity schedules
less motivated to participate than hospital controls
Give examples of disadvantages the controls from family, friends, relatives have.
If the exposure is similar to the one experienced by cases an underestimation of the true association would result
How can you obtain information about exposure status?
Through medical records, interviews, questionnaires, surrogates such as employers, spouses and siblings.
What are the three types of bias in a case-control study?
Selection bias and Observation bias: recall bias and interview bias
What is selection bias?
The relationship between exposure and disease among participants in the study differs from what relationship would have been among individuals in the population of interest.
Another name for Berkson's bias is Hospital-patient selection bias. Describe it.
Consider a hospital-based case-control study assessing the strength of the association between smoking and respiratory diseases. Selecting controls from the hospital will likely underestimate the association between smoking and respiratory diseases because patients with non respiratory diseases in the hospitals are more likely to be smokers than the general population.
Another name for Neyman's bias is Prevalence-incidence bias. Describe it.
It is a form of selection bias in case-control studies attributed to selective survival among the prevalent cases. For example, if cases with coronary artery disease die rapidly, persons available for study are not more severe cases. The association between serum cholesterol (high vs. low) and coronary artery disease will be underestimated.
Almost all studies experience misclassification. Misclassification occurs when either exposure or disease status is inaccurately assigned.What is the difference between differential and non differential misclassification?
Nondifferential misclassification (a.k.a random misclassification) example: If classification of history of hypertension is accurate in 90% of cases and 90% of controls, wherein the level of misclassification is the same between cases and controls.
- differential misclassification example:
- history of hypertension is accurate in 90% of cases and 100% of controls.
Random misclassification in a case-control study will always result in an overestimated or underestimated odds ratio?
Name variables that are routinely considered as potential confounders in epidemiologic research
age, gender, educational level, smoking
What are the strengths in case-control studies?
- 1. effective for rare outcomes
- 2. compared to the cohort study; less $, time and size
- 3. yields the odds ratio (when the outcome condition is rare, a good estimate of the relative risk)
What are the weaknesses in case-control studies?
- 1. Limited to one outcome condition
- 2. Does not provide incidence, relative risk or natural history
- 3. Less effective than a cohort study at establishing a time sequence of events
- 4. Potential recall and interviewer bias
- 5. Potential selective bias
What is the cohort effect?
The cohort effect is the change and variation in the disease or health status os a study population as the population moves through time. Cohort effects include any exposure or influence from environmental effects to societal changes.
What are the measures used in describing cohort data?
cumulative incidence rate- attack rate
incidence density rate- person-time rate
What are the measures of association in cohort studies?
ratio of attack rates (risk ratio)
ratio of person-time rates (rate ratio)
How do you select subjects in a cohort study?
Select people from the population who are at risk of becoming a case.
Who should you exclude when selecting subjects in a cohort study?
People who already have the disease outcome of interest (prevalent cases)
Those who are not at risk (i.e- they had an organ removed so they cannot develop specific cancer to that organ)
What are the two bias in cohort studies?
Loss to follow up and healthy worker effect
How is restriction used in cohort studies?
Restriction is used to limit generalization and improve feasibility and focus
How can you minimize lose to follow up?
- 1. restricting the study participants to those most likely to remain in the study (i.e- excluding those with a highly fatal disease or who are likely to move away from the area)
- 2. collecting personal identifying information
- 3. making periodic contact and incentives
How can you minimize misclassification in a cohort study?
By refining the definition of the exposed and unexposed groups and avoiding exposure classifications that result in differential outcome ascertainment
In a cohort study how can you control confounders at the design level?
- - Restriction to avoid bias due to confounding
- - in double cohort studies, confounding reduced by choosing comparison groups as alike as possible to the exposed population
In a cohort study how can you control confounders at the analysis level?
Collecting data on potential confounders at the beginning of the study makes it possible to adjust for these potential confounders at the analysis level through stratification and multiple regression techniques
Contrast a prospective cohort study to a retrospective cohort study weaknesses.
- A prospective cohort study is more expensive and has a longer duration than a retrospective study and is limited to one exposure.
- In a retrospective cohort study you have less selection of participants and exposure and outcome measures.
What is the definition of a retrospective cohort study?
The investigator identifies a cohort with already available exposure and outcome data.
What is the definition of a prospective cohort study?
The investigator identifies participants, measures exposure status, and follows the cohort over time to monitor outcome events.
What is the definition of a double cohort study?
two distinct populations with different levels of the exposure are followed.
The other name for experimental studies is ____ _____
What are the two types of trials in experimental studies?
controlled trials and community trials
What is the strongest methodological design?
The between- group design
What is the between-group design?
where outcomes are compared between two or three groups of people receiving different levels of intervention
This experimental design is used when the outcome in a single group is compared before and after the assignment of an intervention
What is the unit of analysis in a controlled trial?
What is the unit of analysis in a community trial?
the group or community
Name the function of a random assignment
makes intervention and control groups look as similar as possible
What is the only factor that determines group assignment in a random assignment?
Another name for a non-randomized study is
a convenience sample