Week 7 Flashcards
What type of studys is an analytic study
Ecological
Cross-sectional
Case-Control
Cohort
Is case control and cohort studies observation or experimental
Observational
What are 4 determinants we should consider when creating a cohort study
- Those who enter the cohort study should be outcome-free
- Selecting exposed and comparison groups
- Following up participant
- Determining outcome status
Why is cohort study an observational study
There is no intervention you just follow up who gets a disease and who dosen’t
Is the exposure determined before the outcome happens
Yes
What are some disadvantages of cohort studies
More expensive
Time consuming
Not fesable for many disease
What are some advantages of cohort studies
Better exposure and confounder data
Less vulnerable to information bias (recall bias can still happen)
What is a prospective cohort study
You follow people for 10 years and compare incidence of disease
What is retrospective study
You start at outcome and look back in time to see what caused the disease
What are good ways to collect data for a retrospective study
Military data
Occupational study
What are some advantages of retrospective study
Cheaper
Faster
Efficent with disease with long latenet period
What are some disadvantages to a retrospective cohort study
Exposure and confounder data may be inadequate
More vulnerable to bias
Need an established recording system
What is the incidence rate ratio
a/(a+b)/c/(c+d)
if someone develops breast cancer but then latter dies what is there outcome
It is not death it is breast cancer
What is the concept of person-time
People are in the study for a varied durations
Different times under exposure
Contribute differentialy to the study /.p09
What are advantaged of Cohort studies
Valuable when exposure is rare
Can examine multiple effects of a single exposure
Easier to determine the temporal relationship between exposure and outcome
Allows measurement of incidence
What are some disadvantages of cohort studies
Validity affected by losses to follow-up (selection bias)
Other factors may not be distributed evenly between exposure groups
Inefficient for evaluation of rare disease
Can be expensive and tine consuming
If retrospective they require good records
What are limitations to cohort studies
May need large numbers of subjects to be followed for long periods
Loss to follow up to undermine validity
Not good for rare disease or those with long latency
Not good when exposure data are expensive to obtain
You classify participants according to their outcome status
Cases (+ disease)
Controls(no disease)
Then you need to ask for past exposures if the cases and controls were exposed or not exposed
Is a case-control population real
No they are artifical
If we have a lot of people in a study should we reduce the size of the study
Reducing the size of the study greatly reduces cost with a proper sampling we will get a correct value for the measure of association
What do you calculate when it is case control
Odd provides provides the proper measure of association
What is the odd of exposure
(A/C)/(B/D)
What is the caculation for the not-exposed group
(A/B)/(C/D)
What are some qualifications for selecting cases
Very clear case definition requried
Ideally case selection will invovle direct sampling of cases within a source population
All people in the source population who develop the disease of interest will be included as cases
If someone in the source population developed the disease of intrest would they
What is the definition of controls
A sample of the source population that produced the cases
What is the purpose of cases
To estimate the exposure distribution in source population that produced the cases
What are the two things that matter when selecting controls
- Controls must come from the same source population as the cases
- Controls must be selected independently of exposure
Where can you find controls
- Population based controls
- Nested controls from a cohort population/study
- Hospital or clinic based controls
- Family or friend control
What is the definition of population based controls
Controls selected from the general population most suitable when cases are from well-defined geographic area
What are sources of population-based control
Random digital dialing, cell phone or intrent subsurcibers residence lists, tax lists, voter registration, driver license holders
What is the advantage and the disadvantage of population based control
Controls often come from well-defined source population
It is very time consuming harder to inspire participating may not recall past exposures as we as cases
What is the definition of nested controls
Controls selected from an existing cohort population
What is the advantage and disadvantage of nested controls
Controls come from clearly defined source population already enrolled willing participants
Restricted to memebrs of existing cohort may limit the hypothes
What is the defention of Hospital or Clinic based controls
Controls selected from among patients at a hospital or clinic
What is the requirement of a hospital or clinic based control
Same source population as cases must consider the would criterion and the referral pattern illness should be unrelated to that is independent of the exposure under study
1. Illness that have the same catchment area as the cases
2. Illnesses that have no relation to the risk factors under study
What is the advantage of Hospital or clinic-based control
Easy to identify and access less time and money accuracy of exposure recall comparable to cases more willing to particpate
These controls are not randomly selected this mean that hopsital based controls must be carefully selected to accurately represent the exposure history in source population
What is the defention of the matched design
The process of making a study group and a comparison group comparable with respect to extraneous factors
How can you account for confounding in matched design
Randomization in experimental studies
Matching case to controls
How does matching work
For each case a control with the same characteristics will be selected
Whats the difference between individual matching and frequency matching
Indvidual matching is performed participant by participant and frequency matching providing similar distributions of confounder in groups
What is matching like in cohort studies
Exposed matched to unexposed
Less common
Expensive
Sometimes unpractical
May require control
What is matching in case control studies
After selection of a matching factor for each a case a control with the same characteristics will be selected
A unit is not a person but a pair
Paired anyalsis
What is the measure of association for cross sectional
Prevalence rate ratio
What is the measure of case control
Odds ratio
What is the measure of (Cohort)
Incidence ratio
