Other types of study design Flashcards
What are the 3 other types of study design?
Cross-sectional study
Ecological study
Descriptive study
Where in the heirachy of evidence fo cross-sectional studies lie?
4th out of 6
What is a cross sectional study?
At a single point in time/ snapshot, exposure and/or outcome data are collected
Uses:
measure prevalence of exposure or outcome in population
Assess association between exposure and outcome
What are the strengths of cross sectional studies?
Good quality data
Can detect case which have not presented to medical care
What is the clinical iceberg?
The idea that not all people with a disease are known to health services because some will have symptoms but will not have sought advice and some will be diseased but be unaware as they have no symptoms
What are the weaknesses of cross-sectional studies (3)?
Temporal sequence of exposure and outcome
Confounding
Recall bias
Where in the heirachy of evidence do ecological studies come?
2nd from bottom
What is an ecological study?
Units of analysis are populations rather than individuals
= uses agregate measures for entire population i.e. exposure and outcome measured at district, region or country level
What is the concept of ecological fallacy?
The assumption that average characteristics for populations apply to individuals
Can get:
-ve bias = correct conclusion & underestimate strength of association (association between countries less steeo than association within a country)
+ve bias = correct conclusion & overestimate strength of association (association between each country steeper than association within a country)
Opposite conclusion (population increase in one decreases the other but in individial country increase in one increases the other)
What are the strengths of ecological studies (3)?
- Relatively inexpensive and quick if data availiable
- Logical approach if exposure is environmental/social
- Useful to generate hypotheses to test in individual-level studies higher up the heirachy
What are the weaknesses of ecological studies (2)?
- Results from aggregate level analyses cannot reliably be applied at individual level (ecological fallacy)
- Often heavily confounded
Where in the heirachy of evidence do descriptive studies lie?
Bottom
(because it uses data previosuly collected for another purpose)
What is a descriptive study?
used to describe a variable in terms of time, place and person
n.b. does not assess associations between variables
useful to generate hypotheses (test in different type of study)
may give clues about whihc environmental factors may influence exposure and outcome
What are common sources of data for descriptive studies?
Mortality data (i.e. mortality data deaths)
Hospital activity data
Census data
Regular national surveys
Research surveys
What may be the reasons for a change in incidence over time (6)?
- Chance (variations due to random fluctations)
- Demography (change in age population)
- Ascertainment (change in diagnostic tecniques = more likely to be detected)
- Coding (change in rules by which mortality is encoded)
- Treatment effects (new treatments likely to decrease outcome prevalence)
- True differences (e.g. due to risk factors)
