Lecture 5A, part 1 Flashcards
What is the goal of study designs?
To examine the relationship between exposure and dz (or determinant and outcome) with validity and precision while using a minimum of resources
Steps in the scientific method
Observe and describe health phenomena
Formulate a hypothesis to explain the phenomena
Test the hypothesis in a research study
Replicate the hypothesis test in additional studies
On what does the selection of a design depend?
The research question
Concerns about validity and efficiency
Practical and ethical considerations
Hierarchy of research design (from top to bottom)
Meta-analyses
Systematic reviews
Experimental trials (RCTS)
Cohort studies and case-control studies
Cross-sectional and ecological studies
Case studies or case reports
Types of descriptive studies
Case reports/case studies
Cross-sectional studies
Ecological/correlational studies
Types of analytic studies
Experimental studies
-Field trials, community trials
-Randomized controlled clinical trials, pragmatic trials
Non-experimental studies
-Cohort
-Case-control
How are case reports/case studies useful?
ID’ing new dzs
Who is usually the focus of case reports/studies?
A single pt or small cluster of pts
What do case reports/studies describe?
An unusual or novel outcome
Purpose of a case report/case study
Help form hypotheses for additional study
-Unexpected associations
-Possible pathogenesis of dz
Purpose of cross-sectional studies
Examine the relationship between dzs (or health outcomes) and RFs as they exist in a defined pop at one particular point in time
-“Snapshot”
How are cross-sectional studies measured?
Dz prevalence in relation to exposure prevalence
How are pops selected in cross-sectional studies?
Without regard to exposure or dz status
What do cross-sectional studies include?
Most large-scale surveys of the general pop
Examples of large-scale surveys that do cross-sectional studies
NHANES
BRFSS
NSFG
Strengths of cross-sectional studies
Results can be highly generalizable (especially when based on well-selected sample of the general population)
Low cost
Limitations of cross-sectional studies
More likely to ID prevalent cases of long duration
Temporal sequence between exposure and dz cannot always be inferred
When can temporal sequence between exposure and dz be inferred in cross-sectional studies?
If timing of exps is ascertained relative to measurement of prevalence
When is temporal sequence not an issue in cross-sectional sutdies?
When exposure is an unalterable characteristic (genetic trait)
Purpose of ecological studies
Examines rates of dz by aggregated pops
What are the units of analysis in ecological studies?
Pops or groups rather than individuals
What is a property of the pop in ecological studies?
Exposure status
Types of ecological studies
Completely ecologic
Partially ecologic- some but not all variables are ecologic
Multilevel
What variables are present in completely ecological studies?
All variables (outcome, exposure, covariates)
Description of multilevel ecological studies
Analyses may simultaneously include multilevel and ecological variables on the same construct (also called multi-level modeling, hierarchical regression, or mixed effects modeling)
