Class 1: Descriptive Study Designs Flashcards
Source Population
A group of people with at least one common characteristic (person, place, time, event/exposure)
Fixed population
Defined that an event that happens once; permanent membership
Dynamic population
Defined by state or condition; transient membership
Sample
a subset of the population; ideally selected at random and is representative of the population
Exposure (x)
measurable characteristics that differ across individuals and might affect/be associated with health
Examples of Exposures
demographics, behaviors, environmental factors, policies, health conditions
Health Outcomes (y)
Any measurable disease, disability, injury, infection, syndrome, symptom, biological or subclinical marker, or positive health state
Association
a statistical relationship vs causation which implies that exposure produces health effects
Causation
The exposure produces the effect
Things needed to demonstrate causation
- association
- correct time order
- direction of effect (exposure results in health outcome)
Exposures/Risk Factors/ Determinants
put individuals at higher risk for an outcome
Modifiable Exposures/Risk Factors/ Determinants
behaviors, housing, education, gender identity
Non-modifiable Exposures/Risk Factors/ Determinants
age, sex at birth, race/ethnicity
Surrogate markers Exposures/Risk Factors/ Determinants
for ill defined/unknown/hard to measure exposure
Descriptive Study Design
description of disease patters, hypothesis generation
Types of Descriptive Studies
case reports, case-series, cross-sectional, ecologic
Analytic Design Studies
Compare groups to test hypothesis and evaluate interventions
Types of Analytic Studies
Observational, comparison studies, evaluations
Observational and Comparison Studies
case-control, cohort studies
Evaluation of intervention
clinical trials
Case Report
description of a single individual’s experience with a health outcome
Case Series
description of a group of individuals’ experiences with health outcomes
Strengths of case reports/series
-relatively quick and inexpensive
-often conducted on available data
-recognized and describe new/emerging health problems
-generate hypotheses based on similarities among group
-insight into disease mechanisms
Limitations of case reports/series
-no formal comparison groups
-cannot infer temporal sequence
Ecologic Studies
based on exposure and health outcome at a point in time for more than individuals (counties, states, countries)
Unique Feature of Ecologic Studies
unit of measurement is NOT individuals; the data are averages among different populations
Strengths of Ecological Studies
-relatively quick and inexpensive
-often conducted on available data
-good for early stage of knowledge
-wider range of exposures
-may wish to study ecologic relationships
-only population level data (e.g. good for laws)
Limitations of Ecologic Studies
-ecologic fallacy: an exposure to cause an effect in an individual, the exposure and effect should occur in the same person
-can’t attribute population data to an individual
-little to no control for confounding factors
Cross-Sectional Studies
individuals defined according to current exposure and health out come at a single point in time
can measure prevalence of exposure and outcome at that time
Strengths of a Cross-Sectional study
-relatively quick and inexpensive
-often conducted on available data
-useful for early stage of knowledge
-useful for public health planning
Limitations of Cross-Sectional Studies
-temporal sequence is often unclear
-prevalent cases of long duration may bias results
Continuous Variable
assumes any value between minimum and maximum (infinite)
e.g. blood pressure, viral load
Summary of Continuous variables
-sample mean or median, and standard deviation or interquartile range
Ordinal variables
more than two ranked responses (1,2,3)
Summary of Ordinal Variables
frequencies and percentages
Categorical Variables
grouped and unordered (gender, religion)
Summary of Categorical Variables
frequencies and percentages
Dichotomous Variables
2 response options
Summary of Dichotomous Variables
frequencies and percentages
Discrete
variables that assume only a finite number of values (dichotomous, ordinal, categorical)
