Overview of Epidemiological Studies Flashcards
Lecture Objectives
1) recall the definitions and understand the assumptions underlying epidemiology and epidemiological research
2) understand the differences between experimental research and observational research
3) differentiate between association and causation
4) understand the purpose, characteristics, and measures of associations (if applicable) pertinent to each epidemiological study design
5) identify the advantages, disadvantages, and relative strength of each epidemiological study design
Old definition of epidemiology
study of the distribution and determinants of disease frequency in man
New definition of epidemiology
the study of the distribution and determinants of health-related states or events in specified populations and the application of this study to control health problems
Distribution
Who, where, when “descriptive”
Determinants
What, why, how “analytical”
Population
inhabitants of a given area/region/group
Practical goal of epidemiology
to be used as a tool to improve the health conditions of the public
Names of US population studies
Framingham Eye, Beaver Dam, Baltimore, Los Angeles Latino Eye, Chesapeake Bay Waterman , Wisconsin Epidemiologic study of diabetic retinopathy
Names of Non-US population studies
Barbados, Blue Mountain, Rotterdam, Shihpau, Beijing
Incidence
the number of new problems or cases that develop in a population at risk during a given time period; often quantified as a rate
Prevalence
the proportion of a given population with a problem at a designated time; quantified as a proportion
What are examples of exposure?
actual exposure (toxic chemical or microorganism), a behavior or an individual attribute
The exposure of interest may be associated with…
either an increased or a decreased occurrence of disease or other specified health outcome, and may relate to the environment, lifestyle or inherited characteristics
What are examples of outcome?
the disease state, event, behavior, or condition associated with health that is under investigation
What are epidemiologic assumptions?
disease (and/or risk factors) don’t occur at random; disease is not randomly distributed throughout a population
What are epidemiologic shortcomings?
difficulty in measuring exposure, separating causal contributions of exposure from other exposures, difficulty in assessing role of multiple small risks
What is the impractical goal of epidemiology aka the big epidemiological question?
does exposure A cause outcome B
What are Hill’s causal criteria?
temporality, strength, biological gradient (dose-response), consistency (replication), specificity, plausibility, coherence (compatible), experimental evidence, analogy (alternate explanations
What is the ideal model?
counterfactual model
How does the counterfactual model work?
ideally the population exposed/treated and the population not exposed/treated are the same subjects; you then follow-up and measure incidence
What is the real design?
the two populations are comparable subjects after adjustments, still follow-up and measure incidence
What are the overarching categories of epidemiological studies?
experimental and observational
What is an experimental study?
randomized controlled trial or individual or community
What is the breakdown of observational studies?
analytic and descriptive
What are analytic studies?
case-control and cohort
What are descriptive studies?
case report and ecological and cross-sectional
Experimental
exposure status assigned by researcher; generally described as clinical investigations
Observational
exposure status not assigned by researcher; primarily involves monitoring people in their natural environments
Characteristics of randomized controlled trial
experimental design; population is people w/o the outcome; assign treatment and check for outcome; subjects given standard treatment compared to subjects given experimental treatment; compare outcomes vs no outcome for each group
Randomized control trial of individual/community
measure the disease frequency or exposed/treated vs unexposed/control
Why is randomized control trial the gold standard?
random group assignment, blinded (preferably double), placebo group, treatment only difference between groups
Advantages of randomized control trials?
standard approach to test new treatments, prospective, exposure assigned
Disadvantages of randomized control trial?
time consuming, expensive, same effectivity in real world, non-compliance, ethical issue
Examples of random control trials
DRS, ETDRS, UK prospective DS, DCT, DRCR.net
What are optometry randomized control trials?
CITT and vision in preschoolers study
Analytic
used to help establish cause or risk factors for outcome (what, why, how)
Descriptive
identify patterns among cases and in populations by time, place, person (when, where, who)
What is a cohort study?
a forward or prospective study onset in the present with the future evaluated
Characteristics of a cohort study
observational and analytic; population is people without the outcome; start with exposure and check for outcome; subjects exposed compared to subjects not exposed; compare outcome vs no outcome for each group
Cohort advantages
comparison group, prospective, suitable for rare exposures, can examine a range of outcomes, temporality: exposure before outcome, establish relative risk, measure incidence
Cohort disadvantages
relatively expensive, time consuming (study subject drop-out, prolonged follow-up), not suitable for rare outcomes (disease)
What is a 2x2 contingency table?
compares exposure and outcome
What is an example of a cohort study?
Orinda vision screening study
What is a case-control study?
a backwards or retrospective study; onset in the present with analysis of the past
What are characteristics of the case-control study?
observational, analytic; population: people with the outcome and people without the outcome (control); start with outcome and check for exposure; subjects who have outcome compared to subjects who do not have outcome; compare exposed vs not exposed for each group
Advantages of case-control
comparison group, relatively inexpensive, quick, can study rare diseases, can test multiple hypotheses for a single outcome, generate exposure odds ratio
Disadvantages of case-control
retrospective, recall bias, no temporality: which came first, the outcome or the exposure
Eye related case control study
the lens opacities case-control study; risk factors for cataract
Cross-sectional characteristics
observational, descriptive; populationL people with the outcome and people without the outcome; exposure and outcome are measured simultaneously; measure and compare subjects with exposure and subjects with exposure and outcome
Advantages of cross-sectional study
relatively, inexpensive, quick, establish prevalence
Disadvantages of cross-sectional study
no cause and effect, no temporality, can miss rare exposures/outcomes (diseases), can miss outcomes (diseases) of short duration, not useful for rapidly emerging outcomes
Cross-sectional eye studies
framingham, baltimore, multi-ethnic pediatric eye disease, baltimore pediatric eye disease
Characteristics of ecological study
observational, descriptive; aka correlational study, analyzes groups/populations instead of individuals, try and determine relationship between exposure and outcome variables
Advantages of ecological studies
relatively inexpensive, quick (can sometimes use existing data), big picture (population trends, broad social processes)
Disadvantages of ecological studies
usually cannot establish causation, unable to control for confounding variable, ecological fallacy: can’t extrapolate population data to individual subjects
Example of ecological eye related study
an ecologic study of trends in the prevalence of myopia in chinese adults in singapore born from the 1920s to 1980s
Characteristics of case report
observational, descriptive; case report: documents an unusual medical occurrence, case series: collections of individual case reports
Advantage of case report
first clue/early warning of a potential problem
Disadvantage of case report
anecdotal, non-scientific, the plural of anecdote is not data, no comparison group
Case report eye example
fusarium keratitis outbreak
T/F there are hybrid studies combining study types
true
What is the hierarchy of study strength from basic to scientific?
animal research, case series/case reports, case control study, cohort study, randomized controlled trial, systematic review, meta-analysis
What are grade A study types?
meta-analysis, systematic review, randomized clinical trial, diagnostic studies
What are grade B study types?
weak randomized clinical trial, cohort study, diagnostic study
What are grade C study types?
case control studies, diagnostic studies, studies of strong design, nonrandomized trials
What are grade D study types?
cross sectional studies, case reports/series, reviews, position papers, expert opinion, reasoning from principle
Example of action statement strength of evidence A
individuals should be advised by their health care providers of the risks of smoking and encouraged to quit smoking and/or seek smoking cessation assistance. Evidence grade: A, strong recommendation
