Epidemiological methods Flashcards
What is epidemiology?
Epidemiology is the study of the distribution, determinants, and deterrents of morbidity and mortality in human populations
What are the two main types of epidemiology?
The two main types are descriptive epidemiology and analytic epidemiology
What does descriptive epidemiology describe?
Descriptive epidemiology describes the distribution of determinants, morbidity, or mortality by person, place, or time variables
What is descriptive epidemiology useful for?
Descriptive epidemiology is useful for:
◦ Assessing the health status of a population.
◦ Generating hypotheses.
◦ Examining patterns and establishing plans for public health programs
What does analytic epidemiology study?
Analytic epidemiology studies the associations or causes of disease
What are the two main classifications of study designs?
The two main classifications of study designs are descriptive and analytic.
What are the types of descriptive study designs?
The descriptive study designs are:
* Case series
* Cross-sectional
What does a case series study describe?
A case series study describes the characteristics of a group or cluster of individuals with the same exposure or disease/outcome.
What does a cross-sectional study examine?
A cross-sectional study examines a group of people at one point in time, describing the prevalence of an exposure or disease/outcome
What are the two types of analytic study designs?
The two types of analytic study designs are experimental and observational.
What is a characteristic of experimental studies?
In experimental studies, the investigator intentionally alters one or more factors to study the effects
What is a characteristic of observational studies?
In observational studies, the investigator observes without intervention
What are the types of observational study designs?
The observational study designs are:
* Cohort
* Case-control
* Cross-sectional
Describe a Randomized Control Trial (RCT)
A Randomized Control Trial (RCT) is an experimental study with 4 fundamental steps:
1. Selection of appropriate study sample and baseline assessment.
2. Randomly assign participants into an Experimental Group(s) and a Control Group.
3. Application of intervention.
4. Follow-up assessment(s).
RCT’s reduce bias
What is a cohort study?
A cohort study is a forward-looking study where exposure is ascertained prior to the ascertainment of an outcome. Individuals are followed over an extended period of time
What is a case-control study?
A case-control study is a backward-looking study where the outcome is ascertained prior to the ascertainment of the exposure
What is the timing of data collection in a cross-sectional study?
In a cross-sectional study, exposure and outcome are ascertained simultaneously. It’s a ‘snapshot in time’.
What is the hierarchy of evidence regarding study designs?
Can all relationships be examined with RCTs?
No, not all relationships can be examined with RCTs.
What are the two main categories of measurements in epidemiology?
The two main categories are measures of disease frequency and measures of association
What are some measures of disease frequency?
Incidence and Prevalence
Define prevalence rate
Prevalence rate is the proportion of a population with a given disease or condition at a specified time
How do you calculate prevalence rate?
Prevalence Rate = (Number of cases at a specific time / Total population at that time) x 100
Define incidence rate.
Incidence rate is the proportion of the population at risk that develops a given disease or condition during a specified time period. It includes a time component.
How is incidence rate different than prevalence rate?
Incidence measures new cases over a period of time, while prevalence measures existing cases at a point in time.
How do you calculate incidence rate?
Incidence Rate = (Number of new cases during a time period / Population at risk during that time period) x 100
What are the measures of association?
Relative risk, odds ratio, and prevalence ratio.
Define relative risk (risk ratio)
Relative risk is the ratio of the risk of an outcome in the exposed group to the risk of the outcome in the unexposed group.
What does relative risk estimate?
Relative risk estimates the increased or decreased risk of an outcome due to a particular exposure.
How do you calculate relative risk?
Relative Risk = (Incidence rate in exposed group) / (Incidence rate in unexposed group)
Define prevalence ratio.
Prevalence ratio is used in place of relative risk in case-control and cross-sectional studies. It indicates the proportion of individuals that have the outcome in the two groups (i.e., exposed vs unexposed).
How do you calculate prevalence ratio?
Prevalence Ratio = (Prevalence of outcome in the exposed group) / (Prevalence of outcome in the unexposed group)
Why can’t you use relative risk in case-control and cross-sectional studies?
Because there is no forward time component that determines the outcome in those study designs
Define odds ratio.
Odds ratio is used in place of relative risk in case-control and cross-sectional studies. It indicates the odds of an outcome occurring in the exposed group compared to the unexposed group.
What is the formula for odds ratio?
OR = (a/c) / (b/d) = (ad) / (bc)
using the table:
* Exposed cases (a),
* Exposed controls (b),
* Unexposed cases (c),
* Unexposed controls (d)
What is the difference between prevalence ratio and odds ratio?
Prevalence ratio indicates the proportion of individuals with an outcome in two groups, while odds ratio indicates the odds of an outcome occurring in one group compared to another
When interpreting relative risk and odds ratio, what is the ideal research scenario?
Research questions are posed so that the association is positive and straightforward to interpret, and we can determine the strength of the association
Define causality in epidemiology
An event, condition, or characteristic that preceded the disease event, without which the disease event would not have occurred until some other time
Can causal relationships be assessed with a descriptive cross-sectional study?
No, a descriptive cross-sectional study only gives a snapshot in time.
Briefly describe the Germ Theory and its historical context.
◦ Mid-to-late 1800s and early 1900s
◦ Dominant in clinical medicine and public health
◦ Diseases had a single causative agent
◦ Worked well for infectious diseases
What is the Black Box concept in the history of cause and effect?
◦ Emerged from the 1950s onwards, when chronic diseases became more dominant
◦ Involves understanding risk factors and the disease without necessarily knowing the mechanisms
◦ Represents a period of explosion in knowledge about risk factors for chronic diseases due to advances in epidemiological and statistical methods
Describe the Doll & Hill (1950) studies.
◦ Example of the “Black Box” approach
◦ Studies on smoking and lung cancer
◦ Found more lung cancer patients were smokers
◦ Found a higher proportion of lung cancer patients described themselves as heavy smokers
◦ Helped determine a causal relationship between smoking and lung cancer
Explain the Web of Causation
◦ 1960s
◦ Occurrence can be explained by a complex web of interconnected factors
◦ Pathways can differ from person to person
What is the difference between a risk factor and a cause?
Indirect cause versus a direct cause
Explain why association does not necessarily imply cause and effect.
Confounding variables. A third variable can distort the association between the exposure and the outcome. Correlation does NOT equal causation.
Give examples of confounders in behavioral epidemiology research.
Diet, age, sex, race, SES
Explain the relationship between exposure and outcome for a variable to be considered a cause.
A change in the exposure must result in a corresponding change in the outcome.
What type of study design is related to the definition of cause and effect?
Experimental control study.
List Hill’s Guidelines for Assessing Causation.
- Temporality
- Strength of association
- Consistency
- Specificity
- Dose-Response Relationship (Biological Gradient)
- Biological Plausibility
- Coherence
- Experiment
- Analogy
Define Temporality in the context of causal associations.
The exposure occurred before the disease. It is the most important criterion
Define Strength of association in the context of causal associations.
Strong measure of association (i.e., OR, RR).
Define Consistency in the context of causal associations.
Same results for different people, places, and times. Related to repeatability of the study.
Define Specificity in the context of causal associations.
More commonly related to infectious disease and a single cause leading to an effect (i.e., germ theory).
Define Dose-Response Relationship (Biological Gradient) in the context of causal associations.
As the dose of the exposure increases, the risk for the outcome also increases or decreases in a gradient fashion
Define Biological Plausibility in the context of causal associations.
Existing biological or social model to explain the association (i.e., is it scientifically plausible?)
Define Coherence in the context of causal associations.
Association does not conflict with known facts about the history and biology of disease. Is the mechanism coherent
Define Experiment in the context of causal associations.
Been tested using an experimental design.
Define Analogy in the context of causal associations.
Similarities between the observed association and other associations
