US exam 3 epidemiology Flashcards
Conducting studies
Studies are conducted in an attempt to discover associations between an exposure or risk factor and a health outcome
O1: Epidemiology Study Types
epidemiology study types:
observational: descriptive or analytic
experimental
Descriptive and Analytic Epidemiology
Descriptive epidemiology
- When was the population affected?
- Where was the population affected?
- Who was affected?
Analytic epidemiology
- How was the population affected?
- Why was the population affected?
Distribution
Who: age, sex, race and economic status
When: trends of disease frequency over time (is it increasing?, decreasing?, remaining stable?)
When: tracking an outbreak. Use epidemic curves to plot the number of cases over a period of time.
Where: compares disease frequency in different counties, states or countries or other geographic areas. Can also compare urban vs. rural areas
Distribution
The information gathered on the distribution of disease gives clues about the determinants of disease.
Determinants can be any underlying social, economic, cultural or environmental factors that are responsible for health and disease.
O1: Study Designs
Three study designs are commonly used in epidemiology (Observational)
Cross-sectional
Cohort
Case-control
Experimental studies to test efficacy of medications
Randomized Controlled Studies are used
Will go into further detail next semester in Drug Lit
Determinants
When you have clues to the determinants of disease you can generate a hypothesis
The hypothesis can be tested by formal/systematic epidemiologic studies
These can confirm or disprove the hypothesis
Study Design-Cross Sectional
what does it analyze
how are subjects selected
what does it provide
Analyzes data of variables collected at one given point in time across asamplepopulation or a pre-defined subset.
Subjects are selected because they are members of a certain population at a certain period of time
Provides a “snapshot” of exposures and outcomes
Study Design-Cohort Study
what does it study
what does it start with
what does it provide the best of
what are they in nature and how long does it take to complete it
A study of a group of people, or cohort, followed over time to see how some disease or diseases develop
Study typically begins with healthy subjects and are asked about their exposures
These studies provide the best information about the causation of disease and the most direct measurement of risk of developing a disease
Prospective or retrospective in nature can take years to complete
O2: Cohort Studies
what are the advantages
- how many outcomes can be studied
- what does it provide
- - what is not likely
- how can you directly calculate
what are the disadvantages
- how much money
- how long does it take to complete
- what is it inefficient for
- what may be missing
- what can there be loss in
Advantages
Can study multiple outcomes
Study uncommon & multiple exposures
Provides clearer sequence of events
Selection bias is not likely
Directly calculate disease incidence
Disadvantages
Expensive & time consuming
May take years to complete study
Inefficient for rare outcomes or disease with long latent periods
Data on some confounding variable may be missing
Source of error
Loss to follow-up
Study Design- Case Control
what does it compare
who are the case patients
who are the control patients
how do epidemiologists work
what other studies can it be tested within
A study that compares individuals affected by a disease with a comparable group of persons who do not have the disease to determine possible causes or associations
Those with the disease or condition are case-patients
Those without the disease or condition are control-patients
Epidemiologists work backward from the illness or health
condition (retrospective in nature) to determine any associations - so how did you get the disease
Can be nested within a cohort study-more on this in Drug Lit
Study Design- Case-Control
Cases and controls are compared for the presence or absence of one or more specific exposures or risk factors
Takes a shorter amount of time to complete
All subjects are matched as much as possible (age, race, gender, other factors relative to the disease)
O2: Case Control Studies
advantages
disadvantages
Advantages
Can study multiple exposures
They are efficient for rare diseases or diseases with a long latency period between exposure and disease manifestation
Faster and easier to conduct than other studies
Less expensive
Disadvantages
Bias can be a source of error
Recall Bias
Information Bias
Reporting bias
Selection bias
They are inefficient for rare exposures
Experimental Study
what is this study used for
what are the two groups
how can the two groups be
how are subjects placed into groups
Randomized Controlled Trials (RCT)- Will be covered in detail next semester!
These types of studies are used to test new drugs for safety and efficacy before they are approved for marketing
Start with two groups:
Experimental group-will get the new treatment
Control group-may get placebo
Can be blinded or double blinded
Subjects are randomized into groups
Have an intervention occur (subjects given counseling and drug or no counseling with placebo)
O2: Randomized Controlled Studies
Advantages
Convincing
Can control for confounders (known and unknown)
Disadvantages
Very expensive- millions of $$
Artificial environment
Ethical issues
Difficult to conduct logistically
O3: Measurements used in epidemiology
Frequency:
Rate is the number of cases but this number must be relative to the size of the of the population being studied.
Population at risk (PAR): This number should only include people who are potentially susceptible to the disease being studied.
Measurements used in epidemiology: incidence
- what does it measure
- what does it express
- what must always be included
Incidence is the rate of new cases of a disease in a defined population over a defined period of time
Measures the probability that a healthy person in that PAR will develop that disease during that specific time
Incidence-Expresses the risk of becoming ill.
Must always include a unit of time- such as cases per 10n per day, week, month or year.
prevalence
what is prevalence
how fast to rates change and how useful is it in epidemiology
what does it estimate
how is it expressed
what is mortality rate
Prevalence is the total number of cases of a disease existing in a defined population at a specific time
Rates change slowly-not useful for epidemiologic studies
Useful in assessing social impact of disease-affects planning
Prevalence- Estimates the probability of the population
being ill at the period of time being studied
Prevalence is often expressed as cases per 100 (percentage) or per 1000 population.
Mortality rate is the incidence of death per unit of time (usually per year) in a population, can look at all deaths or a specific cause of death
O4: Measurements of Association
Epidemiology identifies associations between exposures and outcomes
Salt intake → Hypertension (direct association)
Salt in take → Hypertension → Coronary artery disease (in-direct association)
Types of measurements used in epidemiologic studies
Relative Risk
Odds Ratio
Measurements of Association: Relative Risk (RR)
If an association exist, how strong is it?
What is the ratio of the risk of disease in exposed individuals to the risk of disease in unexposed individuals?
RR = 1.0= no association between exposure and the disease
RR > 1.0= shows a positive association (increased risk from exposure of getting the disease)
RR < 1.0 = Shows a negative association (decreased risk from exposure of getting the disease)- sometimes called a “protective effect”
Typically used in COHORT studies
Measurement of Association: Odds ratio (OR)
what does it ask
what is it used in
what can you not directly calculate
what does each mean
OR = 1
OR >1
OR < 1 negative association
What are the odds that the disease will develop in an exposed person? Association of an exposure and a disease
Used in a case-control studies
Can not directly calculate the risk since subjects already started with the disease- it is an approximation of the relative risk
OR = 1 no association, OR >1 positive association, OR < 1 negative association
Measurement of Association: 95% Confidence Interval
The 95% confidence interval (CI) is used to estimate the precision of the OR (also used in RR).
A large CI indicates a low level of precision of the OR, whereas a small CI indicates a higher precision of the OR.
In practice, the 95% CI is often used as a proxy for the presence of statistical significance if it does not overlap the null value (e.g. OR=1).
Will look like: RR=3 (95% CI 1.67-4.2); p<0.5 it is a range of numbers
a range of values so defined that there is a specified probability that the value of a parameter lies within it.
p-value: if we did the experiment 100 times, 95 of the tests will be the same and 5 of them will be different
we are 95% confident that the true value lies between that interval especially if it a small/narrow interval
O5: Sources of Error
Confounding variables:
Is a factor or explanation that may affect a result or conclusion.
Sources of Errors
what is bias
what is selection bias
what study is this a problem for
what si recall bias
Bias- is a systematic error in the design, conduct or analysis of a study that results in a mistaken estimate of an exposure’s effect on the risk of disease.
Selection bias: there is an error in choosing the individuals or groups to take part in a study
This is often a problem in case-control studies
Recall Bias: An error caused by differences in the accuracy or completeness of the recollections retrieved (“recalled”) by study participants regarding events or experiences from the past.
Is an issue in case-control studies
O7: Proving Cause and Effect
Epidemiological studies usually speak of risk factors than causes.
To make the results of a study stronger it is important to show a cause and effect relationship
Elements of Cause and Effect
A study with a large number of subjects is more likely to yield a valid result
The Framingham Heart Study has thousands of participants
This makes the results much more valid than a study of only 30 subjects
Elements of Cause and Effect
The stronger the association measured between the exposure and disease the more likely there is a cause and effect relationship
A RR of 20 in regards to smoking and lung cancer is a much stronger association than a RR of 1
Elements of Cause and Effect
what relationship does exposure and risk have
the more exposure, the more likely what
what are examples
Dose-response relationship between exposure and risk of disease
The more exposure you have the more likely you are to get the disease.
A nurse caring for patients with the Avian Influenza compared to a nurse in a calling station answering patient’s questions. LOTS of exposure to the virus most likely to contract the flu than a the nurse not exposed to the flu virus.
Elements of Cause and Effect
A known biological explanation between exposure and a disease
Is the epidemiological findings consistent with the current biological knowledge?
Rubella causing congenital cataracts
Elements of Cause and Effect
Results are consistent with other investigations
Studies conducted in different populations still have similar results
If an association is found it is expected to see the same results within subgroups of the population and in different populations (unless there is a clear reason to expect differently)
Use in pharmacy practice
Evidence-based medicine is used widely today to make sound medical decisions.
Pharmacoepidemiology studies can be used in the drug therapy decision process.
Studies are published to show that drugs are safe and effective.
Physicians use these studies for treatment guidelines.
All of the following illustrate the purpose of epidemiology in public health EXCEPT
Identifying populations who are at risk for certain diseases
Assessing the effectiveness of interventions
Providing treatment for patients in clinical settings
Determining the importance of the cause of illness
Providing treatment for patients in clinical settings
In March 2021, an outbreak of measles occurred among students at a community college in Boston. This group of cases during this specific time and place is described as a
cluster
HIV/AIDS us one of the worst global diseases in history. It is a/an
pandemic
COVID-19 is a global infectious disease that affected millions. It is a/an
pandemic
The Ebola virus in parts of Africa is in excess of what is expected for this region. The virus is a/an
epidemic
Malaria is present in Africa at all times because of the presence of infected mosquitos. Malaria is ____ in Africa.
endemic
RR vs OR
The relative risk (also known as risk ratio [RR]) is the ratio of risk of an event in one group (e.g., exposed group) versus the risk of the event in the other group (e.g., nonexposed group).
The odds ratio (OR) is the ratio of odds of an event in one group versus the odds of the event in the other group.
PAR simplified
Population Attributable Risk
the proportion of the incidence (the number of case) of a disease in the population that is due to exposure–so the proportion of cases that is due to (or can be attributed to) exposure
how many of the cases are actually due to exposure to the agent? OH OKAY GOTCHA!
incidence example
newly identified cases of a disease or condition per population at risk over a specified timeframe.
[1] An example of incidence would be 795,000 new strokes in the United States, annually.
prevalence example
You talk to all 200 people in your town on a spring day and find 60 of them have allergy symptoms. The point prevalence of allergies in your town would be 30% or 3 in 10 individuals calculated as:
(60 people with allergy symptoms) / (200 people at risk) = 0.3 = 30%
prevalence vs incidence
both have to do with cases in a population
Incidence is a measure of the number of new cases of a characteristic that develop in a population in a specified time period (the number of new cases in a pop. so 14 people in a population have COVID)
prevalence is the proportion of a population who have a specific characteristic in a given time period, regardless of when they first developed the characteristic. (the proportion of cases in a population so 20% of people in a population have COVID)
prevalence
Rates change slowly-not useful for epidemiologic studies
Useful in assessing social impact of disease-affects planning
Prevalence- Estimates the probability of the population being ill at the period of time being studied
Prevalence is often expressed as cases per 100 (percentage) or per 1000 population.
Mortality rate
the incidence of death per unit of time (usually per year) in a population, which can look at all deaths or a specific cause of death. With COVID, measure how many passed form COVID
prevalence
a measure of disease that allows us to determine a person’s likelihood of having a disease
Incidence
the rate of new cases of a disease in a defined population over a defined period
Measures the probability that a healthy person in that PAR will develop that disease during that specific time
Incidence-Expresses the risk of becoming ill.
Must always include a unit of time- such as cases per 10^ n per day, week, month or year
Population Attributable Risk (PAR)
This number should only include people who are potentially susceptible to the disease being studied.
the proportion of the incidence (the number of case) of a disease in the population (exposed and nonexposed) that is due to exposure–so the proportion of cases that is due to (or can be attributed to) exposure
rate
An expression of the frequency with which an event occurs in a defined population
the number of cases in a population
cross-sectional
vs
cohort
both deal with a certain group BUT
Cross-sectional
- measure at ONE time
cohort
- measure over a certain period of time (like over 5 years
The cross-sectional study has an identical structure to the cohort study except that the exposures and outcomes are measured at the same time (i.e. cross-sectionally), whereas in a cohort study outcomes are typically measured after the exposure/s have been measured (i.e. longitudinally).
is the p-value statistically significant?
OR 1.83 (95% CI 1.50 to 2.24); p=0.05
RR 1.66 (95% CI 1.37 to 2.00)
OR 0.83 (95% CI 0.61 to 1.08
RR 0.87 (95% CI 0.66 to 1.15)
RR 2.23; p<0.001
OR 1.83 (95% CI 1.50 to 2.24); p=0.05 - Yes! The 95% CI does not include 1 and the p-value is statistically significant
RR 1.66 (95% CI 1.37 to 2.00) - Yes! The 95% CI does not include 1
OR 0.83 (95% CI 0.61 to 1.08) - No! the 95% CI includes the number 1
RR 0.87 (95% CI 0.66 to 1.15) - No! the 95% CI includes the number 1
RR 2.23; p<0.001 - Yes! The p-value is less than 0.05
RR > 1 positive association
RR < 1 negative association
RR = 1 no association
A study of children age 10-16 in a small community located near high-tension wires for a total of 2 months
Cross-sectional
Case-control
Cohort
cross-sectional
Subjects who have received exercise & nutritional sessions are followed for 5 years to measure health outcomes
Cross-sectional
Case-control
Cohort
cohort
A similar group of subjects with lung cancer are compared to subjects without lung cancer to determine possible associations
Cross-sectional
Case-control
Cohort
case control
case: with lung cancer
control: without lung cancer
what about the epidemic does the epi curve show you
what can it be used for
The curve shows where you are in the course of the epidemic
- Still on the upswing, on the down slope, or after the epidemic has ended.
The curve can be used for evaluation, answering questions like:
- How long did it take for the health department to identify a problem?
- Are intervention measures working?
what is an endemic curve
what does the horizontal axis show
Epidemic curve (epi curve) shows progression of an outbreak over time
The horizontal axis represents the date when a person became ill, also called the date of onset.
what do epi curves show
what does it allow you to distinguish
what does the shape tell you
The epi curve shows the magnitude of the epidemic over time
It permits the investigator to distinguish epidemic from endemic disease
The shape of the epidemic curve may provide clues about the pattern of spread in the population
risk factors
characteristics at the biological, psychological, family, community, or cultural level that precede and are associated with a higher likelihood of negative outcomes.
incubation period
the period between exposure to an infection and the appearance of the first symptoms.
chronic disease
develop over time
Risk factor concept: a particular biologic, lifestyle & social conditions are associated with increased risk for specific chronic diseases
a disease that you have for a long time
which one is statically significant
RR 0.5 p < 0.001
OR 2.5 (95% Cl -2.5 to 5)
RR 2.5 (95% Cl 2 to 5)
OR 1.5 p = 0.8
p < 0.001 is statically significant
OR 2.5 (95% Cl -2.5 to 5) is not because it includes 1
RR 2.5 (95% Cl 2 to 5) is because it does not include 1
OR 1.5 p = 0.8 is more 0.05, has to be less of it to be statistically significant
Subjects who have received exercise & nutritional sessions followed for 5 years to measure health outcomes
A
Cross Sectional
B
Cohort
C
Case Control
B
cohort
Group is followed through time so prospective cohort
A study of children age 10-16 in a community located near high tension wires for 2 months
A
Cross Sectional
B
Cohort
C
Case Control
B
cross-sectional
specific pop. at one point in time
Q
Subjects with lung cancer are compared to subjects without lung cancer
A
Cross Sectional
B
Cohort
C
Case Control
C
case control
Measurement of Association: 95% Confidence Interval
what is the 95% CI used for
what does a high and low CI indicate
what is the 95% CI used for
Is wide or Narrow CI more reliable
what if range includes 1
a range of numbers and we are 95% sure that the true number falls in that range
The 95% confidence interval (CI) is used to estimate the precision of the OR (also used in RR).
A large CI indicates a low level of precision of the OR, whereas a small CI indicates a higher precision of the OR.
In practice, the 95% CI is often used as a proxy for the presence of statistical significance if it does not overlap the null value (e.g. OR=1).
Will look like this: RR=3 (95% CI 1.67-4.2); p<0.5 is a range of numbers
narrow shows a really high precision and shows more confidence than wide
if includes 1 it is not statically significant and we cannot say that there is an association because the true number could be 1 and for OR & RR there is no association
1.67 is higher than 1 so there is a positive assoication, it has to be 1.0 for it to really be 1
how to Design of Case-Control Study
what does it start with
what does it spearate
what are done to both groups
start with population
separate into cases and controls
have both groups each be exposed and not exposed
so in total tested 4 groups:
1 exposed and not exposed for case = 2
1 exposed and not exposed for control = 2
2 + 2 = 4
