Ramdoc - Acpm - Epidemiology

Front 1

What are two ways to measure disease frequency?

Back 1

Counts - # with disease and rates- counts with a denominator (size of population at similar risk). Rates also imply or are associated with a period of time.

Front 2

What is the numerator and denominator for incidence

Back 2

Num = number of NEW cases in a period of time
Den = number of people INITIALLY at risk

Front 3

What are the other names for incidence?

Back 3

Cumulative incidence
incidence rate
attack rate
risk of disease
Probability of getting disease

Front 4

How is incidence density different?

Back 4

The numerator is the same: number of new cases in a time. However, the denominator is "total amount of person-time at risk contributed during the time."

Incidence density accounts for the time someone contributes to the denominator (population at risk for disease) before they become part of the numerator (counts of diseased)

Front 5

Define the numerator and denominator of prevalence

Back 5

Numerator = number of EXISTING cases of the disease at a specific POINT IN TIME
denominator =
size of the population at the SAME POINT IN TIME

If data is for point in time it is point prevalence.
If data is for a specified time. It is period prevalence
Period prevalence is different from incidence because the numerator in incidence includes only new cases where is the numerator in any prevalence is all cases.

Front 6

Define disease specific Mortality Rate

Back 6

Numerator = number of people DYING in a specific time.
Denominator = average number people ALIVE during that period of time

Front 7

Define Case Fatality Rate

Back 7

Numerator = number people who die of the disease
Denominator = total number people who get the disease

Case Fatality Rate measures disease prognosis rather than disease frequency

Some board questions will ask about disease frequency when you are given disease prognosis in the question stem. If this is the case you cannot answer the question with the information given.

Front 8

Define Proportional Mortality

Back 8

Numerator = number of deaths due to a disease in a specific time.
Denominator = total number of deaths during that time.

In Proportional Mortality the denominator is the total number of DEATHS.
In Disease Specific Mortality the denominator is average number of people ALIVE.

Front 9

Define Mortality Rate

Back 9

Numerator = number of people DYING in a specific time.
Denominator = average number people ALIVE during that period of time

Front 10

Define Case Fatality Rate

Back 10

Numerator = number people who die of the disease
Denominator = total number people who get the disease

Case Fatality Rate measures disease prognosis rather than disease frequency

Some board questions will ask about disease frequency when you are given disease prognosis in the question stem. If this is the case you cannot answer the question with the information given.

Front 11

Define Proportional Mortality

Back 11

Numerator = number of deaths due to a disease in a specific time.
Denominator = total number of deaths during that time.

In Proportional Mortality the denominator is the total number of DEATHS.
In Disease Specific Mortality the denominator is average number of people ALIVE.

Front 12

Define Mortality (also known as disease specific mortality rate)

Back 12

Numerator = number of people DYING in a specific time.
Denominator = average number people ALIVE during that period of time

Front 13

Define Proportional Mortality

Back 13

Numerator = number of deaths due to a disease in a specific time.
Denominator = total number of deaths during that time.

In Proportional Mortality the denominator is the total number of DEATHS.
In Mortality the denominator is the avg number of people ALIVE.

Front 14

Define Case Fatality Rate

Back 14

Numerator = number people who die of the disease
Denominator = total number people who get the disease

Case Fatality Rate measures disease prognosis rather than disease frequency

Some board questions will ask about disease frequency when you are given disease prognosis in the question stem. If this is the case you cannot answer the question with the information given.

Front 15

KEY RELATIONSHIP

Back 15

Prevalence = Incidence x Duration

This equation is true when prevalence is less than 10% and incidence and duration are stable over time.

This is important for predicting. Prevalence will increase if incidence or duration increase.

Front 16

KEY RELATIONSHIP

Back 16

Mortality = Incidents x Case Fatality

This is true when incidence and case fatality are stable over time.

Front 17

Define Neonatal Death Rate

Back 17

Numerator = annual number of deaths in the first 28 days of life
Denominator = annual number of live births

Front 18

Define Infant Death Rate

Back 18

Numerator = annual number of deaths in the first year of life
Denominator = annual number of live births

Front 19

Define years of potential life lost (YPLL)

Back 19

It is a measure of premature mortality.

# deaths X (75 - median of age range)

75 is the avg age at death
Median of age range is middle age of the group you are concerned with. ex: 0-10 year olds = 5

So if 2 kids in the age group 0-10 die then 2 (75-5) = 140 YPLL

Front 20

Define years of potential life lost (YPLL)

Back 20

It is a measure of premature mortality.

# deaths X (75 - median of age range)

75 is the avg age at death
Median of age range is middle age of the group you are concerned with. ex: 0-10 year olds = 5

So if 2 kids in the age group 0-10 die then 2 (75-5) = 140 YPLL

Front 21

Be careful of lifetime risk estimates that don't give you that life expectancy on which they were based

Back 21

For example: one in 11 women get breast cancer if they live to 75 (the average life expectancy females)

often quoted is that one in eight women get breast cancer but this is based on a life expectancy of 90 years.

Front 22

What are critical factors when interpreting the disease frequency measures of epidemiology (counts and rates)

Back 22

You must consider the source of your numerator and denominator.
Some numerator data (counts) are limited by quality of reporting and case definition.

Denominator data must condier the kinds of people at risk. Are you including people who are not at risk or only those at high risk?

Factors include gender, location, season, disease cycles, etc.

Front 23

What is the cohort effect?

Back 23

Variation in disease rate between different age cohorts. Example. in 1950 50% of men smoked. Therefore mortality from smoking in men alive in 1950 is higher than men not alive in 1950. There will be a migrating spike over time as this cohort ages.

Front 24

After measuring Disease Frequency (counts and rates) that next common set of epidemiologic measures is the Measure of Excess Risk

Back 24

These rely on 2 x 2 table computations.

They are measures such as relative risk and odds ratio

Front 25

Relative risk (risk, risk ratio, rate ratio) is defined as:

Back 25

RR =
Numerator = incidence in the exposed divided by
Denominator = incidence in the unexposed

This is [a/(a+b)] / [c/(c+d)]

With disease present on the top of the 2x2 table these calculation go horizontal.

Front 26

After measuring Disease Frequency (counts and risks) that next common set of epidemiologic measures is the Measure of Excess Risk

Back 26

These rely on 2 x 2 table computations.

They are measures such as relative risk and odds ratio

Front 27

Relative risk (risk, risk ratio, rate ratio) is defined as:

Back 27

RR =
Numerator = incidence in the exposed divided by
Denominator = incidence in the unexposed

This is [a/(a+b)] / [c/(c+d)]

With disease present on the top of the 2x2 table these calculation go horizontal.

Front 28

When can the relative risk be estimated by the odds ratio?

Back 28

When the disease is rare.

This means that A and C. are very small values compared to B and C.

Front 29

What is the formula for odds ratio:

Back 29

OR = (a/b) / (c/d) = ad / bc

Front 30

What is attributable risk

Back 30

Incidence in the exposed minus incidence in the unexposed.

This is [a/(a+b)] - [c/c+d)]

when you divide this quantity by the first term, [a/(a+b)], and multiply by 100 you get Attribute Risk Percent

Front 31

Attributable Risk and Attributable Risk Percent interpretation:

Back 31

is the absolute number of excess cases because of exposure and percent of excess cases excessive exposure

Front 32

Population Attributable Risk is:

Back 32

Total incidence minus incidence in the unexposed.

This is the proportion of all case of disease that are due to exposure.

Front 33

STUDY DESIGN (concepts of studies)

Back 33

Bias

Study Types

Epidemiologic evidence and causal inference

Confounding effect modification

Disease screening

Infectious disease epidemiology

Front 34

Define bias

Back 34

Bias- result of a systematic error in the design or
conduct of a study resulting in false or misleading
conclusions.

Biases can EITHER falsely raise OR lower the
estimates of risks.

Front 35

Define random versus systematic error

Back 35

Random error is noise in your data is non-directional examples are precision of diagnosis precision of exposure measurement precision of confounder assessment. It does not invalidate a positive result. It may invalidate a negative result.

Systematic error is directional. They are biases. It is an inherent flaw in the study design. It may invalidate either a positive or negative result.

Front 36

Name three major classifications of bias

Back 36

Selection bias -- different probabilities of being included in the study

Information bias -- systematic tendency for individuals to be MISCLASSIFIED into the wrong exposure or outcome categories

Confounding (a specific type of bias)

Front 37

What is recall bias

Back 37

Those without outcome won't remember their exposure. Those with the outcome may exaggerate the amount of their exposure.

Front 38

What is measurement bias

Back 38

An example would be misclassification into an exposed group based on the DEFINITION or interpretation of exposure.

Front 39

What is internal validity

Back 39

The method of the study truly supports the claim of a causal relationship between exposure and outcome.

Front 40

What external validity

Back 40

The strength of the study's generalizability to a population. It can also be thought of as the strength of its inductive reasoning.

Front 41

What is reliability

Back 41

Reliability is the repeatability; the extent to which the results obtained by a test is replicated if the test is repeated (accuracy)

Front 42

What is confounding

Back 42

Confounding is a third variable related to exposure and disease. It biases the estimate of the exposures affect on the outcome. Common confounders or age, gender, tobacco use, BMI, socioeconomic status.

Front 43

Studies have a hierarchy. What are the three main levels to this hierarchy.

Back 43

Descriptive studies followed by observational studies followed by interventional or experiment studies.

Descriptive studies are also described as hypothesis generating

Observational and Interventional Studies Are Considered Hypothesis Testing.

Front 44

Name two types of descriptive hypothesis generating studies

Back 44

Case report
Case series
Descriptive Epidemiologic Study (describes how something varies with the change in a factors such as time, gender, or location)

Front 45

Name four types of observational studies.

Back 45

Correlational or ecological study
Cross-Sectional or prevalence study
Case-Control study
Cohort Study

Front 46

What are the two types of cohort studies

Back 46

Prospective and retrospective

Front 47

What are the two types of interventional studies

Back 47

Randomized controlled trials and croissants experimental

Front 48

Descriptive studies

Back 48

Lack of offices

do not establish causal associations

Front 49

Analytic studies

Back 49

Usually have hypothesis specified in advance

intent on establishing a causal association

The premise is that a variation in exposure is associated with a variation in outcome.