EPI4921: Observational Research Module Tutorial document:
Problem 1: Measures of risk estimates and associations:
➔ Learning goals Tutorial 1:
1. How do we calculate risk?
From gordis epidemiology book:
- Absolute risk: The incidence of a disease in a population is termed the
absolute risk. Absolute risk can indicate the magnitude of the risk in a
group of people with a certain exposure.
- Relative risk: The relative risk can also be defined as the probability of
an event (developing a disease) occurring in exposed people compared
with the probability of the event in unexposed people, or as the ratio of
these two probabilities.
- Risk → always means the probability that an individual in a defined group
will develop a disease or outcome over a specified period of time.
- Risk difference is the same as incidence proportion
- Risk → risk difference between exposed vs. unexposed. This case is an
absolute risk.
2. What is the difference between absolute and relative measures?
- Absolute measures → give you the actual difference or actual occurrence
of the disease. They answer “how much disease (or risk) is there?”, “how
many more cases occur in one group vs. another?”. Absolute measures:
1. Cumulative incidence
2. Risk difference
- Relative measures → This is the ratio of two risks or rates. They
answer “how many times higher (or lower) is the risk in one group
compared to another?”. Relative measures:
1. Relative risk
2. Odds ratio
3. What is the APe, how do we interpret and calculate the APe?
- APe → attributable proportion among the exposed. It tells us what
proportion of cases in the exposed group can be attributed to the exposure.
what does the odds ratio really mean and how does it differ from the
relative risk in short?
- APe only looks at the exposed group
- Interpretation: among those who are exposed, what fraction of their
disease could theoretically be prevented if the exposures were eliminated?
,4. What is the APt, how do we calculate it and interpret it?
- APt = attributable proportion in the total population
- APt is also called the Etiological fraction.
- Apt → looks at the total population
- Interpretation → In the whole population (exposed and unexposed), what
proportion of cases is due to the exposure.
- Both measure the contribution of exposure in causing disease.
5. What is the difference between APt and APe?
6. What is the difference between OR and RR?
- RR: compares the probabilities (risks) directly
- OR: compares the odds of an event happening in one group to the
odds in another group.
7. How are OR and RR related?
- Relation: if the disease is rate, OR=RR.
,8. How to calculate and interpret the OR and RR?
9. What is the incidence rate?
10.What is the cumulative incidence ?
11.How to calculate and interpret the incidence rate and cumulative incidence ?
- Incidence rate interpretation: it's the “speed” or “density” at which new
cases occur in a population.
Example: If the incidence rate is, 0.007 per person year that means:
- On average 7 new cases occur per 1000 person years. So if we
followed 1000 people for one year, we would expect about 7 new
cases. This “events per person-time” view helps us picture it like a
flow of new cases through time, not just a proportion at the end of
a period.
- Cumulative incidence: risk of becoming a case during that time. Its the
probability that a person in the at-risk group will develop the outcome
within the time period.
, Example:
- If we are following 100 people for 1 year to see who gets the flu. At the
end of the year, 20 people got the flu. So the cumulative incidence is
20/100= 0.20. We can also say that there was a 20% chance that a person
in this group would get the flu within 1 year.
12.What is Incidence Rate Ratio ( IRR )?
13.How to calculate and interpret the IRR (Incidence rate ratio)?
- Interpretation of the IRR: how many times higher the risk over time is
in the exposure.
14.What is the cumulative incidence ratio ( CIR ) ?
15.How to calculate and interpret the CIR “Cumulative incidence ratio”?
- Interpretation of CIR: how many times higher the risk over time is in the
exposure.
Why standardisation:
- A good way to compare things. We use standardization if we want to
compare several rates.
- We create a reference population
- These are not real numbers but fictitious numbers
- We can not interpret a standardised rate without having a comparison
group.
➔ Exercises:
ALL ON PAPER!
Problem 1: Measures of risk estimates and associations:
➔ Learning goals Tutorial 1:
1. How do we calculate risk?
From gordis epidemiology book:
- Absolute risk: The incidence of a disease in a population is termed the
absolute risk. Absolute risk can indicate the magnitude of the risk in a
group of people with a certain exposure.
- Relative risk: The relative risk can also be defined as the probability of
an event (developing a disease) occurring in exposed people compared
with the probability of the event in unexposed people, or as the ratio of
these two probabilities.
- Risk → always means the probability that an individual in a defined group
will develop a disease or outcome over a specified period of time.
- Risk difference is the same as incidence proportion
- Risk → risk difference between exposed vs. unexposed. This case is an
absolute risk.
2. What is the difference between absolute and relative measures?
- Absolute measures → give you the actual difference or actual occurrence
of the disease. They answer “how much disease (or risk) is there?”, “how
many more cases occur in one group vs. another?”. Absolute measures:
1. Cumulative incidence
2. Risk difference
- Relative measures → This is the ratio of two risks or rates. They
answer “how many times higher (or lower) is the risk in one group
compared to another?”. Relative measures:
1. Relative risk
2. Odds ratio
3. What is the APe, how do we interpret and calculate the APe?
- APe → attributable proportion among the exposed. It tells us what
proportion of cases in the exposed group can be attributed to the exposure.
what does the odds ratio really mean and how does it differ from the
relative risk in short?
- APe only looks at the exposed group
- Interpretation: among those who are exposed, what fraction of their
disease could theoretically be prevented if the exposures were eliminated?
,4. What is the APt, how do we calculate it and interpret it?
- APt = attributable proportion in the total population
- APt is also called the Etiological fraction.
- Apt → looks at the total population
- Interpretation → In the whole population (exposed and unexposed), what
proportion of cases is due to the exposure.
- Both measure the contribution of exposure in causing disease.
5. What is the difference between APt and APe?
6. What is the difference between OR and RR?
- RR: compares the probabilities (risks) directly
- OR: compares the odds of an event happening in one group to the
odds in another group.
7. How are OR and RR related?
- Relation: if the disease is rate, OR=RR.
,8. How to calculate and interpret the OR and RR?
9. What is the incidence rate?
10.What is the cumulative incidence ?
11.How to calculate and interpret the incidence rate and cumulative incidence ?
- Incidence rate interpretation: it's the “speed” or “density” at which new
cases occur in a population.
Example: If the incidence rate is, 0.007 per person year that means:
- On average 7 new cases occur per 1000 person years. So if we
followed 1000 people for one year, we would expect about 7 new
cases. This “events per person-time” view helps us picture it like a
flow of new cases through time, not just a proportion at the end of
a period.
- Cumulative incidence: risk of becoming a case during that time. Its the
probability that a person in the at-risk group will develop the outcome
within the time period.
, Example:
- If we are following 100 people for 1 year to see who gets the flu. At the
end of the year, 20 people got the flu. So the cumulative incidence is
20/100= 0.20. We can also say that there was a 20% chance that a person
in this group would get the flu within 1 year.
12.What is Incidence Rate Ratio ( IRR )?
13.How to calculate and interpret the IRR (Incidence rate ratio)?
- Interpretation of the IRR: how many times higher the risk over time is
in the exposure.
14.What is the cumulative incidence ratio ( CIR ) ?
15.How to calculate and interpret the CIR “Cumulative incidence ratio”?
- Interpretation of CIR: how many times higher the risk over time is in the
exposure.
Why standardisation:
- A good way to compare things. We use standardization if we want to
compare several rates.
- We create a reference population
- These are not real numbers but fictitious numbers
- We can not interpret a standardised rate without having a comparison
group.
➔ Exercises:
ALL ON PAPER!
