Title: Epidemiological measures
1Epidemiological measures
- Hein Stigum
- Presentation, data and programs at
- http//folk.uio.no/heins/
2Epidemiological measures
- Frequency
- prevalence
- incidence
- Association
- Risk difference
- Risk ratio
- Odds ratio
- Potential impact
- Attributable fraction
How much disease?
3Population types
- Closed Population
- Add none
- Loose only by death
- Open population
- Add immigration, birth
- Loose emigration, death
4Mathematical concepts
5Frequency measures
6Disease frequency
Theoretical concept
Estimator
!
7Disease frequency depicted
t
a
8Prevalence example
9Incidence proportion example
1. If closed population
2a. If open population
2b. If open population
10Incidence rate example
1.
2.
3.
11Incidence rate in small cohort
12Incidence of hip fracture, age 65
Incidence rate pr 10 000 person years
(Lofthus et al. 2001)
13Odds and probability
Jan-10
H.S.
13
14Competing risk
15Prevalence vs incidence
At steady state and no migration
Age specific prevalence? ?
Migration!
16Epidemiology vs survival analysis
Survival
Epidemiology
- Disease
- Lost to follow up
- Incidence rate
- Incidence proportion
- Event
- Censored
- Hazard rate
- 1-SurvivalFailure
Survival analysis is closely linked to
epidemiology Also used for nested case-control
designs
17Confidence intervals proportions
95 confidence interval
18Confidence intervals rates
95 confidence interval
19Adjusted measures
Crude R
Standardized R
20Epidemiological measures
- Frequency
- prevalence
- incidence
- Association
- Risk difference
- Risk ratio
- Odds ratio
- Potential impact
- Attributable fraction
How much disease?
More disease among exposed?
21Association measures
22Association measures
- More disease among exposed?
- Compare frequency among exposed1 and unexposed0
- Difference
- Risk Difference
- Ratio
- Risk Ratio
- Odds Ratio
0no effect
1no effect
1no effect
23RR and RD example
Disease lung cancer Exposure smoking
24Relative risk
Risk ratio rate ratio for short-term risks Both
termed relative risk
25Odds ratio
- Traditional Case-Control study
- No prevalence or incidence of disease
- No RR for disease
- OR (odds ratio)
prisk of disease
If disease is rare
26Odds ratio calculation
OR1, no association
27Bullying, OR example
- Bullying in the nordic countries
- 17 114 children, 2 584 bullied
28(No Transcript)
29RR and OR depicted
Risks and Risk Ratio
Odds and Odds Ratio
30Adjusted measures
Remove the effect of confounders in regression
models
31Epidemiological measures
- Frequency
- prevalence
- incidence
- Association
- Risk difference
- Risk ratio
- Odds ratio
- Potential impact
- Attributable fraction
How much disease?
More disease among exposed?
How important?
32Impact measures
33Potential impact
I incidence d disease
34Attributable fraction, estimation
- From crude relative risk
- From adjusted model
Or use Stata aflogit