Summary Study Design and Interpretation
in Epidemiology and Public Health
2023-2024
Table of content
Week 1 + Week 2 ..................................................................................................................................... 2
Clip| Cohort study ............................................................................................................................... 3
Clip | Case-control ............................................................................................................................... 5
Clip | Prevalence.................................................................................................................................. 6
Clip | Incidence.................................................................................................................................... 6
Clip | Relation between prevalence and incidence ............................................................................. 7
Clip | Ratio measures .......................................................................................................................... 7
Clip | Confounding: what is it? ............................................................................................................ 8
Clip | Confounding: verify criteria ....................................................................................................... 9
Clip | Effect modification: what is it? .................................................................................................. 9
Clip | Introduction to information bias ............................................................................................... 9
Clip | Introduction to selection bias .................................................................................................. 10
Clip | Study designs in analytical epidemiology ................................................................................ 10
E-module: Cohort study..................................................................................................................... 12
E-module | validity ............................................................................................................................ 14
E-module | external validity .............................................................................................................. 14
E-module | selection error ................................................................................................................ 17
E-module | information error............................................................................................................ 30
Chapter 4 Healthy research: study designs for public health p. 114 – 126 ....................................... 36
Chapter 5 Why? Linking exposure and disease p. 156 – 160 ............................................................ 40
Friday reflection week 1 .................................................................................................................... 43
Friday reflection week 2 .................................................................................................................... 45
Week 3 ................................................................................................................................................... 46
Chapter 7 - All that glitters is not gold: the problem of error (p. 187-200) ....................................... 46
Chapter 7 - All that glitters is not gold: the problem of error (p. 200-214). ...................................... 49
Theory reading | Confounding .......................................................................................................... 55
Chapter 8 - Muddied waters: the challenge of confounding p. 217-244 .......................................... 57
E-module | effect modification ......................................................................................................... 67
1
, E-module | Confounding and Effect modification (1) ....................................................................... 68
E-module | Confounding and Effect modification (2) ....................................................................... 69
E-module | confounding exercises .................................................................................................... 69
Friday reflection week 3 .................................................................................................................... 70
Week 4 ................................................................................................................................................... 71
E-module | Study size calculations .................................................................................................... 71
Week 5 + Week 6 ................................................................................................................................... 76
Clip | why causation is relevant......................................................................................................... 76
Clip | causal inference based on counterfactuals ............................................................................. 76
Clip | Rothmans model for causality ................................................................................................. 77
Clip | Causal thinking is not easy: an intriguing question ................................................................. 79
Clip | Relative risks are not a fact of nature but depend on background risk ................................... 79
RR and the role of other component causes ..................................................................................... 81
Lecture systematic reviews and meta-analyses ................................................................................ 81
Chapter 10 – Who sank the boat? Association and causation p. 269-286 ........................................ 86
Chapter 11 – Assembling the building blocks: reviews and their uses p. 288-303 ........................... 91
Friday reflection week 5 .................................................................................................................... 94
Extra | Misconceptions case-cohort and nested case-control .............................................................. 95
Extra | Lecture confounding and effect modification ........................................................................... 96
Extra | Similarities observational studies and meta-analysis .............................................................. 100
The chapters included are from Essential Epidemiology and instruction for students and health
professionals 3rd edition. ISBN: 9781316275948.
Week 1 + Week 2
Refresh knowledge
Clip | Epidemiology: an academic and applied science
Epidemiology in academic setting is about analyzing patterns and causes – to analyse and understand
the patterns. Focus is on generating knowledge on things that are generally applicable.
• Validity and precision – causation
• Relevant to mankind, anytime, anyplace
• The right side of the below figure
Epidemiology in public health setting is also about using this information for prevention in population
– through policies, programs and interventions. This earlier knowledge is applied from the academic
setting.
• Generalizability and importance
2
, • Best practice
• Relevant to me, here and now
• Left side of the below figure
Clip| Cohort study
Biggest advantage – you grow older with your cohort.
Follow the population overtime until disease develops with people who are exposed and
nonexposed. The exposure proceeds the outcome (the disease) – gives a good indicator about the
causality of the relationship.
Purpose of a cohort study:
• To evaluate the occurrence (incidence) of disease in a carefully defined group of people
(monitoring) – prospective.
• To investigate the causes of disease and to establish links between risk factors and health
outcomes (etiology – compare the exposed to the nonexposed). To compare disease
occurrence between exposed and unexposed.
Measurements of exposure and outcome:
• At baseline
− Exposure status
− Covariates: variables that are possible related with the exposure/outcome
3
, • During follow up
− Disease outcome – count the number of disease
− Update information on exposure and covariates
Compare the occurrence of disease in people with and without exposure – incidence proportion or
rate among exposed and incidence proportion or rate among unexposed (incidence proportion or
incidence rate).
Relative risk (RR)
Incidence proportion alcohol consumers = 10/110
Incidence proportion non-alcohol consumers = 15/100
Historical cohort study
The position of the researcher is different. With a cohort, the researcher collects data at baseline and
grows older with the study. With a historical cohort study, the researcher starts at the end – he
identifies a population and observes the incidences as they occur, but goes back to archives to
retrospectively find the previous exposure. Requires good records of past exposures and those long
follow-ups can best be avoided.
Strengths cohort study
• Time sequence can be determined – causality
• Multiple outcomes/risk factors – sub analysis
4
in Epidemiology and Public Health
2023-2024
Table of content
Week 1 + Week 2 ..................................................................................................................................... 2
Clip| Cohort study ............................................................................................................................... 3
Clip | Case-control ............................................................................................................................... 5
Clip | Prevalence.................................................................................................................................. 6
Clip | Incidence.................................................................................................................................... 6
Clip | Relation between prevalence and incidence ............................................................................. 7
Clip | Ratio measures .......................................................................................................................... 7
Clip | Confounding: what is it? ............................................................................................................ 8
Clip | Confounding: verify criteria ....................................................................................................... 9
Clip | Effect modification: what is it? .................................................................................................. 9
Clip | Introduction to information bias ............................................................................................... 9
Clip | Introduction to selection bias .................................................................................................. 10
Clip | Study designs in analytical epidemiology ................................................................................ 10
E-module: Cohort study..................................................................................................................... 12
E-module | validity ............................................................................................................................ 14
E-module | external validity .............................................................................................................. 14
E-module | selection error ................................................................................................................ 17
E-module | information error............................................................................................................ 30
Chapter 4 Healthy research: study designs for public health p. 114 – 126 ....................................... 36
Chapter 5 Why? Linking exposure and disease p. 156 – 160 ............................................................ 40
Friday reflection week 1 .................................................................................................................... 43
Friday reflection week 2 .................................................................................................................... 45
Week 3 ................................................................................................................................................... 46
Chapter 7 - All that glitters is not gold: the problem of error (p. 187-200) ....................................... 46
Chapter 7 - All that glitters is not gold: the problem of error (p. 200-214). ...................................... 49
Theory reading | Confounding .......................................................................................................... 55
Chapter 8 - Muddied waters: the challenge of confounding p. 217-244 .......................................... 57
E-module | effect modification ......................................................................................................... 67
1
, E-module | Confounding and Effect modification (1) ....................................................................... 68
E-module | Confounding and Effect modification (2) ....................................................................... 69
E-module | confounding exercises .................................................................................................... 69
Friday reflection week 3 .................................................................................................................... 70
Week 4 ................................................................................................................................................... 71
E-module | Study size calculations .................................................................................................... 71
Week 5 + Week 6 ................................................................................................................................... 76
Clip | why causation is relevant......................................................................................................... 76
Clip | causal inference based on counterfactuals ............................................................................. 76
Clip | Rothmans model for causality ................................................................................................. 77
Clip | Causal thinking is not easy: an intriguing question ................................................................. 79
Clip | Relative risks are not a fact of nature but depend on background risk ................................... 79
RR and the role of other component causes ..................................................................................... 81
Lecture systematic reviews and meta-analyses ................................................................................ 81
Chapter 10 – Who sank the boat? Association and causation p. 269-286 ........................................ 86
Chapter 11 – Assembling the building blocks: reviews and their uses p. 288-303 ........................... 91
Friday reflection week 5 .................................................................................................................... 94
Extra | Misconceptions case-cohort and nested case-control .............................................................. 95
Extra | Lecture confounding and effect modification ........................................................................... 96
Extra | Similarities observational studies and meta-analysis .............................................................. 100
The chapters included are from Essential Epidemiology and instruction for students and health
professionals 3rd edition. ISBN: 9781316275948.
Week 1 + Week 2
Refresh knowledge
Clip | Epidemiology: an academic and applied science
Epidemiology in academic setting is about analyzing patterns and causes – to analyse and understand
the patterns. Focus is on generating knowledge on things that are generally applicable.
• Validity and precision – causation
• Relevant to mankind, anytime, anyplace
• The right side of the below figure
Epidemiology in public health setting is also about using this information for prevention in population
– through policies, programs and interventions. This earlier knowledge is applied from the academic
setting.
• Generalizability and importance
2
, • Best practice
• Relevant to me, here and now
• Left side of the below figure
Clip| Cohort study
Biggest advantage – you grow older with your cohort.
Follow the population overtime until disease develops with people who are exposed and
nonexposed. The exposure proceeds the outcome (the disease) – gives a good indicator about the
causality of the relationship.
Purpose of a cohort study:
• To evaluate the occurrence (incidence) of disease in a carefully defined group of people
(monitoring) – prospective.
• To investigate the causes of disease and to establish links between risk factors and health
outcomes (etiology – compare the exposed to the nonexposed). To compare disease
occurrence between exposed and unexposed.
Measurements of exposure and outcome:
• At baseline
− Exposure status
− Covariates: variables that are possible related with the exposure/outcome
3
, • During follow up
− Disease outcome – count the number of disease
− Update information on exposure and covariates
Compare the occurrence of disease in people with and without exposure – incidence proportion or
rate among exposed and incidence proportion or rate among unexposed (incidence proportion or
incidence rate).
Relative risk (RR)
Incidence proportion alcohol consumers = 10/110
Incidence proportion non-alcohol consumers = 15/100
Historical cohort study
The position of the researcher is different. With a cohort, the researcher collects data at baseline and
grows older with the study. With a historical cohort study, the researcher starts at the end – he
identifies a population and observes the incidences as they occur, but goes back to archives to
retrospectively find the previous exposure. Requires good records of past exposures and those long
follow-ups can best be avoided.
Strengths cohort study
• Time sequence can be determined – causality
• Multiple outcomes/risk factors – sub analysis
4
