Training Study Designs in Molecular Epidemiology
Please use the paper by Vineis et al. BMJ. 2005 Feb 5;330(7486):277
PART I
1. What is the research question?
What is the association between environmental tobacco smoke and lung cancer, upper
respiratory cancers and death from COPD or emphysema in non-smokers and people who
had not smoked for more than 10 years?
2. What are the main dependent and independent variables?
Dependent variables:
- Age at diagnosis or last contact
Independent variables:
- Smoke exposure (questionnaires)
o Any exposure
o Place of exposure
o Exposure during childhood
- Sex
o Male or female
- Smoking habit
o Former or never
- Country
- Education in four levels
- Energy intake
- Consumption of fruit and vegetables
- Physical activity
3. What is the original design of the European Prospective Investigation into Cancer
(EPIC) study (not the hybrid/combined design)? And schematically give a representation of
this design.
EPIC is a prospective cohort study
EXPOSED DISEASE
NO DISEASE
RECRUITMENT STUDY POPULATION EXPOSURE
(1993-1998) MEASUREMENT
NOT EXPOSED DISEASE
NO DISEASE
, 4. What are advantages and disadvantages of this design compared to other
observational and experimental designs you know (discuss at least three other designs,
not the hybrid/combined designs)?
Cohort vs case-control
Advantages:
- Suitable for rare exposures
- More reliable
Disadvantages:
- Not suitable for rare diseases
- Long time
- Expensive
Cohort vs RCT
Advantages:
- Most of the times ethical approved (whereas RCT might not be ethical)
Disadvantage:
- Long time
Ecological study not feasible because of lack of information
Cross-sectional study feasible but internal validity is compromised due to potential
information bias
Case-control study feasible but internal validity is compromised due to potential
information bias and selection bias
Retrospective cohort study not feasible because of lack of information on exposure
RCT not ethical
5. What other specific designs were chosen by the scientists of this research and why?
The GenAir is a nested case-control study, which studies the relation between air pollution
or environmental tobacco smoke and newly diagnosed cancers of the bladder, lung, oral
cavity, pharynx or larynx o r leukaemia.
Advantages:
- Does only analyse all the relevant data (lower costs and time-consuming)
o All the cases
o Sample of controls
- Cheaper compared to an original cohort study
- High quality biomarker assessment
It is inefficient to assess the biomarkers in the full cohort extremely expensive
Therefore do this in a smaller sample of the cohort.
matching on former smoker would help in the effect of confounder
6. Give a schematic representation of these designs
Please use the paper by Vineis et al. BMJ. 2005 Feb 5;330(7486):277
PART I
1. What is the research question?
What is the association between environmental tobacco smoke and lung cancer, upper
respiratory cancers and death from COPD or emphysema in non-smokers and people who
had not smoked for more than 10 years?
2. What are the main dependent and independent variables?
Dependent variables:
- Age at diagnosis or last contact
Independent variables:
- Smoke exposure (questionnaires)
o Any exposure
o Place of exposure
o Exposure during childhood
- Sex
o Male or female
- Smoking habit
o Former or never
- Country
- Education in four levels
- Energy intake
- Consumption of fruit and vegetables
- Physical activity
3. What is the original design of the European Prospective Investigation into Cancer
(EPIC) study (not the hybrid/combined design)? And schematically give a representation of
this design.
EPIC is a prospective cohort study
EXPOSED DISEASE
NO DISEASE
RECRUITMENT STUDY POPULATION EXPOSURE
(1993-1998) MEASUREMENT
NOT EXPOSED DISEASE
NO DISEASE
, 4. What are advantages and disadvantages of this design compared to other
observational and experimental designs you know (discuss at least three other designs,
not the hybrid/combined designs)?
Cohort vs case-control
Advantages:
- Suitable for rare exposures
- More reliable
Disadvantages:
- Not suitable for rare diseases
- Long time
- Expensive
Cohort vs RCT
Advantages:
- Most of the times ethical approved (whereas RCT might not be ethical)
Disadvantage:
- Long time
Ecological study not feasible because of lack of information
Cross-sectional study feasible but internal validity is compromised due to potential
information bias
Case-control study feasible but internal validity is compromised due to potential
information bias and selection bias
Retrospective cohort study not feasible because of lack of information on exposure
RCT not ethical
5. What other specific designs were chosen by the scientists of this research and why?
The GenAir is a nested case-control study, which studies the relation between air pollution
or environmental tobacco smoke and newly diagnosed cancers of the bladder, lung, oral
cavity, pharynx or larynx o r leukaemia.
Advantages:
- Does only analyse all the relevant data (lower costs and time-consuming)
o All the cases
o Sample of controls
- Cheaper compared to an original cohort study
- High quality biomarker assessment
It is inefficient to assess the biomarkers in the full cohort extremely expensive
Therefore do this in a smaller sample of the cohort.
matching on former smoker would help in the effect of confounder
6. Give a schematic representation of these designs
