Causation & Study Design: HIV & AIDS Flashcards
Study designs
- Which study design is reflected in each? (A-D)

- Which study design is reflected in each? (A-D)
- A = Cross-Sectional
- B = Case-control
- C = Cohort
- D = Randomised Control Trial
Study designs
- Which study design is reflected in each? (A-D)

- Which study design is reflected in each? (A-D)
- A = Cross-Sectional
- B = Case-control
- C = Cohort
- D = Randomised Control Trial
Study designs
- Which study design is reflected in each? (A-D)

- Which study design is reflected in each? (A-D)
- A = Cross-Sectional
- B = Case-control
- C = Cohort
- D = Randomised Control Trial
Study designs
- Which study design is reflected in each? (A-D)

- Which study design is reflected in each? (A-D)
- A = Cross-Sectional
- B = Case-control
- C = Cohort
- D = Randomised Control Trial
Cross-sectional study
- A defined population is surveyed to simultaneously measure
- …/ … status
- …
- Sample selected using inclusion and exclusion criteria
- Different to case-control (sample selected based on outcome status)
- Different to cohort (sample based on exposure status)
- Could be general population or clinic-based
- Prevalence is reported for the population as a whole, and often for subgroups
- A defined population is surveyed to simultaneously measure
- Disease/ condition status (e.g. infertility)
- Exposure (e.g. sedentary lifestyle, alcohol intake)
- Sample selected using inclusion and exclusion criteria
- Different to case-control (sample selected based on outcome status)
- Different to cohort (sample based on exposure status)
- Could be general population or clinic-based
- Prevalence is reported for the population as a whole, and often for subgroups

Cross-sectional study
- A defined population is surveyed to simultaneously measure
- Disease/ condition status (e.g. infertility)
- Exposure (e.g. sedentary lifestyle, alcohol intake)
- Sample selected using … and … criteria
- Different to case-control (sample selected based on outcome status)
- Different to cohort (sample based on exposure status)
- Could be general … or …-based
- Prevalence is reported for the population as a whole, and often for subgroups
- A defined population is surveyed to simultaneously measure
- Disease/ condition status (e.g. infertility)
- Exposure (e.g. sedentary lifestyle, alcohol intake)
- Sample selected using inclusion and exclusion criteria
- Different to case-control (sample selected based on outcome status)
- Different to cohort (sample based on exposure status)
- Could be general population or clinic-based
- Prevalence is reported for the population as a whole, and often for subgroups

Cross-sectional study
- A defined population is surveyed to simultaneously measure
- Disease/ condition status (e.g. infertility)
- Exposure (e.g. sedentary lifestyle, alcohol intake)
- Sample selected using inclusion and exclusion criteria
- Different to case-control (sample selected based on … status)
- Different to cohort (sample based on exposure status)
- Could be general population or clinic-based
- Prevalence is reported for the population as a …, and often for …
- A defined population is surveyed to simultaneously measure
- Disease/ condition status (e.g. infertility)
- Exposure (e.g. sedentary lifestyle, alcohol intake)
- Sample selected using inclusion and exclusion criteria
- Different to case-control (sample selected based on outcome status)
- Different to cohort (sample based on exposure status)
- Could be general population or clinic-based
- Prevalence is reported for the population as a whole, and often for subgroups

Cross-sectional study
- A defined population is surveyed to simultaneously measure
- Disease/ condition status (e.g. infertility)
- … (e.g. sedentary lifestyle, alcohol intake)
- Sample selected using inclusion and exclusion criteria
- Different to case-control (sample selected based on outcome status)
- Different to cohort (sample based on … status)
- Could be general population or clinic-based
- Prevalence is reported for the population as a whole, and often for subgroups
- A defined population is surveyed to simultaneously measure
- Disease/ condition status (e.g. infertility)
- Exposure (e.g. sedentary lifestyle, alcohol intake)
- Sample selected using inclusion and exclusion criteria
- Different to case-control (sample selected based on outcome status)
- Different to cohort (sample based on exposure status)
- Could be general population or clinic-based
- Prevalence is reported for the population as a whole, and often for subgroups

Cohort study
- Used to
- calculate … (new onset cases of a disease/condition/outcome)
- identify … (risk factors) for particular outcome
- Participants selected into exposed vs non-exposed group (no allocation made by the researcher)
- Two groups as similar as possible (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential confounders) at selection
- Exposures of interest are not measured or matched at selection
- At study end, relative numbers of new disease occurrences compared across groups
- Retrospective or prospective
- Used to
- calculate incidence (new onset cases of a disease/condition/outcome)
- identify exposures (risk factors) for particular outcome
- Participants selected into exposed vs non-exposed group (no allocation made by the researcher)
- Two groups as similar as possible (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential confounders) at selection
- Exposures of interest are not measured or matched at selection
- At study end, relative numbers of new disease occurrences compared across groups
- Retrospective or prospective

Cohort study
- Used to
- calculate incidence (new onset cases of a disease/condition/outcome)
- identify exposures (risk factors) for particular outcome
- Participants selected into … vs non-… group
- Two groups as … as possible (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential confounders) at selection
- Exposures of interest are not measured or matched at selection
- At study end, relative numbers of new disease occurrences compared across groups
- Retrospective or prospective
- Used to
- calculate incidence (new onset cases of a disease/condition/outcome)
- identify exposures (risk factors) for particular outcome
- Participants selected into exposed vs non-exposed group (no allocation made by the researcher)
- Two groups as similar as possible (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential confounders) at selection
- Exposures of interest are not measured or matched at selection
- At study end, relative numbers of new disease occurrences compared across groups
- Retrospective or prospective

Cohort study
- Used to
- calculate … (new onset cases of a disease/condition/outcome)
- identify exposures (risk factors) for particular outcome
- Participants selected into exposed vs non-exposed group (no allocation made by the researcher)
- Two groups as similar as possible (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential confounders) at selection
- Exposures of interest are not measured or matched at selection
- At study end, relative numbers of new disease occurrences compared across …
- Retrospective or prospective
- Used to
- calculate incidence (new onset cases of a disease/condition/outcome)
- identify exposures (risk factors) for particular outcome
- Participants selected into exposed vs non-exposed group (no allocation made by the researcher)
- Two groups as similar as possible (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential confounders) at selection
- Exposures of interest are not measured or matched at selection
- At study end, relative numbers of new disease occurrences compared across groups
- Retrospective or prospective

Cohort study
- Used to
- calculate incidence (… … cases of a disease/condition/outcome)
- identify exposures (… …) for particular outcome
- Can be …spective or ….spective
- Used to
- calculate incidence (new onset cases of a disease/condition/outcome)
- identify exposures (risk factors) for particular outcome
- Can be retrospective or prospective

Case-control study
- Used to identify relevant … (not used to calculate …)
- Two groups of participants are selected – one with condition (cases) and one without (controls)
- Controls selected to be as similar as possible to the cases (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential confounders) at selection
- Exposures of interest are not measured or matched at selection
- Always …→Past exposure/s in both groups
- E.g. interview/survey, historical records
- Used to identify relevant exposures (not used to calculate incidence)
- Two groups of participants are selected – one with condition (cases) and one without (controls)
- Controls selected to be as similar as possible to the cases (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential confounders) at selection
- Exposures of interest are not measured or matched at selection
- Always retrospective→Past exposure/s in both groups
- E.g. interview/survey, historical records

Case-control study
- Used to identify relevant exposures (not used to calculate incidence)
- Two groups of participants are selected – one with condition (…) and one without (…)
- Controls selected to be as … as possible to the … (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential confounders) at selection
- Exposures of interest are not measured or matched at selection
- Always retrospective→Past exposure/s in both groups
- E.g. interview/survey, historical records
- Used to identify relevant exposures (not used to calculate incidence)
- Two groups of participants are selected – one with condition (cases) and one without (controls)
- Controls selected to be as similar as possible to the cases (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential confounders) at selection
- Exposures of interest are not measured or matched at selection
- Always retrospective→Past exposure/s in both groups
- E.g. interview/survey, historical records

Case-control study
- Used to identify relevant exposures (not used to calculate incidence)
- Two groups of participants are selected – one with condition (cases) and one without (controls)
- Controls selected to be as similar as possible to the cases (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential …) at selection
- Exposures of interest are … … or matched at selection
- Always retrospective→Past exposure/s in both groups
- E.g. interview/survey, historical records
- Used to identify relevant exposures (not used to calculate incidence)
- Two groups of participants are selected – one with condition (cases) and one without (controls)
- Controls selected to be as similar as possible to the cases (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential confounders) at selection
- Exposures of interest are not measured or matched at selection
- Always retrospective→Past exposure/s in both groups
- E.g. interview/survey, historical records

Case-control study
- Used to identify relevant exposures (not used to calculate …)
- Two groups of participants are selected – one with condition (cases) and one without (controls)
- Controls selected to be as similar as possible to the cases (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential confounders) at selection
- Exposures of interest are not measured or matched at selection
- Always retrospective→… …/s in both groups
- E.g. interview/survey, historical records
- Used to identify relevant exposures (not used to calculate incidence)
- Two groups of participants are selected – one with condition (cases) and one without (controls)
- Controls selected to be as similar as possible to the cases (e.g. age, gender, occupation, stage of illness, etc.)
- Variables not of interest are matched (i.e. potential confounders) at selection
- Exposures of interest are not measured or matched at selection
- Always retrospective→Past exposure/s in both groups
- E.g. interview/survey, historical records

Cohort studys can be retrospective or prospective, but case-control studies are always …
retrospective
RCT
- Used to test the … and …/… of interventions
- Participants selected and then randomly allocated to either receive the intervention of interest or to receive a control (or no) intervention
- Random allocation should ensure that the two groups are essentially identical before provision of intervention or control
- Outcomes are compared across the two groups at the end of the trial
- Used to test the safety and efficacy/effectiveness of interventions
- Participants selected and then randomly allocated to either receive the intervention of interest or to receive a control (or no) intervention
- Random allocation should ensure that the two groups are essentially identical before provision of intervention or control
- Outcomes are compared across the two groups at the end of the trial

RCT
- Used to test the safety and efficacy/effectiveness of interventions
- Participants selected and then … allocated to either receive the …. of interest or to receive a …
- Random allocation should ensure that the two groups are essentially identical before provision of intervention or control
- Outcomes are compared across the two groups at the end of the trial
- Used to test the safety and efficacy/effectiveness of interventions
- Participants selected and then randomly allocated to either receive the intervention of interest or to receive a control (or no) intervention
- Random allocation should ensure that the two groups are essentially identical before provision of intervention or control
- Outcomes are compared across the two groups at the end of the trial

RCT
- Used to test the safety and efficacy/effectiveness of interventions
- Participants selected and then randomly allocated to either receive the intervention of interest or to receive a control (or no) intervention
- Random allocation should ensure that the two groups are essentially … before provision of intervention or control
- Outcomes are … across the two groups at the end of the trial
- Used to test the safety and efficacy/effectiveness of interventions
- Participants selected and then randomly allocated to either receive the intervention of interest or to receive a control (or no) intervention
- Random allocation should ensure that the two groups are essentially identical before provision of intervention or control
- Outcomes are compared across the two groups at the end of the trial

RCT
- Used to test the … and …/effectiveness of interventions
- Participants selected and then randomly allocated to either receive the intervention of interest or to receive a control (or no) intervention
- … allocation should ensure that the two groups are essentially identical before provision of intervention or control
- Outcomes are compared across the two groups at the end of the trial
- Used to test the safety and efficacy/effectiveness of interventions
- Participants selected and then randomly allocated to either receive the intervention of interest or to receive a control (or no) intervention
- Random allocation should ensure that the two groups are essentially identical before provision of intervention or control
- Outcomes are compared across the two groups at the end of the trial

Why do we need to think about causation?

Differences between study designs

Differences between study designs
- List in order (top to bottom) - Cohort, Cross-sectional, Case control, RCT, Case Study


Which study design gives the best evidence for causal association?
RCT
A … is a variable that influences both the dependent variable and independent variable causing a spurious association
A confounder is a variable that influences both the dependent variable and independent variable causing a spurious association
Confounding
- Influences both … and …
- Influences both exposure and disease
