Causality, Bias & Confounding

Question 1

What questions are asked in description?

Answer 1

• what happened?
• who was affected?
• people with X had Y

Question 2

What questions are asked in prediction?

Answer 2

• what will happen?
• who will be affected?
• people with X are more likely to have Y?

Question 3

What questions are asked in causal inference?

Answer 3

• what will happen if…?
• why were they affected?
• if we changed X, how would it change Y?

Question 4

What questions would be asked if it was qualitative?

Answer 4

• what matters?
• why does it matter?
• how can we effectively change x?
• should we change x?

Question 5

The headline “organic food lowers blood and breast cancer risk” is an example of what?

Answer 5

causal nonsense

it is implying that if you eat organic food, you will have a lower risk of contracting cancer

Question 6

What type of approach to causal inference is shown here?

Answer 6

‘causation’ of infectious disease is fairly simple

this is deterministic

Question 7

What is meant by a deterministic relationship?

Answer 7

a deterministic relationship involves an exact relationship between two variables

the deterministic model gives the same exact results for a particular set of inputs, no matter how many times you re-calculate

Question 8

What is an example of a deterministic relationship on a molecular level?

Answer 8

molecular and cellular processes (e.g. laboratory studies) show a deterministic relationship

relaxed myometrial cell + prostaglandin E2 = contracted myometrial cell

Question 10

Why can a deterministic model not be used for the vast majority of health outcomes?

Answer 10

for the vast majority of health outcomes, there are multiple causes

Question 11

What is the difference between a deterministic model and a probabilistic model?

Answer 11

probabilistic models incorporate random variables and probability distributions into the model of an event

a deterministic model gives a single possible outcome for an event

a probabilistic model gives a probability distributon as a solution

Question 12

What type of relationship is shown here?

What problem does this raise with causal inference?

Answer 12

this relationship is NOT probabilistic

how do we identify causes and what works “best” when one thing doesn’t necessarily lead to another?

Question 13

What is the fundamental problem of causal inference?

Answer 13

you can never know what would have happened if you had done things differently

i.e. we cannot observe the counterfactual

Question 14

What do we need to do in order to study how most things work?

Answer 14

to study how most things work, we have to come up with an “estimate” of the counterfactual

i.e. a control

Question 15

What is the problem with estimating the counterfactual?

Answer 15

individual people are very different and have lived very different lives

Question 16

What is meant by exchangability?

Why is it important?

Answer 16

because everyone is different we have to work with groups of people and find ways to ensure our groups are - on average - comparable

Question 17

What is the best way to achieve exchangability?

Answer 17

the easiest way to do this is through randomisation

this produces both the intervention group and the comparison

Question 18

Is randomisation always going to produce exchangability?

Answer 18

NO because randomisation is a blunt tool

the sample needs to be large enough to account for differences

Question 19

What is meant by random sampling error?

Answer 19

the random error in our population estimate (s) that results from chance fluctuations in the profile of our sample

e.g. want 50% blue and 50% green

the sample contains 83% blue and 17% green

Question 20

Without randomisation, does a bigger sample size help to acheive exchangability?

Answer 20

without randomisation, the exposure is assigned by the underlying bio-psycho-social determinants

a bigger sample won’t help to achieve exchangability

Question 21

What is meant by confounding bias?

Answer 21

distortion of the causal association between two variables, due to a common shared cause (a confounder)

confounding does not just generate spurious associations, it can also exaggerate, suppress and entirely mask associations

confounding can result in a distortion in the measure of an association between an exposure and a health outcome

Question 23

How do we reduce confounding?

Answer 23

we reduce confounding by examining like-for-like participants

this is known as conditioning

Question 24

What is shown in this example?

Answer 24

the causal effect of obesity on cancer is confounded by exercise

Question 25

To estimate the unconfounded effect of obesity on cancer, what needs to be done? By which 3 methods can this be achieved?

Answer 25

we would need to **condition** on exercise levels **restriction:** * restrict the sample to a single value of the confounder * e.g. look at the association in people who do zero exercise **stratification:** * calculate category-specific effects for different levels of the confounder * i.e. stratify across exercise levels **covariate adjustment:** * adjust for exercise as covariate in a **regression** of obesity on cancer

Question 26

Why can conditioning not completely remove confounding?

Answer 26

**unobserved confounding:** * because of other confounding variables that we did not measure **residual confounding:** * error in our measure of exercise - imperfect conditioning

Question 27

How does sample size affect random error and bias?

Answer 27

a larger sample size **_REDUCES_ random error** (or 'error') but it has **_NO EFFECT_** on **systematic error** (or 'bias')

Question 28

How does sample size and quality affect precision and accuracy?

Answer 28

*

Question 29

What is the difference between measurement error and measurement bias?

Answer 29

**measurement error:** * error in your measurement due to ***_random factors_*** * e.g. weighting scales vary according to climate (temperature, humidity, etc.) **measurement bias:** * error in your measurement due to ***_non-random factors_*** * ​weighing scales are broken & under-report by 10%

Question 30

When does misclassification error / misclassification bias occur?

Answer 30

when measurement error and measurement bias result in misclassification

Question 31

Is this an example of error or bias?

Answer 31

**BIAS** * the GP could see that John was anxious * she knows that patients have a higher BP when measured * this is a form of **response bias** known as **white coat hypertension** * she was therefore less concerned about the measurement error and knew the result was likely higher than typical

Question 32

What are the 6 categories of bias?

Answer 32

* **confounding bias** - in reading - up on the field * **selection bias -** in specifying and selecting the study sample * **information bias -** in executing the experimental manoeuvre (or exposure) * **experimenter bias -** in measuring exposures and outcomes * **analytic bias -** in analysing the data * **inferential bias -** in interpreting the analysis

Question 33

What is shown by this diagram?

Answer 33

different types of bias are not mutually exclusive a study can be biased in many different ways

Question 34

Why does selection bias occur?

Answer 34

it occurs due to a **systematic difference** between those selected into a study (or analysis) sample and those not selected

Question 35

What are the 3 types of selection bias and why do they occur?

Answer 35

**sampling bias:** * broadly due to faulty sampling by the investigator **participation bias:** * broadly due to behaviour of (potential) participants **attrition bias:** * due to loss of participants from the study

Question 36

Why does sampling bias occur?

Answer 36

a failure to sample evenly across the population resulting in an unrepresentative sample

Question 37

How is this an example of diagnostic bias? *"Oral contraceptive use is associated with endometrial cancer"*

Answer 37

oral contraceptive causes "breakthrough bleeding" this is also a symptom of endometrial cancer there is increased clinical suspicion, leading to referral and diagnosis

Question 38

What is meant by **survivorship bias**? What type of bias can exacerbate this?

Answer 38

* successful people often attribute their success to their actions and behaviours * "take risks!" "be rebellious!" * we do not know about occurrence of these actions and behaviours in non-survivors * this is exacerbated by **attribution bias** * **​**people attributing their success to their actions and behaviours, not their good luck

Question 39

What is meant by **participation bias**?

Answer 39

bias resulting from people having differential preferences (or opportunities) to participate in research willingness and ability to participate in research varies with almost all possible bio-psychosocial factors * health (physical or psychological) * education (interest, curosity, etc.) * beliefs (religious, spiritual, political, etc.) * psychology & personality (self-efficacy, openness, scepticism, etc.) * economics (time and cost, although cost is usually reimbursed)

Question 40

How is this an example of participation bias? *"Increasing levels of education are strongly protective of stillbirth"*

Answer 40

* most cases consent to participate, but consent in controls much lower in women with less education * control group is disproportionately educated * education appears protective

Question 41

Why does information bias occur?

Answer 41

due to **systematic error** in reporting, measurement or recording of information

Question 42

What are the 3 different types of information bias?

Answer 42

**response bias:** * people responding in inaccurate or untruthful ways **recall bias:** * people having different abilities to remember past information **observer effect:** * people behaving and responding differently when they know they are being observed

Question 43

What is meant by **acquiescence bias**?

Answer 43

people prefer yes-, true- or agree-type responses

Question 44

What is meant by social desirability bias?

Answer 44

people downplay undesirable traits and exaggerate desirable ones e.g. people often underestimate how much alcohol they drink

Question 45

How is this an example of the observer effect? * "Physical activity in pregnant women with obesity measured by accelerometer* * Almost all participants were doing \>/= 30 mins moderate / vigorous activity per day"*

Answer 45

women increased their physical activity when observed some also juggled the accelerometer / attached it to their dog

Question 46

What is meant by experimenter bias?

Answer 46

bias due to the behaviours and actions of the experimenter, whether conscious or unconscious beware of conflicts of interest - always check who funded a study!

Question 47

What are the 2 different types of experimenter bias?

Answer 47

**confirmation bias:** * more likely to accept findings that we expect, and refute findings that we don't **systemic bias:** * more likely to chase positive associations (**p-hacking**), seek novel results, or otherwise "find" or publish results that support our career progression and security

Question 48

What are the two main types of error in population-level research?

Answer 48

* random error ('error') * systematic error ('bias')

Question 49

What is the difference between precision and accuracy?

Answer 49

**precision:** * precision increases as ***random error decreases*** * this can be achieved by *increasing sample size* **accuracy:** * accuracy increases as ***systematic error decreases*** * the sample size has no effect on the degree of systematic error

Question 50

What does causal inference in observational data require?

Answer 50

causal inference in observational data requires **causal inference methods** * probability theory * counterfactual reasoning * graphical model theory don't infer causality without directed acyclic graphs