Introduction to Health Research

Question 1

What 3 things make up an epidemiological study?

Answer 1

• the outcome of an interest e.g. cancer death
• the primary exposure of interest e.g. alcohol
• other exposures that may influence the outcome (potential confounders) e.g. age, sex, health behaviours

Question 2

Give 3 examples of some outcomes.

Answer 2

• Death from any cause
• A complication from a disease e.g. retinopathy from
  diabetes
• Use of health service e.g. screening uptake

Question 3

List the 7 elements of an epidemiological study.

Answer 3

• Formulation of the study question and hypothesis
• Selection of study population and study sample
• Selection of indicators of exposure
• Measurement of exposure and disease
• Analysis of relationship between exposure and
  disease
• Evaluation of role of bias
• Evaluation of role of chance

Question 4

Which 2 types of errors can occur in epidemiological studies?

Answer 4

Random error and systematic error - bias.

Question 5

Define bias.

Answer 5

Any factor arising from the design and conduct of a study that skews the data in one direction.

Question 6

How can bias be reduced?

Answer 6

By good study design.

Question 7

What are the 2 main types of bias?

Answer 7

• Selection bias (which affect participants)
• Information bias (which affect measurements;
  outcome, exposure or both)

Question 8

Give 3 questions that are necessary to ask yourself to avoid selection bias.

Answer 8

• Was the cohort representative of the defined population?
• Was there something special about the cohort?
• Was everybody included who should have been
  included?

Question 9

Give 2 examples of selection bias.

Answer 9

• Healthy volunteer e.g. in screening those that attend
  screening are more likely to be healthier
• Respondent bias e.g. those that respond to a survey
  may have different characteristics to those that do not

Question 10

Give 3 questions that are necessary to ask yourself to avoid information bias.

Answer 10

• Was the subject or outcome assessor blinded to the
  exposure?
• Were all subjects classified into exposure groups using
  the same procedure?
• Were subjective or objective measures used?

Question 11

Give 2 examples of information bias.

Answer 11

• Observer bias e.g. knowing that a patient had active
  treatment might affect how an observer interprets a
  patient’s answers to questions about quality of life
• Recall bias e.g. mothers whose children were stillborn
  are more likely to remember drinking alcohol in
  pregnancy than mothers whose babies were healthy

Question 12

Rank types of evidence in order of decreasing internal validity.

Answer 12

1. Systematic reviews and meta-analyses
2. Randomised controlled trials with definitive results
3. Randomised controlled trials with non-definitive results
4. Cohort studies
5. Case-control studies
6. Cross-sectional surveys
7. Case reports

Question 13

What is reverse causality?

Answer 13

Whether the exposure could have caused the outcome.

Question 14

Give an example of reverse causality.

Answer 14

if you see an association between infrequent
exercise and coronary heart disease:
Do people who don’t exercise get coronary heart
disease?
OR
Does coronary heart disease make people less inclined to exercise?

Question 15

Why are epidemiological studies important?

Answer 15

They give measures of disease frequency and measures of effect.

Question 16

What are hypothesis tests?

Answer 16

Hypothesis tests measure the strength of statistical association between exposure and outcomes.

Question 17

How can we tell if the association also shows causation?

Answer 17

If the association meets Bradford Hill criteria for causation.