Sem 2 Pop Sci

Question 1

Describe a population graph, and the four main types of population graphs. (10)

Answer 1

They have % population on the x axis, against age, with one line for male and one line for female.
Rapidly expanding - curved lines, wide at the bottom, narrow at the top.
Expanding - straight diagonal lines to a point at the top.
Stationary - vertical lines which slowly curve off at the top.
Contracting - narrow at the bottom, to vertical lines, curves in at the top.

Question 2

Define public health (3)

Answer 2

The art and science of preventing disease, prolonging life, and promoting health through the organised efforts of society.

Question 3

Describe the three types of prevention used in public health. Give two examples of each. (9)

Answer 3

Primary - prevent the onset of disease by reducing exposure to risk factors - immunisations, reducing smoking.
Secondary - to detect and treat a disease at an early stage to prevent further harm - cervical cancer screening, monitoring bp.
Tertiary - harm limitation of an already established disease - thiamine for alcoholism, steroids in asthma.

Question 4

Describe the three domains of public health. Give an example for each. (6)

Answer 4

Health improvement - sexual health, weight, smoking, mental health
Health protection - screening, immunisations, emergency responses.
Public health in healthcare - prioritisation, research, assessment

Question 5

Define epidemiology. (1)

Answer 5

Study of how exposure affects disease.

Question 6

Explain the difference between chance and systematic errors. (2)

Answer 6

Chance - due to individual differences, random.

Systematic - bias, lowers accuracy.

Question 7

Explain why a study needs both internal and external validity. (4)

Answer 7

Internal validity - means the comparison groups are actually comparable.
External validity - means the results are generalisable because the samples are representative of the whole population.

Question 8

Describe these error types:
Recall 
Observer
Measurement 
(3)

Answer 8

Recall - error common when recollecting the past.
Observer - error due to pre-conceived expectations.
Measurement - due to equipment and interpretation.

Question 9

Define prevalence. (2)

Answer 9

Proportion of people who have a disease at a point in time - diseased / whole population.

Question 10

Define incidence. (2)

Answer 10

Number of new cases of a disease within a certain time frame - number of new cases / population at risk.

Question 11

Describe a case control study design. Give an example. (4)

Answer 11

Find current cases and controls, look for past exposure.

Current asthma and currently no asthma, look for previous exposure to a mouldy house.

Question 12

Describe how from a case control study you would work out the odds of an exposure having an effect or not. (2)
Describe how you would work out the odds ratio. (3)

Answer 12

Exposed = cases (a) + controls (b) 
Unexposed = cases (c) + controls (d)

Odds of exposure having an effect = a / c
Odds of exposure not having an effect = b / d

Odds ratio = (a / c) / (b / d)
= ad / cb

Question 13

Give two issues with case control study designs. (2)

Answer 13

Selection bias

Random error

Question 14

Describe a cohort study design. (2)
Give an example. (2)
Describe the two types of cohort design. (2)

Answer 14

Find unexposed and unexposed people, look for future development of the disease.
Mouldy house vs not mouldy house now, look for asthma in the future.
Can be prospective (exposed now, future outcome), or retrospective (exposed past, outcome now).

Question 15

Describe how you would work out the incidence rate of the exposed population in a cohort study. (2)

Answer 15

IR of exposed = number that developed the outcome in the exposed group / person years at risk.

Question 16

Describe how you would work out the incidence rate ratio in a cohort study. Explain the purpose of this. (3)

Answer 16

IRR = IR of exposed / IR of unexposed.

Shows how much more likely you are to develop the disease if you were exposed.

Question 17

Describe 4 issues with a cohort study design. (4)

Answer 17

Loss to follow up
Survivor bias
Information bias
Random error.

Question 18

Describe a cross sectional study design. Give an example. (2)

Answer 18

Find the number of cases right now in a snapshot form.
2011 = 250,000 asthma cases
2014 = 350,000 asthma cases

Question 19

Describe how to work out the prevalence of a disease. What is another name for the prevalence? (3)

Answer 19

Prevalence = number of people with the disease / total population = absolute risk.

Question 20

Describe the way you work out incidence rate in a cross sectional study. (2)

Answer 20

Number of new cases / sum of person years at risk.

Question 21

Give 3 issues with cross sectional studies. (3)

Answer 21

Sampling bias
Participant bias
Random error

Question 22

Describe an ecological study design. Give an example. (3)

Answer 22

Look at the whole population, separate by characteristic, look at cases in each group.
Men with asthma vs women with asthma.

Question 23

Give 4 issues with ecological studies. (4)

Answer 23

Characteristics could be unclear
Confounding
Random error
Lack of generalisability (dep on sample size)

Question 24

Explain the differences between rate and ratio. (2)

Answer 24

Rate = number of events per population per time. 
Ratio = comparison of rates/odds/anything .

Question 25

Explain the difference between ratio and proportion. (2)

Answer 25

Ratio = one number divided by another. 
Proportion = fraction where the top part has to be included in the bottom part.

Question 26

Explain the difference between ratio and difference. (2)

Answer 26

Ratio = one number divided by the other 
Difference = one number subtract another.

Question 27

Explain the difference between confounding and bias. (4)

Answer 27

Confounding = a situation in which a measure of the effect is distorted by association of the intervention with other influencing factors. Confounder must link to both risk (eg exposure) and outcome. 
Bias = a systematic distortion in allocation / measurement etc.

Question 28

Explain the difference between random selection and random allocation. (4)

Answer 28

Selection = recruitment into a clinical trial, affecting generalisability (external validity) 
Allocation = within the groups in the study (internal validity).

Question 29

Explain the differences between a screening test and a diagnostic test. (3)

Answer 29

A screening test gives you a risk (ie if you screen positive you are have a high change of having the disease - not 100%)
A diagnostic test tells you whether you have the disease or not (100%).

Question 30

Describe the purpose of screening. (3)

Answer 30

Find something before it would normally present with symptoms and treat it at this early stage to achieve a better outcome.

Question 31

Describe the 5 conditions needed for a screening programme to be approved. (10)

Answer 31

Condition - must be important in frequency and severity.
Test - simple, safe, precise, validated. Must be an agreed test population and cut off level.
Intervention - better outcomes achieved by treating sooner.
Programme - opportunity cost is balanced, program is acceptable to public.
Implementation - informed choice, correctly staffed.

Question 32

Explain the difference between a false positive and a false negative result. (2)

Answer 32

False positive - referring healthy people for further investigation
False negative - failure to refer people who have the disease.

Question 33

Define sensitivity of a screening test. (2)

Answer 33

Proportion of the people with the disease who test positive.
Positive test and diseased / all diseased = sensitivity

Question 34

Define specificity of a screening test. (2)

Answer 34

Proportion of the people without the disease that test negative.
Negative test and healthy / all healthy.

Question 35

Define the positive predictive value of a screening test. (2)

Answer 35

Probability that someone who has tested positive actually has the disease.
Diseased and positive test / all positive tests

Question 36

Define the negative predictive value of a screening test. (2)

Answer 36

Probability that someone who has tested negative actually doesn’t have the disease.
Healthy and negative test / all negative tests.

Question 37

Describe the three types of bias that occur with screening programs. (6)

Answer 37

Lead time bias - early diagnosis falsely appears to prolong survival
Length time bias - picking up and treating slow growing things that wouldn’t have caused harm anyway.
Selection bias - those who show up for screening are also those that would have good health behaviours.

Question 38

If this equation was presented to you, which type of study design is it from, and how do you work out the confidence interval? (3)
_______________
e^ { 2 x √ (1/a + 1/b + 1/c + 1/d) }

Answer 38

Case control.

Confidence interval = odds ratio / ef to odds radio x ef.

Question 39

If this equation was presented to you, which type of study design is it from, and how do you work out the confidence interval? (3)
_________
e^ { 2 x √ (1/d1 + 1/d2) }

Answer 39

Cohort

Confidence interval = incidence rate ratio / ef to incidence rate ratio x ef.

Question 40

If this equation was presented to you, which type of study design is it from, and how do you work out the confidence interval? (3)
___
e^ { 2 x √ (1/d) }

Answer 40

Cross sectional

Confidence interval = incidence rate / ef to incidence rate x ef.

Question 41

Explain what it means if the number 1 is within your confidence interval. (3)

Answer 41

1 represents the null hypothesis (no association). If your confidence interval includes the number 1, it means the interval is big enough to include the null hypothesis, and so the null hypothesis is true.

Question 42

Explain how confounding and bias are minimised. (2)

Answer 42

Both minimised by randomisation.

Bias is reduced by blinding.