cross sectional and longitudinal studies

Question 1

what are the two ways results are analysed from a survey

Answer 1

• Descriptive statistics
• Associations between potential causes and effects

Question 2

sample size

Answer 2

bigger is better but also more expensive

Question 3

what is a type 1 and type 2 error

Answer 3

type 1 is a false positive (saying they are diseased when they are not)
type 2 is a false negative (saying they aren’t diseased when they are)

Question 4

what’s power

Answer 4

the ability of a statistical test to find a significant result when it is real
helps you to avoid type 2 errors

Question 5

whats the power calculation

Answer 5

p ±1.96√p(1-p)/n

Question 6

what is a longitudinal study

Answer 6

• also known as cohort studies
• follow a group of people over time
• the group is called a cohort
• cohort is a group of people with similar characteristics

Question 7

example of a longitudinal study

Answer 7

1) pick a group of people
2) measure any factors you consider relevant to your study (age, weight, job type)
3) give them free eye test every year
4) record how many get myopia, hyperopia, glaucoma, cataract
repeat for many years

Question 8

how to calculate incidence rate in 1 year

Answer 8

number of new cases in 1 year / number not already a case at start of year

Question 9

incidence as patient years

Answer 9

new cases/ per 1000 patients/ per 1 year
or
new cases/ per patient year

Question 10

how to work out patient years

Answer 10

number of patient x years followed

100 patients for 1 year (100x1=100)
1 patient for 100 years (1x100=100)
50 patients for 2 years (50x2=100

Question 11

basic cohort study design

Answer 11

• The original cohort must include people who might get the disease
• The original cohort must include people
• Who have been exposed to the risk factor
• Who have not been exposed
• Comparison of disease rates between exposed/not exposed will show association.
• If risk occurs before disease, strong evidence of causation