Lecture 4: Measures of occurrence

Question 1

Epidemiology is the study of how often diseases occur in different groups of people and why. A key feature of epidemiology is the measurement of disease outcomes in relation to a population at risk.

Answer 1

• Clear definitions are needed (‘cancer’ is not precise - consider subtypes - Define your unit of measurement (e.g. whole country, region) - Check that individuals are ‘eligible’ for inclusion

Question 2

What are the different ways that a disease/death can be described?

Answer 2

1. Simple percentages 2. Rates

Question 3

What is the incidence of a disease?

Answer 3

The incidence of a disease is the rate at which new cases occur in a population during a specified period

E.g. If population is 1000 but 200 have already had disease, the population at risk is only 800

Question 4

How is the incidence rate calculated?

Answer 4

Incidence rate = number of new cases in period / number at risk in population MUST include time unit

Question 5

What is ‘prevalence’ defined as?

Answer 5

Defined as the proportion of existing cases (old and new)

Be careful - it is not the risk (or rate) of disease

Question 6

When is prevalence useful?

Answer 6

Useful for public health professionals and commissioners - describes the burden of disease.

Question 7

What is ‘point prevalence’?

Answer 7

The proportion of existing cases in a population at a single point of time

Question 8

How can point prevalence be approximately calculated?

Answer 8

By calculating incidence rate x average duration of disease

Question 9

Point prevalence is the only measure obtainable from cross-sectional studies. What is a cross-sectional study?

Answer 9

A is a type of observational study that analyzes data from a population, or a representative subset, at a specific point in time

Question 10

Incidence and prevalence explained - think of it like a funnel.

How should in-migration, out-migration, death and cures be dealt with?

Answer 10

Population - this is your denominator population

Incident cases - these are new cases

Out-migration, deaths and cures must be exclused (population/cases will decrease)

In-migration must be included (population/cases will increase)

Question 11

What is in-migration?

Answer 11

The process of people moving into a new area in their country to live there permanently

I.e. if people with the disease move there, you have to count them in the population

Question 12

What is out-migration?

Answer 12

The process of people moving out of an area in their country to move to another area in their country permanently.

Question 13

Prevalence vs incidence graph

Answer 13

Question 14

Example rates:

Answer 14

• 18.3/100,000 for non-South Asians
• 13.1/100,000 for South Asians

BUT these are rate for a 16 year period and rates can change over time

Question 15

Age-specific rates example

Answer 15

Different conditions may be more prevalent in different age groups

Question 16

Incidence rate example

Answer 16

Question 17

What is the definition of mortality?

Answer 17

Mortality is the incidence of death from a disease

Question 18

How is mortality rate calculated?

Answer 18

Mortality rate = number of people who die from the disease in period / number of people in the population in period

E.g. 200 / 20000 (200 deaths in population of 20000) in same period

Question 19

What does ‘ascertainment’ mean?

Answer 19

Means that you have counted all the individuals of interest in your study

Missing medical records, poor record keeping or changes in how diseases are classified might mean don’t count individuals, or include the wrong individuals

Question 20

When might ‘ascertainment bias’ arise?

Answer 20

Arises when data for a study or an analysis are collected (or surveyed, screened, or recorded) such that some members of the target population are less likely to be included in the final results than others.

Question 21

Poor ascertainment example:

Answer 21

If you wanted to use death certificates to find out the cause of death in children and young people diagnosed with Type 1 diabetes

• Out of 229 death certificates examined only 54% of all death certificates mentioned diabetes and 21% specifically mentioned Type 1 diabetes
• If you JUST used ONS death certificate data you would MISS nearly half of all deaths in this group – that is poor ascertainment

Question 22

Common denominator examples:

Answer 22

• Population of a country/region – census data
• Number of live births/live births and stillbirths/live births, still births and TOPs – ONS
• Number of patients attending a hospital – Hospital episode stats, individual hospital clinical information systems
• A GP practice population, data from their primary care database provider

Question 23

When looking at a disease you might assess incidence over a long period of time (e.g. 10 years). How does this affect the persons chances of getting the disease?

Answer 23

This means that your population has had a longer time at risk of getting the disease.

In a population of 1,000,000 people over 10 years, their person years at risk is 10,000,000, BUT populations change

Question 24

What is a crude rate?

Answer 24

The total number of events, or count, divided by the mid-year total population of the selected geography and multiplied by a constant, which is a multiple of 10. Typical constants used for public health rates include 100, 1,000, 10,000, or 100,000.

Crude rate relates to the population as a whole

Question 25

What is standardisation used for?

Answer 25

Used to enable comparison between areas with different population structures - mainly by age and sex

(Could compare age-specific rates but standardisation produced one summary rate to compare)

Question 26

What are the 2 methods of standardisation?

Answer 26

Direct and indirect

Question 27

What does direct standardisation involve?

Answer 27

• Direct standardisation involves applying the rates of disease observed in the study group to a ‘standard’ population.

Question 28

What is age standardisation?

Answer 28

A technique used to allow populations to be compared when the age profiles of the populations are quite different.

Question 29

What is indirect standardisation?

Answer 29

Indirect standardisation applies the rates of disease in a ‘standard’ population to the study group.

Uses the population in your study area – if you were looking at the incidence of disease in Scotland by Health Board it makes sense to use the Scottish population data as this is more relevant.

Compares the observed to the expected.

Question 30

How is indirect standardisation usually presented?

Answer 30

As ratios e.g. standardised mortality

SMR = observed number of deaths / expected number of deaths

Question 31

View this link to explain age standardisation:

Answer 31

https://www.statcan.gc.ca/eng/dai/btd/asr

Question 32

Frequently used measures of mortality

Answer 32