Measuring the Occurrence of Disease in Epidemiology

Question 1

What is the Epidemiological method?

Answer 1

1. Defining the disease
2. Counting the frequency of disease in the population (e.g. incidence, prevalence)
3. Describing the occurrence of disease in a population (in time, age, sex)
4. Comparing the disease occurrence between two different sub-populations

Question 2

What is prevalence of disease and how can it be calculated?

Answer 2

-Very simple measure
Cross-sectional study results in prevalence estimates
It can not be smaller than 0% or greater than 100%
P= Number of persons with disease/ number of people checked for presence of disease

Question 3

What are the different ways to express the measured prevalence in a sample?

Answer 3

1. Point prevalence= study can be conducted over a period of few months, but the result is still “point prevalence” – from the perspective of EACH SUBJECT, the information is on one point in time only, and counts only those with active symptoms of disease at that point in time;
2. Period prevalence: if we asked about presence of active symptoms within the past e.g. 6 months or 3 years;
3. Lifetime prevalence: if we asked whether there were ever any symptoms of schizophrenia, regardless of the status of disease in the present (e.g. medication or remission);
4. Lifetime morbid risk: if we followed everyone in the sample until they die, and then added all further new cases to the already noted lifetime prevalence

Question 4

What are the two problems with a test?

Answer 4

In the field study, we can rarely use tests that can definitely establish the diagnosis of disease (e.g. 100% sensitive and specific) – we need to use some method of screening instead, which is never as reliable
In all such cases, we do not measure the number of PERSONS WITH DISEASE, but rather the number of PERSONS WHO TEST POSITIVE for that test

Question 5

What is case definition?

Answer 5

We don’t count the number of disease cases!!!
We count the number of study participants who TEST POSITIVE for a disease of interest, using an internationally agreed CASE DEFINITION for that disease
Try to name a few diseases with perfect case definition – not many!!!
Case definition – in epidemiological studies it has properties of “screening test” – it has its sensitivity and specificity
Moreover, prevalence estimates can be hugely distorted by population choice (e.g. dementia in retirement homes)

Question 6

What is incidence?

Answer 6

Different from prevalence because it is measured in “prospective” (longitudinal) studies –
People in the sample are all healthy at the start, but then followed up to detect new disease cases – it introduces an element of TIME into disease measurement

Question 7

How do we calculate incidence?

Answer 7

Number of new cases of disease/ Number of persons at risk x time of their follow-up

• Measured in new cases per person-years
• INCIDENCE is hugely sensitive on underlying population, age distribution, gender distribution – standardize before comparisons of any two populations (e.g. incidence of malaria in malaria-affected vs. non-malaria affected regions; incidence of cholera in refugee camps)

Question 8

What is the denominator for incidence?

Answer 8

Denominator is person-years (but also: child-weeks, mother-fortnights, adolescent-months)
Denominator is only based on population at risk to develop a disease (important!)

Question 9

Difference between prevalence and incidence

Answer 9

Unlike prevalence, INCIDENCE can be GREATER than 100% (e.g. childhood upper respiratory infections – 4 episodes per child-year = 400% e/cy), but can’t be smaller than 0%

Question 10

What is mortality?

Answer 10

defined and computed in the SAME WAY as incidence – the DIFFERENCE is that numerator has deaths instead of new cases of disease
Another difference in comparison to incidence – mortality can’t be greater than 100% (…at least not yet)

Question 11

What is case-fatality?

Answer 11

mortality divided by incidence; useful - usually for dramatic epidemics of infectious diseases, or other abrupt events with immediate mortality risk (e.g. traffic accidents)

Question 12

What is lethality?

Answer 12

time component somewhat blurred – life-long follow-up, deaths from other causes (e.g. HIV/AIDS)

Question 13

Disease with Low incidence and high prevalence

Answer 13

Dementia

Question 14

Disease with High incidence and low prevalence

Answer 14

Influenza

Question 15

Disease with High case-fatality and low mortality

Answer 15

SARS

Question 16

Disease with Low case-fatality and high mortality

Answer 16

Diarrhoea

Question 17

What is standard population?

Answer 17

We define “standard population” – this is ANY population, as long as it has a defined distribution by age and sex (preferably 5-year age groups)
Standard population – also very useful for standardizing the reported rates of incidence / mortality / prevalence