Classical Epidemiology I

Question 1

What is classical epidemiology?

Answer 1

Epidemiologists study the distribution (frequency, pattern) and determinants (causes, risk factors) of health-related states and events (not just diseases) in specified populations (school, city, state, country).

“Classical” epidemiology = epidemiology in absence of molecular tools.

Question 2

What are some potential questions an epidemiologist might address?

Answer 2

• Communicable disease: what is the burden of Covid in CH?
• Non-communicable disease: do anti-smoking campaigns work?
• Health services research: how does patient satisfaction compare in private and public facilities?

Question 3

What are the 3 main classes of epidemiology?

Answer 3

Descriptive (distribution of disease) - observational
Analytical (determinants of disease) - observational
Intervention (changing the situation)

Question 4

Descriptive epidemiology can help you…

Answer 4

• Identify and classify disease entities
• Describe transmission, distribution and evolution (natural history) of disease

who, what, when, how much

Question 5

Analytical epidemiology can help you…

Answer 5

• Establish determinants and causes of disease
• Define risk factors and/or high-risk groups
• Measure level of risk

How and what (causes, etiology), why (risk factors)

Question 6

What is understood by “population at risk”?

Answer 6

Those that actually have the potential to get a particular disease NOT every single individual (e.g pregnancy-related illnesses could only effect women). With some diseases, the population at risk MAY well be everyone.

It essentially defines who in the population is mainly at risk for getting the disease of interest. Rarely are diseases affecting all age/gender groups.

Covid: everyone
Breast cancer: normally females >30yrs

Question 7

What is the prevalence rate?

Answer 7

A snapshot. The proportion of people affected by a disease at a given point in time.
Expressed as percentage.

existing cases at specified point in time/ total population at risk = x%

Water IN the epidemiology fountain

Question 8

What is the incidence rate?

Answer 8

Number of NEW cases of disease in a given period of time in a population at risk.
Expressed per 100, per 100,000 population etc.

New cases of disease in period of time/ total population at risk

Ex: Incidence of pulmonary TB in Tanzania in 2008: 474 per 100,000 population.

Question 9

What influences the prevalence rate?

Answer 9

Feeding INto the fountain is the INcidence of new infections (number of new cases) and emptying the fountain is cure and disease-related death incidence. The ratio between former and latter will increase or decrease prevalence.

Greater # deaths/ cures, the lower the prevalence.
Greater # new cases, the higher the prevalence.

Question 10

What studies fall within the descriptive epidemiology class and what can they tell us?

Answer 10

cross-sectional
longitudinal

• Frequency and distribution of disease
• GENERATE hypothesis about etiology or risk factors (why one person affected and another not?)

Question 11

What studies fall within the analytical epidemiology class and what can they tell us?

Answer 11

Cohort
Case-control

• TEST hypothesis on etiology or risk factors

Question 12

What studies fall within the intervention epidemiology class and what can they tell us?

Answer 12

Randomised controlled trials
- Allows to test whether an intervention works (vaccine, etc.)

Question 13

Can you measure incidence and prevalence rates from a cross-sectional study?

Answer 13

A cross-sectional study looks at a cross-section of the population at risk at a given point in time (snap-shot). Prevalence can be measured but to obtain incidence of new cases, at least 2 measurements must be taken.

Question 14

What are the two types of cohorts in a longitudinal study?

Answer 14

A longitudinal study follows a certain population over at least 2 time periods and new disease episodes are registered. Incidence can be measured.
Either, new people can (OPEN) or cannot (CLOSED) be added to the cohort over the course of the study period.

Question 15

What is a risk factor?

Answer 15

Every disease or health condition has many “risk factors”, i.e. factors that are associated with the disease of
interest (they cause either an increased or decreased risk).

Impossible to take into account ALL risk factors (might not be aware of all or unable to measure all).

Many factors can influence the disease under investigation and it is hard to look at each factor in isolation. These factors linked to the disease can also be linked to each other and be confounding. Cohort studies with control groups can help elucidate though.

Question 16

What studies might one carry out to investigate potential risk factors?

Answer 16

Cohort and case-control.
Allows for comparison of two groups. Can create situation where you control for other potential risk factors and can have the two groups differ by one factor (e.g. smoking) as much as possible to infer what effect the given risk factor has on disease occurance.

Groups with risk factor compared to group without risk factor. Incidence rates compared.
Relative risk = incidence exposed (Ie)/ incidence unexposed (Io).

Note: these studies observational as there is not experimental intervention.

Question 17

Compare a cohort study to a case-control study.

Answer 17

Both are investigating disease causes and risk factors.

Cohort study:
- compare an exposed and non-exposed group and analyse disease incidence.
- Looking forward in time (prospective)
Work from exposure forwards.
- Investigate one exposure, multiple diseases (outcomes).
- Need large #s of people. disease can’t be rare.
Population -> people without disease -> split into exposed and non-exposed cohorts -> each of these are split into disease and non-diseased.

Case-control.
- Compare group with disease (cases) and comparable group without (controls) in terms of history of exposure to find Rfactors or cause.
- Timeline reversed. Looking backwards in time. (retrospective) From diseased individuals backwards. Fishing for info.
- Investigate multiple exposures, one disease (outcome).
- can work with small #s. Disease can be rare.
- Work from people with disease backwards.
population -> people with disease -> exposed and unexposed.

Question 18

What are two important measures we can obtain from cohort studies?

Answer 18

Difference in incidence rates of disease between both exposure groups.
Relate # of cases of disease to population at risk.

Le and Lo

Relative risk

Le/ Lo = RR = how much does the exposure (or risk factor) increase of decrease risk of getting disease.

If RR = 0 (exposure not risk factor)
If RR more than 1 (exposure leads to increased risk)
If RR less than 1 (exposure leads to decreased risk)

It is relative -> no units.