Lecture 1: Introduction - 09/01/2023
Epidemiology = “study what is upon the people” (epi, demos, logos) → conditions, behaviours,
patterns and illnesses → study of the distribution and determinants of health-related states or
events in specified populations and the application of this study to the control of health
problems → defined by Centres for Disease Control and Prevention (CDC) → elements:
- Distribution:
- frequency = numbers of health events in a population and the relationship of
those numbers to the size of the population.
- pattern = occurrence of health-related events by time (annual/seasonal
diseases), place (urban/rural) and person (age/sex).
- Determinants: causes and other factors that influence the occurrence of disease and
other health-related events → FEX. detecting wheter a factor (smoking) is a cause of a
certain disease (lung cancer).
- Health-related states or events: diseases, injuries and disabilities.
- Application: practices, implications, recommendations based in epidemiological studies.
History of epidemiology: John Snow (father of field epidemiology) → in the 1850s in London there
was a massive Cholera outbreak, because it was a dirty, polluted place → people believed in
Miasma theory = most diseases are caused by bad air and air pollution (bad smells).
→ John Snow was a physician and did not believe in Miasma theory, but instead thought polluted
water was the cause → he looked into an area were there were a lot of deaths (Broad street) and
there was a waterpump many people used and people got ill from drinking this water → John
Snow removed the handle → result; number of cases decreased → also looked into water supply.
1. independent variable (exposure): type of water (supplying company) → is the risk factor,
determinant, independent variable or cause.
2. dependent variable (outcome): death by Cholera → is the dependent variable or effect.
→ look into association and causal relationships to find the effect of exposure on outcome.
5 W-questions of epidemiology: (1) What is the disease/condition being studied? (2) Who has the
disease in question (person)? (3) When is the disease common or rare (time)? (4) Where does the
disease arise (place)? (5) Why did the disease or condition arise?
Uses of epidemiology: public health surveillance - field investigations - analytic studies -
evaluation public health services - policy development.
Study designs:
- Observational: observe and measure a set of
characteristics in a defined sample population.
- Analytical: study of disease determinants →
covers “why?” → it is all about comparing
persons and groups → FEX. are epidemiology
students more likely to have symptoms of
anxiety than theology students? → to
understand risk factors → types (based on temporality):
- Cross-sectional: study within a slice of time (2/3 months FEX.) →
examines the presence/absence of disease and the presence/absence of
an exposure at a particular time/specific point in time → outcome and
exposure are measured at the same time → focus is on prevalence, but
can also include other analyses to investigate associations → to explore
knowledge, attitude, practice (KAP), perceptions, behaviour.
,→ example: acceptance of COVID vaccine in Southeast Asia (in Indonesia).
→ advanatages: (a) inexpensive, (b) no loss to follow up, (c) limited chance of info bias (no recall),
(d) possibility to study several determinants and outcomes in one study.
→ disadvantages: temporal relation between cause and outcome is unclear (causality problem).
- Cohort: prospective (concurrent), so doing a study in real time, start from
now and take the study further → cohort = a group of persons that share
at least one characteristic (FEX. students who completed a course in
2022) → study groups are defined by exposure (exposed vs. non-exposure)
→ these groups are followed for a certain period to assess the proportion
that develops an outcome/disease of interest.
- example: two groups based on exposure to Saccharine (those who
consume it vs. who don’t) -> outcome = pancreatic cancer.
- fixed cohort = once you are in, you cannot leave the cohort →
determined by a certain event/occurrence.
- dynamic cohort = migration occurs, members drop out and new
members join → determined by certain state/situation.
- can also have both prospective and retrospective → FEX. are
women who use assisted reproductive technololgies more likely
to have multiple births (twins)?
- ambidirectional = FEX. are women who use assisted reproductive
technololgies more likely to have twins and ovarian cancer?
→ advantages: (a) sequence between cause and outcome is usually clear, (b) can study multiple
outcomes after a single exposure, (c) useful in study of rare exposures, (d) allow calculation of
incidence rates, RR, CI and other measures, (e) limited chance of info bias (no recall).
→ disadvantages: (a) expensive, (b) selection bias (cannot divide cohorts clearly), (c) not suitable
for rare diseases or with a long latency, (d) there can be loss to follow-up (which can bias results),
(e) over time, the exposure status of study participants can change.
- Case-control: retrospective, look into the past → study groups are
defined by outcome and you look back retrospectively to identify
exposures → influential in understanding risk groups (men who have sex
with men, blood transfusion recipients) and risk factors (multiple sex
partners) for HIV/AIDS.
- sources of cases: patient rosters at medical facilities, death
certificates, disease registries, cross-sectional surveys.
- control selection: (a) controls should be representative of the
source population that produced the cases, (b) controls must be
sampled independent of the exposures (their selection should not
be more/less likely if they have the exposure of interest) → if
these principles are not adhered to, selection bias can result.
→ advantages: (a) inexpensive, (b) can yield important findings in a short time, (c) efficient for
diseases with a long latency period, (d) valuable for rare diseases.
→ disadvantages: (a) info and selection bias (methodological issues), (b) temporal relation
between the cause and outcome is unclear (causality problem).
- Descriptive: study of disease patterns (describing patterns) → covers “who,
where, when?” → FEX. what is the frequency of symptoms of anxiety among
, students? → often the first foray into a new disease or area of inquiry (the first
scientific “toe in the water”) → types:
- Individual: case report (single case), case series report (mix of different
case reports), cross-sectional studies (can be descriptive or analytical).
- Aggregate: ecological studies = looking for associations between
exposures and outcomes in populations → looks into aggregate data →
unit of observation is an entire group/population.
- examples: correlation between COVID and BCG vaccination in 80
countries → European study on death rates from coronary artery
disease correlates with per capita sales of cigarettes.
- limiations: (a) ecological fallacy = data at country level cannot be
used to tell stories about individuals, (b) inability to link exposure
to outcome in individuals, (c) inability to control for confounding.
- Experimental: manipulate a characteristic within a defined sample population with the
purpose of studying the effect this has.
- Randomised controlled trial (RCT): golden standard of clinical research →
resembles cohort study with the important exception of randomisation of
participants to exposures → if properly designed and done, an RCT is likely to be
free of bias → always prospective (starts from now) in design → 2 features:
- (a) control arm = a basic design has 2 arms; intervention and control (drug
vs. placebo, physical therapy vs. surgery) → more arms is possible (old
drugs vs. new drugs vs. placebo).
- (b) randomisation = blinded allocation to one of the intervention groups
based on chance → each person has an equal chance of being assigned to
each treatment, but the treatment to be given cannot be predicted →
used to; (I) minimise selection bias/confounding (groups are equal) and
(II) get rid of the effect of indication for the intervention/prognosis
influences (unconscious) allocation to treatment by doctors → blinded
allocation = single-blinded; only patient -> double-blinded; patient and
researcher don’t know how the treatment is allocated (=the best) →
unblinded trials most likely are biased.
→ advantages: (a) golden standard of clinical research, (b) no selection bias, (c) no confounding
bias (known and unknown), (d) high internal validity (it measures what it sets out to measure).
→ disadvantages: (a) expensive, (b) cannot be used for harmful substances (unethical).
Lecture 2: Frequency, Risks and Associations - 12/01/2023
Measures of frequency:
→ bathtub picture right: there can also be recurrence (person gets infected again).
(1) Prevalence = includes both new cases and those who contracted the disease in the
past and still have the disease → at a specific time →
accumulated amount of cases of a disease.
→ example picture left: amount of new cases (incidence) is
low/ stable, but the accumulated amount of cases of AIDS is
increasing, because people can live longer with AIDS → usually
prevalence and incidence come closer together over time (picture
Epidemiology = “study what is upon the people” (epi, demos, logos) → conditions, behaviours,
patterns and illnesses → study of the distribution and determinants of health-related states or
events in specified populations and the application of this study to the control of health
problems → defined by Centres for Disease Control and Prevention (CDC) → elements:
- Distribution:
- frequency = numbers of health events in a population and the relationship of
those numbers to the size of the population.
- pattern = occurrence of health-related events by time (annual/seasonal
diseases), place (urban/rural) and person (age/sex).
- Determinants: causes and other factors that influence the occurrence of disease and
other health-related events → FEX. detecting wheter a factor (smoking) is a cause of a
certain disease (lung cancer).
- Health-related states or events: diseases, injuries and disabilities.
- Application: practices, implications, recommendations based in epidemiological studies.
History of epidemiology: John Snow (father of field epidemiology) → in the 1850s in London there
was a massive Cholera outbreak, because it was a dirty, polluted place → people believed in
Miasma theory = most diseases are caused by bad air and air pollution (bad smells).
→ John Snow was a physician and did not believe in Miasma theory, but instead thought polluted
water was the cause → he looked into an area were there were a lot of deaths (Broad street) and
there was a waterpump many people used and people got ill from drinking this water → John
Snow removed the handle → result; number of cases decreased → also looked into water supply.
1. independent variable (exposure): type of water (supplying company) → is the risk factor,
determinant, independent variable or cause.
2. dependent variable (outcome): death by Cholera → is the dependent variable or effect.
→ look into association and causal relationships to find the effect of exposure on outcome.
5 W-questions of epidemiology: (1) What is the disease/condition being studied? (2) Who has the
disease in question (person)? (3) When is the disease common or rare (time)? (4) Where does the
disease arise (place)? (5) Why did the disease or condition arise?
Uses of epidemiology: public health surveillance - field investigations - analytic studies -
evaluation public health services - policy development.
Study designs:
- Observational: observe and measure a set of
characteristics in a defined sample population.
- Analytical: study of disease determinants →
covers “why?” → it is all about comparing
persons and groups → FEX. are epidemiology
students more likely to have symptoms of
anxiety than theology students? → to
understand risk factors → types (based on temporality):
- Cross-sectional: study within a slice of time (2/3 months FEX.) →
examines the presence/absence of disease and the presence/absence of
an exposure at a particular time/specific point in time → outcome and
exposure are measured at the same time → focus is on prevalence, but
can also include other analyses to investigate associations → to explore
knowledge, attitude, practice (KAP), perceptions, behaviour.
,→ example: acceptance of COVID vaccine in Southeast Asia (in Indonesia).
→ advanatages: (a) inexpensive, (b) no loss to follow up, (c) limited chance of info bias (no recall),
(d) possibility to study several determinants and outcomes in one study.
→ disadvantages: temporal relation between cause and outcome is unclear (causality problem).
- Cohort: prospective (concurrent), so doing a study in real time, start from
now and take the study further → cohort = a group of persons that share
at least one characteristic (FEX. students who completed a course in
2022) → study groups are defined by exposure (exposed vs. non-exposure)
→ these groups are followed for a certain period to assess the proportion
that develops an outcome/disease of interest.
- example: two groups based on exposure to Saccharine (those who
consume it vs. who don’t) -> outcome = pancreatic cancer.
- fixed cohort = once you are in, you cannot leave the cohort →
determined by a certain event/occurrence.
- dynamic cohort = migration occurs, members drop out and new
members join → determined by certain state/situation.
- can also have both prospective and retrospective → FEX. are
women who use assisted reproductive technololgies more likely
to have multiple births (twins)?
- ambidirectional = FEX. are women who use assisted reproductive
technololgies more likely to have twins and ovarian cancer?
→ advantages: (a) sequence between cause and outcome is usually clear, (b) can study multiple
outcomes after a single exposure, (c) useful in study of rare exposures, (d) allow calculation of
incidence rates, RR, CI and other measures, (e) limited chance of info bias (no recall).
→ disadvantages: (a) expensive, (b) selection bias (cannot divide cohorts clearly), (c) not suitable
for rare diseases or with a long latency, (d) there can be loss to follow-up (which can bias results),
(e) over time, the exposure status of study participants can change.
- Case-control: retrospective, look into the past → study groups are
defined by outcome and you look back retrospectively to identify
exposures → influential in understanding risk groups (men who have sex
with men, blood transfusion recipients) and risk factors (multiple sex
partners) for HIV/AIDS.
- sources of cases: patient rosters at medical facilities, death
certificates, disease registries, cross-sectional surveys.
- control selection: (a) controls should be representative of the
source population that produced the cases, (b) controls must be
sampled independent of the exposures (their selection should not
be more/less likely if they have the exposure of interest) → if
these principles are not adhered to, selection bias can result.
→ advantages: (a) inexpensive, (b) can yield important findings in a short time, (c) efficient for
diseases with a long latency period, (d) valuable for rare diseases.
→ disadvantages: (a) info and selection bias (methodological issues), (b) temporal relation
between the cause and outcome is unclear (causality problem).
- Descriptive: study of disease patterns (describing patterns) → covers “who,
where, when?” → FEX. what is the frequency of symptoms of anxiety among
, students? → often the first foray into a new disease or area of inquiry (the first
scientific “toe in the water”) → types:
- Individual: case report (single case), case series report (mix of different
case reports), cross-sectional studies (can be descriptive or analytical).
- Aggregate: ecological studies = looking for associations between
exposures and outcomes in populations → looks into aggregate data →
unit of observation is an entire group/population.
- examples: correlation between COVID and BCG vaccination in 80
countries → European study on death rates from coronary artery
disease correlates with per capita sales of cigarettes.
- limiations: (a) ecological fallacy = data at country level cannot be
used to tell stories about individuals, (b) inability to link exposure
to outcome in individuals, (c) inability to control for confounding.
- Experimental: manipulate a characteristic within a defined sample population with the
purpose of studying the effect this has.
- Randomised controlled trial (RCT): golden standard of clinical research →
resembles cohort study with the important exception of randomisation of
participants to exposures → if properly designed and done, an RCT is likely to be
free of bias → always prospective (starts from now) in design → 2 features:
- (a) control arm = a basic design has 2 arms; intervention and control (drug
vs. placebo, physical therapy vs. surgery) → more arms is possible (old
drugs vs. new drugs vs. placebo).
- (b) randomisation = blinded allocation to one of the intervention groups
based on chance → each person has an equal chance of being assigned to
each treatment, but the treatment to be given cannot be predicted →
used to; (I) minimise selection bias/confounding (groups are equal) and
(II) get rid of the effect of indication for the intervention/prognosis
influences (unconscious) allocation to treatment by doctors → blinded
allocation = single-blinded; only patient -> double-blinded; patient and
researcher don’t know how the treatment is allocated (=the best) →
unblinded trials most likely are biased.
→ advantages: (a) golden standard of clinical research, (b) no selection bias, (c) no confounding
bias (known and unknown), (d) high internal validity (it measures what it sets out to measure).
→ disadvantages: (a) expensive, (b) cannot be used for harmful substances (unethical).
Lecture 2: Frequency, Risks and Associations - 12/01/2023
Measures of frequency:
→ bathtub picture right: there can also be recurrence (person gets infected again).
(1) Prevalence = includes both new cases and those who contracted the disease in the
past and still have the disease → at a specific time →
accumulated amount of cases of a disease.
→ example picture left: amount of new cases (incidence) is
low/ stable, but the accumulated amount of cases of AIDS is
increasing, because people can live longer with AIDS → usually
prevalence and incidence come closer together over time (picture
