background image

EPIDEMILOGICAL STUDY

DESIGNS

Dr Faris Al-Lami

MB,ChB MSc PhD FFPH


background image

Definition of Epidemiology

"Epidemiology is the study of the

distribution

and

determinants

of

health related states or events in
specified populations, and the
application of this study to

control

of health problems.”

-

Last’

s Dictionary of Epidemiology


background image

USES OF EPIDEMIOLOGIC

RESEARCH METHODS

1. Measure the disease

frequency

and the

pattern

of disease occurrence

2. Identification of disease

determinants

3. Evaluate

validity

of diagnostic screening tests

4. Determining the

natural history

of the disease

and clinical course

5. Determining the

prognostic factors

6.

Testing

new treatments


background image

Epidemiological studies can be

classified into:

I. Descriptive studies which can be

classifies into:

1. Studies conducted on population

(Correlation studies)

2. Studies conducted on individuals:

a- Case report
b- Case series
c- Cross sectional surveys


background image

Epidemiological studies can be

classified into:

II. Analytic studies which can be classified

into:

1. Observational analytic studies:

A-Case control studies
B-Cohort studies

2. Interventional studies (Clinical trials)


background image

Epidemiological Studies

Descriptive studies

Analytic studies

Population

Individuals

Correlation
Studies

-Case Report
-Case series
-Cross Section
Studies

Observation
Studies

Intervention
studies

-Case Control
Studies
-Cohort studies

Experimental
studies
(clinical trials)


background image

The Five W’

s of Journalism /

Epidemiology

• What
• Who

• Where
• When

• Why / How

= Diagnosis

(the disease or condition

being investigated)

= Person

(who is getting the disease,

who is at risk)

= Place

(residence, work, school, etc.)

= Time

(date and/or time of onset

of disease, time of exposure)

= Cause, mode of

transmission, risk factors


background image

The Five W’

s of Journalism / Epidemiology

•What

•Who
•Where
•When

•Why / How

=

Clinical

=

Person

=

Place

=

Time

=

Cause,

=

mode of
transmission,
risk factors

Descriptive
Epidemiology
(Distribution)

Analytic
Epidemiology
(Determinants)


background image

DESCRIPTIVE STUDIES

Advantages:

•They use already

available data

•They are

less expensive

and less time

consuming

•They describe the

pattern

of disease

occurrence

•They

formulate

research hypothesis


background image

DESCRIPTIVE STUDIES

These studies are essential for:

1. Public health administrators:

•To determine which population or subgroup of the

population are most or least affected.

• To decide on efficient allocation of resources and

preventive programs.

2. Epidemiologist:

- To identify risk factors


background image

I. Correlation or Ecological Studies

Based on aggregate measures of exposure
and outcome from several populations.

The

population

is the unit of observation

available for study.


background image

Ecological Study of Homicide Rate by Percentage of

Households with a Gun in 13 Countries, 1990

*per million person-years


background image

background image

I. Correlation Studies

Advantages:
1.

Describe the disease in the entire population in
relation to factor of interest.

2.

Use the Correlation Coefficient (r) to measure the
association between the two variables of interest.

3.

Easy to do, inexpensive, and can be conducted
quickly

4.

Represent the first step in searching for exposure
disease relationship (Generate hypotheses)

5.

Use available data (“

administrative”or other aggregate

data)


background image

Correlational or Ecological

Studies

Exposures:

- What percent of a
population smokes?

- What percent of 1-

year old children are
vaccinated against

measles?
- What percent of a

population has piped
water?

Outcome:

- What percentage of a
population died from
MI?
- What percentage of
children had measles
last year?
- What percentage of
population had
episodes of diarrhea?


background image

Correlational or Ecological Studies

Disadvantages

• Correlation data represent average

exposure level rather than actual
individual values. Data on exposure and
data on outcome are collected
independently

• No assurance that persons with exposure

(risk factor) of interest are the same ones
with the outcome (disease) of interest


background image

Disadvantages

•Inability to link exposure with disease in a

particular individual. Association at the
aggregate level may not reflect association
at the individual level - the

ecologic fallacy

•Unable to adjust for potential confounding

factors.


background image

II. Case report and case series

• The

individual

is the unit of observation.

• They describe the experience of a

single patient

or a small number

of patients with a similar

diagnosis reflecting

unusual

features of a

disease.

• They help in:

Ø

formulation of a hypothesis on etiological

association

Ø

represent the first clues in identification of

new disease or epidemic.


background image

Case reports

Clinical case with

unusual”

clinical picture

(e.g. 40 y.o. female, pulmonary embolism

and no history of CVD, recent use of OC)


background image

Case series

First case report may stimulate compilation of

additional case reports….a case series

(e.g. occurrence of Pneumocystis carinii

among a group of young, homosexual men
with no history of immune deficiency)


background image

Case reports or Case series

Advantages:
ØUse available clinical data
ØDetailed individual data
ØSuggest need for investigation

(hypothesis generation)

Disadvantages:
May reflect experience of one person or

one clinician

No explicit comparison group


background image

III. Cross sectional surveys

• Collection of data on

several individuals

at

one

point”

in time.

• Determines

prevalence

at a point in time

• Therefore, CS is a prevalence study
• The exposure and disease status are assessed

simultaneously among individuals in a well
defined population.

• Snapshot in time

- e.g. - cholesterol measurement and ECG measured

at same time


background image

•Cross sectional studies can be used to

formulate a hypothesis but not to test it
EXCEPT if the exposure is present since
birth as blood group, race, eye colour,


background image

Cross-sectional Surveys

Examples: national census, community
survey, survey of a particular group of
persons (e.g., occupational group,
school children)


background image

III. Cross sectional surveys

Advantages:

•Provide information on frequency and

characteristics of the disease.

•Determine prevalence of disease or any

health phenomenon.

•Standardized data collection tool.


background image

Advantages:

•Able to focus data collection in specific

locations or specific groups of persons.

•Can make comparisons among study

participants.

•Relatively quick to do.
•Can be repeated to get data on trends.


background image

III. Cross sectional surveys

Limitations:

•Inability to determine the temporal

relationship between exposure and
disease.

•May be biased by lack of participation
•Reflects prevalent, not incident cases and

thus results may be related to duration of
disease, or survival with disease (e.g., “
healthy worker effect”

)


background image

Job A (hazardous)

80 healthy

80 well

100 workers

20 ill

10 ill

Job B (non- hazardous)

95 healthy

95 well

100 workers

5 ill

15 ill

Prevalence ill job A: 20/100 = 20%

?

Prevalence ill job B: 5/100 = 5%

?

Prevalence ratio:

4

?

Cross-sectional Studies

10

Time


background image

How to conduct

a Cross-Sectional Study

1. Clear definition of study question. i.e.:

Exposure and outcome

2. Identification of study population:

Small population

All are included

Large population

Sampling


background image

How to conduct

a Cross-Sectional Study

3. Data collection: Exposure and outcome
4. Data analysis: Subdivision of the

population to suspected factor and
compare the prevalence of the outcome
in each subgroup

5. Interpretation: is there a relation

between exposure and outcome?


background image

Design of a C-S Study




رفعت المحاضرة من قبل: Abdalmalik Abdullateef
المشاهدات: لقد قام 20 عضواً و 219 زائراً بقراءة هذه المحاضرة








تسجيل دخول

أو
عبر الحساب الاعتيادي
الرجاء كتابة البريد الالكتروني بشكل صحيح
الرجاء كتابة كلمة المرور
لست عضواً في موقع محاضراتي؟
اضغط هنا للتسجيل