1.

OBSERVATIONAL STUDIES
A. DESCRIPTIVE STUDY
DESCRIBE DIESEASE BY
TIME
PLACE
PERSON
B. ANALYTICAL STUDIES
ECOLOGICAL STUDY
CROSS SECTIONAL STUDY
CASE-CONTROL STUDY
COHORT STUDY
2. EXPEREMENTAL STUDIES
RANDOMIZED CONTROLLED TRIAL (RCT)
FIELD TRIAL
COMMUNITY TRIAL
Cohort Study:

Key Point:

Presence or absence of risk factor

is determined before outcome
occurs.
 Cohort studies
 Rate
 Rate difference
 Rate Ratio (strength of association)

Case control studies

 No calculation of rates
 Proportion of exposure
 Cohort studies

 longitudinal
 Prospective studies
 Forward looking study I
 Incidence study

 starts with people free of disease

 assesses exposure at “baseline”
 assesses disease status at “follow-up”
 Prospective cohort studies

 Framingham study of cardiovascular disease, 1948

 Japanese atomic bomb survivors, 1946
 Colorado Plateau uranium miners, 1950s

 Retrospective cohort studies

 Aniline-dye occupational cohort, 1954

Prospective cohort study

Retrospective (historical) cohort study

Combination of Retrospective and

Prospective cohort study.
 Cohort Study

DZ

E
DZ
Healthy

People DZ
E
-

DZ
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Time
 When there is good evidence of exposure and
disease.
 When exposure is rare but incidence of disease
is higher among exposed
 When follow-up is easy, cohort is stable
 When ample funds are available
 Frame work of Cohort studies
Disease Status
Total Yes No

Study
Exposure
Yes a+b a b cohort

Status
No Comparison
c+d c d cohort

N a+c b+d
exposed

unexposed
General consideration while selection
of cohorts

Both the cohorts are free of the disease.

Both the groups should equally susceptible
to disease
Both the groups should be comparable
Diagnostic and eligibility criteria for the
disease should be defined well in advance.
  Selection of study subjects
 Obtaining data on exposure

 Selection of comparison group

 Follow up

 Analysis
 General population
 Whole population in an area
 A representative sample
 Special group of population
 Select group
 occupation group / professional group (Dolls study )
 Exposure groups
 Person having exposure to some physical, chemical or
biological agent
 e.g. X-ray exposure to radiologists
 Personal interviews / mailed questionnaire
 Reviews of records
 Dose of drug, radiation, type of surgery etc
 Medical examination or special test
 Blood pressure, serum cholesterol
 Environmental survey

 By obtaining the data of exposure we can

classify cohorts as
 Exposed and non exposed and
 By degree exposure we can sub classify cohorts
 Internal comparison
 Only one cohort involved in study
 Sub classified and internal comparison done
 External comparison
 More than one cohort in the study for the purpose of
comparison
 e.g. Cohort of radiologist compared with
ophthalmologists
 Comparison with general population rates
 If no comparison group is available we can compare
the rates of study cohort with general population.
 Cancer rate of uranium miners with cancer in
general population
 To obtain data about outcome to be determined
(morbidity or death)
 Mailed questionnaire, telephone calls, personal
interviews
 Periodic medical examination
 Reviewing records
 Surveillance of death records
 Follow up is the most critical part of the study
 Some loss to follow up is inevitable due to
death change of address, migration, change of
occupation.
 Loss to follow-up is one of the draw-back of the
cohort study.
 Calculation of incidence rates among exposed
and non exposed groups

 Estimation of risk
Calculate
measure of frequency:

 Cumulative incidence
- Incidence proportion
- Attack rate (outbreak)

 Incidence density

end of follow-up
exposed

unexposed
 Disease Status
Yes No Total

Study
Exposure
Yes a b a+b cohort

Status
No Comparison
c d c+d cohort

a+c b+d N
 a
INCIDENCE EXPOSED =---------
a+b

c
INCIDENCE non EXPOSED =---------
c+d
  Absolute measures

Risk difference (RD = AR) Ie - Iue


Relative measures
 Relative risk (RR)
 Rate ratio
 Risk ratio

Ie
Iue

Ie = incidence in exposed
Iue= incidence in unexposed
 Relative Risk
incidence of disease among exposed
RR =
Incidence of disease among non-exposed

` a/a+b
= _________
c/c+d
 Attributable risk (AR)
 AR = the amount of disease incidence that can
be attributed to a specific exposure
 Difference in incidence of disease between exposed and
non-exposed individuals
 Incidence in non-exposed = background risk
 Amount of risk that can be prevented

 Attributable fraction (AF)

 AF = the proportion of disease incidence that can be
attributed to a specific exposure (among those who
were exposed)
 AR divided by incidence in the exposed X 100%
32
 Attributable Risk Fraction
Incidence of disease among exposed –
incidence of disease among non exposed
AF =
Incidence of disease among exposed

a/a+b – c/c+d
AF =
a/a+b
 Attributable Risk( Risk different)

Incidence of dis.exposed – incid. of dis. non exposed

AR = a/a+b – c/c+d

a/a+b – c/c+d
Attrib.Risk Fraction=
( AF) a/a+b

34
 Presentation of cohort data:
Population at risk

Does HIV infection increase risk of developing TB

among a population of drug users?
Population Cases
(f/u 2 years)

HIV + 215 8
HIV - 289 1

Source: Selwyn et al., New York, 1989

 Presentation of cohort data:
Person-years at risk

Tobacco smoking and lung cancer,

England & Wales, 1951

Person-years Cases
Smoke 102,600 133
Do not smoke 42,800 3

Source: Doll & Hill

 Presentation of data:
Various exposure levels
Source: Doll & Hill
 Perlu digaris bawahi :

 Only cohort studies (including clinical

trials) can yield incidence and relative
risk.

 The odds ratio, (a case-control study) will always be

greater than the relative risk.
 For rare diseases, the odds ratio will be close to the
relative risk.
Smoking Lung cancer Total
YES NO
YES 70 6930 7000
NO 3 2997 3000
73 9927 10000

Find out RR and AR for above data

Incidence of lung cancer among smokers

70/7000 = 10 per 1000
Incidence of lung cancer among non-smokers
3/3000 = 1 per thousand
RR = 10 / 1 = 10
 Incidence of lung cancer among smokers
70/7000 = 10 per 1000
 Incidence of lung cancer among non-smokers
3/3000 = 1 per thousand
RR = 10 / 1 = 10
(lung cancer is 10 times more common among
smokers than non smokers)

AF = 10 – 1 / 10 X 100= 90 %
(90% of the cases of lung cancer among smokers are
attributed to their habit of smoking)
Smoking Lung cancer Total

YES NO

YES 70 6930 7000

NO 3 2997 3000

73 9927 10000

Find out RR and AR for above data

42
 Incidence of lung cancer among smokers
70/7000 = 10 per 1000
 Incidence of lung cancer among non-smokers

3/3000 = 1 per thousand

RR = 10 / 1 = 10
(lung cancer is 10 times more common among
smokers than non smokers)

AF = 10 – 1 / 10 X 100
= 90 %
(90% of the cases of lung cancer among smokers are
attributed to their habit of smoking)
43
44
 Excess
Risk
Risk
100

80 AR = Risk among risk

factor positives
60

40
Risk among risk
20 factor negatives
0
+ -
Risk Factor
45
 Risk among - Risk among
risk factor risk factor
positives negatives
AF = X 100%
Risk among
risk factor
positives

46
 Relative risk and odds ratio are important as
measures of the strength of association
 Important for deriving causal inference
 Attributable risk is a measure of how much disease
risk is attributed to a certain exposure
 Useful in determining how much disease can be
prevented
 Therefore:
 Relative risk is valuable in etiologic studies of disease
 Attributable risk is useful for Public Health guidelines and
planning

47
 Strengths
Weaknesses
 We can find out
incidence rate and risk
 losses to follow-up
 More than one disease
 often requires large
related to single sample
exposure  ineffective for rare
 can establish cause - diseases
effect  long time to complete
 good when exposure
is rare
 expensive
 minimizes selection
 Ethical issues
and information bias
exposed

PENELITIAN
BERHENTI

unexposed
55
THANK YOU