Research Methodology

Notes

Faculty of Public Health and Tropical Medicine

Epidemiology Department
2016/2017
FPHTM/EPI/RM/1-T
 What is Research?

Definition
Research is the systematic collection, analysis
and interpretation of data to answer a certain
question or to solve a problem.

FPHTM/EPI/RM/1-T
 Characteristics of Research

1- Controlled: Research procedures are conducted in a

way that minimizes the effects of other factors that
might affect the relationship under the study
(cigarette smoking & cancer lung).
2-Rigorous:The procedures followed to find answers
to research questions are precise and justified (effect
of exercise on blood pressure, guidelines to measure
blood pressure).

FPHTM/EPI/RM/1-T
 Characteristics of Research..
3- Systematic: The research process
follows a certain logical sequence (steps
of scientific research).
4-Valid and verifiable: Research findings
are correct and can be verified by others
(reflect reality).

FPHTM/EPI/RM/1-T
 Characteristics of research..

5-Critical:The process of research must be

perfect and free from defects (sampling).

6-Empirical: This means that research

conclusions are based on evidence that
has been gathered from real life
observations (population field studies).

FPHTM/EPI/RM/1-T
 Drivers for Health Research

1. Curiosity-driven research.
2. Needs-driven research.
3. Profit-driven research.

FPHTM/EPI/RM/1-T
 Types of Research

1. Basic research: generate new knowledge and

technology to deal with unresolved health
problems.

2. Applied research: identify priority health

problems, design and evaluate policies and
programs for solving these problems.

FPHTM/EPI/RM/1-T
 Types of Applied Research
1. Qualitative research : to gain familiarity with a new
phenomenon or experience of individuals.

2. Quantitative research: to identify the

characteristics of a particular population or to
test a hypothesis of a causal relationship between
variables or to test the effectiveness of new drug.

3. Health services research: to improve the

performance and effectiveness health care
services.

FPHTM/EPI/RM/1-T
 Characteristic of Health Services Research

1.Focus on priority health problems.

2. It is action-oriented ( aims to develop solution).

3. It is Multidisciplinary ( involves different disciplines,

epidemiology, health education, environmental
health).

4. It is participatory ( involves policymakers,

researchers, members from community, program
managers).
FPHTM/EPI/RM/1-T
 Characteristic of Health Services Research

1.Focus on priority health problems.

2. It is action-oriented ( aims to develop solution).

3. It is Multidisciplinary ( involves different disciplines,

epidemiology, health education, environmental
health).

4. It is participatory ( involves policymakers,

researchers, members from community, program
managers).
FPHTM/EPI/RM/1-T
 Characteristic of Health Services Research

5. It is short-term research.

6. It is timely research (urgent need for findings).

7.It is cost-effective research (carried by health

personnel during their daily activities).

8. Findings should be presented to decision-maker and

community in a simple understandable way.

9. It has the ability to influence policy, improve services

and ultimately lead to better health.
FPHTM/EPI/RM/1-T
 Criteria of a good research question
The five Essential characteristics of a good research
question

1. Feasible
2. Interesting
3. Novel
4. Ethical
5. Relevant
FPHTM/EPI/RM/1-T
 The Five Criteria of a Good Research question
(FINER Formula)
1. Feasible: can be done and completed.
2. Interesting: to the researcher and experts
3. Novel: contribute new information.
4. Ethical: in selecting & planning the research.
5. Relevant: has important implications.

FPHTM/EPI/RM/1-T
 Good Research Question?

• Feasible • Ethical
– Experts – Social or scientific value
– participants – Safe of participants
– Resources • Relevant
– Management – Advance scientific knowledge?
– Data available – Influence clinical practice?
• Interesting – Impact health policy?
• Novel – Guide future research?
– Confirm or refute previous research?
– New idea
– New setting, new population

FPHTM/EPI/RM/1-T
Steps of conducting scientific Research
1.Formulating the Research Problem
2. Conduct Literature Review
3. Specify the Research Objectives
4. Define research questions and hypotheses
5. Specify Variables of the study
6. Select tools of data collection /questionnaire

FPHTM/EPI/RM/1-T
Steps of conducting scientific Research
7. Determine Population and sampling
8. Select the appropriate Research design
9. Train data collectors
10. Conduct Pretest / pilot study of the questionnaire
11. Collect Data
12. Analyze and Interpret Data
13. Prepare the Final Report

FPHTM/EPI/RM/1-T
RESEARCH PROBLEM

FPHTM/EPI/RM/1-T
 Research problem
Definition (criteria)

1. A discrepancy between what is present and

what should be.

2. The reasons of such discrepancy are not clear .

3. There is more than one answer to the reason of

the problem or its solution.

FPHTM/EPI/RM/1-T
 Research problem
Criteria for prioritizing research problems
1. Relevance: size, severity, who is affected
2. Avoidance of duplication
3. Feasibility: measures of feasibility (3 M)
4. Political acceptability
5. Applicability of research findings & recommendation
6. Urgency for data/ information for decision taking
7. Ethical acceptability of community & study participants
8. Community concern
9. Trend of health problem over years
10. Availability of data

FPHTM/EPI/RM/1-T
 Criteria for prioritizing research problems

1-Relevance: The topic you choose should be a public health

problem. Questions to be asked include:
1-How large or widespread is the problem?
2- How severe is the problem?

2-Trend: health problems with upward trend are usually given

higher priority over those problems that demonstrate download
trend over years.

FPHTM/EPI/RM/1-T
3- Feasibility: Consider the availability of the required
resources and expertise who would help in
conduction of the proposed research.

4- Applicability of the research findings: you should

consider the applicability of the possible research
findings in your community. This mainly depends on
political interest with the searched problem and
availability of the required resources.

FPHTM/EPI/RM/1-T
5. Avoidance of duplication: Before deciding your
research proposal, it is important to ensure that the
proposal has not been searched before, within the
proposed study population.

6. Availability of data: Before starting your research

proposal, ensure that the required data are both
available and accessible.

FPHTM/EPI/RM/1-T
7. Urgency for data: new health problems with
urgency of data are usually given higher priority for
funding.

8. Community concern: It is extremely important to

select a topic that is perceived by community members
as a priority problem.

FPHTM/EPI/RM/1-T
9. Political acceptability: It is advisable to select a research
topic that has the interest and support of the authorities. This
will facilitate approval and funding the proposed research .

10. Ethical issues: research related ethical issues need to be

considered and solved to ensure safety of the study
participants and approval of the research proposal through
research ethical committee.

FPHTM/EPI/RM/1-T
 Analysis of the problem
1. Identify the opinion of : researchers, managers, decision-
makers and community members about the causes of the
problem, through brainstorming.

2. Describe the problem:

a. nature of the problem: discrepancy ……
b. distribution of the problem: time, place, persons
c. size and intensity: prevalence/ incidence, mortality,
disability.
3. Classify possible factors leading to the problem:
a. socio-cultural factors
b. service related factors
c. disease related factors

FPHTM/EPI/RM/1-T
 Analysis of the problem
Classification of possible factors leading to the problem:

a. Socio-cultural factors: age, sex, education, occupation,

awareness about the disease, family size.
b. Service related factors: availability, accessibility,
acceptability, affordability.
c. Disease related factors: severity, duration, complications.

FPHTM/EPI/RM/1-T
LITERATURE REVIEW

FPHTM/EPI/RM/1-T
 Literature review
Sources of literature review:
1- Searching Journal articles through internet
2- Books
3- National health reports and records
4- WHO health reports
5- Experts opinion

FPHTM/EPI/RM/1-T
Why is it important to conduct literature
review when preparing for a research?
1- It prevents duplicating research work that has
been done before.

2- It helps in refining your research problem.

3- It helps you to become more familiar with the

various research methods that might be used in
your study.

4- It should provide you with convincing arguments

for justification of your research proposal.
FPHTM/EPI/RM/1-T
RESEARCH Questions & Hypotheses

FPHTM/EPI/RM/1-T
 Research Questions Must Identify
1. The variables under study
2. The population being studied
3. The testability of the question

FPHTM/EPI/RM/1-T
 Variables of the study
Variable definition:
A characteristic that differ from person to
person. Examples: age, sex, weight, height,
blood pressure, cholesterol level

FPHTM/EPI/RM/1-T
 Variables in Research

• Independent variables • Dependent variables

(exposure) (Outcome)

• Extraneous variables • Confounding variables

FPHTM/EPI/RM/1-T
 Types of variables
A- Qualitative variables : two types
1 - Qualitative nominal variables (categorical):
They are descriptive variables.
Examples: sex ( male, female) - anemia ( present,
absent) – blood groups (A, B, AB, O).
2- Qualitative ordinal variables:
They are descriptive variables that rank individuals.
Examples: grades of malnutrition (mild, moderate and
severe malnutrition) and grades of malignancy (
(grade0, grade1, grade 2, grade3 )

FPHTM/EPI/RM/1-T
 Types of variables
B- Quantitative variables : two types
1 - Quantitative continuous variables:
These variables are measured and have fraction.
Examples: weight of a person 82.5 Kgm and height
110.6 cm.
2- Quantitative discrete variables:
These variables take only integer values. they are
counted. They do not have fraction. Examples:
heart beats , number of patients

FPHTM/EPI/RM/1-T
Dependent & Independent variables

1 - Dependent variable:
The variable that describe or measure the
problem of the study.

2- Independent variables:
The variables that are assumed to cause or
influence the problem.

FPHTM/EPI/RM/1-T
 Dependent & Independent variables

- In studying the relationship between smoking

and cancer lung , what is the dependent
variable and the independent variable?

- Lung cancer is the dependent variable.

- Smoking is the independent variable.

FPHTM/EPI/RM/1-T
 Confounding variable
Definition: a variable that can either strengthen
or weaken the association between the disease
and the possible cause.
Example: a cohort study on effect of smoking on
cancer lung. The study will involve 2 groups.
Smoking group and non smoking group. Age
could be a confounder, if the two groups were
not similar in age.

FPHTM/EPI/RM/1-T
 Extraneous variables
Definition: background variables that might
influence exposure and disease relationship ,
some of the background variables are confounders
Example: a cohort study on effect of smoking on
cancer lung. The study will involve 2 groups.
Smoking group and non smoking group.
Occupation and environmental pollution could be
considered as background variables

FPHTM/EPI/RM/1-T
 Hypothesis
• Statement about the relationship between
two or more variables.
• Why research hypothesis is important? it
enables the researcher to identify the
variables of the study and the data to be
collected.
• Rate of cancer prostate is higher in smokers
than non smokers among males aged ≥50
years in Jazan region.

FPHTM/EPI/RM/1-T
Characteristics of research hypotheses
1. Declarative statement that identifies the
predicted relationship between 2 or more
variables
2. Testability
3. It is Based on sound scientific theory/rationale

FPHTM/EPI/RM/1-T
 Directional versus Non-Directional
Hypotheses
• Directional hypothesis
– Specifies the direction of the relationship between
independent and dependent variables
– Cure rate of drug A is higher than that of drug B
• Non-directional hypothesis
– Shows the existence of a relationship between
variables but no direction is specified
– Cure rates of drug A and drug B are different

FPHTM/EPI/RM/1-T
 Research Hypothesis versus Statistical
Hypotheses

• Research hypothesis: A prediction statement

about two or more variables.
• There are two statistical hypotheses for
testing a research hypothesis
1) null hypothesis - no difference.
2) alternative hypothesis – there is a difference

FPHTM/EPI/RM/1-T
 State Statistical Hypotheses

Null-hypothesis (H0) & alternative hypothesis (Ha)

H0: no difference between drug A & drug B

Ha: there is difference drug: A ≠ drug B (two-sided) or
Ha: drug A < drug B (left-sided) or
Ha: drug A > drug B (right-sided)

FPHTM/EPI/RM/1-T
Hypotheses Testing : Purpose

Purpose of Hypotheses Testing is to decide

whether the observed difference is real
(significant) or due to chance (no significant)
as a result of sampling error

FPHTM/EPI/RM/1-T
 Hypotheses Testing : Steps

A. Select test statistic and calculate the value

C. Convert the value of the test result to P value
D. Compare the P value versus the alpha (α) (0.05)

FPHTM/EPI/RM/1-T
 P-value Definition

- The probability of obtaining the observed difference

or one more extreme, if the null hypothesis is true

- We use this p-value to decide whether to reject or

not reject the null hypothesis.

FPHTM/EPI/RM/1-T
 Interpretation of p-value
Conventions
P > 0.05  non-significant
P  0.05  significant
P  0.01  highly significant

Examples
P =.27  non-significant
P =.01  highly significant

FPHTM/EPI/RM/1-T
Errors of statistical hypothesis testing

- Type I error (a): When a true null hypothesis is

rejected
- Type II error (b): When a false null hypothesis is
accepted

Type I error is considered to be more serious than

a Type II error.

FPHTM/EPI/RM/1-T
Table of hypothesis testing :two
types of Error

False positive rate “sensitivity” or “power”

a
a
False negative rate

b
FPHTM/EPI/RM/1-T
b
RESEARCH OBJECTIVES

FPHTM/EPI/RM/1-T
 Lecture Outline
• Definition of research objective
• Importance of research objective
• Types of research objectives
• Examples of research objectives
• criteria of good research objectives

FPHTM/EPI/RM/1-T
 Research Objectives
Definition of research objectives
• What is to be achieved by the end of the
study.

Importance of research objectives

• Focus the study to the essential points
• Avoid collection of data that are unnecessary
• Help to organize the study methodology

FPHTM/EPI/RM/1-T
 TYPES OF OBJECTIVES

• General objective

• Specific objectives

FPHTM/EPI/RM/1-T
 General Objective
• Characteristics of general objective

– Usually only one general for each study (broad)

– Related to the problem of the study
– Use action verb as: to identify, evaluate, describe,
compare.
– should specify target population and place

FPHTM/EPI/RM/1-T
 Example:

Problem
increased number of diarrhea cases among children in village X

General objective of the study

-

Specific objectives
-
-
-
-
-

FPHTM/EPI/RM/1-T
 Example:
Problem
– Increased number of diarrhea cases among children in
village X.

General objective of the study

- To determine the factors associated with increased number
of diarrhea cases among children in village X.

Specific objectives
1.To find out relationship between breast feeding and
occurrence of diarrhea.
2.To find out relationship between education of mother and
occurrence of diarrhea.
3.To find out relationship between source of drinking water
and occurrence of diarrhea.
4.To determine the nutritional status of the children.
FPHTM/EPI/RM/1-T
 Characteristics of Good Specific Objectives

1. Specific : focusing on one specific point of

the problem

2. Measurable :use of action verbs (determine,

find out, identify, compare, describe)

FPHTM/EPI/RM/1-T
Characteristics of good specific objectives

3. Attainable: at the end of study

4. Realistic: based on the available resources

5. Time bound: can be achieved in the given

time

FPHTM/EPI/RM/1-T
 SMART
Specific
Measurable
Attainable
Realistic
Time bound

FPHTM/EPI/RM/1-T
Population and Sampling

FPHTM/EPI/RM/1-T
 Population versus sample
Target Population

• Study Population

Sample

Sample frame

Target Population: the population we want to estimate a parameter

Study Population: the population to be sampled

Sample: a subset of the study population

Sample frame: list of all sampling units in the study population

FPHTM/EPI/RM/1-T
 Definitions
Target Population:
The population to be studied/ to which the investigator wants
to generalize his results

Sampling Unit:
Basic unit of sampling e.g. person, household, school

Sampling frame
List of all the sampling units from which sample is drawn

Sampling scheme
Method of selecting sampling units from sampling frame

Sample fraction
A ratio between sample size / population
FPHTM/EPI/RM/1-T
 Populations and Sampling

Reasons for studying a sample instead of an entire

population:

• Save time and money

• Sample results may be more accurate than
population.
• Reduce bias
• A study of the entire populations is impossible
(feasibility)
FPHTM/EPI/RM/1-T
 Types of Samples

Samples

Probability Samples
Non-Probability
Samples
Simple
Random Stratified

Convenience Purposive Cluster

Systematic

Multi-stage
quota snowball
FPHTM/EPI/RM/1-T
 Sampling Methods
Non-probability samples
• Advantages:
- Save time
- lower cost
• Disadvantages:
– Does not involve random selection
– sampled are biased (Lack of accuracy)
– results can not be generalized to population
• Use
It is mainly used in qualitative research and social studies

FPHTM/EPI/RM/1-T
 Sampling Methods
Non-probability samples (cont)
• Convenience sampling:
– We use available individuals or volunteers
– It is widely used in research done in health clinics

• Purposive sampling:
– We sample with purpose in mind
– We seek individuals with specified (predetermined)
criteria
– Main types are
1. Quota sampling
2. Snowball sampling

FPHTM/EPI/RM/1-T
 Purposive Sampling
Quota sampling
– We seek individuals with specified (predetermined) criteria
– It is used in political & social studies

Snowball sampling

– It is used when study population is inaccessible (hard to reach)

– Examples: drug addicts, homosexuals & prostitutes
– We begin with the persons known to have the specified criteria
– We ask them to recommend other persons who meet criteria

FPHTM/EPI/RM/1-T
 Probability Samples
Subjects of the sample are chosen based on known probabilities.
Guarantees that every element in the population of interest has the same
probability of being chosen for the sample as all other elements in the
population; “random” selection.

Probability Samples

Simple
Random Systematic Stratified Cluster

FPHTM/EPI/RM/1-T
 Advantage of Probability Samples

1.We can estimate sampling error

2.Results can be generalized from sample to
population using confidence interval
3.We can estimate the size of individuals with
specific characteristics in the population based
on the sample

FPHTM/EPI/RM/1-T
 Representative Sample
1) What is a representative sample?
2) When is it important to get a representative sample?
3) How to select representative sample?
A representative sample is a sample that resembles the
total population. It is important to get a representative
sample when want to understand the characteristics of
a population based on study of a sample (i.e., when you
want to directly generalize from your sample to your
population), through using random sampling

FPHTM/EPI/RM/1-T
 1) Simple Random Sampling

Characteristics :
1. Every individual in population has an equal chance of
being selected
2. The simplest form of random samples
3. Use table of random numbers or computers
programs for obtaining the sample

FPHTM/EPI/RM/1-T
 How to select a simple random sample?
1. Define the population
2. Determine the desired sample size
3. List all members of the population
4. Select the required sample

• tables of random numbers

• computer program

FPHTM/EPI/RM/1-T
 2) Systematic Sampling
Steps of Selection
1) Population size: N
2) Sample size: n
3) Sample fraction ( F) = N/n
4) Select randomly one number within range of sample fraction
5) Select other individuals through consecutive adding of sample
fraction
12345678
First 1- 16
1) N = 200 individual
2) n = 25 in sample 17- 32
3) F = 8 33- 48
4) random number = 3
5) Individuals with serial No 11, 49 - 64
19, 27,35,…....
population

FPHTM/EPI/RM/1-T
 Systematic Sampling (cont)

• Advantage: The sample usually will be easier

to identify than it would be if simple random
sampling was used.

• Dis-advantage: Cyclic periodicity

FPHTM/EPI/RM/1-T
 Steps of Selecting a Systematic Sample
1. Define the study population
2. Prepare sample frame for the study population
3. Determine the desirable sample size
4. Calculate sample interval (k= N/n)
5. Select the 1st person randomly between 1 & K
6. Add sample interval to identify the consecutive
individuals

FPHTM/EPI/RM/1-T
 3) Stratified Random Sampling

Characteristics
- It is the most accurate random sampling method
- It is used when the population is heterogeneous
- the population is divided into homogenous subgroups
(strata) according
to the stratifying variable
- The most important stratifying vaiables are : sex and age

.

FPHTM/EPI/RM/1-T
 Stratified Random Sampling
Steps of selecting stratified sampling

1- Divide the population into subpopulations based on

one or more stratifying variables.

2- Take random samples from each of these subpopulations.

3- The sample sizes from the subpopulations will depend on

whether you are using “proportional allocation sampling” or
“disproportional allocation sampling”

FPHTM/EPI/RM/1-T
 Stratified Random Sampling
 Example:
 We want to take a random sample (100 students) of
Fourth Year Medical students (1000 total students – 750
males and 250 females).

• We either take a random sample of 50 students from

each group (disproportional allocation method) or we
take 75 males and 25 females (proportional allocation
method). The second method is preferred especially if we
want to make generalization on population based on the
sample.

FPHTM/EPI/RM/1-T
 4) Cluster Sampling

 Definition:
It is a random sampling method in which the sampling unit (cluster / school) is
different from the study unit (school)

 Indications of use when :

1- It is difficult to construct the sample frame of the population, or
2- The population to be sampled is dispersed over a wide geographical area

 Main steps:
1) We divide the population into equal homogenous clusters
2) We select some clusters randomly
3) individuals in the selected are studied.

 Examples:
 Sample of primary school children in Jazan ( school)
 Sample of women aged 15-49 in a village (household)
 Sample of hospital Physicians working in Jazan ( hospitals)
FPHTM/EPI/RM/1-T
 Types of Studies

A. Experimental - study factor is manipulated

by the investigator
B. Observational – no manipulation of study
factor by the investigator
• Descriptive
• Analytic

FPHTM/EPI/RM/1-T
 I- Observational Studies
Hypothesis generating
A. Descriptive studies:
 Case report /Case series
 Correlation studies
 Cross-sectional studies
Hypothesis testing
B. Analytic studies:
 Case-control
 Cohort study
FPHTM/EPI/RM/1-T
 II- Experimental Studies
Intervention studies (All are hypothesis testing)
 Clinical trials
 Community trials (quasi-experimental trial)

FPHTM/EPI/RM/1-T
Descriptive Studies

FPHTM/EPI/RM/1-T
1. Case reports (case study)
2. Case series
3. Correlation studies
4. cross sectional studies.
Descriptive studies provides information on various
characteristics of person-place-time.

FPHTM/EPI/RM/1-T
 Cross-Sectional Studies (Prevalence studies)
• An “observational” design that measures existing disease
(D) and current exposure levels (E) at a single point in time
(a cross-section of the population)
• can not determine if really exposure preceded disease or
not. (Chicken egg dilemma)

time
Study only exists at this point in time
FPHTM/EPI/RM/1-T
Cross-sectional Design
factor present
Disease
factor absent
Study
population
factor present
No Disease
factor absent

time
Study only exists at this point in time
FPHTM/EPI/RM/1-T
Main steps of cross-sectional study:

1. Define the objective of the study

2. Define the population to be surveyed
3. Define the diseases or conditions to be studied
4. Specify the sample size and sampling technique

FPHTM/EPI/RM/1-T
Main steps of cross-sectional study (cont.)

6. Define the variables to be collected

7. Develop the study questionnaire
8. Training of interviewers and laboratory personnel
9. Pilot testing of the study tools
10.Collection of the data and laboratory samples

FPHTM/EPI/RM/1-T
Main steps of cross-sectional study (cont.)

11.Data entry and analysis using the appropriate

statistical package for
 overall & specific disease prevalence rates
 overall & specific exposure rates
 personal characteristics of diseased persons

12. Prepare a report of study objectives, activities,

findings and recommendation

FPHTM/EPI/RM/1-T
 Characteristics of Cross-sectional Studies

• “Snapshot Studies”
• Describes what exist at a single point in time
• Each subject is assessed once at point in time.
• Point Prevalence Studies

FPHTM/EPI/RM/1-T
 Uses of Cross Sectional Surveys
1. Estimate prevalence of disease in population
2. identify characteristics of the diseased person
3. Plan health care services
4. Generate hypotheses about risk factors of
disease.

FPHTM/EPI/RM/1-T
 Advantages of Cross-sectional Studies
• Used to study conditions that are relatively frequent
with long duration (chronic conditions)
• Good for generating hypotheses about the cause of
disease
• Can estimate overall and specific disease prevalence
rates
• Can estimate exposure proportions in the population.
• Relatively easy, quick and inexpensive.

FPHTM/EPI/RM/1-T
Disadvantages of Cross-sectional Studies
• Impractical for rare diseases
• Not a useful type of study for establishing causal
relationships

• Does not allow to determine which came first

(Problems with temporal sequence of data)
• Not suitable for studying highly fatal diseases or a disease
with short duration of expression

FPHTM/EPI/RM/1-T
Advantages and Limitations of Cross
sectional Survey
Advantages
• Fairly quick and easy to perform
• Less expensive
Limitations
• Not useful to study disease etiology
• Not suitable for the study of rare disease
• Exposure and outcome examined at the
same time. e.g. Obesity and diabetes

FPHTM/EPI/RM/1-T
1. Assess the presence of diabetes mellitus in
relation to physical exercises.
2. Assess the presence of obesity in relation to
diabetes mellitus.
3. Assess the presence of varicose veins in relation to
the use of oral contraceptive pills.

FPHTM/EPI/RM/1-T
 Summary of Descriptive Studies

Case Report One case of unusual

Disease/injury finding
Multiple cases of
Case Series Disease/injury finding
Correlation studies Entire population
data
Cross sectional Population-based
studies study
FPHTM/EPI/RM/1-T
Cohort studies

FPHTM/EPI/RM/1-T
 I- Observational Studies
A. Descriptive studies:
 Case report
 Case series
Hypothesis generating
 Correlation studies
 Cross-sectional studies
B. Analytic studies:
 Cohort study Hypothesis testing
 Case control
FPHTM/EPI/RM/1-T
(estimate and/calculate risks(
 Basic Question in Analytic Epidemiology

Are exposure and disease linked?

Exposure Disease

FPHTM/EPI/RM/1-T
 Cohort Studies

– An observational design where the investigator

categorizes individuals on the basis of exposure
into 2 groups exposed & unexposed groups

– looking for a difference in the rate of disease between

exposed & unexposed groups

FPHTM/EPI/RM/1-T
 What is a cohort?
• Cohort is a group having a common characteristic
 smokers
 Medical Students
 Infants born in year 2008 (birth cohort)
 Infants borne through C.S
 Low birth weight infants
 Women using contraceptive pills

FPHTM/EPI/RM/1-T
 disease
Factor
Study present no disease
population
free of
disease Factor disease
absent
no disease

future

time
present Study begins here
FPHTM/EPI/RM/1-T
 Types of Cohort Studies
Prospective cohort:

• the disease has not yet occurred.

• Subjects are followed up for disease

Retrospective cohort( historical cohort)

• The disease has occurred before start of study

• Use of Health records of pre-employment exam
• Used in occupational diseases
• Example: records of workers in glass factory from 1998 to
2008
FPHTM/EPI/RM/1-T
 Definitions
• Incidence is the frequency of occurrence of new cases
of a disease in a population at risk during a period of
time

• Prevalence is the proportion existing cases (old+new)

in the population at a given point of time

FPHTM/EPI/RM/1-T
 Strengths of Cohort Studies
1. Can examine multiple effects of a single exposure
2. You are sure that exposure preceded outcome (temporal relation)
3. Can calculate relative risk and attributable risk.
4. Dose response relationship can be done.
5. Less liable for bias than case-control study.
6. Appropriate for studying rare exposure

FPHTM/EPI/RM/1-T
Limitations of Cohort Studies
1. Involve a large sample
2. Take a long time
3. Expensive.
4. They are not suitable for rare diseases
5. loss of persons during follow up (Attrition)

FPHTM/EPI/RM/1-T
Randomized Clinical Trial (RCT)

FPHTM/EPI/RM/1-T
 Clinical trial

• An intervention study that aims to evaluate

safety and efficacy of a therapeutic or a
preventive agent administered to the
individuals involved in the trial

FPHTM/EPI/RM/1-T
 Steps of conducting Randomized
Clinical trials (RCTs)
1. Identify target population: it is the population to which
results of the trial could be generalized
2. Identify study population: it is the population from which
participants will be recruited
3. List inclusion and exclusion criteria of the study
population
4. Get the "informed consent" from the study participants
5. Random allocation of participants to the experiment and
control groups.
6. Follow up of treatment response in experiment and
control groups
7. Compare treatment response in experiment and control
groups using statistical methods.
FPHTM/EPI/RM/1-T
 Phases of Clinical Trial
of a New drug / Vaccine

FPHTM/EPI/RM/1-T
 Phase I
– This phase aims to test effects of the drug on body
functions in volunteers

– This phase is of short duration, usually one or two

months

– It is carried out on a small number of volunteers

FPHTM/EPI/RM/1-T
 Phase II
– It is also carried out on a small group of
volunteers.
– It lasts longer than phase I trials.
– The phase aims to:
1. To assess the effectiveness of the drug.
2. To determine the appropriate dosage, and
possible side effects.

FPHTM/EPI/RM/1-T
Phase III (randomized clinical trial)
– It is called randomized clinical trial
– It involves individuals, who should provide a
consent to participate in the trial..
– this phase aims to assess the efficacy and safety
of the drug.
– Results from phase III trials are used to decide
whether or not the new drug should be licensed
for use in the market.

FPHTM/EPI/RM/1-T
 Phase IV (postmarking surveillance)

• It is done after approval of the Drug use in the

market (i.e., post-marketing studies)
• Require long-term follow-up of subjects
• It aims to
1. Investigate long-term side effects of the drug
2. Re-assess effectiveness, safety and acceptability of
the drug

FPHTM/EPI/RM/1-T
 Blinding
• Types
– Single – blinding (i.e., patient does not know whether he is
taking the new drug or the old drug )
– Double – blinding (both patient and investigator do not
know the new drug from the old drug )
– Triple – blinding (patient, investigator, analyst do not know
the new drug from the old drug )

• Can be impractical
– compare surgical vs. non-surgical treatment

FPHTM/EPI/RM/1-T
 Advantages of Clinical Trials
• Permits manipulation and measurement of
treatment

• Provides the best mean of avoiding bias and

confounding

• It is the best study design to prove cause

effect relationship

FPHTM/EPI/RM/1-T
 Disadvantages of Clinical Trials

• Time consuming and expensive.

• Ethical concerns and problems
• Difficult to implement
• Non-compliance of the participants.
• Drop out of participants during follow-up

FPHTM/EPI/RM/1-T
Human Research Ethics

FPHTM/EPI/RM/1-T
 Fundamental Principles of
Human Research Ethics
• Human research ethics rest on 3 basic
principles that are considered the
foundation of all regulations and guideline
governing research ethics. These
principles are consider universal

1) Respect for persons

2) Beneficence
3) Justice

FPHTM/EPI/RM/1-T
 Respect for persons
• Respect for persons recognizes the capacity and right
of all individuals to make their own choices and
decisions. It refers to the respect of:
1- Autonomy, self-determination
2- Protection of vulnerable groups
—women, children, prisoners, mentally ill
—those with limited education
—the poor
—those with difficult access to health services

3- Informed consent

FPHTM/EPI/RM/1-T
 Beneficence
- Beneficence refers to the principle of no harm.
- Beneficence makes researcher responsible for
the participant’s physical, mental and social
well-being through:
1- Reducing risks to a minimum
2- Protection of well-being of participant is
the primary responsibility of researchers.

FPHTM/EPI/RM/1-T
 Justice
• The researcher’s obligation is to:
1) Recruitment and selection of participants
should be done in an equitable manner
2) Distribute equally the risks and benefits of
participation in the research
3) Special protection for vulnerable groups

FPHTM/EPI/RM/1-T
 Informed consent
• Informed consent is a consent given by a
competent individual who:
1) Has received the necessary information
2) Has adequately understood the information
3) Has arrived at a decision without having
been subjected to coercion , undue
inducement

FPHTM/EPI/RM/1-T
 Essential Elements of Informed
Consent
• Research description
– This is a research study?
– Purpose and objectives of study
– Expected participant’s responsibilities
– Procedures involved
– Study duration
– Explanation of randomization or placebo

FPHTM/EPI/RM/1-T
 Essential Elements of Informed
Consent
⁻ Anticipated risks and benefits
⁻ Compensation for time and transport or injury
⁻ Confidentiality of participant data
⁻ Alternatives treatments and availability
⁻ Participants contacts for research related
questions
⁻ Voluntarism and right to discontinue without
penalty
FPHTM/EPI/RM/1-T
 Elements of Ethical Treatment
• Full Information
• Voluntary participation
• Consent
• Privacy

FPHTM/EPI/RM/1-T
FPHTM/EPI/RM/1-T