Research Designs
Presented by:
Dr.Amira Yahia
Study designs
Non intervention stud.
Intervention stud.
Exploratory stud.
Case stud.
Randomized control Trials
Descriptive stud.
Surveys
Analytical stud.
Case con. Stud.
Cohort study
Quazai experimental stud.
Exploratory study
Exploratory study
It is a small - scale study of relatively short
duration, which is carried out when little is
known about a situation or a problem
Exploratory studies problem is not as may be
descriptive and/ analytical. If the problem is not
well defined , it is always advisable to do an
exploratory study.
Exploratory study
Example:
One community with high and another with low
participation in health activities, to identify the
factors that contribute to community
participation.
DESCRIPTIVE STUDY DESIGNS:
The systematic collection and presentation of
data to give a clear picture of a specific disease
or a health related events
Case studies
A case study describes in depth the characteristics of one
or limited number of cases in its natural environment
a/ Case studies
Case studies
-A case may be a patient, a health centre, a village etc…
- Can provide useful insight into the problem e.g. a new
disease
- Common in clinical medicine, social sciences,
management and administration etc..
- Associated with qualitative data and presentation is in a
narrative way
Case studies
Features:
- Should be well planned and data will be collected
thorough predetermined questions
- Should be flexible to deal with unexpected situations
Case studies
Advantage:
- It permits a holistic approach to the problem under
investigation
Disadvantage:
-Not representative
b/ SURVEYS
SURVEYS
Use :
1. To collect information on demographic characteristics.
Age, sex, education etc…
2. To study characteristics on health related variables.
E.g. MMR, incidence rate, etc….
3.To study attitudes, opinions and beliefs
SURVEYS
Surveys answer the following questions:
- WHEN IS THE DISEASE OCCURRING?
(TIME DISTRIBUTION)
- WHERE IS THE DISEASE OCCURRING?
(PLACE DISTRIBUTION)
- WHO IS AFFECTED? (PERSON DISTRIBUTION)
PROCEDURE
1- Define the problem under study
2- DEFINE THE POPULATION UNDER THE STUDY.
3- DESCRIBE THE DISEASE BY TIME, PERSON AND PLACE.
4- MEASUREMENT OF THE DISEASE.
5- COMPARING WITH KNOWN INDICES.
6- FORMULATION OF AN AETIOLOGICAL HYPOTHESIS.
Cross-sectional designs:
Involve the collection of data at one point in
time: the phenomena under study are captured
during one period of data collection.
Cross-sectional studies are appropriate for
describing the status of phenomena or for
describing relationships among phenomena at
a fixed point in time.
CROSS SECTIONAL STUDY
• PREVALENCE RATE STUDY
• THE RELATIONSHIP BETWEEN THE DISEASE & OTHER
VARIABLES OF INTEREST AS THEY EXIST AT ONE
PARTICULAR POINT OF TIME
CROSS SECTIONAL STUDY OF
HYPERTENSION
1.
2.
3.
4.
5.
MEASURE THE B.P. AMONG THE STUDY POPULATION
(PREVALENCE.)
COLLECT DATA ON AGE, SEX, SOCIAL CLASS ,
OCCUPATION etc...
DETERMINE HOW HYPERTENSION IS RELATED TO
THESE VARIABLES
Compare with findings of others
Draw hypothesis
• Example of a cross-sectional study:
Mindell and Jacobson (2000) assessed sleep patterns and
the prevalence of sleep disorders during pregnancy. With a
cross-sectional design, they compared women who were
at four points in pregnancy: 8 to 12 weeks, 18 to 22 weeks;
25 to 28 weeks; and 35 to 38 weeks.
The main advantage of cross-sectional
designs in such situations is that they are practical:
they are easy to do and are relatively economical
•
LONGITUDINAL STUDY DESIGNS
• OBSERVATIONS ARE REPEATED IN THE SAME POPULATION
OVER A PROLONGED PERIOD OF TIME
THESE ARE USEFUL IN:
• STUDY THE NATURAL HISTORY OF THE DISEASE.
• IDENTIFY THE RISK FACTORS.
• DETERMINE THE INCIDENCE RATE.
Analytical studies
1. Case Control study
Case Control study
In Case Control Study, the investigator compare one group
among which a problem present, with another group (control)
where the problem is absent
Common features
1.
2.
3.
Both exposure and outcome have occurred before the
start of the study
Study proceeds back from effect to cause
Presence of a control group
Basic steps
1.
2.
3.
4.
5.
Selection of cases
Selection of controls
Matching
Obtaining data on exposure
Analysis and interpretation of results
Selection of cases
It involves two specifications:
1. Diagnostic criteria:
Specific diagnostic criteria of the disease, its staging are to
be stated clearly.
2. Eligibility criteria:
Criteria to determine who is allowed to be included in the
study. E.g. Only newly diagnosed cases rather than old
ones
Sources of cases and controls
Cases:
- Patients
- General population
Controls:
-
Patients
Relatives
Neighbor hood
General population
Matching
Is the process by which controls are to be similar to the
study group
Types:
• Individual matching
• Group matching
Analysis:
1. The exposure rate among both the cases and the
controls.
2. Odds ratio: This ratio can be used to estimate the
relative risk as the incidence rate can not be determined
in case-control study design.
•
Thus the case control study is always suggestive of the
etiological association between two variables
Strengths and limitations of the
case-control study design
Strengths:
• Is relatively quick and inexpensive
• Is optimal for the evaluation of rare diseases.
• Can examine multiple etiologic factors for a single
disease
Limitations:
• Is inefficient for the evaluation of rare exposures
• Cannot directly compute incidence rates of disease in
exposed and non- exposed individuals.
• Is particularly prone to bias compared with other
analytic designs, in particular, selection and recall bias.
Analytical studies
2. Cohort study
Cohort study
• A longitudinal a study in which a group of individuals are
followed up for some time.
• A cohort is a group of persons who share common
characteristics or experience within a defined time.
Features
The cohort is identified before the appearance
of the investigated disease
• The study groups are observed over a period of
time
• The study proceeds from cause to effect
Note: The incidence rate can be measured
•
Indications of cohort study:
1.
2.
3.
When there is strong suspicion of association between
exposure and disease
When exposure is rare but the incidence is high among the
exposed
When the attrition of study population can be minimized
Types of Cohort studies
-Historical Cohort study:
Starts from a point in the past to now
-Prospective Cohort study:
Starts from now and study goes on to the future
-Retrospective Prospective Cohort study:
Starts from the past, passing through now and proceeds to
the future
Example of a prospective non experimental
study:
Brook, Sherman, Malen, and Kollef (2000) conducted
a prospective cohort study to examine clinical and cost
outcomes of early versus late tracheostomy in patients
who require prolonged mechanical ventilation. Early
tracheostomy was found to be associated with shorter
lengths of hospital stay and lower hospital costs.
Example of a retrospective study:
• Heitkemper, Jarrett, Taylor, Walker, Landenburger, and
Bond (2001) used a retrospective design in their study of
factors contributing to the onset of irritable bowel
syndrome (IBS). They compared samples of women with
and without IBS in terms of their history of physical
abuse, and found that abusive experiences were more
prevalent among women with IBS
Elements of a cohort study
1- Selection of a study group:
From:- a. a general population or
b. Special group (Such as doctors,Nurses, drivers etc.. or
exposure group)
2- Obtaining data on exposure:
By review of records , medical examination, interview etc…
3- Selection of a comparison group;
-Internal comparisons: According to exposure level
-External comparisons: From out side the study group
Elements of a cohort study- cont.
4- Follow up:
By: a. Periodic medical examination
b. Review of records
c. Review of death certificates
d. Mailed questionnaires
e. Telephone calls etc…
5- Analysis:
a. Incidence rate
b. Relative risk
Analysis
Relative risk:
This is the ratio that measures the strength of association
between suspected cause and effect.
Attributable risk
This is the difference in incidence of disease or death
between exposed & non-exposed group. It is expressed
as a percentage.
It measures the impact that removal of a certain factor
will have on the incidence of the disease.
SMOKING & CA. LUNG
Smoking
status
Developed
Ca lung
70
(a)
Did not
develop Ca
lung
6930
(b)
Yes
No
Total
7000
(a+b)
3
©
2997
(d)
3000
(c+d)
Analysis
• Incidence rate of Ca lung among smokers=
a/a+b=70/7000= 10/1000
• Incidence rate of Ca lung among non-smokers=
c/c+d= 3/3000= 1/1000
Analysis
• Relative risk= Incidence of disease among exposed:
incidence among non-exposed
10/1000
1/1000
• Attributable risk= Incidence rate among exposed-
incidence rate among non-exposed/incidence rate
among exposed= 10-1x100=90%.
Strengths and limitations of the
cohort study design
Strengths:
• Is of particular value when the exposure is rare
• Can examine multiple effects of a single exposure
• Allows direct measurement of incidence of disease in
the exposed and non-exposed groups.
Limitations:
• Is inefficient for the evaluation of rare diseases
• Expensive and time consuming
• Validity of the results can be seriously affected by losses
to follow-up
EXPERIMENTAL STUDY
DESIGNS
EXPERIMENTAL STUDY DESIGNS
In experimental studies the researcher manipulates a
situation and measures its effect after that
Type:
1- Randomized control trials
2- Non-randomized trials
1- Randomized control trials
RANDOMIZED CONTROLLED TRIALS
Those trials are used for assessment of methods of
treatment and prevention.
They include:
- intervention
- control groups and
- randomization
Features of Randomized control trials
1- Manipulation:
The researcher do some intervention (Example provision of new
drug) to one of the study groups.
2- Control:
The researcher introduces one or more control groups to
compare with the experimental group
3- Randomization:
Each subject have an equal chance of being allocated to either
of the two groups( study and comparison groups)
steps:1. Drawing
a protocol.
2. Selecting comparison & experimental groups
3. Randomization.
4. Manipulation (intervention).
5. Follow up.
6. Assessment of the outcome.
THE PROTOCL
One of the essential features of the randomized trials.
The protocol specifies:
1. The objectives.
2. The questions to be answered.
3. The selection criteria.
4. The sample size.
5. The procedures of allocation of the subjects into
experimental and control groups.
THE PROTOCL- cont. …
6. The treatment applied: How, what dose etc...
7. The details of the scientific techniques and
investigations.
Note:
Once the protocol has been evolved, it should be
strictly adhered to through out the study.
RANDOMIZATION
• Randomization is the statistical procedure by
which the participants are allocated into groups
usually called study & control groups to receive
or not to receive intervention or therapeutic
procedure .
• Randomization aims at making the groups
comparable and eliminate bias.
RANDOMIZATION
Randomization ensures that the investigator has no
control over the allocation of the participants to
either the study or control group, thus eliminating
the selection bias.
• Every individual has an equal chance of being
allocated into either group.
• Randomization is best done by using statistical
random table.
•
MANIPULATION
• Manipulation or intervention is usually done by
application or withdrawal of the suspected factor
e.g. drugs, vaccine or dietary factor.
• This manipulation creates an independent
variable (drug, vaccine or new procedure) which
effect is then determined by the measurement of
the final outcome which constitutes the
dependent variable e.g. incidence of disease,
recovery.
FOLLOW UP
• This includes examination of the study & control groups subjects
at defined intervals of time in standard manner under the same
conditions in the same time frame till the final assessment.
• The main difficulties encountered in the follow up process
include:
Attrition from:- death, migration, displacement and loss of
interest etc
ASSESSMENT
The final assessment of the trial is carried in terms of:
• Positive results:
These include the benefits of the experimental study
such as reduced incidence of the disease or severity of
the disease, cost of health services or other appropriate
outcome.
• Negative results:
These include the severity & frequency of side-effects
and complications.
The incidence of positive/negative results is compared in
both groups and the differences are tested statistically.
2- Non-randomized trials
Quasi-experimental
Quasi-experimental:
A study in which subjects are not randomly
assigned to treatment conditions, but the
researcher does manipulate the independent
variable and exercises certain controls to enhance
the internal validates of the results.
Types of quasi-experimental
a. Non-equivalent groups or static groups design
Two receive different treatments, but are not randomly
assigned or matched to conditions. Eliminates history
effects but not subject effects.
b. Time-series design
There is one group of research participants with several
baseline measures, “a treatment and at least one more
measurement eliminates subject effects but not history
effects.
a. QUASI-EXPERIMENTAL STUDY DESIGNS
• At least one of the characteristics of the true
experiment is missing (RANDOMIZATION OR
CONTROL GROUP)
• QUASI-EXPERIMENTAL STUDY DESIGNS ALWAYS
INCLUDES MANIPULATION ( INTERVENTION)
2. Non-randomized trials
b. Before and after study:
- Includes intervention
- No control group
- No randomization
- Observation before and after the intervention
- Test ( analysis)
Bias in experimental studies
• Bias is the systematic difference between observed
results and the actual results.
• Sources of bias:
1. Participant’s bias:
The participants report subjectively that they feel
better or improved if they knew that they were
receiving new treatment.
BIAS – cont. ..
2.Observer’s bias:
The influence of the investigator measuring the
outcome of the trial if he knew beforehand the
particular procedure to which the patient has
been subjected.
How to reduce the sources of bias?
1. Randomization.
2. Blinding.
Blinding is carried in three ways:
1. Single blind trial: The trial is so planned that the
participant is not aware whether he belongs to the
study or control group.
2. Double blind trial: The trial is so planned that
neither the doctor nor the participant is aware of
the group allocation and treatment received.
How to reduce the sources of bias? Cont…
3. Triple-blind trial:
The trial is so planned that the participant, the
investigator& the person analyzing the data are
all blind.
This is the ideal but double-blinding is the most
commonly used.
DISADVANTAGES OF EXPERIMENTAL
STUDIES
• They are costly.
• Ethical problems.
• Feasibility is difficult.
Qualitative design:
Qualitative research: The investigation of
phenomena, typically in an in depth and holistic
fashion, through the collection of rich narrative
materials using a flexible research design.
Qualitative research types
1-Phenomenology: a form of qualitative research
in which the researcher attempts to understand
how one or more individuals experience a
phenomenon.
For example: you might interview 20 widow and
ask them to describe their experiences of deaths
of their husbands
2-Ethnography: is the form of qualitative research
that focuses on describing the culture of a group
of people.
• For example, you might decide to go and live in
a Mohawk communities and study the culture
and their educational practices
• 3-Case study research: is a form of qualitative
research that is focused on providing a detailed
account of one or more cases.
• For example, you might study a classroom that
was given a new curriculum for technology use.
4-Grounded theory: is a qualitative approach to
generating and developing a theory form data
that the researcher collects.
• For example, you might collect data from
parents who have pulled their children out of
public schools and develop a theory to explain
how and why this phenomenon occurs,
ultimately developing a theory of school pull-out
• 5-Historical research: research about events that
occurred in the past.
• For example, you might study the use of
corporeal punishment in schools in the 19th
century.
Mixed research methods:
• It is a general type of research “it’s one of the
three paradigms” in which quantitative and
qualitative methods are mixed in one overall
study.
• i.e. it is research in which the researcher uses
the qualitative research paradigm for one phase
of a research study and the quantitative research
paradigm for another phase of the study.
• For example, a researcher might conduct an
experiment (quantitative) and after the
experiment conduct an interview study with the
participants to see how they viewed the
experiment and to see if they agreed with the
results.