1. Introduction to Study Designs
Dr. Yaser Faden
Asst. Professor, Dept. of OB/GYN
Consultant, Maternal-Fetal Medicine
Director, RTP in OB/GYN
2. What is the basis of a good
research study?
AN APPROPRIATE STUDY
DESIGN
3. Learning Objectives
By the end of this session, you will be able to:
Distinguish between observational and experimental studies
Describe the key characteristics of experimental, cohort,
case-control, cross-sectional, and ecologic studies
List the advantages and disadvantages of each type of study
design
Identify the design of a particular study by reading an
abstract
Discuss the factors that determine when a particular design
is indicated
4. Diagnostic
Puzzler
You are a practicing physician in the 1970’s.
Your patient is a very ill 24 year old woman
who is hospitalized for fever, low blood
pressure, and a rash, including peeling of the
hands. Your review of symptoms is positive
only for menstruation. You treat her in the
intensive care unit. You feel fortunate that she
survived, but are uneasy because you never
really knew what was wrong with her.
5. Categories ofCategories of
Epidemiologic StudiesEpidemiologic Studies
Epidemiologic studies
Observational Studies Experimental Studies
Descriptive Studies Analytic Studies
Case report
Case Series
Cross sectional studies
Case control
Cohort
RCTs
Ecological Studies
Retrospective Cohort Prospective Cohort
6. Categories of Epidemiologic
Studies
Observational Studies
Investigators collect, record and analyze
data on subjects as they naturally divide
themselves by potentially significant
variables ( i.e. case-control, cohort)
Experimental Studies
Involve some sort of control by the
investigators (i.e. RCT’s)
7. Epidemiologic Study Designs
Type of observational studies based on:
– Type of sampling from population
Based on exposure and/or disease
– Temporal sequence of observation
One time point, forward, backwards
8. Exposure and Outcome
Exposure
– Refers to the potential risk factor
• Can be exposure such as tobacco smoke
• Can be behavior (e.g.. sedentary lifestyle)
• Can be attribute ( e.g.. SES)
Outcome
– Is the disease or other health related problem
which is being studied
9. Descriptive Studies
Are a class of epidemiologic studies which focus
on characterizing morbidity or mortality of
populations by person, place or time variable and
have no a priori hypotheses.
Examples Include
– Case Report
– Case Series
– Some cross-sectional studies
– Some ecologic studies
10. Case Reports
Detailed presentation of a single case
Generally report a new or unique finding
– Previous undescribed disease
– Unexpected link between diseases
– Unexpected new therapeutic effect
– Adverse events
11. Case series
Experience of a group of patients with a
similar diagnosis
Cases may be identified from a single or
multiple sources
Generally report on a new/unique condition
May be the only realistic design for rare
disorders
12. Case reports and series
Case report: describes an observation in a
single patient.
– “I had a patient with a cold who drank lots of
orange juice and got better. Therefore, orange
juice may cure colds.”
Case series: same thing as a case report,
only with more people in it.
– “I had 10 patients with a cold who drank orange
juice….”
13. Case Reports / Case Series
Pros
– Useful for hypothesis generation
– Informative for very rare diseases with few
established risk factors
– Easy to understand
– Can be written up in short period of time
Cons
– Cannot study cause and effect relationship
– Cannot assess disease frequency
14.
15.
16. Cross-Sectional Studies
Assess both exposure and outcome at
the same time “snapshot”
These are generally surveys or
interviews
Used to determine the prevalence of a
condition (prevalence study)
Used to identify possible causative
factors in disease
18. Strengths
One stop, one time (snapshot)
Relatively easy, quick, and
inexpensive
Estimates disease prevalence
Useful for planning services
Good design for hypothesis generation
Rely on questionnaires and no follow-
ups are required
Cross-Sectional StudiesCross-Sectional Studies
19. Weaknesses
•Only representative of participants
•Impractical if disease is rare
•May not be possible to establish temporal
relationship
•Not a useful study for establishing causal
relationships
Cross-Sectional StudiesCross-Sectional Studies
20.
21. Analytic Studies
Unlike descriptive studies, analytic studies are
designed to test hypotheses about an exposure of
interest and a particular outcome
Exposure Outcome
?
24. Back to our diagnostic puzzler
You make an inquiry to the CDC about patients
with these types of symptoms
Yes, they have collected a few other cases like
this. All were menstruating women.
You have a keen interest in this new syndrome
and work with the CDC and other doctors to
publish a case series.
You notice that one common characteristic of all
of the affected women is tampon use. Is this just
chance, or could it be related?
25. Case-control studies
Attempt to make inference from existing
observations (retrospective)
Compares patients with outcome/disease
with those without and attempts to
identify factors that influenced that
outcome (or caused that disease)
Important concept: start with the result
(disease) and work backwards for the
cause
27. Back to our diagnostic puzzler
How would you design a case-control
study to test the theory that
menstruation (or perhaps tampon use)
is somehow connected with this new
illness, which some people have started
to call “toxic shock syndrome”?
29. Strengths of case control studies
Rare diseases
Several exposures
Rapidity
Low cost
Small sample size
Available data
No ethical problem
30. Limitations of
Case-Control Studies
Cannot compute directly relative risk
Not suitable for rare exposure
Temporal relationship exposure-disease difficult
to establish
Biases +++
– control selection
– recall biases when collecting data
Loss of precision due to sampling
31. Cohort studies
Studies whether exposure to a “risk
factor” is associated with a subsequent
“outcome”
Select two populations who seem the same
except for the hypothesized risk factor
Follow them ahead in time and see how
many have the outcome or disease
Important concept: Start with the risk,
then look for the outcome
35. Back to our diagnostic puzzler
How would you design a prospective
cohort study to test the theory that
tampon use by menstruating women
is somehow connected with “toxic
shock syndrome”?
37. Cohort Studies
Prospective cohort studies start with the
exposure, then follow patients over time
Retrospective (or historical) cohort studies
start with an exposure that happened some
time ago, then look at the outcomes today
Important point: Even though this is
retrospective, it starts with the exposure or
risk and then measures the outcome
38. Strengths of cohort studies
Can directly measure
– incidence in exposed and unexposed groups
– true relative risk
Well suited for rare exposure
Temporal relationship exposure-disease is clear
Less subject to selection biases
39. Weaknesses of cohort studies
Large sample size
Lost to follow
Exposure can change
Multiple exposure = difficult
Ethical considerations
Cost
Time consuming
41. Descriptive studiesDescriptive studies
Examine patterns of disease
Analytical studiesAnalytical studies
Studies of suspected causes of diseases
Experimental studiesExperimental studies
Compare treatment modalities
Epidemiologic Study Designs
42. Randomized Control Trial
(RCT)
Gold standard of all studies
Prospective
Two or more groups assigned by randomization
Baseline measurements on all groups
Give different treatments
Measure outcome
43. Types of Clinical Trials
Treatment trials test experimental treatments,
behavioral therapies, new combinations of drugs,
or new approaches to surgery or radiation
therapy
Prevention trials look for better ways to prevent
disease in people who have never had the disease
or to prevent a disease from returning
– These approaches may include medicines, vitamins,
vaccines, minerals, or lifestyle changes
44. Randomization
Assigned to groups by method similar to
“flipping a coin”
If randomization works, groups will be the
same/comparable
The larger the sample, the greater the likelihood
of equal groups
Results should show that the demographic
characteristics between groups are similar
If groups are similar, do not need to control for
extraneous variables
45. Randomization
Sometimes we cannot randomize people (e.g., cross-
contamination or “system” interventions)
Can randomize hospitals, or units instead
– For example, testing clinical reminder systems
Once randomized, always randomized
Subjects are treated as part of that group, even if they
die, are lost to follow up, or withdraw
46. Randomization
Randomization
Two kinds of randomization:
– Random sampling
• Every person in a population must have
an equal chance of getting into the sample
– Random assignment
• Each person in a sample must have an
equal chance of getting into the
experimental and control group
• That is, they are randomly placed in one
of the groups
47. Randomization
Researcher must actually go through some
randomization process
– For example, number each potential subject, and then pull
numbers from a box or use a random table to determine
assignment to a group
Randomization is a very strong and positive control
method
Randomization can always strengthen a study
48. Homogenous Sampling
Trying to make your sample as much alike is
helpful in studies
– because it can minimize the possibility extraneous
variables have affected the results
However, it also has limitations
– Because it makes generalization more difficult since the
study population is smaller and applies to fewer people
– It can also make it more difficult to get enough people
in the study
So, homogenous sampling increases internal
validity, but decreases external
49. Blinding
Un-blinded: Everyone knows treatment
Single Blinded: Researcher or patient does not know
treatment
Double Blinded: Neither researcher or patient knows
treatment
Why blinding?
– Many people believe they feel better if they are given
something
– This is the placebo effect
50. Double Blind Example
Patient:
– Patient agrees that he will be randomized to one of 4
smoking cessation treatments
– None of these 4 smoking cessation treatments are
known to be better than the other
Provider:
– Providers do not know that patients are assigned to
groups
– Hire different people to run each group and do not
tell them about the study
52. Phases of Clinical Trials
Phase I trials (a pilot study): Researchers test an
experimental drug or treatment in a small group of
people (5-60 subjects) for the first time to
– Evaluate its safety
– Determine a safe dosage range
– Identify side effects
Phase II trials (a larger pilot study): The experimental
study drug or treatment is given to a larger group of
people (100 subjects) to see if it is effective and to further
evaluate its safety
53. Phases of Clinical Trials
Phase III trials (RCT): The experimental study drug or
treatment is given to large groups of people (200-3,000
subjects) to
– Confirm its effectiveness
– Monitor side effects
– Compare it to commonly used treatments
– Collect information that will allow the experimental drug or
treatment to be used safely
Phase IV trials (implementation research):
– Post marketing studies
– Delineate additional information, including: the drug's risks,
benefits, and optimal use
54. Non-Randomized Comparison
Group
Next best thing to RCT
Used when we cannot randomize our
subjects
– For example, due to cross-contamination, or facility-
or community-level interventions
Make sure groups are as similar as
possible
56. RCT Advantages
– The “gold standard” of research designs.
They thus provide the most convincing
evidence of relationship between exposure and
effect.
– Example:
• trials of hormone replacement therapy in
menopausal women found no protection for
heart disease, contradicting findings of
prior observational studies
57. RCT Advantages
Best evidence study design
No inclusion bias (using blinding)
Controlling for possible confounders
Comparable Groups (using randomization)
59. Epidemiologic study designs
What type of study to choose depends on:
What is the research question/ objective
Time available for study
Resources available for the study
Common/rare disease
Type of outcome of interest
Quality of data from various sources
Often there are multiple approaches which will all work
Choosing an established design gives you a huge head start
in design, analysis and eliminating biases
62. Study Design
Examples
1. A study examines 200 women with
cervical cancer and 200 controls. They
determine that there is an increased
risk of cervical cancer with smoking
Groups by Disease
Case Control
63. Study Design
2. A study started in 1990 and followed 1000
consecutive women who smoked in
pregnancy and 2000 consecutive non
smoking pregnant women. The study was
completed five years after inception. They
determined that there is an increase in
stillbirth in smokers.
Groups determined by risk factors ie smokers
Retrospective Cohort
64. Categories ofCategories of
Epidemiologic StudiesEpidemiologic Studies
Epidemiologic studies
Observational Studies Experimental Studies
Descriptive Studies Analytic Studies
Case report
Case Series
Cross sectional studies
Case control
Cohort
RCTs
Ecological Studies
Retrospective Cohort Prospective Cohort
THANKTHANK
YOUYOU
Notes de l'éditeur
Not all prospective trials are placebo-controlled, however. A non-controlled trial might identify potential subjects, give them all a treatment, and then see how they do. Such open-label single arm trials cannot control for placebo effects or experimenter biases, and again results should be considered preliminary. Open or uncontrolled trials are not useless, however. The outcome of subjects in such trials can be compared to historical controls, and if a significant result is apparent (along with safety) can be used to justify a larger and more rigorous trial. Controlled trials have one or more comparison groups in the trial itself – different groups of subjects receive different treatments or no treatment. All subjects can be followed in same manner. Control groups allow the experimenter to make sure that all the subjects have the same disease or symptoms, that they receive known treatments, and many variables (such as other treatments they may be receiving, severity at inclusion, age, sex, race, etc.) can be accounted for.