Study Design in Medical Research  Know the study in order to form a
standard protocol
STATISTICS
Derived from the Latin word status, meaning
“manner of standing” or “position”
Involves using statistical methods that
summarize data & use statistical procedure to
reach certain conclusion that can be applied to
patient care or public health planning
 A manner of study or position that
involves using statistical methods,
summarizes data, statistical procedures
to reach certain conclusions that can be
applied to patient care or public health
planning
 For us to understand statistics, we must
know the statistical methods
 The title of the study already reveals
what study design was used
BIOSTATISTICS & BIOMETRICS
Application of statistics in health-related fields.
 More on the clinical trials
 More on the health fields
 Allows the health professional to know
the exact time and place of an epidemic
10 Reasons to Learn Biostatistics
1. Evaluate the literature
 Identify true facts from fake news/data
 Know the credibility of the source of
data or the data itself
2. Applying study results to patient care
 Know the results of the study/treatment
of the patient in order to give quality
patient care
3. Interpret Vital Statistics
 Must know how to interpret data and
results in order to give information and
advice to patients about the correct
treatment and correct situations in the
society’s health.
4. Understand the epidemiologic problem
5. Interpret information about drugs and
epidemiology
 Be able to compute and interpret the
effectivity of a medicine based on
research
6. Using diagnostic procedures
7. Being informed
8. Guidelines
 Guidelines cannot be made without
being informed and without statistics
9. Protocols and articles
10. Research projects
EPIDEMIOLOGY
Refers to the study of health and illness in
human populations, or to the patterns of health
or disease and the factors that influence these
patterns
Based on the Greek word “upon” (epi) and
“people” (demos)
Used to understand the cause of the disease,
determine public health policy & plan
treatment.
 The application of the population in a
specific field; is based on the
information of decision making
 Refers to the study of health and
wellness in human population or
patterns of health or diseases, factors
that influence this patterns
 Asks the reason why the event occurred
ex. Outbreak of dengue
 Asks the causes, factors
 Used to understand the cause of the
disease, determine public health
policy, and plan treatment
 Knowing epidemiology could help
patients treat their diseases
Study Design in Medical Research
Classification of study design
Observational studies
 One or more group are observed
 Characteristics about the patients are
analyzed
 Focuses more on data
 Ginkikita la, no intervention
Experimental studies
 Involve an intervention
 An investigator-controlled maneuver
 Execution of procedures
Observational studies
A. Descriptive or case series
B. Case control studies (retrospective)
1. Causes & incidence of disease
2. Identification of risk factors
C. Cross sectional studies, surveys Case control study (Retrospective)
(prevalence)

1. Disease description
2. Diagnosis and staging
3. Disease Process and mechanism
D. Cohort studies (prospective)

1. Causes and incidence of disease

2. Natural history, prognosis
3. Identification of risk factors
E. Historical cohort studies
Experimental studies
 Begins with the absence or presence of an
A. Controlled trials outcome and then look backward in time to
try to detect possible causes or risk factors
1. Parallel concurrent controls that may have been suggested in case-
a. Randomized series report.
b. Non randomized  CASES: individuals selected on the basis of
disease or outcome
2. Sequential controls  CONTROLS: individuals without disease or
a. Self-controlled outcome
b. crossover  Backward looking
 Knowing the factors of the disease
3. External controls (including historical)  Looking if exposed/not exposed
B. Studies with no controls  Answers: What happened?
 Background check
Meta Analysis  Starts from the outcome and goes back to
I. OBSERVATIONAL STUDIES the origin (risk)
End Start
Case series study
Risk Outcome
 Group (or series) of patients (or cases) are
described in a published report
 Involve patient seen in a relatively amount
of time Cross-sectional study
 Do not include control subjects, person
who do not have the disease or condition
being described,
 Evaluate two methods
 Based on results; has no control
Example:

40 patients who had been refereed evaluation

of stroke, transient ischemic attack or carotid
bruit

What to compare 2 methods to w/c better

predict peak systolic velocity
Relationship between the both is strong

 Cross sectional study/surveys.

Epidemiological study & prevalence study
 Analyze data collected in a group of
subjects at one time rather than over a
period of time
 Answers: What is happening?
 Goes to the area and asks about the outcome
 Short period of time
 Diagnosing and staging a disease
 Evaluating different methods of doing the Outcome Assessment
same thing
To determine if the right patient care is given
 Establishing norms
to the patients.
 Surveys/ interviews
 Functional status
 Quality of life
 Patient satisfaction
 Cost effective & cost benefit analysis
Historic Cohort Study

possible if the record on follow-up are

complete and details, & ascertain the current
status of the patient.

Historical information is used.

Cohort Study (Prospective study)
 Group of people who have something in
common and who remain part of a group
over an extended period of time
 Subjects are selected by some defining
characteristics suspected of being a
precursor or risk factor for a disease or
health effect.
 Often examine what happens to the disease
overtime
 Answers: What will happen?
 You choose the subject
 Will determine if the risk factors are
exposed
 Long period of time
Start End
Risk Outcome
Cost effectiveness: evaluate economic
outcome:
Eg. Cost & benefit housing policy
Compared children w/ lead poisoning
increase. 46dollars saved per building if these
structures were brought to compliance.
Gives policy makers & health or providers
critical data to make informed judgments
about interventions.
COHORT VS CASE CONTROL
Henderson & colleagues (1997) study:
Study to look at the risk factor for depression in
elderly.
After an interview to collect information on
potential risk factors, they interviewed the
subject 3-6years later.
Case control:
Takes the outcome as the starting point
COHORT STUDY:

Starts with a factor or exposure & looks at the

consequences
II. EXPERIMENTAL STUDIES (Clinical
Trials)

Easier to identify compared to observational

studies

Researchers select the subject at the onset of
the study
Then determine whether they have the risk
factor or have been exposed
All subjects are followed over a period of time
to observe the effect and the exposure
 Involves humans
 There is intervention
 Determines if the intervention has a
difference
Clinical trials
 Experimental studies in medicine that
involve humans.
 Purpose is to draw conclusions about a
particular procedure or treatment.
Controlled trials (Placebo)
 Studies in which the experimental drug or
procedure is compared with another drug
or procedure
 Studies with controls detect whether the
difference is due to experimental treatment
or to some other factor.
 Giving a known effect of drug (controlled)
to a patient 1 and giving a different drug
(uncontrolled) to patient 2 with the same
effect as to the drug given to patient 1.
Uncontrolled trials (Experiment)
 Studies in which investigators’ experience
with the experimental drug or procedure is
described, but treatment is not compared
with another treatment.
 Based on the outcome only
 Effective but unsure how
Controlled studies are viewed as having far
greater validity in medicine than
uncontrolled studies
DETERMINE WHETHER THE
INTERVENTION (TREATMENT) makes a
difference.
1. Trials with independent concurrent
controls
CONCURRENT CONTROL:
Experimental group and control group plan
interventions for the same period in the same
study
Execute the same procedure for the same
study
To ensure that subjects are treated similarly
DOUBLE BLIND TRIALS:
Neither subject or investigators know whether
the subject is the treatment or the control
group
To reduce the chances that subject or
investigators see what they expecting to see
Blind trial:
When only the subject is unaware.
A. Randomized controlled trials
 Provides the strongest evidence for
concluding causation
 Provides best insurance that the result was
due to intervention.
 Randomly chooses patients with control
Example:
22k healthy people receive aspirin and
followed over an average of 60 months
How to assign patients to the experimental
condition & to the other control condition.
E.g. whether aspirin in low doses reduces the
mortality rare from CVD
RESULT LESS PHYSICIAN EXPERIENCE mi W/
aspirin
B. Non randomized trials
 Clinical trials or comparative studies, with
no mention of randomization
 They do nothing to prevent bias in patient
assignment
 Patients are treated without at any plan
 Kumuha la hin patient without plan
 Not controlled; will not know the bias
happening
Example:
Stronger patient receives the more aggressive
treatment & the higher risk patients who are
treated conservatively
Patient are treated within big blocks of time.
2. Sequential controls
TRIALS WITH SELF CONTROL
Same group of subjects for both experimental
and control group
HAWTHORNE EFFECT
People change their behavior & sometimes
improve simply because they receive special
attention by being in a study & not because of
the study intervention.
E.g. Patient underwent cholecystectomy,
follow-up after 3mos detect change abdominal
pain, dyspepsia etc.
SIMILAR TO COHORT EXCEPT FOR THE
INTERVENTION OR TREATMENT IS
INVOLVED.
Crossover study
1 GROUP EXPERIMENT|1 GROUP PLACEBO
AFTER A TIME, PLACEBO & EXP WITHDRAW:
WASHOUT PERIOD/ NO TREATMENT
EXCHANGE TREATMENT
3. Trials with external controls
Result of another investigator’s research is
used as comparison.
Historical controls
 Used to study disease for which cures do
not yet exist
 Controls are patients the investigator has
previously treated in another manner
Researcher should evaluate whether other
factors may have changed since the time the
historical control were treated
Any differences may be due to these other
factors & not to the treatment.
UNCONTROLLED STUDIES
 More likely to be used when the
comparison involves a procedure than
when it involves a drug.
 Investigators assume that the procedure
used and described is the best one.
E.g.
Trial of radiotherapy
 Determining the length of time, a patient
had no recurrence of the tumor as well as
how long the patient survived
 Result of radiotherapy for prostate
carcinoma in which patient are followed at
least 12 and as long as 70mos.
Not all Studies involving interventions have
controls, and by strict definition they are not
really experiments or trials
Result: long term survival in patients who had
different tumor classification (scores that
measure the severity of the tumor)
1 particular treatment is recommended and
then discounted after a controlled clinical trial
is undertaken
Research. New meds try to other patients/
gatas-gatas as effective.
III. META-ANALYSIS
 Does not fit both on observational or
experimental.
 Collect previous studies and makes a
conclusion out of it
 Uses published information from other
studies and combines the results so as to
permit an overall conclusion.
 Similar to review articles
 Includes a quantitative assessment and
summary of the findings.
Eg study whether catheters impregnated with
antiseptic were preventing in catheter related
bloodstream infection, compared untreated
catheters
Method: 12 randomized trials address this
question and combines the results statistical
manner to reach an overall conclusion & their
effectiveness.
CLINICAL TRIALS
ADVANTAGE
 Objective is to establish the efficacy of
a treatment or a procedure
Historical trials /controls maybe useful in
preliminary studies are needed when
researcher are dealing with late treatment for
an intractable disease.
DISADVANTAGE
 Great expense
 Long duration
 Randomized trial comparing treatment
for carcinoma … follow the subject for a
long period of time
COHORT STUDIES
Advantage
 Follow a cohort of patients forward
through time
 Possess correct time sequence
 Provide strong evidence for possible
causes & effects
DISADVANTAGE:
 Events occurring in the intervening
period may have been affected.
 Do not involve intervention (causation
cannot be proved)
 Extended time (costly)
Group of people something in common and
who remain part of a group over an extended
time.
(causation cannot be proved)
Length of time depends on the disease being
studied
Observational only
CASE CONTROL STUDY
ADVANTAGE:

 Quickest & least expensive

 Appropriate for rare disease
DISADVANTAGE

 Have largest possible biases & errors

 Depend completely on high quality

existing records
 Selection of an appropriate control
group
Some statisticians have recommended the use
of 2 control groups:

1 control group similar in some ways to the

cases (same admission same time) and another
healthy subjects
CROSS SECTIONAL
ADVANTAGE

 Quick to complete
 Inexpensive

 Best for determining status of the

condition (prevalence)
DISADVANTAGE

 Misleading information due to snapshot

in time
 Survey decline
80mm Hg STAND MEAN BP

HOWEVER, PATIENT FOLLOWED SEVERAL

YEARS INCREASED DIASTOLIC BP IN ELDERS
CASE SERIES STUDY
Advantage

 Easy to write

 Observations may be investigators

designing a study to evaluate
explanations of the observation
Disadvantage

 Prone to biases
VIEW IT AS HYPOTHESIS GENERATION BUT
NOT AS CONCLUSIVE