The document discusses different types of experimental study designs used to evaluate medical treatments, including randomized controlled trials, non-controlled experimental studies, and controlled experimental studies using parallel, crossover, and time series (before-after) designs. It explains key aspects of each design such as use of control groups, randomization, blinding, advantages, and disadvantages. An example is provided for each type of design to illustrate its application.

1. Measurement System for Impact
of OBE through Examination
and Evaluation.
Dr Rajendra Talware

2. Experimental Study Design
 Best design for determining efficacy of treatment:
Randomized, controlled, double blinded,
experimental designed study.
 Experimental study can be controlled or non-
controlled.
 Randomized or non-randomized
 Double blinded, single blinded or no blinding
 A randomized controlled trial (study) may be
described as prospective, longitudinal, follow-up, and
experimental.

3. Randomized Controlled Trial
 Patients are assigned to one of two or more groups
that receive a treatment or intervention and are
followed over time for a measured outcome.
 Some patients are designated as the control group
which serves as the comparison group and which a
placebo or sugar pill or a comparison drug is given
 Both groups are compared on the outcomes and
treatment effectiveness is determined.

4. Randomized Controlled Trial
 Advantages
 Utilizes a control group (strongest evidence)
 Bias is minimized
 Extraneous factors being responsible for
outcomes is minimized.
 Most reliable technique for evaluating
treatments
 Most statistically powerful study design

5. Randomized Controlled Trial
 Disadvantages
 Expensive
 Patient enrollment can be difficult (time length)
 Drop out rate is higher because of long term
follow-up (required by FDA)
 Ethical problems in testing new therapies in
humans
 Time necessary to perform study can be long.
 Most complex to interpret.

6. Experimental Controlled Study
Example
 To determine whether Retin A cream can
reduce wrinkle formation, investigators
recruited 100 healthy subjects from 65-70 y.o.
and randomized them to receive Retin A or
placebo. They applied the cream 2x daily for
1 year. A dermatologist rated the degree of
facial wrinkles using a scale from 1-10 at the
beginning and end of the study for all
participants and compared the two groups.

7. Randomized Controlled Trial
(RCT) Summary
 The RCT design is suitable for most types of
pharmaceutical research.
 It is considered to be the most statistically
powerful study design.
 The major difference between the clinical trial
and other designs is the ability of the
investigator to actively intervene, rather than
simply observe.

8. Non-controlled experimental Study
 Does not utilized a control group.
 The drug treatment procedure to be studied is
administered to a single group of patients.
 Outcome measures are determined in this
group.

9. Non-controlled experimental Study
Example
 A study to determine the efficacy of Zofran in
prevention and treatment of Cisplatin induced nausea
and vomiting (N&V) was done. 50 patients receiving
Cisplatin for ovarian cancer were given Zofran at the
start of Cisplatin therapy. The number of episodes of
nausea and vomiting were recorded. Only 10% rated
the N&V as severe. Investigators concluded that
Zofran is effective in minimizing Cisplatin induced
N&V.

10. Design Types of Controlled
Experimental Studies
 Parallel design
 Crossover design
 Time series design (Before and After)

11. Differences between types of
controlled experimental designs
 Different outcomes can be measured more
accurately with different designs.
 Different statistical tests are applied to
different study designs.
 Different bias risks are inherent in each
design

12. Parallel Design
 A parallel design includes independent study groups and each
group receives a different treatment regimen or intervention
 Randomized Controlled Trials are often parallel design
 Parallel design is more useful for studying conditions which are
prone to change over time (pain, acute exacerbations of a
disease, remissions)
 Example:In a study to evaluate the efficacy of beta blockers for
hypertension, 24 patients are randomized into two groups of 12
patients. One group is then treated with a beta blocker and the
other treated with placebo.

13. Crossover Design
 A crossover design may have just one study
group that receives all of the treatments (ie.
Drug and placebo).
 It is more statistically sensitive and efficient,
using fewer patients.
 Fewer patients can lead to a more
homogenous group with less variability in
measurement. (Less variability between groups implies a
measured difference is more likely to be due to treatment effects
instead of interpatient variability.

14. Cross-Over Design Example
 In a study to evaluate the efficacy of beta
blockers for hypertension patients, 12
patients would be enrolled into the study and
6 patients would be assigned to treatment
with the beta blocker first, followed by
placebo treatment and the other 6 patients
would receive the same treatments in reverse
order, all having a washout period in-between
treatments.

15. Cross-Over Design
 Advantages
 Smaller number of patients are required since the same
patient groups receive both treatments
 The ability to analyze patients both within groups as well as
between groups
 Within groups: baseline factors (age, gender differences)
which could influence the results are eliminated because
patients serve as their own control group.
 Between groups: Evaluate the effect of time on the
results (we can see what the patient does during the
placebo time period as well as what he does during
treatment period)

16. Cross-Over Design
 Disadvantages
 Time involved for a crossover design is longer
than other design types
 More drop outs because of time involved.
 Study design is very sensitive to drop outs
since small number of patients involved.
 Period effects
 Sequence effects
 Carry over effects

17. Period Effects
 Differences between treatment groups over
passages of time.
 Period effects occur because patients are
observed at least twice and their condition
may change between the first and second
observation.
 Period effects increase within-person
variability, which reduces the power of the
design and decreases the advantage of a
cross-over design study

18. Period Effect Variables
 Depressed patients may be less depressed during
the 2nd treatment period simply because depression
tends to improve over time.
 Learned effects
 Development of tolerance or resistance
 Changes in the disease state
 Psychological variables: pain syndromes,
exacerbation of exzema, multiple sclerosis acute flare
ups, etc.

19. Sequence Effects
 Changes in the effectiveness of the drug
treatment produced by the order in which the
drugs were administered.
 Appear statistically as interactions.
Interactions affect the interpretation of the
results because the magnitude of the
treatment differences is not consistent.

20. Carry-over effect
 When the effects of the drug given during the
first period persists into the second period.
 Carryover effects only affect the treatment
response in the 2nd time period.
 Can be eliminated by using a washout period
between treatments. This allows the patient to
return to baseline levels before the 2nd
treatment is started.

21. Carry-over effects
 The ability to remove the influence of carryover
effects through the use of a washout period
differentiates carry-over and period effects.
 Period effects represent long term or permanent
changes in the subject that are unlikely to be
eliminated with a washout period.
 Carry-over effects represent temporary changes
secondary to continued presence of the drug in the
system, such as for a drug with a long half-life
persisting into the 2nd treatment period, but the
effects of the drug with a short half-life not persisting

22. Cross-Over Design
 One risk is that this design is not powerful enough to
detect a clinically important interaction of period,
sequence or carry-over effect.
 If the interaction is clinically significant, then you must
transform the crossover study into a parallel design to
do the statistical analysis. (defeats purpose of using
less subjects and saving money)
 Types of studies good for cross-over design are:
 bioavailability studies (interactions would be less likely
since the patients serve as their own controls)

23. Time Series (Before and After)
Design
 Patients are studied before the experimental
drug is given. After the drug is given for a
certain amount of time, the patients are
evaluated again to determine the effects of
the drug.
 More than one drug can be tested with this
type of study design by continuing to
administer drugs in sequence.

24. Times Series (Before and After)
 Advantages
 Certain factors which could influence the study
are eliminated (age, gender differences, etc)
 Patients serve as their own controls so smaller
number of patients are needed (as compared
to parallel design)

25. Times Series (Before and After)
 Disadvantages
 The disease/condition being treated can change
over time, unrelated to the drug treatment.
 Carry-over effects could occur.
 Things that cause a carry-over effect
 Drugs with a long elimination half-life
 Drugs with active metabolites (esp. active
metabolites with long half-lives)
 Drugs whose effects on the disease state being
treated persist after the drugs themselves are
eliminated from the body (lipophillic drugs)