Experimental Study Designs
Dr. lamya Alnaim
Introduction




The primary method for testing the effectiveness of new
therapies and other interventions, including innovative
drugs.
By the 1930s, the pharmaceutical industry had adopted
experimental methods.
In the 1960s, the controlled clinical trial became the
standard for doing pharmaceutical research and
measuring the therapeutic benefits of new drugs.
Introduction


The drug regulations of the 1960s also reinforced the
importance of controlled clinical trials by requiring that
proof of effectiveness for new drugs be made through use
of these research methods.
In pharmacoepidemiology, the primary use of
experimental design is in performing clinical trials, most
notably randomized, controlled clinical trials.
Experimental Design

An experiment is a study designed to compare benefits of
an intervention with standard treatments, or no
treatment, such as a new drug therapy or prevention
program, or to show cause and effect.


Performed prospectively.
Subjects are selected from a study population, assigned to the
various study groups, and monitored over time to determine
the outcomes that occur and are produced by the new drug
therapy, treatment, or intervention.
Experimental Design
Advantages.





Randomization tends to balance confounding variables across
the various study groups, especially variables that might be
associated with changes in the disease state or the outcome of
the intervention under study.
Detailed information and data are collected at the beginning to
develop a baseline and at specified follow-up periods
throughout the study.
The investigators have control over variables such as the dose
or degree of intervention.
The blinding process reduces distortion in assessment.
This design is the only real test of cause–effect relationships.
Experimental Design
Disadvantages
 Subject participation criteria that may limit generalizability
of findings.


Restrictive criteria for inclusion or exclusion of subjects may
produce a very homogeneous study population that restricts
application of the results to patients with other characteristics.
May require years of follow-up and prolonged
observation to determine treatment outcomes.



Higher costs,
increased likelihood that patients will be lost to follow-up
delayed treatment recommendations.
Experimental Design
Disadvantages
 Large sample sizes are typically required to demonstrate
differences among study groups.




Increasing the study population raises the cost of the trial
may make it difficult to locate a sufficiently large pool of eligible
patients.
Ethical concerns
subjects may not comply with the treatment and
assignment.
Classification
(1)

(2)

(3)

Therapeutic trials
Therapeutic agents or procedures are given to patients in an
attempt to cure the disease, relieve symptoms, or prolong
survival
Intervention trials
The investigator intervenes before the disease has developed
in individuals with certain characteristics that increase their
risk of developing the disease
Prevention trials
To determine the efficacy of a preventive agent or procedure
among people who do not have the disease but may be at
greater risk for developing it.
The basic design of a randomized, controlled
clinical trial
Clinical Drug Trials


The development of the clinical trial is a product of the
application of the modern scientific method to medicine.
Experiments on human populations have inherent
difficulties not found in the laboratory.


laboratory experiments involve controlled environments and
the manipulation of just a small number of variables,
Control of human subjects and their environments, with a
variety of other factors, is much more difficult.
Clinical Drug Trials


The basic idea is to "try out" (trial) a new drug in clinical
practice on sick patients.
The goal is to determine the therapeutic benefits of the
new drug in humans after it has passed safety testing in
animals as well as safety, dosage range, and effectiveness
studies. (phases I and II).
Clinical Drug Trials
The design of drug trials varies by certain
components
 the specific treatments being evaluated and compared
 how patients are selected for the different study
groups
 who knows which patients are receiving the
experimental treatment
 which ones are receiving other treatments or no
treatment
Table 4-1. Components of a Clinical Drug Trial
Experimental study: A research method in which one group of subjects receives a new treatment
(e.g., drug, device, surgery), and they are compared with one or more other group(s) who receive
different standard treatments or placebo. These groups are studied over the same time period using
the same measures of safety and effectiveness.
Experimental group: The group of patients that receives the drug under investigation.
Control group: The group of patients that receives a different type of treatment, either a traditional
one (already approved and used in therapy) or no treatment (a placebo).
Randomization: The process of assigning individual patients to different treatment groups in such a
way that each patient has the same chance, equal to and independent of every other patient, of
being selected for any particular study or treatment group. The idea is to make all study groups as
equal as possible at the beginning of the experiment.
Blinding: The process of ensuring that almost everyone involved in the drug trial is unaware of who
is receiving the experimental drug and who is receiving a traditional drug treatment, or a placebo,
throughout the duration of the study. In experimental studies, lack of knowledge about which
patients are receiving which treatments may be limited to the patients (single-blind); patients and
treating clinicians (double-blind); or patients, treating clinicians, and the scientific investigators
(triple-blind).
Placebo: An inactive form of treatment, usually an inert sugar pill, received by patients in the
control group. This treatment provides the basis for a trial group (the controls) to receive no
(beneficial) treatment so that a good comparison can be made with the results of the experimental
group. The use of placebo in most clinical trials is declining due to ethical concerns.
The "clinical inquiry"
a simple, nonexperimental method in which
only one group of patients receives a new drug
 data are collected and analyzed to see if it had
any effect;
 no control or comparison group is used.

Controlled clinical trial

The control group consists of participants not from the
original pool of potential study patients or patients whose
past cases involved receiving other treatments in regular
health care or research situations.
The randomized, controlled
clinical trial



The best method for determining the true therapeutic
benefit of a new unknown treatment
A clinical trial is often performed at more than one
facility to increase the number and diversity of the
subjects who choose to participate in it, leading to more
powerful and generalizable results.
This may also shorten the time of the study.
Variables
Types of variables

Variable


A characteristic that changes or has different values
for different individuals (EX: height, level of anxiety,
temperature of room)
Independent Variable (IV)


Variable that is changed or manipulated by the
experimenter
Dependent Variable (DV)

Variable that used to measure the change or
effect of the IV
Types of variables

Control Variables:



held constant by the experimenter to eliminate them as
potential causes.
useing only research participants who have problems with
anxiety or depression, this diagnosis would be a control
variable.
Random Variables:


allowed to vary freely to eliminate them as potential
causes.
Characteristics of the research participants, as long as
they really do vary freely. Examples: age, personality
type, or career goals.
Types of variables

Confounding Variables:
vary systematically with the independent
variable
 May be a cause.
 Good experimental designs eliminate them.

Types of variables

Confounding Variables:
 If we divide the research participants into two groups,
the experimental group the control group
 A systematic difference between these two groups, it
will not be a fair test.
 If those in the experimental group know they are
getting a new treatment and therefore expect to get
better, the expectations will be a confounding variable.
 If the experimental group does improve, we will not
know whether it was because of the therpay itself (the
Independent Variable) or because of the participants'
expectations (a Confounding Variable).
Reliability And Validity
Reliability And Validity
1. Reliability



Are the results of the experiment repeatable?
If the experiment were done the same way again,
would it produce the same results?
Reliability is a requirement before the validity of the
experiment can be established.
Reliability And Validity
II. Internal validity





Accuracy or truth-value
Does the research design lead to true statements?
Did the independent variable cause the effects in
the dependent variable?
In experimental research, this usually means
eliminating alternative hypotheses.
whether the therapy really was the causal factor in
improving participants' state.
Reliability And Validity
III. External validity
 Generalizability
 Can the results can be applied in another
setting or to another population of research
participants?
Hypotheses
Hypotheses

Hypothesis
 Makes a prediction about how the
manipulation of the independent variable
will affect the dependent variable
OR
 Makes a prediction about the relationship
between two variables (a correlation)
Types of Hypotheses




Experimental research methods revolve around
hypotheses, educated guesses.
We start with a hypothesis about how the results will
turn out, i.e., that there is an effect and it is due to
the independent variable.
This first hypothesis is the research hypothesis.
Then we hold the possibility that there is no effect of
the independent variable on the dependent variable
or that the differences observed are due to chance
only. This hypothesis is the null hypothesis.
Types of Hypotheses
The first step in experimential research, then,
is ruling out chance.
 we set up an experimental design that will
allow us to reject the null hypothesis.
 If we can confidently reject the null
hypothesis, then we gain confidence in the
research hypothesis.
 At this point, another group of hypotheses
comes into play, the alternative hypotheses

Types of Hypotheses



If there is an effect beyond chance, it may be due to
the independent variable or it may be due to a
number of other factors, so-called extraneous
variables or confounding variables.
We use experimental designs to allow us to eliminate
alternative hypotheses.
The bottom line for experimental designs is this:
 THE PURPOSE OF EXPERIMENTAL
DESIGNS
IS TO ELIMINATE ALTERNATIVE
HYPOTHESES.
Types of Hypotheses
Null hypothesis
 states that differences are due to chance or
that there are no differences between
treatments (used in statistical analysis).
 the null hypothesis is that the new form of
therapy is no better than either no therapy
or conventional therapies.
Types of Hypotheses
Alternative hypotheses
 suggest that results are due to factors other
than IV.
 These factors, rather than the independent
variable, may cause the improvements.
Types of Experimental
Design
Pre-experimental or faulty
designs
One group of subjects gets one treatment.
 a pre- and post-test or just a post-test.
 May eliminate chance, otherwise eliminates no
alternative hypotheses.

I-Pre-experimental or faulty
designs
Example
 Research participants who receive the new form of
therapy are tested afterward, or participants are
measured before and after the therapy.
 This may be useful in showing that there is some
reason to believe the new therapy works,
 we cannot draw any conclusions about why there is
improvement.
 It should be considered a pilot test at best and
followed up with a better research design.
II-Quasi-experimental Designs
Eliminate some, but not all, alternative
hypotheses.
 They are especially useful in applied settings
where real-life constraints make it undesirable
or impossible to control the research setting.


II-Quasi-experimental Designs
Example
 if the new therapy is being used in a mental health
center or a private practice. Rather than
compromise the needs of clients to eliminate
alternative hypotheses, we would be willing to
allow some alternative hypotheses. This is a choice
of relevance and external validity over control and
internal validity.
 Ideally, such a design would be paired with others
to allow us to draw stronger conclusions.
II-Quasi-experimental Designs
1-Non-equivalent groups or static groups
design
 Two groups receive different treatments, but
are not randomly assigned or matched to
conditions.
 Eliminates history effects but not subject
effects.
II-Quasi-experimental Designs
Example
 Participants may be given the choice of which therapy
to receive.
 The participants most likely to benefit from the
new therapy are assigned to that condition.
 Intact (already-existing) group may be used
 all the clients in an existing therapy group
 If the group receiving the new therapy improves
more than the control group, we can be somewhat
more confident in the benefits of the new therapy.
II-Quasi-experimental Designs
2. 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.
II-Quasi-experimental Designs
Example
 One group of research participants is selected for
the study.
 Endpoint is measured each month for several
months. Then given the new therapy and
measured again.
 If they improve after the therapy, we are more
confident the new therapy helps.
 Used most often to evaluate public policy changes
which affect a large group of people.
II-Quasi-experimental Designs
3. Multiple
time-series design
 Two or more groups (not randomly assigned)
receive several pre-treatment measures and at
least one post-treatment measure.
 Can eliminate history and most subject effects.
 considered a "strong quasi-experimental
design."
II-Quasi-experimental Designs
Example:
 After taking a series of measurements at two
different Clinics, all clients at one clinic are
given the new therapy.
 If those clients improve more than the clients
at the "control" clinic we can be more
confident of the new therapy.
III-True Experimental Designs
These designs attempt to eliminate most
alternative hypotheses, especially those related
to time (history, maturation, and regression)
and those related to make-up of the groups
(selection effects).
 Such control may be at the expense of making
the situation too artificial.

III-True Experimental Designs
1. Randomized groups design, betweengroups design
 Each research participant is randomly assigned
to one group and gets only one level of the
indepndent variable.
 There may be pre-tests and post-tests or only
post-tests.
 This design can eliminate selection, history, and
maturation effects.
III-True Experimental Designs
2. Repeated measure design, withinsubject design
 Each research participant gets all levels of the
IV.
 Treatment orders must be counter balanced to
eliminate order effects.
III-True Experimental Designs
3. Mixed model designs or complex
designs
 These designs combine randomized groups
and repeated measdures designs.
 For instance, there may be two IVs, one
measured between groups and one measured
within groups.
Back to Validity
Validity Threats
Subject effect, selection effect:
 results are due to systematic differences in
research participants assigned to different
treatments.
 Example: If the research participants who receive
the new form of therapy are different from those
in a control
 One group could be healthier, more motivated,
or more experienced with psychotherapy.
 Common solution: Matching or random
assignment to groups
Validity Threats
History effect:
 results are due to events outside the experiment.
 Example: if there is one group of research
participants who are being measured at several
points in time.
 Some event that is not part of the research, say
something traumatic like a natural disaster,
which occurs at the same time as the treatment
could affect the results.
 Common solution: A control group which will be
exposed to the same history but not the new
form of therapy.
Validity Threats
Maturation effect:
 results are due to changes within subjects over time,
e.g., growth, warm-up, fatigue, learning to learn.
 This is a problem in research that measures a DV
over a period of time and especially in research with
repeated exposures to the IV.
 Example: If one group of participants, have
improvments over time without the new form of
therapy.
 Common solution: A control group which is
measured over the same period of time but does not
receive the new therapy.
Validity Threats
Experimenter expectancy effect, Experimenter
bias:
 results are due to the experimenter's actions or
expectations.
 A number of studies have shown that researchers
tend to find the results they are looking for.
 The causes for this range from overt cheating to very
subtle influences on data collection and interactions
with research participants.
 Experimenters are not always aware of the extent of
these influences.
Validity Threats
Example
 If the researcher is the one to assess the research
participants‘ dependent variable, he or she may
distort the assessments in the direction of the
research hypothesis.
 Common solution: Use independent judges or more
objective measurements of the dependent variable.
Validity Threats
Demand characteristics, Hawthorne
effect:
 Results are due to subjects' expectations of
desired behavior in the research setting
 Called "demand" because participants may
perceive a demand to behave or report on
themselves in a certain way.
 Called the Hawthorne Effect after a famous
series of experiments at a manufacturing plant
in Hawthorne, Ohio.
Validity Threats
researchers selected a group of factory
workers and changed various conditions such
as lighting to see what would increase
performance.
 They found that any change increased
performance, suggesting that research
participants were responding to the general
expectation that they would perform better
and to the social dynamics of being observed
closely.

Validity Threats
Example
 The researcher communicates his or her
expectations to the research participants
which in turn influences their responses.
 Common solution: Blind and double-blind
designs help avoid these problems.
 Also, using a control group which is measured
the same way without the treatment.
Validity Threats
Testing effect, reactivity:
 results are due to the data gathering procedures, e.g.,
being influenced by the test or learning from one test
administration to the next.
Example
 Measuring the participants' mental health could get them
thinking about their lives, thus improving them.
 Improvements would then be due to the data gathering,
not the therapy itself.
 Common solution: Use a control group which is also
measured, but without the therapy or with an alternative
form of therapy.
Validity Threats
Regression artifact, regression-to-the-mean
 results are due to extreme scores moving toward the
mean over time.
 Example: If a group is made up of those with the
worst mental health scores over time they are likely
to improve without therapy.
 This may be mistakenly attributed to the therapy.
 Common solution: Use a control group which has
similar charcteristics but which does not receive the
new therapy
Validity Threats
Instrumentation
 results are due to an aberration in measuring
tools, either mechanical instrument or test.
 Example: The dependent variable may be
measured by a poor test.
 Common solution: Select or develop a better
measure.
Validity Threats
Halo effect
 the researcher's expectations about certain
subjects based on some subject characteristics.
 Example: Judges rating the mental health of the
participants (the dependent variable) may
ascribe better mental health based on other
characteristics.
 Common solutions: Random assignment, blind
judges, more objective measures.
Validity Threats
Attrition or mortality effect
 When subjects drop out of an experiment, it
can bias the results.
 Especially true when more subjects drop out
of one treatment condition than another.
 The study is no longer a fair test.
 This leads to a kind of subject effect because
the subjects in the different groups are no
longer equivalent.
Validity Threats
Attrition or mortality effect
 If a study consists of a new therapy group, a conventional
therapy group, and a no-therapy group.
 If more subjects in the new therapy group drop out of the
study, it may be because the new therapy was not
appropriate for them.
 This leaves only those who benefitted most, making the
therapy look better than it really is.
 Common solution: There is no way to force research
participants to stay in the study, but if attrition looks like
a problem, find out why participants dropped out.
 This can sometimes give important clues about the study.
Randomized, Controlled
Clinical Trials
Randomized, Controlled Clinical Trials




the most scientific and ethical method for
investigation of new therapies
The main objective of this method is to make certain
that, after the trial is over, the better (or best) of the
studied treatments is identified.
The key components in a randomized clinical trial is
the use of a control group and randomization of
subjects into the study groups.
Not all clinical trials, however, are of this type.
Randomized, Controlled Clinical Trials


In some circumstances, the randomization process
may not be justified.
 This issue has arisen in recent research on drugs
for treatment of AIDS. Patients with AIDS believe
that they all should be in the clinical trial and in the
new investigational drug study group. No one
should be denied an experimental treatment
because they were randomly selected to be in the
control or placebo group.
Patients should always be informed of any
randomization process that is being used to influence
their treatment choices.
Randomized, Controlled Clinical Trials
still are far from being a perfect method.
 Clinical trials answer only questions that have
been asked specifically.
 the experimental conditions of the trial often
differ so much from the conditions
encountered in clinical practice that the trial's
results may not be applicable to real-life
situations.
 Concerns about the quality of clinical trials.

Randomization



It maximizes the probability that the two groups are
as similar as possible in terms of background
characteristics, especially factors and variables that
may influence response to therapy or primary
outcome measure.
In a randomized study, treatment group assignment is
based on probability alone and is not influenced by
the physicians' or patients' preferences.
Every patient in the original patient pool, or group,
must have a chance equal to every other patient of
being assigned to a particular treatment group.
Blinding
A process of keeping individuals involved in the
study unaware of assignment of subjects to
different study groups.
 Single-blind keeping the subjects unaware
 double-blind: keeping the investigators and
subjects unaware
 triple-blind: data analysis is done by outside
evaluators independent of the investigators.

Blinding
When a study is double-blinded and
randomization of patients was used, most
people involved in the study should not know
who is getting which treatment, or no
treatment, and a more objective assessment of
the new therapy can be made.
 The blinding process can break down in
studies where some patients are not receiving
any treatment, and other patients are taking a
very active drug with intensely noticeable
effects.

Intention-to-Treat and Interim
Analyses


Comparison of the entire experimental group with
the entire study group regardless of whether subjects
completed the full course of their study treatments.
Loss to follow-up is an issue.
 Investigators often have more control over patient
involvement throughout the clinical trial because of
the patient's investment in seeking a new, better
treatment.
 The key issue is whether the probability of loss to
follow-up is related to exposures, treatments, the
disease, and other study outcomes.
Intention-to-Treat and Interim
Analyses
The major advantage of ITT analysis is that it
maintains the randomization scheme, as
subjects do not randomly drop out of studies.
 It is necessary to develop guidelines for
deciding whether a trial should be modified or
stopped before originally scheduled.
 This decision is usually based on interim
analyses.

Intention-to-Treat and Interim
Analyses
Interim analyses.


To ensure that the welfare of the participants is
protected, and to provide useful new therapies to the
medical community, study result should be monitored
at various points by a group that is independent of
the investigators conducting the trial.
If the data indicate a clearly significant benefit in
terms of the primary endpoint, or if one treatment is
clearly harmful, then early termination of the trial
must be considered
Intention-to-Treat and Interim
Analyses
The decision to terminate a study early is
based on a number of complex issues and
must be made with great caution.
 The first requirement for considering
termination or modification is the observation
of a sustained statistical association that is
extreme and highly significant, virtually
impossible to arise by chance alone.

Intention-to-Treat and Interim
Analyses
Statistical test results should not be used as
the sole basis for the decision to stop or
continue a trial.
 Observing significant associations must be
considered in the context of the totality of
evidence, including postulated or known
biological mechanisms that might explain the
results, the results of other clinical trials or
observational studies.

Process Of Performing A
Clinical Trial
Process Of Performing A Clinical
Trial

The first step is to formulate the major
research question.
 usually is referred to as a hypothesis.
 determines the importance of selected
independent variables, such as the types of
interventions or treatments to be compared
and the nature of the dependent variables,
endpoints, or outcomes to be evaluated.
Process Of Performing A Clinical
Trial


The number and nature of subjects for the study as
well as eligibility and exclusion requirements for
subject participation in the study.
Endpoints must be determined in terms of clinical
importance, which ones can be measured in a
reasonable manner, and which ones can be studied in
the given population with the constraints and
resources available to the study.
 More than one endpoint can be measured to assess
treatment efficacy.
Sample Size Determination And
Data Analysis
The number of subjects who should be
enrolled in the clinical trial is known as the
sample size.
 should be determined soon after the primary
research question or outcome has been
formulated.
 At the conclusion of an experiment, the data
are analyzed and a statistical decision is made
either to accept or to reject the hypothesis.

Sample Size Determination And
Data Analysis
The decision to accept or reject a hypothesis
is based on probabilities, and the study results
based on these decisions may not truly reflect
what is occurring in "reality.“
 Reality, or "the truth," can be thought of as the
results of the intervention when they are
applied correctly to all possible patients (i.e.,
anyone with the clinical condition).

Sample Size Determination And
Data Analysis



Using only a sample of the entire patient population
for the trial, the investigator hopes to use the trial's
results to generalize about all patients.
The problem is that the investigator is intervening
only on the study sample and cannot intervene with
the whole population of patients.
There is always a risk of arriving at a mistaken
conclusion because the trial sample did not represent
all types of patients with the same disease
The relationship between clinical trial
results and reality
REALITY
STUDY
RESULTS
Outcome different
Outcome Not
different
Outcome different
Outcome not different
A
B
Study result valid (true
positive)
Type I error (false positive)
C
D
Type II error (false
negative)
Study result valid (true
negative)
The relationship between clinical trial
results and reality

Cells A and D represent a relationship
between study results and reality that is
correct.
 The study results are valid and truly reflect
what would happen to other patients in the
population if they were given the same
treatment as those in the trial sample.
The relationship between clinical trial
results and reality

Cells A and D represent a relationship
between study results and reality that is
correct.
 In cell A, the study results and reality both
indicate that the outcomes are different (e.g.,
the new drug is more effective than a
placebo). This is also known as a true-positive
result.
The relationship between clinical trial
results and reality

Cells A and D represent a relationship
between study results and reality that is
correct.
 In cell D, the study results and reality both
indicate that the outcomes are not different
(e.g., the new drug is not more effective than
a placebo). This is known as a true-negative
result.
The relationship between clinical trial
results and reality
Two types of errors can be made in interpreting
study results.
1. The results might indicate that there is a
difference in outcomes or treatments when
in reality there is no difference (cell B).
 This is a type I error;
 under these circumstances, the study results
are not valid but are falsely positive.
The relationship between clinical trial
results and reality
Two types of errors can be made in interpreting
study results.
2. if the study failed to find a difference in
outcomes or treatments when there actually
was a difference, a type II error has occurred
(cell C)
 under these circumstances, the study
results are not valid but are falsely negative.
The relationship between clinical trial
results and reality
Falsely positive or falsely negative study results
can occur because of
 faulty methodology, chance occurrence, or for
both
 some errors can be minimized by careful
attention to study design
 errors due to chance can never be completely
eliminated.

Such errors, however, can be estimated.
The relationship between clinical trial
results and reality
The notation used to describe the likelihood
of a type I error (i.e., the difference observed
between treatment groups is not a true
difference but due instead to chance) is the
alpha level.
 The notation used to describe the likelihood
of a type II error (i.e., the study did not find a
difference between treatment groups when
there is indeed a difference) is called the beta
level.

The relationship between clinical trial
results and reality

Researchers specify the levels when planning a
study.
 an alpha level may be specified at 0.05, a 5%
risk of committing a type I error, falsely
concluding that the treatment groups differ
when in reality they do not.
 Or specify a level, or risk of committing a
type II error, of 0.1, a 10% chance of missing a
true difference between the treatment
groups is being allowed
The relationship between clinical trial
results and reality

The statistical power, or ability of a study to
detect a true difference between groups, is 1beta .
 For a beta level of 0.10, the study would
have a 90% chance of detecting the true
difference in outcomes between treatment
groups.
The relationship between clinical trial
results and reality
Once the alpha and beta levels have been
established, must specify an extremely
important study parameter for determining
sample size—the magnitude of the difference
in outcome between treatment groups that
the study will be designed to detect.
 Should be selected on the basis of clinical
information.

The relationship between clinical trial
results and reality

In deciding on the level of outcome difference for
detection, the following may be considered
 The difference in outcome that would be important
to clinicians treating this type of patient
 The difference that would be meaningful to a
patient who may suffer the consequences of the
disease,
 the difference in outcome that would justify use of
the more effective treatment despite greater
expense or greater side effects
Clinical Trials And The U.S.
Drug Approval Process
Patient Compliance In Clinical
Drug Research
Patient Compliance In Clinical Drug
Research



The failure of patients to take drugs as prescribed,
called noncompliance, has been documented, and its
implications for patient care are well known to health
professionals.
The impact noncompliance can have on the drug
development process, from clinical trials to marketing,
has not been well appreciated by pharmaceutical
manufacturers and regulators.
Noncompliance occurs for a variety of reasons, and
many determinants have been identified through
research
Patient Compliance In Clinical Drug
Research

Noncompliance by participants in clinical trials
affects the evaluation and approval of new
pharmaceutical agents at two stages.
Patient Compliance In Clinical Drug
Research

First, during the pre-approval drug
development phases, noncompliance may
alter results of clinical investigations that
determine the optimal dose for package
insert labeling.
Patient Compliance In Clinical Drug
Research

Second after the drug has been approved and
is marketed, noncompliance, especially if it is
widespread, may lead to the emergence of
unexpected problems, such as irregular
dosing, overprescribing, failures to achieve
patient care outcomes, and toxicities.
Patient Compliance In Clinical Drug
Research
Some drug approval delays can be attributed to
noncompliance and the failure of clinical
investigators to recognize it.
 There may be difficulty in demonstrating a
drug's effectiveness in intention-to-treat
analyses, which assumes that patients follow
the regimens to which they were assigned.
 Noncompliance also can interfere with
determination of therapeutic dosage ranges.

Patient Compliance In Clinical Drug
Research
Most trials that now report the safety and
efficacy of a drug do not reflect the individual
patient's response to a prescribed dosage.
 They reflect a composite of the responses of
patients who fail to comply, those who partially
comply, and those who fully comply with the
therapeutic regimen.

Patient Compliance In Clinical Drug
Research
It is necessary to take into account the impact
of known compliance or noncompliance on
estimates of a drug's safety and efficacy.
 Giving equal weight to compliant patients who
took the drug and to patients who never took
the drug, took the wrong drug, or took
inadequate amounts increases the probability
of overestimating the therapeutic dose range.

Patient Compliance In Clinical Drug
Research
 Adverse reactions then could occur when
patients take the recommended dose in
practice.
 Additional trials may be necessary for
approval, and the problem resides with patient
noncompliance, not the drug product's dose or
formulation.
 Noncompliance also affects statistical analysis
of clinical trial and postmarketing data.
Patient Compliance In Clinical Drug
Research



Clinical trial research, particularly on chronic diseases,
must include new endpoints.
Priority should be given to subjective, in addition to
objective, endpoints that imply a focus on risk factors
and symptomatic treatments over cures. Such
research needs to emphasize the psychosocial
component as equally or more important than the
biological component.
These areas of study are exploring the emergent
quality-of-life concepts and new ethical priorities in
health care.