EPI4922 Tutorials
Task 1 Random trail designs
Learning goals:
1 Link type of study to correspondent box
- Parallel group design (= Conventional RCT) Study 7
- Randomized-group design (= Cluster randomized design) Study3
- Cross-over design Study 5
- N-of-1 trial design (= N=1 design, single case study, etc) Study 2, looking at every
patient as an individual.
- Factorial trial design Study 5
- Random consent design (= Pre-randomization design, Zelen design) Study
- Equivalence trial / Non-inferiority trial study 1 comparing new formula to old one,
looking at constipation.
Study 1; cross-over/parallel group yough age
Study 4: Zelen design randomization will be shown after the study and consentens will
be asked after sending postcards
Study 5: factorial 3 kinds of interventions and combinations
Study 6: non-equivalence
Study 7: parallel
2 Find and analyze literature regarding equivalence trial/non-inferiority trial (take an
example to discuss in tutorial sessions).
This occurred in the Comparison of Age-Related Macular Degeneration Treatments Trials
(CATT), where four groups (one standard therapy— monthly administration of intravitreal
injections of ranibizumab—and three unproven therapies—as needed injections of
ranibizumab and monthly and as needed injections of bevicizumab) were compared using a
noniferiority design [ The CATT Research Group. Ranibizumab and bevacizumab for
neovascular age-related macular degeneration. N Engl J Med 2011;364:1897–1908)
Study https://www.bmj.com/content/bmj/325/7376/1323.full.pdf
- Nursing intervention.
,3 Understand the methodology of equivalence and non-inferiority trials
In studies of equivalency, the objective is to test whether a new intervention is equivalent to
an established one.
Noninferiority trials test whether the new intervention is no worse than, or at least as good
as, some established intervention.
The control treatment (or standardized treatment) must have shown conclusively to be
effective. These requirements also mean that the trials that demonstrated efficacy of the
standard should be recent and properly designed, conducted, analyzed, and reported.
The evidence that demonstrated the benefits of the control must be available so that a
control group event rate can be estimated. The response variable used in the new trial must
be sensitive to the postulated effects of the control and intervention.
The intervention is claimed to be “the same” as the control (in an equivalence trial) or no
worse than the control (in a noninferiority trial).
The relative risk of the new intervention compared to the control should be close to 1,
Therefore the RR is set in the range of 1.2-1.4. in the worst case, the new intervention
may be 20-40% inferior to the standard treatment and yet be considerd equivalent of
noninferior.
Non-inferiority study (one-sided)
Equivalence (two sided)
Look at the variance and the differences, not only at the P-value (H0)
testing more then one Hypothese type 1 error and bonferroine als correction
Point-estimate says if the study is inferiority study.
The more the difference is close to the alpha the more N you need
Equivalence trial/non-inferiority trial
Equivalence trial two interventions are similar in characteristics (within a pre-defined threshold);
absence of significant difference
Non-inferiority trial new treatment is not worse than a standard one
Definition of non-inferiority limit dependent of many factors; ideally, margin would be defined by
physician (maximization of wellness for the patient)
Results of trial:
o Non-inferior
o Inconclusive
, o Inferior
If not inferior, the new intervention tested may have advantages such as:
o Better outcome
o Less invasive procedure or easier application
o Less side effects for the patient
o Fewer drug interactions
o Economical differences
Lower costs
Less of a burden for healthcare system
o Example: Methadone vs fentanyl
4 Compare main variances between non-inferiority and superiority trials
As mentioned in the introduction to this chapter, traditionally, most trials were designed
to establish whether a new intervention on top of usual or standard care was superior to
that care alone (or that care plus placebo). If there were no effective treatments, the
new intervention was compared to just placebo. As discussed in Chap. 8, these trials
were generally two-sided. That is, the trial was designed to see whether the new
intervention was better or worse than the control.
With the development of effective therapies, many trials have been designed to
demonstrate that a new intervention is not worse than the intervention previously
shown to be beneficial, i.e., an active control, by some prespecified amount. As noted
earlier, the motivation for such a question is that the new intervention might not be
better than standard treatment on the primary or important secondary outcomes, but
may be less toxic, more convenient, less invasive and/or have some other attractive
feature, including lower cost. The challenge is to define what is meant by “not worse
than.” This has been referred to as the “margin of indifference,” or δ, meaning that if
the new intervention is not less effective than this margin, its use might be of value
given the other features. In the analysis of this design, the 95% upper confidence limit
would need to be less than this margin in order to claim noninferiority. Defining δ is
challenging and will be discussed in Chap. 5.
The question in a noninferiority trial is different than in a superiority trial and affects
both the design and conduct of the trial. For example, in the superiority trial, poor
adherence will lead to a decreased ability, or power, to detect a meaningful difference.
For a noninferiority trial, poor adherence will diminish real and important differences
and bias the results towards a noninferiority claim. Thus, great care must be taken in
defining the question, the sensitivity of the outcome measures to the intervention
being evaluated, and the adherence to the intervention during the conduct of the trial.
5 Understand the methodology concerning a Zelen design
Zelen proposed a modification of the standard randomized control study Zelen, therefore,
suggested randomizing eligible participants before informing them about the trial not
blinded
major criticism of this controversial design centers around the ethical concern of not
informing participants that they are enrolled in a trial
Task 1 Random trail designs
Learning goals:
1 Link type of study to correspondent box
- Parallel group design (= Conventional RCT) Study 7
- Randomized-group design (= Cluster randomized design) Study3
- Cross-over design Study 5
- N-of-1 trial design (= N=1 design, single case study, etc) Study 2, looking at every
patient as an individual.
- Factorial trial design Study 5
- Random consent design (= Pre-randomization design, Zelen design) Study
- Equivalence trial / Non-inferiority trial study 1 comparing new formula to old one,
looking at constipation.
Study 1; cross-over/parallel group yough age
Study 4: Zelen design randomization will be shown after the study and consentens will
be asked after sending postcards
Study 5: factorial 3 kinds of interventions and combinations
Study 6: non-equivalence
Study 7: parallel
2 Find and analyze literature regarding equivalence trial/non-inferiority trial (take an
example to discuss in tutorial sessions).
This occurred in the Comparison of Age-Related Macular Degeneration Treatments Trials
(CATT), where four groups (one standard therapy— monthly administration of intravitreal
injections of ranibizumab—and three unproven therapies—as needed injections of
ranibizumab and monthly and as needed injections of bevicizumab) were compared using a
noniferiority design [ The CATT Research Group. Ranibizumab and bevacizumab for
neovascular age-related macular degeneration. N Engl J Med 2011;364:1897–1908)
Study https://www.bmj.com/content/bmj/325/7376/1323.full.pdf
- Nursing intervention.
,3 Understand the methodology of equivalence and non-inferiority trials
In studies of equivalency, the objective is to test whether a new intervention is equivalent to
an established one.
Noninferiority trials test whether the new intervention is no worse than, or at least as good
as, some established intervention.
The control treatment (or standardized treatment) must have shown conclusively to be
effective. These requirements also mean that the trials that demonstrated efficacy of the
standard should be recent and properly designed, conducted, analyzed, and reported.
The evidence that demonstrated the benefits of the control must be available so that a
control group event rate can be estimated. The response variable used in the new trial must
be sensitive to the postulated effects of the control and intervention.
The intervention is claimed to be “the same” as the control (in an equivalence trial) or no
worse than the control (in a noninferiority trial).
The relative risk of the new intervention compared to the control should be close to 1,
Therefore the RR is set in the range of 1.2-1.4. in the worst case, the new intervention
may be 20-40% inferior to the standard treatment and yet be considerd equivalent of
noninferior.
Non-inferiority study (one-sided)
Equivalence (two sided)
Look at the variance and the differences, not only at the P-value (H0)
testing more then one Hypothese type 1 error and bonferroine als correction
Point-estimate says if the study is inferiority study.
The more the difference is close to the alpha the more N you need
Equivalence trial/non-inferiority trial
Equivalence trial two interventions are similar in characteristics (within a pre-defined threshold);
absence of significant difference
Non-inferiority trial new treatment is not worse than a standard one
Definition of non-inferiority limit dependent of many factors; ideally, margin would be defined by
physician (maximization of wellness for the patient)
Results of trial:
o Non-inferior
o Inconclusive
, o Inferior
If not inferior, the new intervention tested may have advantages such as:
o Better outcome
o Less invasive procedure or easier application
o Less side effects for the patient
o Fewer drug interactions
o Economical differences
Lower costs
Less of a burden for healthcare system
o Example: Methadone vs fentanyl
4 Compare main variances between non-inferiority and superiority trials
As mentioned in the introduction to this chapter, traditionally, most trials were designed
to establish whether a new intervention on top of usual or standard care was superior to
that care alone (or that care plus placebo). If there were no effective treatments, the
new intervention was compared to just placebo. As discussed in Chap. 8, these trials
were generally two-sided. That is, the trial was designed to see whether the new
intervention was better or worse than the control.
With the development of effective therapies, many trials have been designed to
demonstrate that a new intervention is not worse than the intervention previously
shown to be beneficial, i.e., an active control, by some prespecified amount. As noted
earlier, the motivation for such a question is that the new intervention might not be
better than standard treatment on the primary or important secondary outcomes, but
may be less toxic, more convenient, less invasive and/or have some other attractive
feature, including lower cost. The challenge is to define what is meant by “not worse
than.” This has been referred to as the “margin of indifference,” or δ, meaning that if
the new intervention is not less effective than this margin, its use might be of value
given the other features. In the analysis of this design, the 95% upper confidence limit
would need to be less than this margin in order to claim noninferiority. Defining δ is
challenging and will be discussed in Chap. 5.
The question in a noninferiority trial is different than in a superiority trial and affects
both the design and conduct of the trial. For example, in the superiority trial, poor
adherence will lead to a decreased ability, or power, to detect a meaningful difference.
For a noninferiority trial, poor adherence will diminish real and important differences
and bias the results towards a noninferiority claim. Thus, great care must be taken in
defining the question, the sensitivity of the outcome measures to the intervention
being evaluated, and the adherence to the intervention during the conduct of the trial.
5 Understand the methodology concerning a Zelen design
Zelen proposed a modification of the standard randomized control study Zelen, therefore,
suggested randomizing eligible participants before informing them about the trial not
blinded
major criticism of this controversial design centers around the ethical concern of not
informing participants that they are enrolled in a trial
