Sponsored by the National Eye Institute National Institutes of Health US Department of Health and Human Services 1 Background Bias in Clinical Trials Minimizing bias in a randomized clinical trial is a critical consideration during trial design ID: 503159
Download Presentation The PPT/PDF document "An Evaluation of Study Participant Maski..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
An Evaluation of Study Participant Masking of Intraocular Injections in a Randomized Clinical TrialSponsored by the National Eye Institute, National Institutes of Health, U.S. Department of Health and Human Services
1Slide2
BackgroundBias in Clinical TrialsMinimizing bias in a randomized clinical trial is a critical consideration during trial designRandomization alone does not ensure unbiased experimentTypes of Bias:Non-differential bias: Noise affecting outcomes of treatment groups equallyDifferential bias: Systematic differences in outcomes between treatment groups
2Slide3
Participant masking is one method of minimizing differential biasThe decision on whether to mask study participants to treatment should be based on:
Subjectivity of trial outcomes
Type of treatmentDisease type
Background
Minimizing Differential Bias
3Slide4
PurposeDescribe a trial design which attempted to mask participants to treatment allocationDiscuss rationale for masking in the trialEvaluate success of masking study participant to treatment assignment4Slide5
Study Design
5
Ranibizumab Injection
+Prompt
Laser
Ranibizumab Injection
+Deferred
Laser
Sham Injection
+Prompt
Laser
Triamcinolone Injection
+Prompt
Laser
Objective:
Evaluate efficacy of 4 treatment groups on
diabetic macular edema (measured by visual acuity)
Visits every 4 weeks through 52 weeks
52 Week Visit Primary Outcome
Randomized Clinical Trial (multi-center)Slide6
Participants with 1 study eye; randomized to 1 of 4 treatment groupsParticipants with 2 study eyes; one eye randomized to 1 of the 3 study drug groups and sham+prompt laser in the other
Study Design
6
Sham + Laser
Ranibizumab + Prompt Laser
Ranibizumab + Deferred Laser
Triamcinolone + Laser*
Injection Intervals (Weeks)
4W
4W
4W
Triam
: 16W
Sham: 4W-intermittently
Injection Type
Always Sham
Always Real
Always Real
Sometimes sham and sometimes real
Median (Quartiles)
# of injections/13 possible
11 (8,13)
8 (6, 10)
9 (6, 11)
Triam
: 3 (2, 4) Sham: 5 (3, 7)Slide7
Participants were masked through the 1-year primary outcome except those in the ranibizumab + deferred laser group in either eye DRCR.net investigators believed that a successful sham laser treatment was not feasible.
Investigators were not masked
.
Masking
7Slide8
Injection ProcedurePre/post injection procedures for sham and intraocular injections were identicalProvidone iodine prep of the conjunctivaPre-, peri
-, or post-injection antibiotics were at investigator discretion. Comparing real injections with sham injections, antibiotics were given:
Pre- and post-injection 34% vs. 27% of the timeOnly pre-injection 9% vs. 12% of the timeOnly post-injection 21% vs. 14% of the timeFor sham, the hub of a syringe without a needle was pressed against the
conjunctival surface
8Slide9
9Slide10
Why MaskVisual acuity testing requires a study participant response and knowledge of treatment group could affect test performance if he/she believes that the better or worse treatment was receivedImproving diabetes control can have an effect on diabetic macular edema and subsequently visual acuity. Knowledge of the treatment group could affect participant’s behavior in a way that could influence the course of the disease and the primary outcome measureInvestigators believed masking would work
10Slide11
Assessing Masking SuccessAt the completion of the 1-year visit, a questionnaire was administered to each participant by the study coordinator; Do you think the injections you have been getting in the [left/right] eye during the study have been:
real injections into the eye,
sham injection, meaning that a needle has not been injected into the eye, or
sometimes real and sometimes sham?
11Slide12
Results: Masking QuestionnaireParticipants with 1 Study Eye
12
Sham + Laser
N=105
Ran + Prompt LaserN=105
Ran +
Deferred Laser
N=112
Triam
+
Laser
N=118
Refused to answer question
4%
5%
5%
2%
Response
N=101
N=100
N=106
N=116
Always Real
72%
88%
90%
55%
Sometimes real/ sometimes sham
18%
12%
7%44%
Always Sham10%04%
1%Slide13
Results: Masking QuestionnaireParticipants with 2 Study Eyes13
Ranibizumab
+ Prompt Laser
N=42
Ranibizumab
+ Deferred Laser
N=42
Triamcinolone
+ Laser
N=37
Refused to answer question
4%
21%
5%
N=44
N=33
N=35
Injected eye
Correct response
84%
88%
31%
Incorrect response
16%
12%
69%
Sham eye
Correct response
30%
27%
14%
Incorrect response
70%
73%
86%Slide14
Results - Subgroups14Correctness of response within treatment group did not appear to differ by: Prior injection for DME
Prior laser for DME
Baseline visual acuity Age Gender
Race and ethnicity
Slide15
Relationship of Masking and Visual Acuity – Sham + Laser Group15
Change in Visual Acuity
Believed
Sham Injections Always Given
Believed
Real Injections were Always or Sometimes Given
Distribution of visual
acuity change
Improved ≥ 5 letters (≥ 1
line): N=60
10%
90%
Changed
± 4 letters: N=29
7%
93%
Worsened ≥ 5 letters (≥ 1
line): N=12
17%
83%
Mean
± Standard Deviation
3.6
± 8.1
6.0
± 9.4Slide16
DiscussionThe results suggest that very few participants (4%) believed they received only sham injections in a randomized trial comparing sham ocular injections with real ocular injections. Even when visual acuity outcomes did not improve or worsened, the study participant’s perception as to whether a sham or real injection was received did not appear to be influenced. 16Slide17
SummaryCompared with the participants who received sham injections only, participants in the treatment group that received both real and sham injections in the same eye more often thought that a sham injection was at least sometimes given (28% vs. 45%, P-value = 0.009). This may imply that a participant is more likely to identify a sham injection when also receiving a real injection in the same eye.It appears that participants with two study eyes believe sham injections were always given in the sham eye (24% in this study) compared with participants with only one study eye (10% in this study).17Slide18
ConclusionSham intraocular injections can mask participants to treatment group assignments successfully when a procedure that attempts to mimic a real injection is followed.Although still successful, the success of masking potentially is reduced for participants who receive real injections in 1 eye and sham injections in the other, or receive both a sham injection and a real injection in the same eye.This study did not assess the bias in measuring visual acuity that may occur when participants receiving intraocular injections are not masked to treatment assignment.18