Unblinding: Difference between revisions
No edit summary |
No edit summary |
||
| Line 6: | Line 6: | ||
==Unblinding sponsors== | ==Unblinding sponsors== | ||
In a trial, the word “sponsor” can be designated as an individual, company, institution or organization which takes responsibility for the initiation, management, and/or financing of a clinical trial <ref name="urlwww.fda.gov">{{cite web |url=http://www.fda.gov/downloads/RegulatoryInformation/Guidances/ucm127073.pdf |title=Guidance for Clinical Trial Sponsors |format= |work=|accessdate=2013-04-11}}</ref>. The main sponsors of clinical trials are pharmaceutical and biotechnology companies. Blinding (sometimes called masking) refers to keeping key persons, such as participants, health care providers, and sponsors, unaware of the treatment administered or interventions after inclusion of participants into the study <ref name="pmid11827510">{{cite journal |author=Schulz KF, Chalmers I, Altman DG |title=The landscape and lexicon of blinding in randomized trials |journal=Ann. Intern. Med. |volume=136 |issue=3 |pages=254–9 |year=2002 |pmid=11827510 |doi= |url=}}</ref><ref name="pmid11308438">{{cite journal |author=Devereaux PJ, Manns BJ, Ghali WA, ''et al.'' |title=Physician interpretations and textbook definitions of blinding terminology in randomized controlled trials |journal=JAMA |volume=285 |issue=15 |pages=2000–3 |year=2001 |month=April |pmid=11308438 |doi= |url=}}</ref> | In a trial, the word “sponsor” can be designated as an individual, company, institution or organization which takes responsibility for the initiation, management, and/or financing of a clinical trial <ref name="urlwww.fda.gov">{{cite web |url=http://www.fda.gov/downloads/RegulatoryInformation/Guidances/ucm127073.pdf |title=Guidance for Clinical Trial Sponsors |format= |work=|accessdate=2013-04-11}}</ref>. The main sponsors of clinical trials are pharmaceutical and biotechnology companies. Blinding (sometimes called masking) refers to keeping key persons, such as participants, health care providers, and sponsors, unaware of the treatment administered or interventions after inclusion of participants into the study <ref name="pmid11827510">{{cite journal |author=Schulz KF, Chalmers I, Altman DG |title=The landscape and lexicon of blinding in randomized trials |journal=Ann. Intern. Med. |volume=136 |issue=3 |pages=254–9 |year=2002 |pmid=11827510 |doi= |url=}}</ref><ref name="pmid11308438">{{cite journal |author=Devereaux PJ, Manns BJ, Ghali WA, ''et al.'' |title=Physician interpretations and textbook definitions of blinding terminology in randomized controlled trials |journal=JAMA |volume=285 |issue=15 |pages=2000–3 |year=2001 |month=April |pmid=11308438 |doi= |url=}}</ref><ref name="urlwww.cochrane.org">{{cite web |url=http://www.cochrane.org/handbook/891-rationale-concern-about-bias |title=Cochrane Handbook |format=}}</ref>;. | ||
In general, sponsors are unblinded after the completion of the research trial. Unblinding the sponsors during the study process would introduce false estimates to the study, since the sponsor would be aware of the results and the level of treatment while the trial is ongoing. In other words, the sponsor will have access to which group is taking the medication and to the results, therefore the unblinded sponsor is able to modify the course of the trial. | In general, sponsors are unblinded after the completion of the research trial. Unblinding the sponsors during the study process would introduce false estimates to the study, since the sponsor would be aware of the results and the level of treatment while the trial is ongoing. In other words, the sponsor will have access to which group is taking the medication and to the results, therefore the unblinded sponsor is able to modify the course of the trial. | ||
Revision as of 19:59, 12 April 2013
Editor-In-Chief: C. Michael Gibson, M.S., M.D. [1] Associate Editor-In-Chief: Marcelo R. Zacarkim, M.D. [2]
Overview
Blinding is the best way to investigate a new drug or any type of new intervention in clinical research, especially, if related to a placebo-controlled trial. However, sometimes blinding is challenging and when it becomes not possible, bias is more difficult to prevent.
Unblinding sponsors
In a trial, the word “sponsor” can be designated as an individual, company, institution or organization which takes responsibility for the initiation, management, and/or financing of a clinical trial [1]. The main sponsors of clinical trials are pharmaceutical and biotechnology companies. Blinding (sometimes called masking) refers to keeping key persons, such as participants, health care providers, and sponsors, unaware of the treatment administered or interventions after inclusion of participants into the study [2][3][4];.
In general, sponsors are unblinded after the completion of the research trial. Unblinding the sponsors during the study process would introduce false estimates to the study, since the sponsor would be aware of the results and the level of treatment while the trial is ongoing. In other words, the sponsor will have access to which group is taking the medication and to the results, therefore the unblinded sponsor is able to modify the course of the trial.
Unblinding participants
After the completion of the research trial, each participant will be required to return to a consult to be unblinded. This option allows them to know what medication they were taking during the trial. Usually, unblinding of participants during conduct of a clinical trial is not allowed unless there are compelling severe adverse effects [1].
Risks of unblinding
- Biasing the protocol, since it would appears more interesting to continue the trial (falsifying data/results for patients known to receive a given treatment, in hopes of making the treatment look better)[5];
- Changes of protocol by scientific developments (new tests, approval of new products, announcement of results of other trials);
- Changes of protocol by financial considerations, production problems, enrollment problems and missing data) [3];
- Unavoidable bias by changing an endpoint;
- Adversity in interpreting the results;
- Change statistical methodology during trial based on trends seen so far;
- Hard to prove bias did not enter study;
- Adversity in interpreting the results;
- Unavoidable bias by changing an endpoint.
Bias
Bias is a systematic error that causes false estimates [6], which affects the measure in a constant way; for instance, studying the effectiveness of a new drug for any type of disease, where the pharmacokinetic properties, such as absorption, distribution, metabolism, and excretion of the drug are critically influenced by the route of administration makes blinding not feasible leading to biases and, consequently, jeopardizing the results.
Who can be unblinded?
- Study Treatment Manufacturer [4];
- Pharmacist;
- Medical Monitors (in case of patient having a serious safety issue)[1];
- Data Safety Monitoring Board (DSMB);
- Statisticians who generated the randomization schedule used by pharmacists/manufacturers.
Summary
In summary, all clinical trials should be double-blinded and sponsor/study personnel should be blinded even if is an “open label” trial.
References
- ↑ 1.0 1.1 1.2 "Guidance for Clinical Trial Sponsors" (PDF). Retrieved 2013-04-11.
- ↑ Schulz KF, Chalmers I, Altman DG (2002). "The landscape and lexicon of blinding in randomized trials". Ann. Intern. Med. 136 (3): 254–9. PMID 11827510.
- ↑ 3.0 3.1 Devereaux PJ, Manns BJ, Ghali WA; et al. (2001). "Physician interpretations and textbook definitions of blinding terminology in randomized controlled trials". JAMA. 285 (15): 2000–3. PMID 11308438. Unknown parameter
|month=ignored (help) - ↑ 4.0 4.1 "Cochrane Handbook".
- ↑ Montori VM, Bhandari M, Devereaux PJ, Manns BJ, Ghali WA, Guyatt GH (2002). "In the dark: the reporting of blinding status in randomized controlled trials". J Clin Epidemiol. 55 (8): 787–90. PMID 12384193. Unknown parameter
|month=ignored (help) - ↑ Gluud LL (2006). "Bias in clinical intervention research". Am. J. Epidemiol. 163 (6): 493–501. doi:10.1093/aje/kwj069. PMID 16443796. Unknown parameter
|month=ignored (help)