Unblinding: Difference between revisions
Appearance
Content deleted Content added
Wikiman2718 (talk | contribs) sourcing |
Wikiman2718 (talk | contribs) better redirect Tag: Redirect target changed |
||
(40 intermediate revisions by 4 users not shown) | |||
Line 1: | Line 1: | ||
#REDIRECT [[Blinded experiment#Unblinding]] |
|||
'''Unblinding''', sometimes called '''code-break''', occurs in a [[blinded experiment]] when information becomes available to one from whom it has been masked. Unblinding that occurs before the conclusion of an [[experiment]] is a source of [[bias]]. In blinded experiments, some degree of premature unblinding is common.<ref>{{cite journal |vauthors=Bello S, Moustgaard H, Hróbjartsson A |title=Unreported formal assessment of unblinding occurred in 4 of 10 randomized clinical trials, unreported loss of blinding in 1 of 10 trials |journal=J Clin Epidemiol |volume=81 |issue= |pages=42–50 |date=January 2017 |pmid=27555081 |doi=10.1016/j.jclinepi.2016.08.002 |url=}}</ref> When a blind is imperfect, its success is judged on a [[spectrum]] with [[open-label trial|no blind]] (complete failure of blind) on one end, perfect blinding on the other, and poor or good blinding between. Thus, the common view or studies as blinded or unblinded is an example of a [[dichotomy|false dichotomy]].<ref>{{cite journal |vauthors=Schulz KF, Grimes DA |title=Blinding in randomised trials: hiding who got what |journal=Lancet |volume=359 |issue=9307 |pages=696–700 |date=2002 |pmid=11879884 |doi=10.1016/S0140-6736(02)07816-9 |url=}}</ref> |
|||
Success of blinding is assessed by questioning study participants about information that has been masked to them (e.g. did you receive the drug or [[placebo]]?). In a perfectly blinded experiment, the responses should be consistent with no knowledge of the masked information. However, if unblinding has occurred, the responses will indicate the degree of unblinding. Since unblinding cannot be measured directly, but must be [[latent variable|inferred]] from participants' responses, its measured value will depend on the [[choice modeling|nature of the questions asked]]. As a result, it is not possible to measure unblinding in a way that is completely objective. Nonetheless, it is still possible to make informed judgments about the quality of a blind. Poorly blinded studies rank above unblinded studies and below well-blinded studies in the [[hierarchy of evidence]].<ref name=Kolahi2009>{{cite journal |vauthors=Kolahi J, Bang H, Park J |title=Towards a proposal for assessment of blinding success in clinical trials: up-to-date review |journal=Community Dent Oral Epidemiol |volume=37 |issue=6 |pages=477–484 |date=2009 |pmid=19758415 |pmc=3044082 |doi=10.1111/j.1600-0528.2009.00494.x |url=}}</ref> [[Metascience (research)|Meta-scientific research]] has relieved high levels of unblinding in pharmacological trials. In particular, [[antidepressant]] trials are poorly blinded. Some researchers believe that the entire observed effect of antidepressant drugs is due to failure of blinding, and that they have no genuine value in the treatment of depression.<ref name=Ioannidis2008 /> [[CONSORT]] standards and [[good clinical practice]] guidelines recommend that all studies assess and report quality of blinding. In practice, this rarely occurs.<ref name=Bello2014 /> |
|||
==Causes== |
|||
===Post-study unblinding=== |
|||
'''Post-study unblinding''' is the release of masked data upon completion of a study. In [[clinical studies]], post-study unblinding serves to inform subjects of their [[treatment and control groups|treatment allocation]]. Removing a blind upon completion of a study is never mandatory, but is typically performed as a courtesy to study participants. Unblinding that occurs after the conclusion of a study is not a source of bias, as data analysis is complete at this time.<ref>{{cite journal |vauthors=Dinnett EM, Mungall MM, Kent JA, Ronald ES, McIntyre KE, Anderson E, Gaw A |title=Unblinding of trial participants to their treatment allocation: lessons from the Prospective Study of Pravastatin in the Elderly at Risk (PROSPER) |journal=Clin Trials |volume=2 |issue=3 |pages=254–259 |date=2005 |pmid=16279148 |doi=10.1191/1740774505cn089oa |url=}}</ref> |
|||
===Premature unblinding=== |
|||
'''Premature unblinding''' is any unblinding that occurs before the conclusion of a study. In contrast with post-study unblinding, premature unblinding is a source of bias. A [[code-break procedure]] dictates when a subject should be unblinded prematurely. A code-break procedure should only allows for unblinding in cases of emergency. Unblinding that occurs in compliance with code-break procedure is strictly documented and reported.<ref>{{cite journal |last1=Quittell |first1=Lynne M. |title=The Scientific and Social Implications of Unblinding a Study Subject |journal=The American Journal of Bioethics |date=3 October 2018 |volume=18 |issue=10 |pages=71–73 |doi=10.1080/15265161.2018.1513589 |url=https://www.tandfonline.com/doi/full/10.1080/15265161.2018.1513589?scroll=top&needAccess=true |accessdate=20 April 2019 |issn=1526-5161}}</ref> |
|||
Premature unblinding may also occur when a participant infers from experimental conditions information that has been masked to them. A common cause for unblinding is the presence of side effects (or effects) in the treatment group. In pharmacological trials, premature unblinding can be reduced with the use of an [[active placebo]], which conceals treatment allocation by ensuring the presence of side effects in both groups.<ref>{{cite journal |last1=Double |first1=D. B. |title=Placebo mania. Placebo controlled trials are needed to provide data on effectiveness of active treatment. |journal=BMJ : British Medical Journal |date=19 October 1996 |volume=313 |issue=7063 |pages=1009 |pmid=8892442 |url=https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2352320/ |accessdate=21 April 2019 |issn=0959-8138}}</ref> However, side effects are not the only cause of unblinding; any perceptible difference between the treatment and control groups can contribute to premature unblinding. |
|||
A problem arises in the assessment of blinding because asking subjects to guess masked information may prompt them to try to infer that information. Researchers speculate that this may contribute to premature unblinding.<ref>{{cite journal |last1=Rees |first1=Judy R. |last2=Wade |first2=Timothy J. |last3=Levy |first3=Deborah A. |last4=Colford |first4=John M. |last5=Hilton |first5=Joan F. |title=Changes in beliefs identify unblinding in randomized controlled trials: a method to meet CONSORT guidelines |journal=Contemporary Clinical Trials |date=February 2005 |volume=26 |issue=1 |pages=25–37 |doi=10.1016/j.cct.2004.11.020 |url=https://www.sciencedirect.com/science/article/pii/S1551714405000157 |accessdate=24 April 2019}}</ref> Furthermore, it has been reported that some subjects of clinical trials attempt to determine if they have received an active treatment by gathering information on social media and message boards. While researchers counsel patients not to use social media to discuss clinical trials, their accounts are not monitored. This behavior is believed to be a source of unblinding.<ref>{{cite web |last1=Ledford |first1=Heidi |title=A question of Control |url=https://www.nature.com/magazine-assets/d41586-018-07351-8/d41586-018-07351-8.pdf |website=nature.com |publisher=Nature Magazine |accessdate=24 April 2019}}</ref> [[CONSORT]] standards and [[good clinical practice]] guidelines recommend the reporting of all premature unblinding.<ref name=":0">{{cite journal |last1=Moher |first1=David |last2=Altman |first2=Douglas G. |last3=Schulz |first3=Kenneth F. |title=CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials |journal=BMJ |date=24 March 2010 |volume=340 |pages=c332 |doi=10.1136/bmj.c332 |url=https://www.bmj.com/content/340/bmj.c332 |accessdate=24 April 2019 |language=en |issn=0959-8138}}</ref><ref>{{cite web |title=E6(R2) Good Clinical Practice: Integrated Addendum to ICH E6(R1) Guidance for Industry |url=https://www.fda.gov/downloads/Drugs/Guidances/UCM464506.pdf#page=24 |website=fda.gov |accessdate=21 April 2019}}</ref> In practice, unintentional unblinding is rarely reported.<ref name=Bello2014>{{cite journal |vauthors=Bello S, Moustgaard H, Hróbjartsson A |title=The risk of unblinding was infrequently and incompletely reported in 300 randomized clinical trial publications |journal=J Clin Epidemiol |volume=67 |issue=10 |pages=1059–1069 |date=2014 |pmid=24973822 |doi=10.1016/j.jclinepi.2014.05.007 |url=}}</ref> |
|||
==In clinical research== |
|||
Studies of blinded pharmacological trials across widely varying domains find evidence of significant levels of unblinding. Unblinding has been shown to effect both patients and clinicians. This evidence challenges the common assumption that blinding is highly effective in pharmacological trials. Unblinding has also been documented in clinical trials outside of pharmacology.<ref>{{cite journal |last1=JUL 2009 |first1=The Pharmaceutical Journal31 |title=An example of problems that arise from clinical trials and how to avoid them |journal=Pharmaceutical Journal |date=31 July 2009 |volume=283 |pages=129-130 |url=https://www.pharmaceutical-journal.com/opinion/comment/an-example-of-problems-that-arise-from-clinical-trials-and-how-to-avoid-them/10973550.article?firstPass=false |accessdate=24 April 2019 |language=en}}</ref> |
|||
===Pain=== |
|||
A 2018 [[meta-analysis]] found that assessment of blinding was reported in only 23 out of 408 [[randomized controlled trials]] for chronic pain (5.6%). The study concluded upon analysis of pooled data that the overall quality of the blinding was poor, and the blinding was "not successful." Additionally, both pharmaceutical sponsorship and the presence of side effects were associated with lower rates of reporting assessment of blinding.<ref>{{cite journal |last1=Colagiuri |first1=Ben |last2=Sharpe |first2=Louise |last3=Scott |first3=Amelia |title=The Blind Leading the Not-So-Blind: A Meta-Analysis of Blinding in Pharmacological Trials for Chronic Pain |journal=The Journal of Pain |date=September 2018 |doi=10.1016/j.jpain.2018.09.002 |url=https://www.jpain.org/article/S1526-5900(18)30603-5/abstract |accessdate=22 April 2019 |language=English |issn=1526-5900}}</ref> |
|||
===Depression=== |
|||
Studies have found evidence of extensive unblinding in [[antidepressant]] trials: at least three quarters of patients are able to correctly guess their treatment assignment.<ref>{{cite journal |last1=Perlis |first1=Roy H. |last2=Ostacher |first2=Michael |last3=Fava |first3=Maurizio |last4=Nierenberg |first4=Andrew A. |last5=Sachs |first5=Gary S. |last6=Rosenbaum |first6=Jerrold F. |title=Assuring That Double-Blind Is Blind |journal=American Journal of Psychiatry |date=1 March 2010 |volume=167 |issue=3 |pages=250–252 |doi=10.1176/appi.ajp.2009.09060820 |url=https://ajp.psychiatryonline.org/doi/full/10.1176/appi.ajp.2009.09060820 |accessdate=23 April 2019 |issn=0002-953X}}</ref> Unblinding also occurs in clinicians.<ref>{{cite journal |title=Side effects and the "blindability" of clinical drug trials |journal=American Journal of Psychiatry |date=December 1992 |volume=149 |issue=12 |pages=1730–1731 |doi=10.1176/ajp.149.12.1730 |url=https://ajp.psychiatryonline.org/doi/abs/10.1176/ajp.149.12.1730?journalCode=ajp |accessdate=22 April 2019 |issn=0002-953X}}</ref> Better blinding of patients and clinicians reduces [[effect size]]. Researchers concluded that unblinding inflates effect size in antidepressant trials.<ref>{{cite journal |last1=Hardy |first1=Rebecca |last2=Wessely |first2=Simon |last3=Moncrieff |first3=Joanna |title=Meta-analysis of trials comparing antidepressants with active placebos |journal=The British Journal of Psychiatry |date=1998 |volume=172 |issue=3 |pages=227–231 |doi=10.1192/bjp.172.3.227 |url=https://www.cambridge.org/core/journals/the-british-journal-of-psychiatry/article/metaanalysis-of-trials-comparing-antidepressants-with-active-placebos/1222DAAE4794E07C1D38A70D8D557945 |accessdate=23 April 2019 |language=en |issn=0007-1250}}</ref><ref>{{cite journal |last1=Greenberg |first1=Roger P. |last2=Bornstein |first2=Robert F. |last3=Greenberg |first3=Michael D. |last4=Fisher |first4=Seymour |title=A meta-analysis of antidepressant outcome under "blinder" conditions. |journal=Journal of Consulting and Clinical Psychology |date=1992 |volume=60 |issue=5 |pages=664–669 |doi=10.1037/0022-006X.60.5.664 |url=https://psycnet.apa.org/record/1993-06735-001 |accessdate=22 April 2019 |language=en}}</ref> Some researchers believe that antidepressants are not effective for the treatment of depression, and only outperform placebos due to [[observational error|systematic error]]. These researchers argue that antidepressants are just [[active placebo|active placebos]].<ref name=Ioannidis2008>{{cite journal |vauthors=Ioannidis JP |title=Effectiveness of antidepressants: an evidence myth constructed from a thousand randomized trials? |journal=Philos Ethics Humanit Med |volume=3 |issue= |pages=14 |date=2008 |pmid=18505564 |pmc=2412901 |doi=10.1186/1747-5341-3-14 |url=}}</ref> |
|||
===Zinc=== |
|||
One clinical trial on the efficacy of zinc supplementation in the treatment of the common flue experienced failure of blinding due to the strong metallic after-taste of zinc.<ref>{{cite journal |last1=Hróbjartsson |first1=A |last2=Boutron |first2=I |title=Blinding in Randomized Clinical Trials: Imposed Impartiality |journal=Clinical Pharmacology & Therapeutics |date=12 October 2011 |volume=90 |issue=5 |pages=732–736 |doi=10.1038/clpt.2011.207 |url=https://ascpt.onlinelibrary.wiley.com/doi/full/10.1038/clpt.2011.207 |accessdate=22 April 2019 |language=en |issn=1532-6535}}</ref> |
|||
===Acupuncture=== |
|||
[[File:Real Needle vs. Sham Needle.png|thumb|Real Needle vs. Sham Needle|500px|<center>Sham needles do not penetrate the skin]] |
|||
While the possibility of blinded trials on [[acupuncture]] is controversial, a 2003 review of 47 [[randomized controlled trial]]s found no fewer than four methods of blinding patients to acupuncture treatment: 1) superficial needling of true acupuncture points, 2) use of acupuncture points which are not indicated for the condition being treated, 3) insertion of needles outside of true acupuncture points, and 4) the use of placebo needles which are designed not to penetrate the skin. The authors concluded that there was "no clear association between type of sham intervention used and the results of the trials."<ref>{{cite journal |last1=Dincer |first1=F |last2=Linde |first2=K. |title=Sham interventions in randomized clinical trials of acupuncture—a review |journal=Complementary Therapies in Medicine |date=December 2003 |volume=11 |issue=4 |pages=235–242 |doi=10.1016/S0965-2299(03)00124-9 |url=https://www.sciencedirect.com/science/article/pii/S0965229903001249 |accessdate=23 April 2019}}</ref> |
|||
A 2018 study on acupuncture which used needles that did not penetrate the skin as a sham treatment found that 68% of patients and 83% of acupuncturists correctly identified their group allocation. The authors concluded that the blinding had failed, but that more advanced placebos may someday offer the possibility of well-blinded studies in acupuncture.<ref>{{cite journal |last1=Vase |first1=L |last2=Baram |first2=S |last3=Takakura |first3=N |last4=Takayama |first4=M |last5=Yajima |first5=H |last6=Kawase |first6=A |last7=Schuster |first7=L |last8=Kaptchuk |first8=TJ |last9=Schou |first9=S |last10=Jensen |first10=TS |last11=Zachariae |first11=R |last12=Svensson |first12=P |title=Can acupuncture treatment be double-blinded? An evaluation of double-blind acupuncture treatment of postoperative pain. |journal=PloS one |date=2015 |volume=10 |issue=3 |pages=e0119612 |doi=10.1371/journal.pone.0119612 |pmid=25747157 |url=https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4352029/ |accessdate=23 April 2019 |issn=1932-6203}}</ref> |
|||
==Significance== |
|||
Bias due to poor blinding tends to favor the experimental group, resulting in inflated effect size and risk of [[type I error]].<ref name=":0" /> Success or failure of blinding is rarely reported or measured; it is implicitly assumed that experiments reported as "blind" are truly blind.<ref name=Bello2014 /> Critics have pointed out that without assessment and reporting, there is no way to know if a blind succeed. This shortcoming is especially concerning given that even a small error in blinding can produce a [[statistical significance|statistically significant]] result in the absence of any real difference between test groups when a study is sufficiently [[Power (statistics)|powered]] (i.e. statistical significance is not robust to bias). As such, many statistically significant results in randomized controlled trials may be caused by error in blinding.<ref>{{cite journal |last1=Siegfried |first1=Tom |title=Odds are, it's wrong: Science fails to face the shortcomings of statistics |journal=Science News |date=2010 |volume=177 |issue=7 |pages=26–29 |doi=10.1002/scin.5591770721 |url=https://onlinelibrary.wiley.com/doi/abs/10.1002/scin.5591770721 |accessdate=22 April 2019 |language=en |issn=1943-0930}}</ref> Some researchers have called for the mandatory assessment of blinding efficacy in clinical trials.<ref name=Kolahi2009 /> |
|||
== References == |
|||
{{reflist}} |
|||
==External links== |
|||
*[https://adc.bmj.com/content/98/9/656 Blinding in pharmacological trials: the devil is in the details] |
|||
{{Medical research studies}} |
|||
{{Experimental design}} |
|||
{{DEFAULTSORT:Unblinding}} |
|||
[[Category:Medical statistics]] |
|||
[[Category:Design of experiments]] |
|||
[[Category:Statistical hypothesis testing]] |
|||
[[Category:Clinical research]] |
Latest revision as of 06:15, 11 June 2019
Redirect to: