The effects of industry funding and positive outcomes in the interpretation of clinical trial results: a randomized trial among Dutch psychiatrists.
Conflict of interest
Credibility
Industry funding
Industry funding disclosure
Methodological quality
Perceived clinical relevance
Perceived credibility
Positive outcome bias
Psychiatrists
Journal
BMC medical ethics
ISSN: 1472-6939
Titre abrégé: BMC Med Ethics
Pays: England
ID NLM: 101088680
Informations de publication
Date de publication:
18 09 2019
18 09 2019
Historique:
received:
29
12
2017
accepted:
03
09
2019
entrez:
20
9
2019
pubmed:
20
9
2019
medline:
14
4
2020
Statut:
epublish
Résumé
Most studies are inclined to report positive rather than negative or inconclusive results. It is currently unknown how clinicians appraise the results of a randomized clinical trial. For example, how does the study funding source influence the appraisal of an RCT, and do positive findings influence perceived credibility and clinical relevance? This study investigates whether psychiatrists' appraisal of a scientific abstract is influenced by industry funding disclosures and a positive outcome. Dutch psychiatrists were randomized to evaluate a scientific abstract describing a fictitious RCT for a novel antipsychotic drug. Four different abstracts were created reporting either absence or presence of industry funding disclosure as well as a positive or a negative outcome. Primary outcomes were the perceived credibility and clinical relevance of the study results (10-point Likert scale). Secondary outcomes were the assessment of methodological quality and interest in reading the full article. Three hundred ninety-five psychiatrists completed the survey (completion rate 45%). Industry funding disclosure was found not to influence perceived credibility (Mean Difference MD 0.12; 95% CI - 0.28 to 0.47, p?) nor interpretation of its clinical relevance (MD 0.14; 95% CI - 0.54 to 0.27, p?). A negative outcome was perceived as more credible than a positive outcome (MD 0.81 points; 95% Confidence Interval (CI) 0.43 to 1.18, p?), but did not affect clinical relevance scores (MD -0.14; 95% CI - 0.54 to 0.27). In this study, industry funding disclosure was not associated with the perceived credibility nor judgement of clinical relevance of a fictional RCT by psychiatrists. Positive study outcomes were found to be less credible compared to negative outcomes, but industry funding had no significant effects. Psychiatrists may underestimate the influence of funding sources on research results. The fact that physicians indicated negative outcomes to be more credible may point to more awareness of existing publication bias in the scientific literature.
Sections du résumé
BACKGROUND
Most studies are inclined to report positive rather than negative or inconclusive results. It is currently unknown how clinicians appraise the results of a randomized clinical trial. For example, how does the study funding source influence the appraisal of an RCT, and do positive findings influence perceived credibility and clinical relevance? This study investigates whether psychiatrists' appraisal of a scientific abstract is influenced by industry funding disclosures and a positive outcome.
METHODS
Dutch psychiatrists were randomized to evaluate a scientific abstract describing a fictitious RCT for a novel antipsychotic drug. Four different abstracts were created reporting either absence or presence of industry funding disclosure as well as a positive or a negative outcome. Primary outcomes were the perceived credibility and clinical relevance of the study results (10-point Likert scale). Secondary outcomes were the assessment of methodological quality and interest in reading the full article.
RESULTS
Three hundred ninety-five psychiatrists completed the survey (completion rate 45%). Industry funding disclosure was found not to influence perceived credibility (Mean Difference MD 0.12; 95% CI - 0.28 to 0.47, p?) nor interpretation of its clinical relevance (MD 0.14; 95% CI - 0.54 to 0.27, p?). A negative outcome was perceived as more credible than a positive outcome (MD 0.81 points; 95% Confidence Interval (CI) 0.43 to 1.18, p?), but did not affect clinical relevance scores (MD -0.14; 95% CI - 0.54 to 0.27).
CONCLUSIONS
In this study, industry funding disclosure was not associated with the perceived credibility nor judgement of clinical relevance of a fictional RCT by psychiatrists. Positive study outcomes were found to be less credible compared to negative outcomes, but industry funding had no significant effects. Psychiatrists may underestimate the influence of funding sources on research results. The fact that physicians indicated negative outcomes to be more credible may point to more awareness of existing publication bias in the scientific literature.
Identifiants
pubmed: 31533704
doi: 10.1186/s12910-019-0405-7
pii: 10.1186/s12910-019-0405-7
pmc: PMC6749641
doi:
Types de publication
Journal Article
Randomized Controlled Trial
Langues
eng
Sous-ensembles de citation
IM
Pagination
64Références
CMAJ. 2004 Feb 17;170(4):477-80
pubmed: 14970094
N Engl J Med. 2008 Jan 17;358(3):252-60
pubmed: 18199864
Cochrane Database Syst Rev. 2009 Jan 21;(1):MR000006
pubmed: 19160345
BMJ. 2015 Sep 16;351:h4320
pubmed: 26376805
J Gen Intern Med. 2000 Dec;15(12):881-4
pubmed: 11119185
JAMA Psychiatry. 2015 May;72(5):500-10
pubmed: 25806940
BMJ. 2014 Nov 21;349:g6501
pubmed: 25416499
PLoS Med. 2012;9(3):e1001189
pubmed: 22448149
N Engl J Med. 2012 Sep 20;367(12):1119-27
pubmed: 22992075
Tijdschr Psychiatr. 2011;53(7):393-403
pubmed: 21748704
PLoS One. 2013 Jul 05;8(7):e66844
pubmed: 23861749
J Med Ethics. 2010 May;36(5):265-70
pubmed: 20448003
J Clin Epidemiol. 1997 Oct;50(10):1129-36
pubmed: 9368521
Arch Intern Med. 2010 Nov 22;170(21):1934-9
pubmed: 21098355
N Engl J Med. 2015 May 7;372(19):1860-4
pubmed: 25946288
JAMA. 2004 May 26;291(20):2457-65
pubmed: 15161896
Ann Intern Med. 2010 Aug 3;153(3):158-66
pubmed: 20679560
Cochrane Database Syst Rev. 2014 Oct 01;(10):MR000035
pubmed: 25271098
World Psychiatry. 2011 Jun;10(2):155-9
pubmed: 21633695
JAMA. 2012 Feb 15;307(7):669-70
pubmed: 22337676
CNS Drugs. 2013 Jun;27(6):457-68
pubmed: 23696308
BMJ. 2003 May 31;326(7400):1167-70
pubmed: 12775614
PLoS Med. 2008 Nov 25;5(11):e217; discussion e217
pubmed: 19067477
Cochrane Database Syst Rev. 2012 Dec 12;12:MR000033
pubmed: 23235689
JAMA. 2008 Sep 3;300(9):1069-71
pubmed: 18768418
Am J Psychiatry. 2005 Oct;162(10):1957-60
pubmed: 16199844