Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Why current statistics of complementary alternative medicine clinical trials is invalid

Pandolfi, Maurizio LU and Carreras, Giulia (2018) In Journal of Clinical Medicine 7(6).
Abstract

It is not sufficiently known that frequentist statistics cannot provide direct information on the probability that the research hypothesis tested is correct. The error resulting from this misunderstanding is compounded when the hypotheses under scrutiny have precarious scientific bases, which, generally, those of complementary alternative medicine (CAM) are. In such cases, it is mandatory to use inferential statistics, considering the prior probability that the hypothesis tested is true, such as the Bayesian statistics. The authors show that, under such circumstances, no real statistical significance can be achieved in CAM clinical trials. In this respect, CAM trials involving human material are also hardly defensible from an ethical... (More)

It is not sufficiently known that frequentist statistics cannot provide direct information on the probability that the research hypothesis tested is correct. The error resulting from this misunderstanding is compounded when the hypotheses under scrutiny have precarious scientific bases, which, generally, those of complementary alternative medicine (CAM) are. In such cases, it is mandatory to use inferential statistics, considering the prior probability that the hypothesis tested is true, such as the Bayesian statistics. The authors show that, under such circumstances, no real statistical significance can be achieved in CAM clinical trials. In this respect, CAM trials involving human material are also hardly defensible from an ethical viewpoint.

(Less)
Please use this url to cite or link to this publication:
author
and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Bayesian statistics, Complementary alternativemedicine (cam), P-value, Scientific plausibility
in
Journal of Clinical Medicine
volume
7
issue
6
article number
138
publisher
MDPI AG
external identifiers
  • scopus:85114275906
ISSN
2077-0383
DOI
10.3390/jcm7060138
language
English
LU publication?
yes
id
c1d80236-b8eb-402d-b5f4-f4b9b6ee2d88
date added to LUP
2021-10-05 12:18:51
date last changed
2022-04-27 04:48:37
@article{c1d80236-b8eb-402d-b5f4-f4b9b6ee2d88,
  abstract     = {{<p>It is not sufficiently known that frequentist statistics cannot provide direct information on the probability that the research hypothesis tested is correct. The error resulting from this misunderstanding is compounded when the hypotheses under scrutiny have precarious scientific bases, which, generally, those of complementary alternative medicine (CAM) are. In such cases, it is mandatory to use inferential statistics, considering the prior probability that the hypothesis tested is true, such as the Bayesian statistics. The authors show that, under such circumstances, no real statistical significance can be achieved in CAM clinical trials. In this respect, CAM trials involving human material are also hardly defensible from an ethical viewpoint.</p>}},
  author       = {{Pandolfi, Maurizio and Carreras, Giulia}},
  issn         = {{2077-0383}},
  keywords     = {{Bayesian statistics; Complementary alternativemedicine (cam); P-value; Scientific plausibility}},
  language     = {{eng}},
  number       = {{6}},
  publisher    = {{MDPI AG}},
  series       = {{Journal of Clinical Medicine}},
  title        = {{Why current statistics of complementary alternative medicine clinical trials is invalid}},
  url          = {{http://dx.doi.org/10.3390/jcm7060138}},
  doi          = {{10.3390/jcm7060138}},
  volume       = {{7}},
  year         = {{2018}},
}