Journal of Orofacial Sciences

EDITORIAL
Year
: 2021  |  Volume : 13  |  Issue : 2  |  Page : 83--84

It is High Time that Journals Underscore the Indispensable Nature of Statistics in Scientific Research


Viswa Chaitanya Chandu 
 Department of Public Health Dentistry, Government Dental College and Hospital, Vijayawada, Andra Pradesh, India

Correspondence Address:
Viswa Chaitanya Chandu
Department of Public Health Dentistry, Government Dental College and Hospital, Vijayawada, Andra Pradesh, Postal Code: 520004
India

Abstract

The role of biostatisticians in scientific research is poorly defined. Majority of manuscripts submitted to scientific journals do not receive statistical advice at as many stages as it requires and neither do they get subjected to a thorough statistical review as an integral part of the peer-review mechanism. This raises concerns about the validity of research findings, especially in light of the increasing reports on the incidence of statistical misuse in health-care research. It is high time that journals emphasize on statistical review of research submissions to ensure scientific validity.



How to cite this article:
Chandu VC. It is High Time that Journals Underscore the Indispensable Nature of Statistics in Scientific Research.J Orofac Sci 2021;13:83-84


How to cite this URL:
Chandu VC. It is High Time that Journals Underscore the Indispensable Nature of Statistics in Scientific Research. J Orofac Sci [serial online] 2021 [cited 2022 Jan 18 ];13:83-84
Available from: https://www.jofs.in/text.asp?2021/13/2/83/335848


Full Text



In the era of evidence-based dentistry, a multitude of factors determines the level of evidence of a study. Those times are elapsed where research studies were assessed for their level of evidence exclusively based on the design. As the researchers ascend the hierarchy of levels of evidence, it is incumbent on them to avail the existing evidence, to incorporate strategies that augment the strength of the study, to precisely execute the study in the planned manner, and to pertinently analyze the study data. In the course of keeping up with the aforementioned objectives, it is inevitable that informed statistical support is sought as necessary. However, it is disheartening to acknowledge that poor coordination between health-care researchers and biostatisticians remains an area of significant concern.[1] What is more alarming is that a substantial proportion of scientific journals does not require the submitted manuscripts to be subjected to an independent statistical review; the peer review process in these journals exclusively refers to review by subject experts of the concerned discipline for whom statistical insights need not be an instinctive competence.[2]

There are numerous examples for the publication of exaggerated results glorifying the intervention in reputed scientific journals.[3],[4] Erroneous choice of statistical tests, misinterpretation of statistical outcomes, and undue emphasis on P-values has become common in health sciences research.[5] Though dereliction in seeking statistical advice could be one of the primary reasons for this phenomenon, discriminatory reporting of study results, making questionable assumptions in analyzing the study data, and intentional posthoc shifts in predetermined end points of the trials in favor of the administered interventions have also been vastly reported with the latter being more common in industrial research and the former bearing ready relevance with research in academia. These findings necessitate all the scientific journals to deploy strategies that serve as gatekeepers to ensure scientific validity; a few such strategies are: making data availability mandatory; inclusion of qualified biostatisticians in the editorial boards of scientific journals; ensuring a thorough and rigorous review process. Though it has become increasingly common that journals require authors to provide a data availability statement, it is obvious that mere mandates for stating data availability could not ensure availability of data, and policies relating to data availability must be cautiously drafted and strictly adhered to in order to serve the primary purpose of disseminating valid scientific findings.[6] In addition, the inclusion of biostatisticians into editorial boards should not be a mere formality, but the conduct of an independent statistical review of manuscripts must evolve as a standard. As a first step, journals can provide subject experts with an opportunity to recommend a manuscript for an independent statistical review if required. Such practices increase the likelihood of researchers’ adherence to statistical discipline and go a long way in ensuring validity.

Research calls for insights from multiple domains. It would be wise of us to underscore this fact and integrate statistical advice as necessary at various levels of research for it only magnifies the potential of researchers in health care in making innovative and well-grounded observations.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1Gardner MJ, Altman DG, Jones DR, Machin D. Is the statistical assessment of papers submitted to the “British Medical Journal” effective? Br Med J 1983;286:1485-8.
2Hardwicke TE, Goodman SN. How often do leading biomedical journals use statistical experts to evaluate statistical methods? The results of a survey. PLoS One 2020;15:e0239598.
3Schork NJ. Personalized medicine: time for one-person trials. Nature 2015;520:609-11.
4Hanin L. Why statistical inference from clinical trials is likely to generate false and irreproducible results. BMC Med Res Methodol 2017;17:127.
5Makin TR, Orban de Xivry JJ. Ten common statistical mistakes to watch out for when writing or reviewing a manuscript. Elife 2019;8:e48175.
6McGuinness LA, Sheppard AL. A descriptive analysis of the data availability statements accompanying medRxiv preprints and a comparison with their published counterparts. PLoS One 2021;16:e0250887.