In Detail (Has Detail)
From The Embassy of Good Science
Here you can go in more detail and even add subheaders to give more relevant background information.
- ⧼SA Foundation Data Type⧽: Text
[
The main method for assessing the efficacy and safety of novel medications and other medical technology is clinical trials. The results of these research, when published in peer-reviewed publications, not only large medical community a scientific foundation for making treatment decisions, but also fund further research. However, pharmaceutical firms with large financial interests in the items being tested finance the majority of clinical trials. Additionally, they frequently pay the scientists who plan, carry out, evaluate, and report clinical trials in the form of wages or consultancy fees. This raises important questions like, should those who have a financial interest in clinical trial results be so invested in its execution? What potential sources of bias may the industry funding of these research have? How are reliable and impartial trial report data ensured by medical journals? Also, it brings a certain stigma and mistrust from the general public.
When searching the literature notes and guidelines can be found on proper steps when publishing a research with funding from external sources. All research articles should have a funding acknowledgement statement included in the manuscript in the form of a sentence under a separate heading entitled ‘Funding’ directly after your Acknowledgements and Declaration of Conflicting Interests, if applicable, and prior to any Notes and your References. The funding agency should be written out in full, followed by the grant number in square brackets, see following example (the text in bold is mandatory, unless specified otherwise by the journal).
A comma and space should be used to separate multiple grant numbers. When more than one agency contributed to funding a study, a semicolon should be used to separate each agency, and the word "and" should come before the last funder.
Industry-funded studies will only be taken into consideration by SAGE publications if sponsorship is explicitly disclosed in the manuscript (where applicable). The funding source's involvement in the study's design, data analysis, and interpretation must also be disclosed by the authors.
According to some researchers, to assure impartiality in clinical research industry-academia partnerships should only continue if academic medical centers take full ownership of the planning, execution, analysis, and reporting of clinical trials. There is an idea of setting up conflict-of-interest committees at academic institutions to maintain a close eye on the financial motivations of both clinician-investigators and institutional decision-makers. Such approaches may aid in reducing the possibility of bias in industry-sponsored research by introducing checks and balances for academic-industry collaborations.
The term HARKing was coined in a seminal article by Kerr<sup>[1]</sup> and is usually used synonymously with accommodational hypothesizing<sup>[2]</sup> and presenting post hoc hypothesis as a priori (PPHA).<sup>[3]</sup> Kerr identified twelve potential costs of HARKing:
1. Translating Type I errors into hard-to-eradicate theory.
2. Propounding theories that cannot (pending replication) pass Popper’s disconfirmability test.
3. Disguising post hoc explanations as a priori explanations (when the former ted also be more ad hoc, and consequently, less useful).
4. Not communicating valuable information about what did not work.
5. Taking unjustified statistical licence.
6. Presenting an inaccurate model of science to students.
7. Encouraging “fudging” in other grey areas.
8. Making us less receptive to serendipitous findings.
9. Encouraging adoption of narrow, context-bound new theory.
10. Encouraging retention of too-broad, disconfirmable old theory.
11. Inhibiting identification of plausible alternative hypotheses.
12. Implicitly violating basic ethical principles.
While Kerr’s article initially was not widely cited, this changed in the wake of the replication crisis and empirical studies into the prevalence and drivers of detrimental research practices and research misconduct. The surge of interest in HARKing worryingly showed that it indeed is a rather prevalent practice. Various studies on the prevalence of detrimental research practices found that a sizeable proportion of researchers (up to 58% in one study) from different disciplines (most notably psychology) did engage in HARKing in the past.<sup>[4]</sup>
To identify measures to reduce HARKing, it is necessary to understand its causes. A key driving factor of HARKing most likely is publication bias: it is much more difficult to publish negative findings than positive findings, and confirmatory research seemingly following a hypothetico-deductive model is generally higher valued than exploratory research, at least in most fields of research. The number of publications, however, still is one of the most important metrics commonly used in researcher evaluation. As a result, researchers have an incentive to publish as much as possible, while the publication system rewards analyses that (seemingly) yield positive findings derived from hypothesis testing research.
One pathway to reduce HARKing thus is changing the incentives for researchers by, for example, evaluating the quality rather than the quantity of publications and recognizing the value of replication studies. The latter also would be facilitated by a comprehensive move towards open science and a recognition of the value of open science practices. Another pathway to reduce HARKing is preregistration because it helps tying the hands of researchers before the data analysis. If researchers decide to preregister a study, they submit a time-stamped paper describing the rationale of their study, the experimental and analytical methods they will use, and their hypotheses. This document cannot be changed at a later stage so that HARKing would be easily detectable and lead to inconsistencies in the line of argument. If the pre-registered study is reviewed, publication is guaranteed if the registered protocol is followed, regardless of the results. Consequently, preregistration and changes in the incentive system are potentially mutually reinforcing.
However, it is worth noting that it is in principle possible to preregister studies after the results are known (PARKing) and thereby reap the reputational benefits coming with what seems to be a commitment to methodological rigor without actually following the practice.<sup>[5]</sup>
Although pure HARKing is unquestionably a detrimental research practice because it misportrays the research process, tends to bias results and ultimately deceives readers, the same cannot necessarily be said about other forms of post hoc hypothesizing. Using the fictional example of a group of epidemiologists conducting a drug trial to cure a new life-threatening disease, Hollenbeck and Wright argue that HARKing is not detrimental to science if it is done transparently and informed by theory, a practice they call THARKing (transparently hypothesizing after the results are known).<sup>[6]</sup> In their example, the epidemiologists initially find no effect of the tested drug, but know of cases where it apparently worked. Discussing about these cases, they recognize that all cured patients they know of are female, yet a reanalysis of the data turns out insignificant, even though the effect size for women is larger than for men. They continue discussing if gender could be an important factor and, drawing on their implicit theoretical knowledge, develop the hypothesis that estrogen levels (that peak at certain ages) might be a crucial moderating variable. A reanalysis of their data corroborates their hypothesis. They publish an article summarizing their study, noting in the discussion section that the age-by-gender interaction was the result of an exploratory analysis conducted after the main effects turned out to be insignificant. Other research teams replicate their study, and eventually a meta-analysis confirms their findings. Hollenbeck and Wright argue that THARKing, unlike secretly hypothesizing after the results are known, SHARKing or pure HARKing), is justifiable if readers are transparently informed that a hypothesis is post hoc rather than a priori in the discussion section of an article (in other words, the introduction in their view should only include a priori hypotheses).
In general, pure HARKing is a detrimental research practice and hampers scientific progress. It can be disincentivized by changes in the research system, such as changes in researcher assessment and increasing preregistration of studies. Transparent post hoc hypothesizing, by contrast, seems justifiable if the exploratory nature of results is clearly stated.
'''References'''
[1] Kerr, N. (1998). HARKing: Hypothesizing After the Results are Known. ''Personality and Social Psychology Review, 2''(3), 196-217. doi:[https://doi.org/10.1207/s15327957pspr0203_4 10.1207/s15327957pspr0203_4]
[2] Hitchcock, C., & Sober, E. (2004). Prediction versus Accommodation and the Risk of Overfitting. ''The British Journal for the Philosophy of Science'', ''55''(1), 1–34. http://www.jstor.org/stable/3541832
[3] Leung, K. (2011). Presenting Post Hoc Hypotheses as A Priori: Ethical and Theoretical Issues. ''Management and Organization Review, 7''(3)'','' 471-479. doi: [https://doi.org/10.1111/j.1740-8784.2011 10.1111/j.1740-8784.2011]
[4] An overview of different studies on the prevalence of HARKing can be found in Table 1 in Rubin, M. (2017). When does HARKing hurt? Identifying when different types of undisclosed post hoc hypothesizing harm scientific progress. ''Review of General Psychology, 21,'' 308-320''.'' doi: [[10.1037/gpr0000128]]
[5] Yamada, Y. (2018). How to Crack Pre-registration: Toward Transparent and Open Science. ''Frontiers in Psychology, 9:1831.'' doi: [https://doi.org/10.3389/fpsyg.2018.01831 10.3389/fpsyg.2018.01831]
[6] Hollenbeck, J. R., & Wright, P. M. (2017). Harking, Sharking, and Tharking: Making the Case for Post Hoc Analysis of Scientific Data. ''International Journal of Qualitative Methods'', ''43''(1), 5-18. [https://doi.org/10.1177/1609406920947600 10.1177/1609406920947600]
Part of the problem is scholarly journals which are prone to only publishing positive results. Changes in publishing policies and fees, especially in the era of digital, publicly available databases and journals, could provide a climate for publishing negative results. Pre-registrations of trials/research can only solve a problem if complete results are published after completion. +
<br />
'"`UNIQ--references-00000289-QINU`"' +
References:
1. Integrity in research collaborations: The Montreal Statement. ''Lancet''. 2013;382(9901):1310. doi:10.1016/S0140-6736(13)62126-1
2. S G, B N, K D. Differing Perceptions Concerning Research Integrity Between Universities and Industry: A Qualitative Study. ''Sci Eng Ethics''. 2018;24(5):1421-1436. doi:10.1007/S11948-017-9965-4
3. RM R, A T, M DC, et al. Challenges of non-commercial multicentre North-South collaborative clinical trials. ''Trop Med Int Health''. 2013;18(2):237-241. doi:10.1111/TMI.12036
4. The European Code of Conduct for Research Integrity. Accessed October 11, 2021. www.allea.org
5. 3rd World Conference on Research Integrity. Montreal Statement on Research Integrity in Cross-Boundary Research Collaborations. 2013;(May):2013. '"`UNIQ--nowiki-000006E7-QINU`"' Statement English.pdf
6. Albert T, Wager E. How to handle authorship disputes: a guide for new researchers. Published online September 1, 2009. doi:10.24318/COPE.2018.1.1 +
References:
1. The European Code of Conduct for Research Integrity. Accessed October 11, 2021. www.allea.org
2. TP C. Authorship matrix: a rational approach to quantify individual contributions and responsibilities in multi-author scientific articles. ''Sci Eng Ethics''. 2014;20(2):345-361. doi:10.1007/S11948-013-9454-3
3. Vasilevsky NA, Hosseini M, Teplitzky S, et al. Is authorship sufficient for today’s collaborative research? A call for contributor roles. '''"`UNIQ--nowiki-000006E5-QINU`"'''. 2020;28(1):23-43. doi:10.1080/08989621.2020.1779591
4. 3rd World Conference on Research Integrity. Montreal Statement on Research Integrity in Cross-Boundary Research Collaborations. 2013;(May):2013. '"`UNIQ--nowiki-000006E6-QINU`"' Statement English.pdf +
#Kearney, Margaret H. Predatory Publishing: What Authors Need to Know. Research in Nursing & Health 2015; 38 (1): 1–3. doi:10.1002/nur.21640. '"`UNIQ--nowiki-000001F8-QINU`"'.
#Bohannon J. Who's Afraid of Peer Review? Science 2013;342 (6154): 60–5. doi:10.1126/science.342.6154.60
#Sorokowski P. "Predatory journals recruit fake editor". Nature 2017; 543 (7646): 481–483.
#Eriksson, Stefan; Helgesson, Gert (7 October 2016). "The false academy: predatory publishing in science and bioethics". Medicine, Health Care and Philosophy. 20 (2): 163–170. doi:10.1007/s11019-016-9740-3
+
Preclinicaltrials.eu is an international registration platform dedicated to animal studies that was launched in 2018 by a team of Dutch researchers. With several features, the platform permits a fast and efficient (pre)registration, data sharing, and collaboration:
*'''Free''' to access and use, with a '''short and focussed''' form to facilitate preregistration.
*Preregistration under embargo is available to protect '''privacy''' and '''intellectual property.''' Protocols also receive a time-stamp to prove '''authenticity'''.
*Track changed amendments are available to enable '''flexibility'''.
*Anonymised personal details and required login to view protocols ensure '''confidentiality'''.
*Protocol get a persistent identifier to use in grants or papers, which promotes '''FAIR data.'''
*Researchers from the same field or working on similar topics can reach each other via encrypted messages and form '''collaborations'''.
Moreover, preclinicaltrials.eu provides several resources to guide researchers with their preregistration.
To know more about this registry, you may check their introduction video: https://www.youtube.com/watch?v=xYjLvDBTsV4
More knowledge on how to use the platform can be seen on this walkthrough: https://www.youtube.com/watch?v=qLu1fXYumyk
For further information on preregistering at Preclinicaltrials.eu, you may check the registry's website: www.preclinicaltrials.eu or e-mail the Preclinicaltrials.eu team at info@preclinicaltrials.eu +
'''The start of Superb Supervision'''
“Some people, when you let them swim, come out really well. It is a bit a of a matter of luck and not everyone has it. A little thing can cause unfortunate struggles where a supervisor could have given you the support that you deserve.” (Tamarinde Haven, translated from Dutch)
The motivation to start a course for supervisors goes a long way back, according to PhD-candidate and one of the initiators Tamarinde Haven. During her PhD project on academic research climate, a survey within AmsterdamUMC research institutes revealed inadequate supervision to be a major concern. To improve this situation, 'Superb Supervision' courses for Junior and for Senior researchers were developed. The courses aim to improve supervision skills and to encourage responsible research.
'''Resources'''
Superb Supervision first started with a pilot version, which was very positively evaluated. The current program has been running since January 2020. The course developers (Joeri Tijdink, Lex Bouter, Louise Mennen and Marlies Stouthard) come from different universities in Amsterdam. The course is promoted through the webpage of AmsterdamUMC. Superb Supervision hopes to be sustainable through course fees alone. Whilst this might demotivate researchers to register, a positive side-effect is that people usually not drop out.
'''Evaluation'''
After each day of the course, an evaluation takes place in which the relevance, topics, and duration of the course are discussed. Reactions so far have been positive. Supervisors who sign up for the Superb Supervision course are however usually already aware of the importance of good supervision. The course is voluntary, so it is debatable if the 'bad supervisors' are reached. +
