Evaluation of Excess Significance Bias in Animal Studies of Neurological Diseases

Tsilidis, Konstantinos K.; Panagiotou, Orestis A.; Sena, Emily S.; Aretouli, Eleni; Εvangelou, Εvangelos; Howells, David W.; Salman, Rustam Al-Shahi; Macleod, Malcolm; Ioannidis, John P. A.

doi:10.1371/journal.pbio.1001609

Cited by 250 publications

(195 citation statements)

References 48 publications

Supporting

Mentioning

184

Contrasting

Order By: Relevance

“…Serious concerns have, however, been raised about an excess of false positive results contami nating the neuroscience literature [1][2][3][4] . Controlling the false positive rate is critical, since theoretical progress in the neuroscience field relies fundamentally on drawing correct conclusions from experi mental research.…”

Section: P E R S P E C T I V Ementioning

confidence: 99%

“…Various recent publications force neuroscientists to acknowledge the possibility that the harvest of their hard labor is contaminated by an abundance of false positive effects [1][2][3][4] . Nested designs are ubiq uitous in neuroscience, and an increased awareness of the problem of nesting in both researchers and reviewers will prevent costly and time consuming quixotic pursuits of spurious effects, thereby assist ing progress in the understanding of the nervous system.…”

Section: Box 3 Estimating the Power To Detect An Experimental Effectmentioning

confidence: 99%

See 1 more Smart Citation

A solution to dependency: using multilevel analysis to accommodate nested data

et al. 2014

View full text Add to dashboard Cite

Section: P E R S P E C T I V Ementioning

confidence: 99%

Section: Box 3 Estimating the Power To Detect An Experimental Effectmentioning

confidence: 99%

A solution to dependency: using multilevel analysis to accommodate nested data

et al. 2014

View full text Add to dashboard Cite

“…This 'publish or perish' pressure may also lead investigators to neglect findings, not conform to their hypothesis and instead to go for the desired outcome, may bias authors to publish positive, statistically significant results (Tsilidis et al 2013) and to abandon negative results that they believe journals are unlikely to publish (the file-drawer phenomenon; Franco et al 2014). This pressure to publish may even entice investigators to make post hoc alterations to hypotheses, data or statistics (Motulsky 2014;O'Boyle et al 2014), so that there is a more compelling story to tell, essentially transforming uninteresting results into top-notch science (the chrysalis effect; O'Boyle et al 2014).…”

Section: What Are the Issues At Hand?mentioning

confidence: 99%

Preclinical data reproducibility for R&D—the challenge for neuroscience

Steckler

2015

Psychopharmacology

View full text Add to dashboard Cite

“…Results of animal biomedical experiments have vastly failed to translate into human clinical trials which are mostly attributed because of differences in the underlying biology between humans and animals to shortcomings in the experimental design or to bias in the reporting of results from animal studies [28]. Animal studies have been commented widely to be methodologically weak [29] which likely continues to provide the biological basis for epidemiological investigation but substantial improvement is needed in how it has to be conducted and synthesized to improve the predictability of animal studies for the human condition.…”

Section: Preclinical Studies Overall Scenariomentioning

confidence: 99%

Preclinical Research: A Rise or Dawn

Manjunath¹

2018

PPIJ

View full text Add to dashboard Cite

IntroductionComplex processes are involved in developing a new drug along with its time and costs consumption. Prior to launching, it requires variety of evidences to show its potential activity [1]. Key step in drug discovery is considered to be the target discovery which ranges from molecular entities to biological phenomenon [2]. As there are urges in translating scientific discoveries into practical importance, preclinical studies using animal models play an important role in evaluating the efficacy and safety under clinical scenario [3]. After enormous debates, it is concluded that the benefits of animal research is worth to manage public health and medical research [4]. Subsequently, animal studies are justifiable if done in a way to minimize pain in animal models and wherever applicable alternative methods are explored and avoiding unnecessary cruel treatment to animals [5]. However, various reports have enforced on the failures of trails that has never capitalized after thorough observation of safety and efficacy in animal models which mainly portraits the requirement of robustness in preclinical studies [6][7][8][9][10]. For instance, reports say that, almost 95% of drugs that enter clinical trials related to stroke or septic shock do not get launched in the market though enormous expenditures have been dealt with the basic studies and drug discoveries [11] and similarly, the average success rate in clinical cancer trials is less than 8% after preclinical estimation [10].Still it is recommended to circumvent preclinical studies because of the scientific, practical and ethical reasons, to conduct studies in humans [6], and thus prior to this attempt, preclinical drug combination studies in vitro and/or in animals has to be carried out to obtain the base for studies in human [12]. Though the debate over use of the animals in medical experiments has a long and unpleasant history, eminent researches along with Nobel Prize winners have signed a declaration provided by the Research Defense Society has made claim that experiments on animals have made important contribution to advances in medicine and surgery [13]. Research focused on pain research has shown enormous rate of failures and increased costs of drug development this is because of phenotypic differences and complexity in behavioral alterations along with the changes in daily activities and psychological disturbances [14,15]. Moreover, the test and models of nociception has been questioned in the aspect of appropriateness between tests, models and procedures [16]. Thus in order to develop robustness in pain management and to effectively gather required evidences, it is better to evaluate pain states in higher order preclinical species where we can predict whether biological mechanisms and specific compounds have relevance for clinical pain better than rodent species. The pain models in non-rodent species may enhance the speed, and in addition reduce costs along with the increased probable results for a successful analgesic development [17]. Hiccups in...

show abstract

Evaluation of Excess Significance Bias in Animal Studies of Neurological Diseases

Abstract: The evaluation of 160 meta-analyses of animal studies on potential treatments for neurological disorders reveals that the number of statistically significant results was too large to be true, suggesting biases.

Cited by 250 publications

References 48 publications

A solution to dependency: using multilevel analysis to accommodate nested data

A solution to dependency: using multilevel analysis to accommodate nested data

Preclinical data reproducibility for R&D—the challenge for neuroscience

Preclinical Research: A Rise or Dawn

Contact Info

Product

Resources

About