Animal Studies Relating to Neurological Disorders and

¶ … animal studies relating to neurological disorders and how they are often ostensibly biased is the subject of the article covered for this brief report. The article was published earlier in 2013 and appeared in PLOS and on the internet (Tsilidis, 2013). The article did a study of a series of prior studies that used animal testing and research to look at neurological disorder. In total, a total of nearly 4,500 data sets were looked at.

The diseases that were followed the most included Alzheimer disease, experimental autoimmune encephalitis, focal ischemia, intracerebral hemorrhage, Parkinson's disease, and spinal cord injuries in general. The study focused on the significance bias of these studies. To that end, it found that 919 out of the 4,445 results were expected vs. The 1,719 that were observed. This excess significance was found across all of the disorders mentioned above and were not limited to just one or several of the disorders listed (Tsilidis, 2013).

The study continues by stating that they found eight interventions with strong and significant statistically significant benefits in animal models without corresponding evidence of small-study effects or excess significance errors in the study results. This result is important, as noted in the next paragraph, because there is a common thread in a lot of studies whose results are then extended to human trials because the results hardly ever hold as being as good or beneficial in the human trials.

However, the study queries itself and the listener by asking whether focusing on studies without excess significance and with an ample enough study size would yield better results in human studies as the quality of result would be better, in theory. Even so, this particular study was up front in that it had specific reservations about the limitations of the study.

Among them, they note that excess significance can mean that there is a problem with the study, up to and including bias, but excess significance alone absolutely does not mean that there is definitely bias. Rather, it just means there could and might be, not that it's proven or that it actually exists. They also note that given the relatively small size of some of the samples, this makes parsing or verification of any of the relevant results quite difficult to pull off. Indeed, the median sample size was 16.

As such, such a study would only allow for eight pairs of two each and that is on the small side if being definitive and sure is the goal. Conclusion In the end, the study reviewed for this report was clearly focuses on a few facets of prior studies and how they offer an incorrect if not incomplete picture of what is being research. Indeed, only doing.