Lacking quality in research: Is behavioral neuroscience affected more than other areas of biomedical science?

Abstract

There are many reasons why novel therapeutics fail in clinical trials but these failures are often attributed to lacking quality of preclinical data. These problems are not limited to any specific therapeutic area, academic or industrial research and are due in large part to several generic factors influencing research quality (e.g., related to definition of pre-specified endpoints, principles of study design and analysis, biased reporting, and lack of proper training). Yet, neuroscience drug discovery is often said to be affected more than the other fields. Within neuroscience, behavioral studies are the most blamed for being poorly designed, underpowered and mis-reported and there are indeed several factors that may be rather unique for behavioral research, such as a multitude of environmental conditions that are difficult to control and that are often not reported, ethical concerns about in vivo research and the pressure to reduce animal numbers, contributing to under-powered studies, and the complexity of study design and analysis, creating too much room for post hoc data massaging and selective reporting. Also, the blood-brain barrier as a frequently neglected complicating factor has to be considered in CNS research. The importance of these factors is increasingly recognized and urgent efforts are needed to demonstrate that behavioral methods of preclinical neuroscience research deliver results that can be as robust as with the non-behavioral methods Until this goal is achieved, behavioral neuroscience and neuroscience in general may be losing young talent, CNS drug discovery may lack the needed investment and this field may indeed be amongst the most affected by the current preclinical data quality crisis.

Keywords: Behavioral neuroscience; CNS drug discovery; Reproducibility; Research quality.