Fashionable scientific areas induce a bandwagon effect, making it more likely that investigators will be keen to take part and more likely that their approach will be less critical than it should be; they will, therefore, concentrate on trying to confirm the findings of others rather than trying to falsify them. Negative findings may be less likely to be published (see Publication bias). In such cases, the positive predictive value becomes progressively smaller as more studies accumulate, making it less likely that the published findings are true; this may also be partly due to regression to the mean.
Furthermore, the current popularity of a topic can affect how much publicity is given to it. Current excitement about a particular topic can lead to the inappropriate publication of research findings (or opinions) earlier, or in greater volume, or with greater prominence, or with less scrutiny than would otherwise be the case in the normal course of events.
Similarly, topics that are regarded as controversial may be given more publicity by editors who are keen to draw attention to their journals, even though they may not be aware of it or may subsequently deny that it was so, and by journalists looking for an interesting copy.
Controversial topics may lead to a heated debate about the relative merits of a hypothesis or intervention. Examples include the current debate about the use of statins, the long-standing debate about the efficacy of antidepressants, such as the SSRIs, and previous debates about the use of hormone replacement therapy and the falsification of a link between vaccination and autism. Often the most critical question – what is the benefit to harm balance – is neglected in such debates.
The impact of hot stuff bias has not been formally investigated and must come from one’s knowledge, experiences and observations of how the research environment and human behaviour interact.
Researchers should resist the seductive effect of fashionable science. They should adhere to stringent standards for reporting studies; they should publish research protocols in advance, and their publications should state when and in what ways they have deviated from published protocols. Adherence to study protocols or otherwise full documentation of changes to study protocols reduces the likelihood that biases will affect research outputs.
Protocols should include, whenever relevant, guidelines for premature termination of studies and release of results.
Funders should cap the amounts of support they are willing to give to fashionable topics compared with important non-fashionable research.
Editors of journals should strive for excellence in the review and editorial processes, to ensure that they publish high-quality research. They should consider the implications of publishing simply because a topic is fashionable or controversial. They should ensure that authors adhere to stringent standards for reporting studies, and reject studies whose protocols have not been published.