Structuring overall goals for biomedical research groups -- thinking about some general options

A recent discussion was raised within our research group on whether we should go from here in terms of recruiting new researchers and how exactly we wanted to move our research forward. Of course there are no right or wrong answers, but here are some general thoughts to elicit discussion:

1. Working with novice vs. established researchers: One of the first points was on whether we should focus our attention on training clinicians (e.g., students from Medicine and other clinical areas, as well as residents, and fellows), graduate students (undergoing MSc and PhD degrees in areas related to biomedical research), or simply try to attract individuals who already have research as a focus in their careers. There are obviously pros and cons to each approach. Clinicians in training have the advantage of being available in very large numbers, also bringing a lot clinical insight into the research process since they act as translational agents. Translational agents have direct contact with patients in their own fields and thus bring to research a real-world, hot off the press perspective of what is happening in practice. The downside associated with this group is the usual lack of previous formal training as well as the high churn rate -- most will have a single interaction with research and will never come back. The latter decreases the cost-benefit of their training since a lot of time and effort goes into something that will have a single point in time return. Graduate students have a more formal training and the advantage of having already decided for themselves that research is an important aspect of their careers, ultimately decreasing the churn rate. At the same time, the non-clinically active students have less translational exposure and, although having more formal research training than clinicians, their research skills are still not completely mature. Finally, established researchers have the advantage of a mature research training, although they are available in relatively small numbers and usually are fully committed to a certain topic that might be different from where our group might want to go.

2. US-based vs. global collaborations: The question of whether to work in local vs. global research networks is somewhat more subtle, in that the differences between the two are currently decreasing. That said, there are still some significant differences. US-based researchers are usually evaluated at their institutions using the same metrics we are (primarily publications and funding), and so the existence of common cultural values facilitates the exchange of ideas and goals. A common language is also an advantage, in that all researchers are fluent in English and can therefore communicate in a seamless manner. The disadvantages are related to the high costs of conducting research in the US when compared to other countries, as well as the scarcity of researchers in certain areas since most are already fully committed to their own lines of research. The advantages of global networks are the synergy and variety of resources -- what is difficult in Asia might be simpler in the US and vice-versa, e.g., a data analysis method or access to a certain patient population. Disadvantages are obviously language and cultural differences, as well as communication mechanisms. Although today it is much easier to pick up the phone and make a 3 cent/minute call to anybody in Asia while sharing your computer screen with this person, communication is still not perfect.

3. Generating information vs. knowledge: A somewhat deeper question that researchers usually fail to ask until later in their careers is whether their purpose should be to generate information or knowledge. Before we embark on this discussion, an analysis of their definitions is probably warranted. In information sciences researchers usually distinguish among data, information, and knowledge. Data is what is contained within, for example, a spreadsheet with the results of a clinical study. In and of themselves, data points are meaningless and do not convey any information. After data analysis and interpretation of their results in light of previous findings, data becomes information. Notice that there is not necessarily a 1:1 relationship between data and information, in that the same data can be interpreted in multiple forms, generating multiple different pieces of information. Knowledge, the last step in this chain, is information applied to practice. For example, after generating information on whether a certain treatment A is better than treatment B, a researcher would work to promote its implementation in the community, advocate for its payment in professional associations or the congress, or found a company that will manufacture the means necessary to deliver the treatment. Back then to our original question: Should researchers stop at the information step, since this is what most academic institutions will ask us to do, or should we proceed and get involved in knowledge generation? The formal answer we get from academic institutions in the US is that information generation alone will suffice, but this does not mean we should stop there.

Again, there are no right or wrong answers, and the options above are by no means exclusive of each other or static over time. What do you think?