Knowing What You Don’t Know
A Bottom-Up vs Top-Down Approach to Analysis
More than a century ago, the scientific thought leaders of Europe initially dismissed Albert Einstein as inconsequential. Given his academic performance, they intuitively knew not to waste their time on him. Later, we know, his prolific publication record and revolutionary ideas gained their attention – and the world’s. Today, he’s regarded as one of the foremost physicists of all time. Clearly, intuition can be wrong.
Are you making the same type of mistake?
Einstein, you might say, was a victim of top-down analysis. Experts saw what they expected to see, and that didn’t include an obscure patent clerk.
It was only by publishing four papers in 1905 that the sheer quantity of his work forced the recognition of the scientific community. You could consider that recognition the result of bottom-up analysis.
Current Medical Affairs practices are overlooking modern geniuses, too. The proliferation of scientific literature and the global wealth of experts makes it impossible to rely on experience and intuition as the sole determinants of who to invest your time and share your data with. Conversely, are you spending too much time with established experts? If these experts shift their focus, how will you know in time to respond strategically?
Analytics designed specifically for Medical Affairs – Medical Affairs 2.0 – make those analyses fast, easy and accurate. Modern tools let you search by areas of research to identify, for instance, the most active researchers in the area of BRAF inhibitors. As a result, you can identify prolific researchers who may not be on your radar screen. Trend lines based on years of activity can show you who’s interest is growing in an area and, just as importantly, whose focus is waning (Chart 1, below).
Clearly, relying on intuition and experience to reach the right balance isn’t always accurate.
Expecting data to turn out a certain way can take you down a dangerous path. The true benefit of collecting and normalizing massive quantities of data, removing ambiguity and putting it all into context is learning something you don’t know and probably didn’t suspect.
Traditionally, Medical Affairs professionals search a plethora of data and sources, to identify all the relevant data about one specific – a leading author in a certain disease or the leading Centers of Excellence, for example. They’re using common knowledge and intuition to choose the search topic. If the search results and their gut instincts agree, they tend to stop gathering information and start using it.
This “top-down” approach is powerful, but it creates confirmation bias – the tendency to search for data that confirms your own beliefs or hypothesis. As a result, you are encouraged to select the intuitive answer rather than the best answer, simply because you don’t know what you don’t know.
A “bottom-up” approach, in contrast, relies on data rather than instinct or experience to identify the most relevant meetings, the most influential opinion leaders, the best journal… It’s counterintuitive, but actually can be more beneficial because it can reveal new information you didn’t know to look for.
Consider this example. You want to present new data at several neurology conferences, so you comb the Internet using the search term “neurology conferences.” That’s a logical approach for locating pizza parlors, so you assume it also will be successful when applied to Medical Affairs. In reality, a keyword search for “neurology conferences” actually returns, at best, only about 25 percent of the meetings where neurology data is presented.
The discrepancy occurs because other meetings use other terms. There are no standards for the meta tags that conference organizers, professional societies and journals use to categorize their content. Consequently, the content of your initial pool of meetings depends upon the expertise of the person inputting the meta tags that help searchers find that information on the Internet, and the lack of industry standards for search engine optimization.
So, while the Annual Meeting of the American Academy of Neurology may be your intuitive choice, a top-down approach may omit options like the Brain Tumor Biotech Summit or the Keystone symposia, Neuroinflammation: Concepts, Characteristics, Consequences.
A similar bias is evident when selecting scientific imperatives during the run-up to product launches. In the past, Medical Affairs professionals used marketing data and opinions from focus groups to determine which scientific features to emphasize in new product launches. But opinions are inherently biased and often aren’t based on comprehensive information.
Taking a “bottom-up,” data-driven approach provides objective evidence that either supports or refutes the choice of imperatives. For example, when one major pharmaceutical company analyzed recent product launches, the data the Medical Affairs team found caused the company to redesign its scientific communications strategy and focus on different scientific imperatives.
When Medmeme and a Pharma Partner analyzed the output of leading scientific contributors and their interactions with the company’s field medical team, they found that resources often were concentrated on the wrong people. Some Field Medical professionals were very engaged with the least active researchers (figure 1), while all but ignoring some of the most productive (figure 2). And, they overlooked the up-and-coming researchers simply because they weren’t on their radar screen.
A bottom-up analysis can identify those gaps, helping you stop mistakes before they happen. In Einstein’s time, a bottom-up analysis involved a handful of important journals. Now, however, the number of relevant journals and conferences and the ways of analyzing them are too great for manual examination.
This type of bottom-up analysis shows what “you don’t know you don’t know.” It’s akin to a GPS system for your car. You drove without one for years, but having it gives you more data that makes you more informed so you can choose the best routes and avoid traffic jams. That, essentially, is what Medical Affairs 2.0 analytics does, too. It helps you identify researchers able to speed you along and spend less time and resources with researchers who are less interested.
Medical Affairs 2.0 analytics doesn’t replace intuition and experience, it augments them, helping you discover what you don’t know you don’t know.
Because you’re immersed in data every day, it’s easy to become mired. As my grandmother used to say, “You can’t see the whole picture when you’re standing in the frame.” Getting extra, objective data developed specifically for Medical Affairs helps you step out of the frame, see the entire picture and ground your thinking.
SHARE THIS POST