Opinion Leader Identification: Big Data to KOL Short List

Ariel Katz
Feb 9, 2021

For a pharmaceutical company the decision to work with a certain KOL is very important. Engaging the right KOL is critical in establishing credibility among physicians for a new drug or device, guiding the design of clinical studies, or providing real world evidence - to name but a few. For Medical Science Liaisons (MSLs) identifying, engaging and maintaining the relationship with KOLs is the most important aspect of their job.

Accessing and analysing information of 100,000s of potential KOLs - physicians, researchers, as well as other experts – is not possible for an MSL. They rely on Internet searches, extended networks as well as information, such as the affiliation of a prospective KOL with a renowned hospital or institute, a prestigious award they won or the number of papers they published, to establish a list of candidates. That list then gets prioritized, passed through filters and plotted in matrices to arrive a short list of KOL candidates. The problem with this approach is, that the starting list is incomplete and already biased towards people the MSL knows, are affiliated with institutions the MSL thinks of highly, or people who attend the same conferences, in short: the “usual suspects”. Qualified candidates, especially up and coming experts in their fields, or local/regional influencers without a national reputation might never make it onto the original list and will therefore not be considered at all.

Modern technologies allow MSLs to access unprecedented amounts of data about thought leaders turning the process of key opinion leader identification into more of a science than an art. This wealth of data, however, creates a new problem: how can that big data be reduced to a manageable KOL short list in a reasonable amount of time, without breaking the bank. More critically even: what criteria and what weights should be applied to pare down the original big list.

Every Case is Different

As often, the answer to the question about the perfect KOL selection criteria is: it depends. KOLs play an important role throughout the drug development process: during the research and pre-clinical phases KOLs help identify unmet medical needs and pharmacological targets, during the clinical trials KOLs are involved in designing protocols, answering medical questions, and reviewing and analyzing data, while past approval the role of KOLs shift to building product awareness and educating their colleagues, regulators and payers.

As different as these roles are, as different are the criteria that need to be applied to identify a KOL. The clinical researcher, who one needs to engaged in the pre-clinical phase, has a very different profile from the thought leade,r who will best explain the advantages of the new drug or device to their colleagues. While published papers and posters, academic awards and grants are valuable indicators for a researcher, the ideal KOL for the post-approval phase will stand out because of their years of relevant experience, their practice size, their high-level speaking engagements, or role on advisory committees.

Paring Down the Big Data

To identify the best KOLs and avoid costly mistakes a good process looks like this:

·   Start with the entire universe of possible KOLs such as captured in H1.Insight’s Curie database

·   Quickly reduce the number of possibilities by eliminating all KOLs outside the therapeutic area of interest

·   Further pare down the list by applying high-level criteria, e.g. if finding a regional KOL is the goal, a geographic search will provide only KOLs in the target geography

·   Key word searches and filters, such as for specific clinical trials or payments received, can help to further hone in on the list of thought leaders that are good KOL candidates

·   Curated indicators in the H1.Curie database make identifying specific types of KOLs easy. Looking for an up and coming KOL, the “Rising Star”-ranked KOLs are worth looking at. Searching for an outstanding researcher, the “Research Scholar” indicator highlights researchers with an excellent publication record, the “Clinical Investigator” has an exceptional list of clinical trials and the “Industry Collaborator” is a thought leader who has worked closely with the industry.

This process allows an MSL to start very broad but boil huge amounts of information down to a manageable short list quickly and repeat the process if their needs change. Periodically repeating the search will allow an MSL to review whether new experts have emerged that should be engaged and allows for flexibly managing and adjusting the KOL roster.