How to make your medical insights more quantitative
Many MSL teams focus on collecting medical insights in free-text form. Then manually weed through these to create a report to present to leadership. The goal of the report is to identify medical trends and help leadership make decisions about the Medical Strategy. Often the report misses the mark because it is hard to digest and doesn’t help decide which actions to take next.
One of the top complaints we hear from MSL leads is how awful it is to create reports from free-text insights (check out this actionable insights guide to help make the analysis of free-text easier). At the same time, Medical leaders wish reports were easier to read and more quantitative. Creating easy to read and quantitative insights reports shouldn't be painful. Keep reading to learn how a structured data approach plus free-text insights save time and lead to more meaningful outcomes.
How medical insights reports are born
Medical insights often result from a scientific exchange between MSLs and thought leaders. Insights typically arrive back to the organization as free-text statements. Depending on the number of MSLs and therapeutic area, this could mean 1000's of insights of per month. So now what?
It typically falls to an MSL lead to create an insight report. They then organize, categorize, collate, and make graphs in Excel to create a final report.
This process is time-consuming and painful. Who wants to spend hours reading through text like, "This physician sees many patients with this disease state and reports problems with self-administration of the drug" or "Dr. so-and-so believes there is an unmet need in patients with comorbidities"? Why should an experienced and highly-paid MSL spend hours or even days on this drudgery?
Additionally, it is hard for senior leaders to digest endless rows of free-text insights and decide what action to take. Insights reports should be easy to read so therapies can get to the right patients, at the right time.
Free-text insights: the good and the bad
Free-text insights are good for identifying opinions and uncovering blind spots. They can also help uncover nascent trends and confirm existing trends. On the flip side, they come with limitations:
Free-text insights can be hard to analyze
It is difficult to know if the insight is significant or subject to sample bias
Analysis of insights can be subject to human bias
There is a better way: combine free-text insights with a structured data approach.
But wait, aren’t medical insights inherently qualitative?
You may be thinking insights are the perspectives of HCPs and are qualitative in nature. How can you make them quantitative? Here’s an example:
Look at the following insights collected about a complete response letter (CRL):
“The specialist was aware of the CRL and did not think it was a significant issue. “
“Met with the KOL at a conference and while disappointed about the CRL, she was not overly concerned because the efficacy data looked good.”
“The physician was disappointed by the CRL but thinks it will not impact its ability to be approved. Asked to be updated as updates become available.”
“The HCP was not aware of the CRL and was still excited about the drug getting approval.”
“This physician specialist was neutral about the CRL because the FDA did not request more clinical trials. Wants to be notified when it gets approval. “
“This KOL was not concerned about the CRL because more trials were not requested.”
“Not concerned about the CRL press release and still wants to try the drug when it becomes available.”
“The physician specialist wasn’t too concerned with the CRL/delay. Made a comment that this type of thing happens all the time. “
“The KOL was upset about the CRL: she wants to use this medication as soon as it gets approved.”
“The physician thought it was unfortunate and requested to be updated.”
Notice any patterns? Do any of these seem repetitive? Instead of listing slide after slide of free-text, these insights can be collated to create graphs:
Example of how free-text insights can be converted into a survey question (i.e. using a structured data approach) to quantify KOL feedback. The free-text insights about a complete response letter above were converted into a simple chart. With this approach, you can quickly glance over the graph to understand how key opinion leaders perceived the complete response letter. This avoids having to read through repetitive free-text insights and trying to identify trends in your head.
Example of how free-text insights can be converted into quantitative data using a structured data approach (see text below to learn more about structured data approaches). The free-text insights about a complete response letter above were converted into a simple chart. In this example, the why behind the key opinion leader’s (KOL) perspective was quantified. With the structured data approach Medical leaders can quickly see that 46% of the KOLs visited were still interested in the product and 31% were not concerned because the FDA did not request additional clinical trials. With this type of information, Medical leaders can use hard numbers to decide on what next steps should be taken.
Humans are visual creatures. We can more easily and quickly digest information in graphs and charts than in text form. When free-text insights are condensed into graphs, we can quickly get an understanding of general trends. Added bonus: most leaders love charts!
This is often what MSL leads do manually when they analyze and collate free-text insights. A structured data approach bypasses the manual labor by collecting information in a way that is easy to graph from the beginning. Imagine automatically having a graph instead of having to make one yourself!
What is a structured data approach?
A structured data approach means collecting insights in the form of numbers. Or in other words, making your insights quantitative. It allows you to test perceptions and determine if an insight is representative or just one person’s opinion.
Collecting free-text insights and structured data are symbiotic. A pattern or trend is identified in the free-text (ideally by a machine. Learn more with our ultimate guide to AI and MSL insights.) and then a structured data approach tests and quantifies this trend. This way, you can determine if the pattern/trend is accurate or an outlier. The result is a more educated decision on next steps and which actions to take.
“Without data you are just another person with an opinion.” W.E. Deming
Structured data approach: how it works
How does an MSL team start to apply a structured data approach for insights? It means building out survey questions to guide MSLs. Many organizations already provide guiding questions to help MSLs structure meetings with KOLs and focus on collecting info relevant to the Medical Strategy.
Check out this infographic describing the main steps for setting up a structured data approach:
You might be thinking it's hard to come up with possible answers to your questions. Give it a try! It's surprising how easy it is and how it quickly becomes second nature.
Many MSL teams use tools like kernel to create surveys and analyze the data. Get in touch if you want to learn more. Collecting insights as both free-text and structured data is our bread and butter.
Structured data questions can still be reactive
A misconception about a structured data approach is that it encourages MSLs to proactively bring up topics in meetings. With training and practice, MSLs learn to naturally weave questions into a conversation. The result is that they are able to collect more info relevant to your Medical Strategy.
Benefits of a structured data approach
A structured data approach helps avoid vanity metrics
A structured data approach helps MSL teams to quantify insights. This enables leadership teams to get away from metrics like the number of KOL visits the MSL team had in a quarter. Reporting activites is important but do not help make decisions about the Medical Strategy
This structured data approach provides hard numbers behind what KOLs are thinking. Medical leadership teams can make decisions based on information like this:
‘All of the KOLs visited were aware of the clinical trial results and 57% were concerned about the safety profile presented in phase 3 results. So we should take XYZ action.’
Quantitative insights help show the MSL team's value because it ties their actions to meaningful outcomes.
Conclusion
Free-text insights and a structured data approach go hand-in-hand. Free-text insights help identify a pattern/trend and structured data quantifies it.
Structured data allows you to figure out the facts and free-text insights help determine the "why" behind a KOL's feedback. Would a study only measure how well the patients say they feel? Or is it prudent to measure other endpoints as well? A good medical insights process tracks both qualitative and quantitative data.
Free-text insights + structured data = more quantitative data = best in class insights process
A combination of free-text insights and structured data gives you better data in less time. It can help your organization understand why HCPs act a certain way. It will help you test your Medical Strategy and bring your therapy to the right patients, at the right time.
Insights analysis Ebook
A free Ebook with a simple 4 step process to analyze insights.
Actionable insights report template
Check out this free template to create insights reports that motivate your stakeholders to act!
Need help setting up a structured data approach for your MSL team? Reach out! We are here to help.