We live in a world in which “feelings” symptoms carry far less gravitas than those that are physical. Though physical traits will always be paramount in disease tracking and treatment, we have realised that acknowledging how patients feel is just as important.
Why is feelings data important to us?
We believe that feelings data can add context to the symptoms and side effects experienced by patients. Patients are the people with the most information about how it feels to have a particular condition or be on a certain medication. Therefore, by tracking these individuals feelings, there is a greater chance of earlier diagnosis, optimised dosing and improved reporting on symptom evolution.
How do we collect feelings data?
We know that different people share information in different ways. This is why we collect feelings data from different sources to ensure all demographics are accounted for.
- Text mining in Social – We collect data directly from first-hand patient accounts on Reddit and Twitter. During COVID-19 we began annotating a novel dataset in order to train machine learning models to identify and extract feelings and other relevant information such as symptoms and medicines from the personal accounts of COVID-19 patients. We are building our knowledge as we move across therapy and disease areas of medical need with the aim to be able to track and predict trends.
Figure 1: Word cloud illustrating how patients with cardiovascular health conditions are feeling. (NB: The larger the word, the greater the amount of times this term appeared in posts). Source: Talking Medicines Social feed_13July2020
- Smart Surveys – We also engage with patients concentrated cohorts. These groups of people generally share a common trait whether it be their health condition of medication they take. Through asking the right questions we gain granular insight into how patients feel with regards to their medicines, support received and patient confidence in knowing how their medicines work.
- Connected device – Our Medsmart® App is also used to connect with patients and how they feel. We class this data as high grade due to the close engagement across our patient groups. We ask our App users questions and track their feelings daily as they take their medicines. In time we hope to report back more feelings data to our users so they can be an informed part of the discussion, adding value to healthcare appointments.
By blending data collected through our tools we are able to provide a unique perspective. We are experienced in patient engagement and medicines, and it is this expertise combined with our data processing knowledge that sets us apart from others.
Through listening and engaging with people we capture and structure the experiences and feelings people are having with the medicines they are taking – why and how they are taking their medicines, to provide patient centric data intelligence service to our pharmaceutical customers.
