The pharmaceutical sector is highly-regulated with strict controls over how companies can communicate with health care practitioners (HCPs) and what data can be shared. Marketing remains labour-intensive and expensive, with field sales-force visits to explain new products, changes in indication and new clinical studies one of the main routes-to-market. This overhead serves to increase the cost of products while diverting funds away from R&D and patient access to medication.
To address this, quantLab brought together a team of AI, big data and domain experts to explore how an AI-based engagement recommendation solution could be deployed. This would optimise the HCPs each sales representative should visit, when was the most appropriate time to do so, what topics they should discuss and relevant content to share. The solution would need to be capable of working at a global level, but still have local country salience.
Data on prescriptions written by each practitioner is available, together with CRM data from pharmaceutical manufacturers, but this was found to be limited when looking to tailor engagement to the specific needs and behaviours of an HCP. Disease prevalence and demographics were added to predict patient load in any given week and to model the number of prescriptions that can be written. Where the actual level matches this model, no sales visit is required, but where it is lower, the cost of direct engagement can be justified.
Using AI and neural networks, it became possible to understand the relationships between different data sets, for example that an outbreak of influenza in a given city will lead to an upswing in demand for respiratory products six weeks later. This insight can be used to schedule representatives’ visits at the right time in the demand curve.
The expectation is that mathematical modelling will now help to drive the digital transformation of this global sector.