To be, or not to be connected - the pharma supply chain

Ongoing drug shortages across many therapy areas continue to shine a light on the pharma supply chain. You would expect the supply chain for critical medication would be slick, connected and truly end-to-end, which in this case would be from raw materials to patient delivery. Sadly, it’s not.

The cause of shortages is multiple, although not the subject of this article. The aim here is to shine a light on the patient end of the pharma supply and highlight the connectivity opportunities, using the UK as an example.

The current situation

Typically, pharma companies manufacture against forecasts, which are universally accepted not to be as accurate as they could be. While it is more problematic to estimate expected demand on new products, established products should be much easier by using real data from across the entire supply. However, the data currently used has lots of gaps in it.

The industry has visibility of what is sent to the first ‘customer’, typically a wholesaler. Once supplies have been delivered to the wholesaler this visibility disappears. How many drugs are sent where, and when, is managed by the wholesaler based on requests from their customers and isn’t linked back to the manufacturer.

Compare this to the UK food industry which uses data to track the movement of goods from the factory to the point of purchase as well as to monitor inventory levels across the supply chain and forecast demand. In addition, data analytics are also used to optimize supply chain operation, reduce waste, and improve efficiency. There is a lot that pharma can learn from them.

Following its journey to the wholesaler, the next step is to dispense the drugs to a patient. Again, this is managed in isolation typically by the pharmacy or maybe at a pharmacy chain level (Boots, Lloyds, Well, etc.). Here is another opportunity to do things differently and share real-time data.

In a typical busy pharmacy, there is no time to look ahead and forecast upcoming needs, even though over 80% of the 1.14 billion prescriptions dispensed in the UK in 21/22 were for items that were issued on a repeat basis. For example, Atorvastin (given to lower cholesterol and prevent heart attacks) alone accounted for 53.4 million prescriptions, all dispensed month after month. If you did the same for all the 334m cardiovascular treatment prescriptions, you have access to a lot of data. If you then extend it into other disease areas such as diabetes, osteoporosis and you add another 119m data points. Now we are starting to build a lot of useful data.

This data provides an opportunity to forecast requirements, not only providing data back up the chain, but also providing visibility of what will be needed next month. Over 1.2 million patients use the electronic repeat prescription service, routinely having their prescription sent to the same pharmacy. Imagine being able to accurately understand what stock will be needed next month.

What needs to change?

Behaviours and thinking need to change, both at pharmacy and industry level. Creating pharmacy systems that forecast demand based on the last dispensing for a patient and their expected next visit (like how IRT works in clinical trials) is needed, along with the relevant training for pharmacy staff. This would capture most drugs dispensed, not the one-off (antibiotics etc.) or first-time prescriptions. In respect to one-off and first-time prescriptions, a traditional