Transfers of value: the disclosure trap

On 5 July 2024, Novo Nordisk (NN), received its second public reprimand from the PMCPA in 18 months. The reprimand relates to a voluntary admission of undisclosed transfers of value. That disclosure showed a wholesale failure in NN's compliance function, with £7.8 million of undisclosed transfers of value (c500 transfers to >150 vendors) over a three year period. The unsurprising outcome was a finding of breach of the ABPI code - indeed of nine clauses, some more than once, including the most egregious, a breach of clause 2.

According to NN, this wholesale failure was a result of incorrect “tagging” of vendors in their systems. Once a vendor was tagged on initial entry into the system, this was never checked.

Even a basic cross-check of total payments made against those disclosed would have shown this glaring omission.

Lessons learned

Even with a sophisticated system for logging transfers of value for disclosure, such as SAP that NN used, errors, often human, can creep in. Such errors are compounded when applied on “rinse and repeat”, mushrooming to the large proportions that we see here. There is no commentary in the report on whether there was any deliberate disregard of rules or a “pure” human error and individual vendors are not identified. NN have promised to provide more resourcing around their compliance systems.

In the UK only companies signed up to the ABPI Code are required to disclose transfers of value. If the new labour government continues the work of the previous one, in line with the findings of the 2020 Cumberledge report, we can expect transparency reporting to become law for all pharmaceutical and medical device companies in the next one to two years.

Given the length of time it takes to put in place systems to allow reporting, companies would be well advised to set up IT systems that will allow them to track and report on transfers of value to HCPs, HCOs, patient organisations and also journalists in the UK. In building those systems, adding functionality to allow cross-checking, perhaps with the assistance of AI, but certainly also some human oversight might allow inevitable fallibilities to be discovered long before such vast discrepancies can occur.