Using a simple web search, it is now easy to retrieve databases from multiple sectors. These databases can be downloaded as a flat file or directly integrated into an information system by connecting via an API. Many organisations publish all or part of their data with free and open access (open data), particularly public bodies. The practice of utilising external data is becoming widespread, alongside growth in the number of data sources available.
Using external data: numerous uses for insurers
These open databases can enable insurers to gain better knowledge of their customers’ health habits by region, type of service, healthcare professional, etc. They therefore allow insurers to adapt their coverage by geographical region or consumer profile, for example.
Open data in health could also enable real-time monitoring of changes in medical expenses so that customer needs can be anticipated, for example in the event of a pandemic.
How to navigate this burgeoning and unstructured jungle?
These multiple data sources offer numerous opportunities to enrich knowledge. Nevertheless, it is becoming difficult to identify the relevant data sources that are reliable over the long term. Effectively utilising this data means implementing a structured approach and suitable technical resources. The rise in the number of data sources available, both public and private, is prompting the emergence of database aggregators. These aggregators are websites that list a large number of databases and provide tools for structuring an open data approach.
Publishing and sharing your data: a new business model?
Although external data sources can be easily used by health insurers, open data also provides the ability for them to share their own data with third parties.
Insurers have a large amount of customer data in their possession. Although this data is less extensive than the data a bank has, an insurer could look to introduce a new win-win model to gather more data. Contrary to what we might think, sharing personal data is not a hurdle for customers if they perceive the added value this can bring. If the insurer handles all of the unforeseen events linked to the policyholder’s life, by communicating about the end benefit (personalised protection) rather than the product/service offering, the policyholder would agree to provide access to their data in return. This is what Ping An does by offering smart devices for patients. As well as being designed for patients (adult, child and senior models), these devices measuring physical activity, blood glucose and blood pressure are extremely easy to use. There is a clear benefit for patients: their medical records are enriched and can be used as a data source when preparing a more accurate online diagnosis, a medicine prescription or other items.
This win-win situation would allow the customer’s lifestyle to be analysed in greater depth. In group insurance, data on company employee claims by sector could be shared with organisations specialising in preventing psychosocial risks and musculoskeletal risks for example. The idea is not to monetise personal data but anonymised data, in compliance with GDPR. Although the data is anonymised, analyses of averages could be used: average expense rates, consumption by geographical sector, type of sick leave by sector of activity and other items.
However, this new business model does not seem to be resonating with insurers. The first reason is the legal framework and in particular the General Data Protection Regulation (GDPR), which is seen as highly constrictive for the retention of data storage, and even more so its use. The second is rooted in the insurance model itself: data is at the heart of an insurers’ business and is considered a strategic asset, and they don’t yet seem ready to share it.
In addition, in the eventuality that insurers’ attitudes on the subject change, they will still have one challenge to face: creating a genuine data culture, which they have not fully acquired so far, unlike other players such as the GAFAMs (Google, Apple, Facebook, Amazon and Microsoft). Insurers’ internal data still largely remains underutilised and may prove poorly documented and not necessarily up to date. Added to this is the fact that insurers with data that can be utilised will need to have the algorithms required to obtain useful information.
If you are interested in this topic, the Cegedim Insurance Solutions team is at your disposal to answer to all your questions.