#FutureReadyHealthcare

Who We Are
Investor Relations
News
Careers
Indegene

Pharma industry consensus: fix data strategy before pushing artificial intelligence and machine learning applications

18 Sep 2019

'Data is the new oil', but only analytics-ready data is a real business asset and everyone knows this. However, absence of a cohesive strategy to incorporate cloud platform and data lakes, integrating and consolidating data warehouses, data hubs and databases as a single source of data, can easily send organisations into a tailspin. A huge volume of data is currently being generated from transaction applications, social media and operational devices and processes. Heterogeneous data sources and diversity of data management technology can further make it a management nightmare.

Top executives managing global digital and marketing operations from the pharma industry agree that advanced data quality and analytics governance on data practices is the stepping stone for current and futuristic use cases of artificial intelligence (AI) and machine learning (ML) applications.

According to John McCarthy, Principal Consultant, DT Associates, and former VP Global Digital, AstraZeneca, "as medicines become more expensive, clinical utilisation will be more and more dependent on diagnostic testing or even generic testing. We are going to have to be better at data capture and utilisation that allows ML or AI to engage with the right patients and be more efficient in the way you are engaging broad patient populations. Finding those patients and bringing them to the right doctors will be the new trend in terms of how we utilise data through ML or AI. The old way of working will just not help."

Thomas Thestrup-Terp, Vice President, Commercial Operations, Novo Nordisk, agrees unequivocally that organisations lack the good clean commercial data they need before starting to create any form of commercial AI. "I see these technologies playing out actively in the next five years. We need to do our homework on how do we capture, store and derive insights from data before we invest too much in machine learning or AI," said Thomas.

Raakhi K Sippy, Global Head of Marketing Operations & Third-Party Partnerships, GSK, also stresses the importance of AI for marketing. "We started to harness AI really well on the regulatory and evidence generation side, but where we have not leveraged AI, which is a trend that is picking up pace, is in marketing. Now you've got AI to drive the actions of our content. We need to absolutely adapt to drive efficiencies around marketing operations and automation to help with speed to market of our content. We are already beginning to run pilots on the learnings from clinical evidence, regulatory, etc. and their application to commercial actions," said Raakhi.

Jeff French, Vice President and Chief Digital Officer, ViiV Healthcare, puts forth compliance and understanding as more fundamental issues for AI and ML applications. "A lot of people go after ML and AI and think that they can do it tomorrow. They forget that ML is a learning exercise that takes time. It requires patience to train the machine. The second part is compliance - how do you get it to make the right calls? The decisions that it makes might become a compliance issue from pharma's perspective. Health chatbot companies can help facilitate getting somebody to care, but they cannot be the caregivers themselves," said Jeff.

According to Marc Schwartz, Global Multi-Channel Marketing Lead, Sanofi: "Everyone is talking about AI and ML, but generally there is not enough concreteness or specificity yet. The good news is that pilots are happening in organisations and competencies are being embedded. There is no question that AI and ML will play a critical role and will impact all pharma functions in the long term. It will play a critical role in meeting the needs of our customers. This will transform all healthcare. It’s just not there today."

Organisations should exploit AI/ML for data augmentation, data catalogue, metadata management, master data management, data quality, profiling, cleansing, linking and identifying to make data management self-driven. Organisations need to utilise smart and active metadata, discovery and collaboration platforms to dynamically connect, optimise and automate data integration processes to reduce time to data delivery and most importantly, actionable insights.

According to Uwe Dalichow, Head of Global Marketing Operations, Bayer: "Data analytics is the springboard for meaningful applications of next-gen AI and ML technology. I think if you do your homework in getting that kind of data in order, and then make use of information that we often already have, this can make a huge difference in reaching out to customers. But it all needs to be put into a consumable and adjustable format, and connected from different sources. Then it is possible to programme IT systems on how to advise in order to improve our business and better meet customer needs. Sales reps can get better equipped to hone in their skills on customer engagement strategy."

For more details, please refer to:

https://www.pmlive.com/pharma_thought_leadership/pharma_industry_consensus_fix_data_strategy_before_pushing_artificial_intelligence_and_machine_learning_applications_1301901