Artificial intelligence in market access

Artificial Intelligence in market access.


What are the future applications and challenges?

Artificial intelligence (AI) technologies are becoming increasingly relevant in many fields, such as business and marketing and most recently in the healthcare sector. AI is starting to impact the healthcare sector in many aspects, from drug discovery using algorithms to predict new molecules to treat diseases, to diagnostic and prescription using trained neural networks1,2. However, one of the key questions that pharmaceutical companies and decision-makers have is if there is a role to play by AI in the increasingly challenging pricing, reimbursement, and market access landscape of new medicines.

Why now is a good time for Artificial intelligence in healthcare?

The healthcare industry is well suited for the implementation of AI due to large amounts of data available in electronic health records, clinical trials and patient/disease registries and health surveys. For example, the size of the worldwide health care data in 2012 was estimated to be 500 petabytes and is expected to reach more than 25000 petabytes available this year3. This together with the general shift to value-based healthcare, where decision making requires the analysis of multiple and extensive data sources, creates a favourable environment to apply AI solutions.

Potential applications of Artificial intelligence in market access

One of the main applications of AI in market access will be to speed up the acquisition and analysis of medical evidence, clinical trials data and market information, allowing to reduce the hours of human-intensive analysis and obtain further insights with the currently available data. For instance, machine learning models can be trained with anonymized real-world evidence to forecast patient outcomes for a given treatment or to define clinically meaningful patient subgroups through clustering methods1,4, two key challenges faced nowadays by companies. Incorporating these AI-derived insights into the health technology assessment submissions would further support the evidence obtained by traditional methods during the price negotiations and the reimbursement process with payers. Moreover, with the capacity of handling big data, machine learning models could be used to predict physician’s uptake of a new treatment using current prescribing patterns across diseases and patient groups. This will not only allow pharma companies to predict revenues more accurately at launch and build robust budget impact models but also to better allocate resources to promote their product focusing in the most relevant regions or hospitals of a specific market.

Which brings us to the other relevant aspect where the pharmaceutical industry can benefit using AI: the identification of key opinion leaders (KOLs), whose support is crucial to ensure adoption of new treatments in local markets. With natural language processing algorithms, the large volume of unstructured KOLs’ details data on the web could be translated into structured datasets from which the most relevant KOLs for a specific disease in a specific market could be identified. This would allow companies to better allocate resources in the medical affairs, sales and marketing activities focusing them on building relationships with the most relevant KOLs in the target markets.

Challenges that Artificial intelligence will need to face

The adoption of AI is just starting, and with its implementation, many barriers and challenges are expected to appear from both an ethical and a technical perspective5. The gathering and use of private data have dramatically increased across many different platforms. As privacy issues are especially sensitive with medical and other health records, regulations on how medical data can be obtained, used, and shared are catching-up and constantly evolving. Therefore, it is still unclear how much of the medical data AI will be allowed to handle in the future and thus the scope of its use. From a technical standpoint, AI prediction’s accuracy is expected to face challenges due to the current quality and heterogenicity of medical data4. Today health organizations and systems produce and manage data in different ways and physicians do not always consistently input electronic health record data lead to hidden errors that could be difficult to identify. Consequently, efforts will need to be made towards strengthening healthcare databases to obtain the full potential value of AI.


There is no doubt that AI is changing our daily lives, and it is likely to take a more active role in healthcare in the future, from the development and prescription of medicines to their access to the market. Although the adoption of AI is just starting in healthcare, ethical/regulatory and technical challenges have emerged. However, as AI is adopted in more aspects of our lives and more resources are devoted to further develop the mathematical methodologies and the hardware required for AI, these technologies are expected to expand its uses in healthcare. With the increasing amount of medical data available, companies that manage to exploit the AI potential will gain competitive advantage by rapidly generating insights which traditional human-intensive analytics are unable to access. Through algorithms able to model patient outcomes, prescription rates, patient subgroups and identify KOLs, AI will play a role in the pricing, reimbursement, and market access of medicines.


  1. Pinal-Fernandez, I. & Mammen, A. L. On using machine learning algorithms to define clinically meaningful patient subgroups. Annals of the Rheumatic Diseases (2019) doi:10.1136/annrheumdis-2019-215852.
  2. Vamathevan, J. et al. Applications of machine learning in drug discovery and development. Nat Rev Drug Discov 18, 463–477 (2019).
  3. Bodas Sagi, D. J. & Labeaga, J. Big Data and Health Economics. Opportunities Challenges and Risks. International Journal of Interactive Multimedia and Artificial Intelligence, (2017).
  4. Is Artificial Intelligence the Next Big Thing in Health Economics and Outcomes Research? ISPOR | International Society For Pharmacoeconomics and Outcomes Research
  5. Artificial Intelligence in healthcare. Academy of Medical Royal Colleges (2019).

Stay in the know, subscribe to our newsletter

Be the first to receive exclusive content on the latest from the pharmaceutical and market access sector.