Healthcare is a knowledge industry. Its successful delivery is predicated on the effective application of clinical knowledge in the context of an individual patient or service. The problem is that not all knowledge is equal. Learnt experience and professional consensus is an important part of the story, but scientific knowledge is the boss. In the context of medicine, we apply science through the discipline of evidence-based practice (EBP). A similar approach can be used to decide on what health technology to use, clinicians need to find those health technologies which are supported by evidence that their use produces increased performance, impact, and value. This is why in IBM Watson Health ‘Leading with Science’ is so important.

EBP has become mainstream during my clinical career, established as received wisdom and is now the foundation of clinical practice. But it wasn’t always like that. I can remember discussions early in my career when EBP was resisted as something that might undermine professional autonomy and devalue individual judgment. The argument was eventually overcome by demonstrating the consequences of unexplained variation in care and examples of established clinical practices that were of little value or, in fact, actually harmful. In short, following the evidence leads to better outcomes. As a result, EBP became hard to resist. We have now reached a more mature state where the clinical value of EBP is fully recognised. Evidence is seen as the foundation of a consultation, but is also paired alongside the human aspects of professional judgment, instinct and critically, the patient’s own perspective. It therefore found its place in clinical practice in a balanced way.

As a young clinician early in my career, I was a champion of EBP and its application at the point of care. However, over the years, two major challenges have emerged …

The first was simply my inability to remember all the relevant evidence. Clinicians need to stop pretending that we can always remember everything for every patient. A gap has emerged between the requirement in modern clinical practice to process the available knowledge and the finite capacity of the human brain to manage such large volumes of information. Medical knowledge has been expanding exponentially. Whereas the doubling time was an estimated 50 years back in 1950, it accelerated to 7 years in 1980, 3.5 years in 2010, and a projected 73 days by 2020 1 . Unfortunately, there has not been doubling of clinician brain power to the same degree. The mismatch is clear. There has never been a better time to augment clinicians with contextually relevant decision support.

The second challenge was that the patient in front of me usually bore no resemblance to the neat cohorts in the randomised controlled trials. My patients reflected real life and tended to have multiple conditions, be older and have the usual complex interplay between social, medical and psychological factors rather than fit the neat cohorts of a randomised controlled trial which had screened out the more complicated patients. A luxury that is largely unavailable in clinical practice.

Since its beginning, IBM has been working in the health field, Watson Health was formed because we recognized the opportunity to help address challenges like these and make improvements in the industry with technology. For example, the ability to process and contextualize large volumes of unstructured data and present it in the context of an individual patient. The papers we have presented at ASCO and ESMO are building on the body of evidence for this. Watson Health can also help us rethink and invigorate how evidence is generated. By using tools like Watson for Clinical Trial Matching we can help close the gap between research and direct patient care – helping to make trial recruitment an everyday part of direct care. We know the majority of patients want to take part in research and we also know that less than 5% of patients actually do. 2,3 This is a big opportunity for us to accelerate the generation of knowledge in healthcare. Additionally, with the emergence of modern data management and advanced analytics using sophisticated linkage and normalization techniques we can start identifying evidence from actual populations rather than research cohorts – so-called real-world evidence. Not only will this mean that complex patients would be included in research it also means that clinicians better understand what works in the real situations of complex healthcare delivery.

My generation has built the case for evidence-based practice but thanks to technologies offered by Watson Health, the next generation of clinicians will be able to apply it at scale. I believe we will see a shift in clinical practice to one that is much better connected to the use of evidence for clinicians, managers, and patients. 

  1.  Densen P. Challenges and opportunities facing medical education. Trans Am Climatol Assoc. 2011; 122:48-58.
  2.  Moorcraft SY, Marriott C, Peckitt C, Cunningham D, Chau I, Starling N, Watkins D, Rao S. Patients’ willingness to participate in clinical trials and their views on aspects of cancer research: results of a prospective patient survey. Trials. 2016; 17:17.
  3. Unger JM, Cook E, Tai E, Bleyer A. The role of clinical trial participation in cancer research: barriers, evidence and strategies. Am Soc Clin Oncol Educ Book. 2016; 35:185-98.