No-one knows anything. Or, everything...

Atul Gawande frequently asks the question: how do we manage when the knowledge needed to succeed does not exist in any one person’s head - when it would be impossible for anyone to know everything? I admit to a middle-aged man crush on Gawande. Turns out, even his music taste is great. But, if brilliant and thoughtful physicians like Atul Gawande are asking the question of medical practice, it seems only fitting that we should turn that question to pharma.

Let me ask you: how forgiving is pharma of admitting you don’t know everything?

Gawande’s answer for medicine is intelligently-designed healthcare systems, but I think more importantly an argument for interdependent systems that are not just multi-disciplinary but inter-disciplinary. No one physician could possibly hold everything that was ‘true’ at medical school in her head, never mind stay abreast of the developments since. Gawande talks, with reference to important work on knowledge, of ‘ignorance and ineptitude’ as the two drivers of getting things wrong. Ignorance means you (and/or ‘they’) never knew something; ineptitude means that the answer was known, but not applied in this case…

I am reminded of the wonderful author and screenwriter William Goldman’s tenet (about Hollywood): ‘no-one knows anything.’ He was writing (with evidence) about the role of the oracle-type luminaries of the film studios. Yet, even in Hollywood, the idea that there are sages and visionaries, gifted with foresight into what people will pay to see, and how to make those films that people will pay to see, remains alive (he said it around 30 years ago, and it seems the only thing that has changed is that there is now a business model called The Hollywood Model which describes spreading risk around uncertain cheap projects instead of into sure-fire expensive ones…). However (Pixar aside…), even machines like Marvel were wrong about Deadpool, and wrong about the Fantastic Four, in the past year… The parallels in the music industry are well known - a read through the excellent Cowboys and Indies reveals how serendipity remained the biggest force behind the discovery and development of The Beatles, Elvis and, well…, almost all of the great bands.

Of course, in pharma, we are in this situation. No-one knows, or could know, which product will make it to market, through the thickets of clinical trials and real, live, human biology, or, if it does, whether it will be successful in the market and in real patients. Yet, that is not how we behave. We behave as if prediction is both possible, and important. We behave as if the task is to present the runes in front of the chosen ones, so that they can make decisions about what to progress, how to progress it, and how successful we will be when we do. Yet, the index of Productive Innovation shows that the difference between companies that are good at launching innovation, and those that really can’t, is still only counted in a handful of great new medicines per year.

I started to muse recently on the opposite of ‘certain’. It probably isn’t ‘uncertain’, but ‘curious’… As I have written before, borrowed from (or inspired by) smarter folks, there is a world of difference between ‘I think’, ‘I am confident’, ‘I know’ and ‘I am certain’, although they’re often presented in the same tone of voice, or as interchangeable. Knowing your knowns and unknowns is a good way to start to move in a different direction.

We have some scarily smart folks at IDEA, but the rule remains that they’re not right alone. They can’t be. As soon as you accept, or believe, that it is impossible for anyone to know everything, you have to abide by the rules. We encourage insatiable curiosity, but the second that acquired knowledge becomes pomposity, we reach for the rubber chickens (long story), or worse…

You may remember from biology the idea of the ‘edge effect’, where two distinct elements meet and thrive - think of the ocean and the shore, or the field and its hedgerow, or the river and the sea. The edges typically produce most of the life, and the diversity. What if we embraced the edge effect in pharma - that the scientists could learn something from interacting with the clinicians, that the clinicians could gain from butting up against the marketers, and that regulators, manufacturing and project management each have something to learn from everyone else…? And I don’t mean the way that it is done now, where folks ‘have a seat at the table’ and were ‘there’ when the crucial decisions were made, but couldn’t have stopped it.

So, I mused on the problems of those multidisciplinary team meetings that we see. Why haven’t they led to better decisions in pharma? And the simplest answer (not, despite Occam’s razor, necessarily the best) is that they may be multi-disciplinary, but they’re hardly interdependent, rarely inter-disciplinary. They’re rarely designed to be, despite the era of ‘high-performing teams’. Back in phase II, the most common phrase is ‘we’re not studying that’ or ‘that’s not what we’ll get out of the current CDP’ or ‘let me tell you why that won’t work…’ Which, even the most optimistic marketer will admit, isn’t a very open door for challenge… Is risk aversion a bigger driver than gain? 

It seems that interdisciplinarity is an age-old idea (according to Wikipedia):

Interdisciplinarity involves the combining of two or more academic disciplines into one activity (e.g., a research project). It is about creating something new by crossing boundaries, and thinking across them. It is related to an interdiscipline or an interdisciplinary field, which is an organizational unit that crosses traditional boundaries between academic disciplines or schools of thought, as new needs and professions emerge. 

How good does that sound? Instead of silos dominated by arrogant smart physicians, scientists who want to be left alone at their bench, people who describe the FDA as a hallowed place only special ones can be granted access to, and people who have sat with a ‘value dossier’ in front of real payers (and have the scars to prove it), we had a team who regarded their role as to offer up opportunity to their companions - their collective companions? What if we had ‘consilience’ (the ‘jumping together’ of ideas and challenges)? What if we didn’t start with desirable, and then slowly whittle it away to the overlap in the Venn of what everyone feels is feasible?

I remember a long while back researching the reason that South Park (and Comedy Central) gets away with taking such huge digs at celebrities, where the rest of the entertainment industry would run screaming with alarm at the memo from Legal. It turns out that it is simple: legal were invited to help, rather than to say ‘yes’ or ’no’ afterwards. They offered up creative ways to get done what the writers wanted to get done. They were involved. Creativity from Legal? Turns out all you need to do is ask…

I remember the most brilliant IP attorney, about to leave the industry’s largest company, asking me once ‘why doesn’t IDEA ideate with IP?’ My assumption had always been that IP was a fixed and unchangeable thing. It turns out, as so often, that my assumptions were doing me no favours. With an open mind, and a goal, IP can be flexible, can provide competitive advantage, and is fascinating. Left to the lawyers, not so much… Turns out all you need to do is ask…

So, surely, you say, someone in pharma must combine expertise in IP, clinical, regulatory, marketing, pricing, biology, manufacturing techniques, packaging, digital, …? In one head. Some people have huge brains, and stay up all night reading The Lancet - those people? Well, I have met hundreds of outstanding people in pharma, but unless they’re hiding away in the ‘oracle’ department, I am yet to meet one of those interdisciplinary individuals. Even Atul Gawande is not one of those people. My heroes - Steven Pinker, Steven Johnson, Michael Schrage, Stephen Jay Gould (why so many Stevens?) - all seem capable of encyclopaedic knowledge, but prefer(ed) to remain curious, and open to their own ‘edge effect.’ Curiosity may have killed the cat, but it also keeps alive huge intellects that offer up huge possibilities to all of us.

 So, if that person doesn’t exist, who is making all the decisions? And, why are they getting most of them wrong? And what are we going to do about it?

If the very idea of predictability doesn’t work (and it doesn’t - even McKonsultancies admit that), what then is the opposite of ‘certain’? Well, of course, it is curiosity.

And then, that same day, a random comment from a cancer physician struck me: “despite all the advances of the past 30 years, we’re still largely empiricists.” Well, of course you are - you have to be. We, in pharma, have provided few, if any, reasons not to be. Back to Wikipedia, and I learn that the Empirical movement was a reaction

Empiricism is a theory that states that knowledge comes only or primarily from sensory experience. One of several views of epistemology, the study of human knowledge, along with rationalism and skepticism, empiricism emphasizes the role of experience and evidence, especially sensory experience, in the formation of ideas, over the notion of innate ideas or traditions;[2] empiricists may argue however that traditions (or customs) arise due to relations of previous sense experiences.

“Emphasizes the role of experience and evidence”, along with rationalism and skepticism… Isn’t that where we’re supposed to be?

It seems we have a fundamental disconnect. A ’smarter than thou’ culture, a deterministic consultancy-friendly process, with teams made up of individuals who report through their discipline silos, yet it is all set against an unknown - the intractable challenge in human biology. We can either be convinced that the way we’re doing it is the best way, bar a tweak or two, or we can think differently.

As soon as you embrace the thought that the question of whether drug x will work in disease y, and work well enough to do well on market, is not just unknown but unknowable until you experiment, you begin to think differently. The industry’s predictions have been horrible for many years now (I’d argue that, over the years, serendipity has produced more great drugs than targeted, directed Discovery and Development has - too much focus on improving translational science, rather than accepting that extrapolating from mice to man may never answer these questions?), so what are we going to do about it? 

Well, we’re going to experiment. We’re going to develop several working hypotheses for each molecule, and each disease. Then, we’re going to take the step of finding out. I remain unconvinced that you, personally, know whether P13K is a good idea (for the reasons stated above), whether one immunotherapy agent will combine successfully with another, whether beta-amyloid will modify Alzheimer’s (or whether, if it does, it will then present a commercially viable product). But I know that, if together we reframe the challenge to be ‘would you like to find out?’ we’ll know lots of ways to find out.

So, let me come back to the central question: how do we manage when the knowledge needed to succeed does not exist in any one person’s head - when it would be impossible for anyone to know everything? The answer, like many other industries, is to stop hoping we can find that person, and to redesign how we work together. We know this is possible, that there are teams who are better at this, that there are companies that are better at this. It is actually a simple change, but not one everyone is ready for.