Blockchain for Medical Research

Accelerating Trust in Healthcare

Sean Manion, Yaël Bizouati-Kennedy

• Editorial: Productivity Press
• Año: 2020
• ISBN:
  9781000045482
• Edición: 1
• Categoría: Medicine
• Categoría: Health Care Delivery

It takes 17 years on average to bring new medical treatments ideas into evidence-based clinical practice. The growing replicability crisis in science further delays these "new miracles." Blockchain can improve science and accelerate medical research while bringing a new layer of trust to healthcare.

This book is about science, its value to medicine, and how we can use blockchain to improve the quality and impact of both. The book looks at science and medicine from an insider's perspective and describes the processes, successes, shortcomings and opportunities in an accessible way for a broad audience. It weaves this a non-technical look at the emerging world of blockchain technology; what it is, where it is useful, and how it can improve science and medicine. It lays out a roadmap for this application to transform how we develop knowledge about health and medicine to improve our lives.

In the first part, Blockchain isn't Tech, the authors look at blockchain/distributed ledger technology along with critical trade-offs and current explorations of its utility. They give an overview of use cases for the technology across industries, including finance, manufacturing and healthcare, with interviews and insights from leaders across government, academia, and tech/health industry both big and start-up.

In the second part, Science is Easy, the authors look at science as a process and how this drives advancement in medicine. They shed a light on some of science's shortcomings, including the reproducibility crisis and problems with misaligned incentives (i.e. publish or perish). They apply a breakdown of critical components to the functional steps in the scientific process and outline how the open science movement is looking to improve these, while highlighting the limit of these fixes with current technology, incentives and structure of science.

In the third part, DAO of Science, the authors look at how blockchain applied to open science can impact medical research. They examine how this distributed approach can provide better quality science, value-based research and faster medical miracles. Finally, they provide a vision of the future of distributed medical research and give a roadmap of steps to get there.

III

The DAO of Science

A distributed autonomous organization (DAO) allows networked collaboration without hierarchy and intermediaries, and may be the perfect solution for improving science.

Chapter 10

Distributing Science

How does this all come together? We have seen that science is valuable; how research contributes to evidence-based medicine; and how this knowledge is translated into practice. We have also explored how this process is often slow and imperfect, with many opportunities for improvement. We have looked at blockchain, a new tool that may be applied to these problems; what it is; what it isn’t; and where it is and can be used. We have covered some of the industries where this technology is advancing most rapidly, and where it is gaining footholds in health and medicine. Now we look at how it all fits together to improve the engine of science in order to make science better and to accelerate the actionable advancement of medical treatment for improved health outcomes and saving lives; in other words, how it can help achieve faster medical miracles.

Because of the differences in data complexity, people and processes in science, you can’t simply plug the fintech blockchain applications into all areas of health research and expect success or even compliance. We first need to dissect science, identify the key areas for application of the technology, compile this vision and figure out how we are going to get there.

The first step is to recognize that science has scaled and specialized but has done so while maintaining a system that compartmentalizes most of the work into silos headed by principal investigators (PIs). Despite massive changes in scope of the enterprise and detail of the execution since the 1950s, this siloed PI structure has been deemed necessary for trust. Nowhere is this more evident than in human-focused research, where in the United States and elsewhere this role and its associated responsibility have been codified in related human research protection laws and regulations.

With blockchain as a new tool to facilitate rapidly auditable trust, it may give us the opportunity to de-silo and hence distribute more efficiently this historically micromanaged trust and research execution.

Beyond One Basket

Traditionally, the majority of biomedical research areas and many other branches of science have been conducted in small teams of individuals headed by a PI. This is the person ultimately responsible for each project from a scientific, ethical, regulatory and fiduciary perspective. They have almost always earned a PhD in their field along with subsequent work (i.e., post-doctoral training) to establish their bona fides in both subject matter knowledge as well as mastery of all phases of the scientific process. Along with the overall responsibility, they also get most of the credit, with others working under guidance doing many of the tasks and receiving credit that somewhat corresponds to their roles. Junior faculty, post-docs, senior graduate trainees and senior research assistants may execute and oversee some or all the aspects of the project as assistant investigators. Junior graduate trainees along with research assistants and technical support team members are those who put in the rest of the work. In the end, it is the PI who will pull everything together: ideation, research design, funding proposals, regulatory approvals, study execution and data collection, data analysis, interpretation and dissemination of findings.

The PI role and the academic training system to achieve it are the foundation of the academic research system. Someone with the end-to-end mastery of all phases of science is considered critical as a position of trust at the lead of each project. It is roughly the equivalent of the head of a car-manufacturing company being required to have demonstrated mastery of developing a concept car, design and specifications, manufacture, safety check, maintenance, sales and distribution.

Most businesses do not run like this, at least not anymore. Scaling and diversification of tasks requires that top leadership and even upper management focus more on aligning specialists across the relevant areas in order to achieve the desired output with maximum quality and efficiency. But science is not a system to produce a known quantity, it is a unique system created to deliver new knowledge. The end-stage products are pieces of new knowledge that can be scrutinized by others, but for which there is no immediate quality test available as there would be for a car or a widget.

In this science and its immediate output, peer-reviewed papers are more like works of art than manufactured products. Unlike a painting or a novel, however, scientific papers and their results are not stand-alone items to be appreciated or ignored, they are the building blocks for future exploration in their field and in quantity become the foundation for evidence-based application of medicine or policy. Because of this, science requires a trusted individual, the PI, to sit at the helm of every project and allow their track record to validate the results.

This dynamic of the trusted PI was once largely aligned with the size and mission of the decentralized scientific enterprise. The main driver for going into science was curiosity, with its exacting tedium and often limited funding. Those who were involved were more often collaborators than competitors, with the vastness of the unknown and the unique focus and specialization of each area of exploration requiring shared curiosity to advance. Individual competition did occur, often notable and vicious, but most fields were able to come together with the lead researchers collaborating at all phases to advance the whole.

Science has scaled considerably over the last 75 years, with rapid expansion in the United States since World War II with Vannevar Bush’s direct influence of federal funding for science. The rapidly rising cost of healthcare along with societal impacts of an aging population have led to further increases in the funding and attention to biomedical research over the past several decades. With this expansion in research has come an advance in many areas of health, but also some complex secondary effects that appear to be slowing the rate of return on this research investment.

The number of new PhDs in biomedical sciences has outstripped the available positions for them to achieve PI status. Where the majority of those who completed doctoral training used to move into tenure-track research faculty roles and extended careers as PIs, now just 10%–15% can hope to achieve that goal. This overabundance of trained talent has led many otherwise qualified potential PIs into other areas beyond research, while also creating a new system of extended post-doctoral training. These “perma-docs” are often stuck in post-doc or non-tenure track faculty positions for a decade or more. This has resulted in the development of mega-labs across many fields of biomedical research, with sometimes dozens of otherwise qualified potential PIs working as AIs or Co-PIs under the lead lab PI. This results in research that is much more homogenous than would be the case if these perpetual post-docs led their own labs.

These labs and their incremental but low risk advance in research are more effective at getting large grant funding then the newer, smaller labs that do appear. While the smaller labs bring novel perspectives and approaches to health research, the risk involved in funding new approaches is also higher. The funding systems, especially in the federal government, are particularly risk-averse, and so the systems for selecting and awarding new funding have been shifting toward the mega-labs over the past couple of decades. Given that the universities housing these researchers receive nearly 50% overhead administrative funds for those research dollars awarded to their labs big and small, they have been incentivized to maintain the mega-labs at the cost of fewer smaller labs, and hence, less innovation.

Adding to this competitive dilemma, the rapid increase in science funding, and the administrative dollars it brings to the university, has led to expanded programs for training new PhD students even as the space for them in the traditional PI roles has not kept pace. These keep the large labs supplied with highly motivated, highly skilled labor for a ridiculously low cost. Biomedical grad students are usually given tuition reimbursement and below minimum wage stipend for what can usually be expected to be an 80+ hours work week for 5–7 years (often having to apply for and receive their own funding in the latter years). This dynamic along with the general perception, and probably reality, that attaching oneself to these mega-labs makes you more likely to gain success in your career has created a system that is giving us bigger and bigger islands of less and less innovation.

These islands are also competing for funding, which makes the collaboration that used to occur across fields now happen primarily only at the latter phases of research when obtaining funding or publishing incremental findings is no longer in jeopardy by talking to your competitors. The competition for funding and findings, and the fear of being scooped has created isolation in an endeavor that thrives on collaboration, and steadily degraded both the economic impact of our research investment and the innovative impact of some of our most innovative minds.

New Model Science

By breaking the scientific process into its component parts using a mission essential task list (METL) method, science can be approached in a new way. This will allow for an increased engagement of scientists across current silos. The emerging blockchain technology can provide the framework for trust for these scientists engaged via a platform marketplace for gig science, the growing trend of freelance scientific research. This will provide both direct network effects from the blockchain network as it grows, along with indirect network effects from the platform model for pairing scientists with funding and ideas with those scientists with skills, availability and time. In the health sciences, this will lead to better science, cheaper research and faster miracles.

Science is thought of and sometimes even described by its practitioners as more artful or intuitive than it actually is. There are certainly aspects of creativity in developing new ideas and appropriately interpreting the results of an experiment, but much of it is a complex series of very simple processes. Each of these is describable, trackable and measurable (though metrics will differ by discipline). Understanding this is a key aspect in understanding science, the challenges with its current execution, and opportunities to improve it with blockchain and distributed ledgers in general.

Mission Essential Task List

A mission essential task list or METL is a general term for an outline of the basic tasks necessary to complete a specific mission. It is utilized in the U.S. military to focus limited resources and ensure details of critical steps necessary for mission are handed off when there is staffing turnover.

The METL used by the military, especially the U.S. Army, aims to enhance mission execution and success in combat situations when resources may be strained, and timeframe may be short. It is a critical focusing agent in a high-stress and life-threatening situation. It is also useful in a slower paced but complex situation like the bureaucracy in Washington, D.C. How to focus resources and actions when there are competing priorities can be a helpful guide.

In 2014–2015, this METL approach was used by the research leadership at the Defense and Veterans Brain Injury Center, a military and veteran traumatic brain injury research network. This was a helpful framework to implement, as the network includes more than a dozen sites in the United States and Europe and approximately 100 researchers working on 60+ research studies, all in the context of clinical care and education mission priorities. There was resistance from the researchers, as many feel their work and process are too complex and intuitive to be broken down in a project planning style task-by-task way. This simply wasn’t true. Despite having a diversified research portfolio with studies ranging from simple one person retrospective data analyses to a 15-year prospective, multi-track, longitudinal study with a staff of 50, we were able to align similar tasks, define them down to sub-task, assign and project workload and staffing requirement, describe areas of hard vs. soft sched...

Índice

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• List of Figures
• List of Tables
• Preface
• Acknowledgments
• Authors
• Introduction
• I Blockchain Isn’t Tech
• II Science Is Easy
• III The DAO of Science
• Notes
• Index