The Pandemic-inspired Case for an Open-Source Medical Hardware Ecosystem

thumbnail

The co-authors of this article, along with Robert L. Read, are:

The Present Situation

COVID-19 has created nothing, if not confusion. Confusion precedes enlightenment.*

The pandemic has been horrible; but it has created beautiful human stories. One such story recounts tens of thousands of persons self-organizing into unpaid teams to address the pandemic by successfully making and distributing personal protective equipment (PPE). A parallel but no less daunting challenge has been undertaken by thousands of technologists, managers, medical professionals, and makers struggling to address ventilator and therapeutic oxygen shortages. Dale Dougherty of Make Magazine has dubbed these community efforts “Plan C” and highlighted many examples.

The ventilator shortage problem evolved over time as industry stepped up production (usually for pay) of ventilators at the same time that doctors learned how to better treat the disease. The term “pandemic ventilator” progressed from meaning originally “emergency invasive ventilator for intubated patients” to “noninvasive respiratory support machine capable of invasive ventilation.” In August 2020, for patients in the wealthiest nations, it is safe to say that a shortage of ventilators specifically is now unlikely to cause death. The pandemic demonstrated that if medical systems are overwhelmed in general, outcomes degrade, which means more people die.

It is inspirational that many of the teams, mostly populated by citizens of wealthy nations, are still working, largely as volunteers, to provide ventilators for less wealthy nations with more overwhelmed medical systems.  Ventilators are necessary but insufficient to address COVID-19. One also needs ICU beds, trained staff, and therapeutic oxygen. This highly contagious disease, if not controlled by social policy, can overwhelm any medical system, most devastatingly one with a low resource-to-population ratio.

Of more than 100 teams that sought to build ventilators, six to ten teams are now near ready to be deployed. They are in the process of satisfying the safety requirements of the US FDA or other regulatory bodies. Some teams have advanced to manufacturing and financing. If we count March 2020 as the beginning of the crisis response, it appears that it has only taken six-to-nine months for free open-source hardware and software solutions to be ready for deployment, and begin saving lives. The authors personally deem this a medium-term success to match the short-term success of the open source medical supply community that made simpler PPE.


Screenshot of the Analysis of Open Source Ventilator Projects

The Long-term Future

Without sapping energy from those still working furiously to address the current crisis, it is perhaps time to reflect on a long-term future. Just as the immune system learns how to fight threats that it has encountered, humanity must take stock of what we have been taught by the pandemic and learn how to better prepare for the future of COVID-19 and other crises. A poignant story in only six words is sometimes attributed to Hemingway:

“Baby shoes for sale, never used.”

COVID-19 has now written a horror story in a two-word noun phrase:

“Starter pandemic.”

The entire system of employer-sponsored health care came to a crashing halt when tens of millions United States citizens found themselves without jobs at the beginning of the crisis.  Not only were these individuals facing death in the form of a pandemic, but other forms of medical care simply became unavailable to them and to their families because their employers had not prepared emergency funds to float their businesses through a crisis (or were not willing to spend the funds necessary to do so).  Through no fault of their own, millions of people in the United States found themselves incapable of providing basic health care to their families.

Hospitals in the United States were not immune to problems. As a result of hospital dependence on employer-sponsored insurance for services reimbursement, as well as an increasing reliance on market competition as a form of acquiring customers, hospitals are struggling financially. As the hospitals and health systems work to keep their doors open, there is little to no monetary incentive to help another hospital that has run out of critical supplies as each attempts to keep their revenue generating services open.  One hospital’s loss is another hospital’s gain, and the patients who would ultimately suffer from this latent competition and reliance on employment-based insurance are simply lost in the shuffle.  When the coronavirus hit, groups like  Open Source COVID19 Medical Supplies  and GetUsPPE.org were founded when it became clear that hospital administrators simply had no solution for the drastic PPE shortfall.  This systemic flaw, coupled with an inadequate response from government agencies tasked with supplying PPE to hospitals, has resulted in thousands of preventable deaths.

Capitalism is poorly prepared to deal with acute crises that come and go quickly and largely affect poor people; there is no money in helping people who cannot pay for the help. Investing  in the cost of planning today for unpredictable crises of tomorrow has little return on investment. In an acute worldwide crisis, no one firm is likely to have the production capacity to meet demand; if the next crisis requires distribution of thousands of dialysis machines, we will find ourselves back in our current situation. By their nature, a rapidly transmitted disease which reaches pandemic proportion overwhelms local production capacity. As we have already seen, other firms can be recruited to increase production rapidly compared to a normal steady-state production schedule, but not rapidly enough. Furthermore, the firms which can do this, such as Ford Motor Company in the US, may have no incentive to do so for peoples and nations that cannot pay for the machines, and may even be forbidden by national laws from doing so. All of these failures are completely predictable from economic theory.

Not necessarily predicted by classic economic theory is another weakness: supply-chain fragility. Our global system of trade has allowed nations to exploit their competitive advantages to provide local specialization which is then globally distributed. In general, this arrangement is advantageous to society as a whole, but creates single-points-of-failure that can result in disastrous local reductions of inventory that cannot be rectified while supply chains are disrupted. In a global emergency, one nation may be the only source of an essential part, or at least a significant fraction of that global supply.  Supply-chain failures coupled with the just-in-time inventorying practices of US hospitals resulted in critical shortages of supplies. An example is air flow-sensors, an absolutely critical part for ventilators. Between April and July (at least) there was a severe shortage of off-the-shelf flow sensors, despite well-intentioned efforts by at least one well-regarded firm, Sensirion, to amplify production.  Several teams were compelled to simply design their own flow sensor from scratch, at significant design and testing effort.  N95 masks are another critical example—most of the worldwide production of N95 material is in China, but once trade with China was disrupted, demand for masks made from this material skyrocketed.

The current system heavily utilizes closed-source designs of medical devices.  Makers of medical devices often obtain patents, representing variable degrees of true innovation, in order to limit competition and to protect themselves from patent claims made by other firms.  This system creates artificially large barriers to entry to the market by incentivizing extensive legal patent coverage for any new devices, which in turn inflates the prices of devices and creates an anticompetitive environment.  The authors are eager to experiment with more competitive alternatives that emphasize agility and performance over device-type, notably illustrated by the SILA Standard, an open source communication structure for laboratory systems breaking from a single profit-driven motivation that characterizes closed-sourced ecosystems. The success metric for open-source standards is often multifocal, encompassing effective deployment, broad application, and a regenerative feedback loop that sparks further innovation.

This closed-source medical device ecosystem has three consequences beyond increasing the barrier to market entry for new manufacturers:

  • First, the closed market creates exquisite reliance on the supply chain. A single device manufacturer with their own supply chain fragility places another point of failure in the delivery of medical care on all downstream purchasers of that manufacturer.  Should any step fail, all downstream customers are affected, as shown by the current crisis.
  • Second, the closed market allows for colonial attitudes towards new markets. Rather than encouraging local manufacturing of devices, a manufacturer would want to be a sole supplier for a market and maintain a monopoly position, placing the citizens in the market into a direct dependence on the supplier, regardless of whether or not the market has some form of socialized medicine.  Local manufacturing would mean that local device manufacturers could participate directly in the local economies and perhaps be resilient to supply chain disruptions.
  • Third, the closed market does not encourage sharing of new innovations with other device manufacturers, as doing so would be to lose an edge in the market.  An open device standard enables those developing innovations to bring those innovations back to the rest of the community using the same specifications.

Where do all of these considerations leave us?  In the longer term, many of these problems are systemic and require much larger societal change beyond the scope of this essay.

However, we humanitarian engineers can improve some of the foundations of a health care system that has become so dependent upon closed medical device specifications to provide patient care.  We can create medical device designs that address real needs of medical care providers, and give those designs away.  By open sourcing the designs, health care practitioners can create their own devices, or service those devices, without requiring payouts to centralized corporations.  On the other side of the spectrum, for those hospitals that have requirements around devices and corporate liability, open source designs allow for medical device manufacturers to create their own versions of the device and charge hospitals for those devices, while not preventing other manufacturers from entering the market so that there can be real price competition between manufacturers.  These devices can be serviced and maintained by anyone, even if the original manufacturer has gone out of business or has otherwise left the market.

In late 2020, we as a society are far better prepared to ramp up ventilator production than we were in March. If COVID-19 or another pandemic requires ventilators, we have free-libre open source hardware designs which ANY nation can manufacture and deploy. If we could turn back time and have this foreknowledge, we could have saved lives. That is of course, always true. The question is not how to mitigate the last crisis, but how to prepare for the next crisis and how to better address the next six months of the COVID-19 pandemic.

A Proposal

If we think of ourselves as a loosely organized humanitarian engineering community, we do not have infinite resources, but we are far from powerless. Philanthropists have already shown themselves to be generous in addressing this problem, at least on a small scale, and are probably more willing to fund preparedness than ever before. Additionally, successful techniques for organizing large teams without monetary incentives have been previously pioneered by the free and open-source software communities. Since March, these techniques have been expanded to free and open source hardware teams, at least in the realm of ventilators and respiration support devices. Organizations such as Public Invention, Helpful Engineering, and others have stepped up to provide a loose organization of teams, small amounts of money and large amounts of knowledge. There has been a tremendous transfer of knowledge from medical and regulatory professionals to engineers. These efforts have been aided by many selfless volunteers who are not necessarily engineers but are writers, artists, managers, financiers, and executives.

Furthermore, we have many open source designs. We have gone from essentially two or three prescient and seminal pre-pandemic ventilator prototype designs to a few dozen highly tested fully open, replicable, manufacturable, and in some cases, FDA-cleared or FDA-clearable designs. We have a few dozen teams that are actively working on improving the designs of open source ventilators, oxygen concentrators, testers, and related equipment. Many of these teams are attempting to negotiate with for-profit manufacturers in a variety of ways to get these devices deployed. These necessary business negotiations are chaotic, decentralized and dispersed, creating a lot of additional overhead necessary for building business relationships under normal circumstances, but consuming valuable time and energy during a time-critical crisis.

What humanity needs is a non-profit firm that does not quite exist today (though some organizations may be close.) Humanity needs Open Source Medical Device manufacturing non-profit corporations (let us call these “OSMD firms”). The charter of such non-profit firms will be:

  1. To obtain US FDA and other regulatory approval of open source designs.
  2. To coordinate and support extant open-source teams that are producing designs by sharing knowledge of designs.
  3. To encourage and insist upon practices and documentation that are transparent and reusable to support US FDA and other regulatory approval by curating designs and extensive design documentation.
  4. To organize emergency manufacturing capability across the world, even in the absence of need, so as to be prepared for emergencies.
  5. To relieve design teams of the need to negotiate manufacturing directly by centralizing knowledge of how to negotiate manufacturing contracts.
  6. To make it easier for philanthropists to know their gifts will not only result in designs, but in deployable designs.
  7. To take upon themselves a majority of liability for manufacturing, to the relief of design teams that do not have expert legal advice.
  8. To ease purchasing of non-monopolized open source medical devices by coordinating testing, approval, financing and manufacturing.
  9. To keep prices competitively low by offering universally usable designs which prevent monopolies on medical devices.
  10. Strongly encourages share-alike hardware, software, and documentation licenses which constantly build long-term communal capabilities.
  11. To allow local manufacturers to create their own versions of devices, lower the chances of continued colonial relationships between richer and poorer nations.
  12. To provide liability shields for both purchasers and manufacturers such that there are legal protections for the producers and consumers of devices.
  13. To provide a mechanism for sharing design improvements based on further discovery of medical knowledge across any and all manufacturers producing devices.

Helpful Engineering, a large organizer of engineering teams, recently received 501(c)3 designation and has begun to develop processes to assist in FDA clearance and other approval by other regulatory bodies. However, it has specifically decided not to engage in manufacturing. Open Source Medical Supplies likewise acts as a nexus of knowledge about medical supplies (especially Personal Protective Equipment), but does not directly manufacture devices. Both may some day come to serve for medical devices the role that the Apache Foundation and the Free Software Foundation serve for software: that is, they may serve as incubators, from which new companies are founded based on top-level projects within the portfolio of Apache projects.  Open-source projects such as Spark and Kafka have companies that provide professional support for these projects, which allows corporations to pay for support contracts and have confidence that their open source projects will be supported.  Tetra Bio Distributed has been founded to be an offshoot project from Helpful Engineering following this Apache model.  Public Invention likewise develops and curates open source designs, but does not intend to manufacture on a large scale.  Helpful Engineering, Public Invention, and Open Source Medical Supplies could be called open-source medical design organizations, but not OSMD manufacturing firms.

On Non-Monopolization

Before motivating further the need for Open-source Medical Devices (OSMD) firms, let us explain what they do not do. They do NOT:

  1. ask any team to close-source its design so that it can make more money.
  2. preclude for-profit competitors from using these same designs and design documentation, either for regulatory approval or manufacture.
  3. preclude other non-profits from using these same designs and operating procedures
  4. preclude any nation from setting up its own version or versions of OSMD.

However, they DO encourage the use of reciprocal, share-alike licenses, sometimes called “copyleft” licenses, which incentivize firms and teams that make improvements to designs to share those designs back to the community, from which it, and other firms unrelated to it will benefit.

To those with experience in the free-libre open source software world, these positions may be clear, but may require some explanation and examples.

Example 1: A For-profit Firm Uses The Designs

Suppose:

  1. It is January 1, 2021, and the pandemic continues to grow unabated.
  2. An OSMD firm has studied the Respiraworks and SmithVent ventilators, which require slightly different clinical infrastructure, and decided to manufacture both.
  3. The OSMD firm has used their good documentation, perhaps reused from Helpful Engineering, to obtain FDA clearance of its ability to manufacture and control the quality of these devices through well-documented sub-contracting and quality assurance procedures, in combinations with significant testing.
  4. Through a combination of philanthropy and negotiation, it is supplying 500 such ventilators a week to nations in Africa, which are paying approximately twice the cost of the parts needed to manufacture the ventilators, or approximately $3000 per ventilator, about a tenth of what top-of-the-line commercial ventilators cost.

Now, imagine that a firm, let’s call it LAX, that has significant business experience, contacts, and financing in Latin America perceives a need and wants to make money using exactly the same designs. They are welcome to use the freely published designs of the Respiraworks and SmithVent team. However, the FDA does not authorize designs; it authorizes firms to make ventilators. Let’s say the nation of Peru wants to purchase 10,000 SmithVent-based ventilators made from LAX, but demands that they get FDA authorization or the equivalent in Peru. LAX may freely reuse the tests and documentation that the OSMD firm published when they got their original authorization, but must still apply for authorization itself, because of course it is using its own manufacturing and subcontracting process.

Now let’s say LAX does convincingly hurdle these requirements  and get regulatory approval. LAX is  free to sell devices for whatever amount they see fit at a profit to Peru if they can; the OSMD firm does not get in the way.

Of course, the Peruvian government knows that the OSMD firm is already producing and selling these at USD$3000 in Africa, but the OSMD firm may not be able to supply 10,000 such ventilators. Nonetheless, the existence of the OSMD firm may exert downward pressure on the price LAX may be able to negotiate, because Peru can always say, “We will not pay $10,000 for these, we think we can get them at a third of that from an OSMD and it is a similar product.”  Alternatively, the Peruvian government may allow for a higher price point to have local manufacturing of the device; either way, that multiple firms can produce the exact same device creates true competition in the market, and individual participants in the market can make rational decisions about how to proceed with their operations.

Thus the OSMD firm does not prevent firms from making a profit; in fact, LAX possibly could never have afforded to duplicate the work that the OSMD firm did to prove the machines are safe. (The OSMD firm may very well have reused designs, tests, and documentation produced by Helpful Engineering or some other design-oriented organization to assist them.) Thus the OSMD firm created a profit opportunity that did not previously exist for LAX, though at the same time perhaps limiting the maximum per-unit profit they can hope to make.

A Graphical Process Representation

An image representation of how money, devices, designs, documentation, and clearance is depicted below in a Plectica map. Note that in the current ecosystem, strong organizations are focusing on design but few OSMD manufacturing firms are manufacturing new products. The hypothetical firm LAX would be and OSMD Manufacturer in the graphic below.

 

Example 2: A Nation (P) uses the designs without approvals

Suppose:

  • A poor nation (P) has little hard currency and, due to regional conflict, has difficulty trading.
  • It is struck by an acute disaster (not COVID-19), either man-made or natural. For example, they might suffer from a disease, drought or fire which creates respiratory distress, and need nebulizers, humidifiers, or BPAP machines.
  • (P) has a pressing need measured in days or weeks for a device which an OSMD firm has in its library of curated designs, and for which OSMD has been given US FDA approval.

Although (P) does not have time to obtain FDA or similar authorization for these devices, by marshalling many local resources they are able to manufacture them. Although their manufacturing process is not as controlled as one might desire, they have confidence in the design because they can copy a proven, documented, approved design, making only modest changes to the device due to supply-chain disruption. The design may have come from one team, and the OSMD firm may have developed manufacturing techniques around that design; both are openly reusable by nation (P). They may be able to save many lives starting with devices curated by the OSMD firm with no interaction or communication with the OSMD firm whatsoever.

An Example OSMD Firm: Tetra Bio Distributed

Although by the principle of non-monopoly there may someday be many organizations doing this, one corporation, Tetra Bio Distributed, has already begun to fulfill this mission by creating and curating one particular emergency medical device.

Tetra Bio Distributed is a collection of diverse individuals who banded together with a shared aim to lend their skills towards supporting those most impacted by the COVID-19 crisis. In the early days of the pandemic, this group looked at all the many ventilator projects, and realized that this aspect of clinical care was well covered, but that there still existed a risk of a surge in cases leaving first responders incapable of rendering medical assistance to those in need of ventilation. Their namesake “Project Tetra” was created to ameliorate that risk by allowing multiple patients to be temporarily ventilated from a single ventilator.  The American Society of Anesthesiologists put forth a list of reasons why a ventilator should never be split; the engineers on Project Tetra used that list as a series of design constraints.  If Tetra could not address those concerns, then the project would not be useful to front line clinicians.


Project Tetra – Most recent exploded part list for the inspiratory complex. (Photo by Mark M. Roden)

Tetra Bio Distributed has always been a humanitarian effort focused on both saving lives and on demystifying the medical device creation process.  All of Tetra’s efforts are readily available on GitHub, including meeting notes, design specifications, bills of material, documentation, and the entire software suite.  The main goal of the project is to create an open source medical device that can be manufactured and serviced anywhere in the world using simply sourceable components and access to 3D resin printers, printed circuit board manufacturing, Raspberry Pis, and information to build and maintain the splitter.  The secondary goal of the project is to produce a fully manufactured and serviceable device to be sold and maintained by Tetra. This manufacturing distinguishes Tetra Bio Distributed from design organizations, making it an OSMD.  All proceeds and profits from this secondary goal will be used to invest in further splitter development and improvements, and to seed the creation of other open source medical devices.

More information about Tetra Bio Distributed can be found at https://tetra.bioTetra Bio Distributed has applied for 501(c)3 status.


Deciding where to glue components. (Photo: Mark M. Roden)

Summary

The open source medical device community has accomplished much, but the fulfillment of these accomplishments requires the creation of one or more Open Source Medical Device manufacturing firms. These firms must combine designs, finance, and, hardest of all, careful attention to the quality assurance and risk assessment needed to satisfy the US FDA and other regulatory agencies. This combination of skills is rarely found in one person, therefore management and executive leadership are highly demanded.

Call to Action

In order to make open source medical devices a life-saving reality in the future, we need to forge a community including firms such a Tetra Bio Distributed and a greatly expanded library of open source medical devices. As a community, we can learn how to best meet the technical and financial needs over the coming months. Whether that can be accomplished by one or many overlapping and cooperating organizations will be communally learned as we grow the set of curated, safe, well-documented open-source designs.

You can help by:

  • If you have money, donate to Tetra Bio Distributed, Public Invention, Open Source Medical Supplies, or Helpful Engineering.
  • If you have writing, legal, business, graphic art, manufacturing, medical or engineering skills, choose one of these organizations and volunteer your time and talent.
  • If you have an entrepreneurial or organizational spirit, consider creating a new OSMD-class manufacturing organization to curate and manufacture some set of open source medical device designs.

About Public Invention

Public Invention is a US 501(c)3 public charity that invents “In the public, for the Public.” The medical devices curated by an OSMD firm need not be new inventions but practical open source designs of well-understood ideas. During the pandemic, Public Invention has been working on pandemic-related response, whether it rises to the level of new invention or not. Many of the volunteers who work on OSMD projects will likely have an interest in Public Invention, but Public Invention and OSMDs will remain separate organizations.

About Open Source Medical Supplies

Open Source Medical Supplies is a 501(c)3 non-profit and trusted resource in the open source movement. Having brought together a global network of over 70,000 makers, fabricators, community organizers, and medical professionals, the community works together to meet the unprecedented medical supply challenges stemming from the COVID-19 pandemic. To support this community, OSMS provides curated open-source designs and requirements for over 110 open source medical supply designs across 38 design categories.

About Helpful Engineering

Helpful Engineering is a 501(c)3 non-profit that organizes 18,000 global volunteers via a Slack team anyone can join. They support a large number of open source volunteer projects, from PPE projects and ventilators. It has become the main communication hub of the COVID-19 volunteer engineers.

*“Enlightenment is always preceded by confusion.” — Milton H. Erickson

Photos: Mark M. Roden, Project Tetra, Tetra Bio Distributed.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Back To Top