[82lsju]

You can probably strip down a vent to have a power switch and four nobs: the percent oxygen (20 - 100%), the mode (whether the vent delivers a constant pressure or fixed volume), the backup respiratory rate, and the pressure or volume delivered, and you build it to recognize and alarm if there is an obstruction and if the tube gets disconnected.



BC

yes you probabley can, and this gets to the issue of what's the fastest product you could make that would be "useful", the MVP you have mentioned previously, maybe not economically viable in normal times given limited functionality, but maybe needed now, and maybe thrown away in 1-2 years

the supply chain question of what parts you could get still exists, and/or if you could make them with non-standard parts (maybe new 3D printed parts) that would be "good enough" for 3 or 6 or 9 months before failing to get over the hump......

[82lsju]

You can probably strip down a vent to have a power switch and four nobs: the percent oxygen (20 - 100%), the mode (whether the vent delivers a constant pressure or fixed volume), the backup respiratory rate, and the pressure or volume delivered, and you build it to recognize and alarm if there is an obstruction and if the tube gets disconnected.



BC

yes you probabley can, and this gets to the issue of what's the fastest product you could make that would be "useful", the MVP you have mentioned previously, maybe not economically viable in normal times given limited functionality, but maybe needed now, and maybe thrown away in 1-2 years

the supply chain question of what parts you could get still exists, and/or if you could make them with non-standard parts (maybe new 3D printed parts) that would be "good enough" for 3 or 6 or 9 months before failing to get over the hump......

[Goose]

There would be advantages to engineers working to strip out unnecessary modes and simplify (and cheapen) ventilators.  In normal times, you might build a general purpose ventilator that can deal with any possible issue that comes up and can perform all sorts of complicated ventilation modes.  But in this case, you are building for speed and cost, so you strip out all unnecessary modes, all complicated UI's and include only the bare essentials (plus you want to make it easy to use in case you are short respiratory therapists in a crisis).


You can probably strip down a vent to have a power switch and four nobs: the percent oxygen (20 - 100%), the mode (whether the vent delivers a constant pressure or fixed volume), the backup respiratory rate, and the pressure or volume delivered, and you build it to recognize and alarm if there is an obstruction and if the tube gets disconnected.



BC

All True. But none of that will matter in the next two weeks. Getting the paperwork out to the assemblers and the new test instructions to QA will take longer than that. I also imagine there are some at least minimal FDA device approvals in play. I have had many instances in my career as an engineer where we had great ideas to simplify a device. When we went to the production people they said it would cost more and take longer to remove the extra functions than to just build what we had. If you have a large quantity and some time, yes. Two weeks, even probably a month, it is just going to slow you down.

[Goose]

There would be advantages to engineers working to strip out unnecessary modes and simplify (and cheapen) ventilators.  In normal times, you might build a general purpose ventilator that can deal with any possible issue that comes up and can perform all sorts of complicated ventilation modes.  But in this case, you are building for speed and cost, so you strip out all unnecessary modes, all complicated UI's and include only the bare essentials (plus you want to make it easy to use in case you are short respiratory therapists in a crisis).


You can probably strip down a vent to have a power switch and four nobs: the percent oxygen (20 - 100%), the mode (whether the vent delivers a constant pressure or fixed volume), the backup respiratory rate, and the pressure or volume delivered, and you build it to recognize and alarm if there is an obstruction and if the tube gets disconnected.



BC

All True. But none of that will matter in the next two weeks. Getting the paperwork out to the assemblers and the new test instructions to QA will take longer than that. I also imagine there are some at least minimal FDA device approvals in play. I have had many instances in my career as an engineer where we had great ideas to simplify a device. When we went to the production people they said it would cost more and take longer to remove the extra functions than to just build what we had. If you have a large quantity and some time, yes. Two weeks, even probably a month, it is just going to slow you down.

[BostonCard]

I’ll defer to your expertise here. I can see that for a factory already tooled for ventilators, and experienced factory workers, changing designs would require time and effort and you may be better off cranking out more of what you’ve been cranking out. But I wonder, for a factory whose experience is making cars or vacuum cleaners, it strikes me that handing them a design stripped of all complexities will be easier to execute than the design that has all the bells and whistles.

BC

[BostonCard]

I’ll defer to your expertise here. I can see that for a factory already tooled for ventilators, and experienced factory workers, changing designs would require time and effort and you may be better off cranking out more of what you’ve been cranking out. But I wonder, for a factory whose experience is making cars or vacuum cleaners, it strikes me that handing them a design stripped of all complexities will be easier to execute than the design that has all the bells and whistles.

BC

[Goose]

I’ll defer to your expertise here. I can see that for a factory already tooled for ventilators, and experienced factory workers, changing designs would require time and effort and you may be better off cranking out more of what you’ve been cranking out. But I wonder, for a factory whose experience is making cars or vacuum cleaners, it strikes me that handing them a design stripped of all complexities will be easier to execute than the design that has all the bells and whistles.

BC

If the paperwork is already done, and it is correct, then a simpler product could be easier and cheaper to build. However, any new factory or builder, even a very good one, will have significant start-up time. They have to learn to interpret the drawings made by another company who has tons of corporate knowledge about the product. The new company/contract manufacturer does not. Some of the processes and terms are a complete mystery to them. They don't have the SOPs that the original manufacturer has, so when it says "install in accordance with SOP XXXX", the assembler has no idea what that means. They have to go look it up and figure out how to actually do it. The QA people have to be qualified to inspect and test the devices. They know how to inspect car doors. Ventilators are a whole different beast. Can they learn? Sure, but all this takes time.
Potentially more important, any robotics used by the original manufacturer will have to be re-programmed for the new, simpler configuration. The new builder would need to have robotics that are compatible with that programming. That can take a long time to sort out. So it is not so simple. Once up and running, all will be easy, but it will probably take a month best case and maybe significantly more.

[Goose]

I’ll defer to your expertise here. I can see that for a factory already tooled for ventilators, and experienced factory workers, changing designs would require time and effort and you may be better off cranking out more of what you’ve been cranking out. But I wonder, for a factory whose experience is making cars or vacuum cleaners, it strikes me that handing them a design stripped of all complexities will be easier to execute than the design that has all the bells and whistles.

BC

If the paperwork is already done, and it is correct, then a simpler product could be easier and cheaper to build. However, any new factory or builder, even a very good one, will have significant start-up time. They have to learn to interpret the drawings made by another company who has tons of corporate knowledge about the product. The new company/contract manufacturer does not. Some of the processes and terms are a complete mystery to them. They don't have the SOPs that the original manufacturer has, so when it says "install in accordance with SOP XXXX", the assembler has no idea what that means. They have to go look it up and figure out how to actually do it. The QA people have to be qualified to inspect and test the devices. They know how to inspect car doors. Ventilators are a whole different beast. Can they learn? Sure, but all this takes time.
Potentially more important, any robotics used by the original manufacturer will have to be re-programmed for the new, simpler configuration. The new builder would need to have robotics that are compatible with that programming. That can take a long time to sort out. So it is not so simple. Once up and running, all will be easy, but it will probably take a month best case and maybe significantly more.

[BostonCard]

Apparently we tried the "cheap ventilator" route before.

https://www.nytimes.com/2020/03/29/busin...rtage.html

The government shouldn't have let Covidien off the hook.  The ventilator contract was something that they should have known about when they did due diligence to acquire Newport, and if Newport was saddled with a bad contract with BARDA, maybe Covidien should have offered less.  Of course it is entirely possible that Covidien acquired Newport specifically so it wouldn't fulfill the contract:

Government officials and executives at rival ventilator companies said they suspected that Covidien had acquired Newport to prevent it from building a cheaper product that would undermine Covidien’s profits from its existing ventilator business.

BC

[BostonCard]

Apparently we tried the "cheap ventilator" route before.

https://www.nytimes.com/2020/03/29/busin...rtage.html

The government shouldn't have let Covidien off the hook.  The ventilator contract was something that they should have known about when they did due diligence to acquire Newport, and if Newport was saddled with a bad contract with BARDA, maybe Covidien should have offered less.  Of course it is entirely possible that Covidien acquired Newport specifically so it wouldn't fulfill the contract:

Government officials and executives at rival ventilator companies said they suspected that Covidien had acquired Newport to prevent it from building a cheaper product that would undermine Covidien’s profits from its existing ventilator business.

BC

[Goose]

Unfortunately this has been the trajectory of doing business with the US Government for a while. Look no further than the number of defense contractors that existed in the Bay Area in 1980 vs the number that exist today. A small company with low overhead and good people can make a product that the Government wants to buy at a reasonable price. The company is very successful and does quite well for a couple of years. Then the contract is over, the Government has what it needs for the next 5 years or more. What now? No big volume business and really no hope of such business. If the small company is fortunate, it can book some foreign sales, but it quickly becomes clear the market is very limited and quite fragile. A better funded competitor can "buy in" to the business and easily wipe out the small company at any time.

If the small company is smart, they will figure this out before it actually happens to them. Often the small company is bought up by a larger company who thinks the big backlog and professional staff makes it a good buy. The small company puts lots of lipstick on the pig to further that notion. After the purchase is complete, the buyer's remorse sets in. They discover that with the "drag" induced by the "big company" overhead and systems, they can't make the product for anywhere near as low a cost as the small company could. Their "due diligence" was done by a bunch of consultants that know nothing about actually building anything. The spreadsheets said it would work. Problem is, the numbers were wrong.

What happens next varies, but it generally means the small company is gone and the Government has to deal with it's own mirror image in the big company. The government actually prefers that, because it is familiar and "safe". It may not be efficient or low cost, but it is predictable and has people to fill out all the right forms in a timely manner. There may be repeated delays and various problems, but they will be well documented so nobody is "at fault". Welcome to 21st century government procurement.

[Goose]

Unfortunately this has been the trajectory of doing business with the US Government for a while. Look no further than the number of defense contractors that existed in the Bay Area in 1980 vs the number that exist today. A small company with low overhead and good people can make a product that the Government wants to buy at a reasonable price. The company is very successful and does quite well for a couple of years. Then the contract is over, the Government has what it needs for the next 5 years or more. What now? No big volume business and really no hope of such business. If the small company is fortunate, it can book some foreign sales, but it quickly becomes clear the market is very limited and quite fragile. A better funded competitor can "buy in" to the business and easily wipe out the small company at any time.

If the small company is smart, they will figure this out before it actually happens to them. Often the small company is bought up by a larger company who thinks the big backlog and professional staff makes it a good buy. The small company puts lots of lipstick on the pig to further that notion. After the purchase is complete, the buyer's remorse sets in. They discover that with the "drag" induced by the "big company" overhead and systems, they can't make the product for anywhere near as low a cost as the small company could. Their "due diligence" was done by a bunch of consultants that know nothing about actually building anything. The spreadsheets said it would work. Problem is, the numbers were wrong.

What happens next varies, but it generally means the small company is gone and the Government has to deal with it's own mirror image in the big company. The government actually prefers that, because it is familiar and "safe". It may not be efficient or low cost, but it is predictable and has people to fill out all the right forms in a timely manner. There may be repeated delays and various problems, but they will be well documented so nobody is "at fault". Welcome to 21st century government procurement.

[82lsju]

Ford, in collaboration with GE Healthcare, will leverage the design of Airon Corp.’s FDA-cleared ventilator to produce in Michigan

Ford expects to produce 50,000 of the ventilators within the next 100 days, with the ability to produce 30,000 a month thereafter as needed

The simplified ventilator design – licensed by GE Healthcare from Florida-based Airon Corp. – is responsive to the needs of most COVID-19 patients, and operates on air pressure without the need for electricity

https://media.ford.com/content/fordmedia...-days.html

Early on Friday, March 20, the G.M. team sat down with Ventec executives to learn how the ventilators are made, and what parts are required. Ventec had already started a push to ramp up production to 1,000 a month. The group concluded that with G.M.’s resources, 20,000 a month would be possible, four people familiar with the talks said.

The next day, G.M. emailed its suppliers specifications of Ventec parts, asking if any could produce them in high volumes. Mr. Kienle’s team quickly zeroed in on Kokomo as a location to assemble the machines, a person familiar with the matter said. The plant, unlike much grittier car assembly factories, has the type of clean room needed for making medical devices.

On the evening of Sunday, March 22, G.M.’s purchasing chief, Shilpan Amin, emailed Ms. Barra and other top executives to let them know that the company and Ventec had secured commitments from suppliers to produce 95 percent of the needed parts, according to three people familiar with the email.

By last Tuesday, G.M. and Ventec had the details of their collaboration hammered out, which they discussed publicly early in the week. G.M. would operate as a contract manufacturer for Ventec, which would sell and distribute the machines. Ventec also plans to increase production at its plant in Washington State.

https://www.nytimes.com/2020/03/30/business/gm-ventilators-coronavirus-trump.html

[82lsju]

Ford, in collaboration with GE Healthcare, will leverage the design of Airon Corp.’s FDA-cleared ventilator to produce in Michigan

Ford expects to produce 50,000 of the ventilators within the next 100 days, with the ability to produce 30,000 a month thereafter as needed

The simplified ventilator design – licensed by GE Healthcare from Florida-based Airon Corp. – is responsive to the needs of most COVID-19 patients, and operates on air pressure without the need for electricity

https://media.ford.com/content/fordmedia...-days.html

Early on Friday, March 20, the G.M. team sat down with Ventec executives to learn how the ventilators are made, and what parts are required. Ventec had already started a push to ramp up production to 1,000 a month. The group concluded that with G.M.’s resources, 20,000 a month would be possible, four people familiar with the talks said.

The next day, G.M. emailed its suppliers specifications of Ventec parts, asking if any could produce them in high volumes. Mr. Kienle’s team quickly zeroed in on Kokomo as a location to assemble the machines, a person familiar with the matter said. The plant, unlike much grittier car assembly factories, has the type of clean room needed for making medical devices.

On the evening of Sunday, March 22, G.M.’s purchasing chief, Shilpan Amin, emailed Ms. Barra and other top executives to let them know that the company and Ventec had secured commitments from suppliers to produce 95 percent of the needed parts, according to three people familiar with the email.

By last Tuesday, G.M. and Ventec had the details of their collaboration hammered out, which they discussed publicly early in the week. G.M. would operate as a contract manufacturer for Ventec, which would sell and distribute the machines. Ventec also plans to increase production at its plant in Washington State.

https://www.nytimes.com/2020/03/30/business/gm-ventilators-coronavirus-trump.html

[82lsju]

Five years ago, the U.S. Department of Health and Human Services tried to plug a crucial hole in its preparations for a global pandemic, signing a $13.8 million contract with a Pennsylvania manufacturer to create a low-cost, portable, easy-to-use ventilator that could be stockpiled for emergencies.

This past September, with the design of the new Trilogy Evo Universal finally cleared by the Food and Drug Administration, HHS ordered 10,000 of the ventilators for the Strategic National Stockpile at a cost of $3,280 each.

But as the pandemic continues to spread across the globe, there is still not a single Trilogy Evo Universal in the stockpile.

Instead last summer, soon after the FDA’s approval, the Pennsylvania company that designed the device — a subsidiary of the Dutch appliance and technology giant Royal Philips N.V. — began selling two higher-priced commercial versions of the same ventilator around the world.

https://www.propublica.org/article/taxpa...ial#181338

[82lsju]

Five years ago, the U.S. Department of Health and Human Services tried to plug a crucial hole in its preparations for a global pandemic, signing a $13.8 million contract with a Pennsylvania manufacturer to create a low-cost, portable, easy-to-use ventilator that could be stockpiled for emergencies.

This past September, with the design of the new Trilogy Evo Universal finally cleared by the Food and Drug Administration, HHS ordered 10,000 of the ventilators for the Strategic National Stockpile at a cost of $3,280 each.

But as the pandemic continues to spread across the globe, there is still not a single Trilogy Evo Universal in the stockpile.

Instead last summer, soon after the FDA’s approval, the Pennsylvania company that designed the device — a subsidiary of the Dutch appliance and technology giant Royal Philips N.V. — began selling two higher-priced commercial versions of the same ventilator around the world.

https://www.propublica.org/article/taxpa...ial#181338

[magnus]

Saw this update on ventilators


https://apnews.com/e08621567fd8758c89b0c5aed5ac5d72

Sounds like we now have a surplus? (and for sure will have one later in the year (unless a 2nd wave gets much much worse) when we're scheduled to amass 200K new ones).

Anyone hear how many NY actually ended up needing?  I'm curious how close to Cuomo's projection the number got.

[magnus]

Saw this update on ventilators


https://apnews.com/e08621567fd8758c89b0c5aed5ac5d72

Sounds like we now have a surplus? (and for sure will have one later in the year (unless a 2nd wave gets much much worse) when we're scheduled to amass 200K new ones).

Anyone hear how many NY actually ended up needing?  I'm curious how close to Cuomo's projection the number got.

[M T]

Anyone hear how many NY actually ended up needing?  I'm curious how close to Cuomo's projection the number got.

NYC's data is at
  https://www1.nyc.gov/site/doh/covid/covid-19-data.page
The closest I see is daily hospital admissions (not the number in the hospital on any particular day).
Elsewhere in the data, I see ever-hospitalized numbers (44K total to now).

The numbers at
  https://covid19.healthdata.org/united-st...a/new-york
look pretty bad, in terms of beds, ICU beds, and number of ventilators, but I don't know that these are accurate, or what NYC did about a shortfall of 7K beds, 4.5K ICU beds.

[M T]

Anyone hear how many NY actually ended up needing?  I'm curious how close to Cuomo's projection the number got.

NYC's data is at
  https://www1.nyc.gov/site/doh/covid/covid-19-data.page
The closest I see is daily hospital admissions (not the number in the hospital on any particular day).
Elsewhere in the data, I see ever-hospitalized numbers (44K total to now).

The numbers at
  https://covid19.healthdata.org/united-st...a/new-york
look pretty bad, in terms of beds, ICU beds, and number of ventilators, but I don't know that these are accurate, or what NYC did about a shortfall of 7K beds, 4.5K ICU beds.

[Mick]

Anyone hear how many NY actually ended up needing?  I'm curious how close to Cuomo's projection the number got.

NYC's data is at
  https://www1.nyc.gov/site/doh/covid/covid-19-data.page
The closest I see is daily hospital admissions (not the number in the hospital on any particular day).
Elsewhere in the data, I see ever-hospitalized numbers (44K total to now).

The numbers at
  https://covid19.healthdata.org/united-st...a/new-york
look pretty bad, in terms of beds, ICU beds, and number of ventilators, but I don't know that these are accurate, or what NYC did about a shortfall of 7K beds, 4.5K ICU beds.

I don't know about ventilators, but as of April 23, the navy hospital ship departed with comparatively little use.  Only one of the five field hospitals ever had any use.

https://www.usatoday.com/story/news/nati...010754001/

[Mick]

Anyone hear how many NY actually ended up needing?  I'm curious how close to Cuomo's projection the number got.

NYC's data is at
  https://www1.nyc.gov/site/doh/covid/covid-19-data.page
The closest I see is daily hospital admissions (not the number in the hospital on any particular day).
Elsewhere in the data, I see ever-hospitalized numbers (44K total to now).

The numbers at
  https://covid19.healthdata.org/united-st...a/new-york
look pretty bad, in terms of beds, ICU beds, and number of ventilators, but I don't know that these are accurate, or what NYC did about a shortfall of 7K beds, 4.5K ICU beds.

I don't know about ventilators, but as of April 23, the navy hospital ship departed with comparatively little use.  Only one of the five field hospitals ever had any use.

https://www.usatoday.com/story/news/nati...010754001/