Published in


Oxygen Concentrators and Affordability

A write up from our Oxygen CoLab meet up series

Oxygen affordability is a big topic at the moment given COVID-19. The Oxygen CoLab started asking affordability questions out of a need to align with an evolving market. To understand the areas where innovation is needed, what people are working on, and how the total cost of ownership can be kept low, whilst improving device quality for Low Resource Settings.

This article covers some of the key insights and learnings from an Oxygen CoLab meet up on Oxygen Affordability. On April 15th 2021 we were joined by over 20 Oxygen experts from different ends of the Medical Oxygen ecosystem. A huge thank you to those that joined us and contributed to a rich discussion. The size of this topic warrants much more time for further debate, nonetheless here is what emerged during the short time we shared.

To get us warmed up, we asked participants what they wanted to focus on from several topics. The group aligned on wanting to discuss the total cost of ownership of Oxygen Concentrators, so we started with a rough equation.

Next we crowdsourced some of the collective knowledge in the room. We asked our audience of Oxygen experts to tell us; What is the largest part of this total cost of ownership equation over a fixed period (7 years) in relative terms?

What we learnt is that the energy cost for powering the device and the maintenance of the device are seemingly the largest areas of the total cost of ownership equation and therefore the largest targets for potential innovation on affordability.

Firstly, we discovered that we need to update our fixed period time frame of 7 years. Over seven years, it’s likely that a concentrator in low and middle income countries (LMICs) will have to be completely rebuilt. There aren’t so many compressors that can be used for more than 30,000 hours without needing to be replaced and repaired. Typically a sieve bed will need replacing within three years. The standard device warranty is also three years with some even stretching to five years. Few concentrators, even from larger providers, have very robust maintenance programmes.

We learnt that the cost equation question is a lot more complex than it seems, because there are key considerations within the equation to breakdown, such as;

  • Quality is crucial. A cheap concentrator is probably going to break faster than a more expensive one.
  • Energy costs. These will vary a lot depending on whether you’re using a diesel generator or on the grid. CoLab meetup participant Gerry Douglas from Open O2 did a quick on-the-fly calculation for the group, based on 13.9 cents per kwh and a concentrator using 400 watts, run for 10 hrs per day, 365 days a year, for seven years, is $14,120. This is 2–3 times the basic price of the concentrator (excl shipping).
  • Cost of training. This is an important component but difficult to quantify on a per concentrator comparison basis. Training a biomedical engineer may put many more concentrators back into the system, which becomes more cost effective than buying new ones.
  • Backup system. If the power’s out for a long time and you don’t have a backup and need to use cylinders, or if your concentrator breaks and sits in a store for a year. There’s no direct costs from a backup system but the indirect cost must be factored in.

The ecosystem is key to the cost

Concentrators have generally experienced a lack of maintenance. Bev Bradley, Technical Specialist from UNICEF’s Supply Division said that “there is a total shortage of Biomedical Engineers and much needed support in LMICs. We need more groups that can do maintenance as a long term solution. We need to really start thinking differently about maintenance”.

Elana Robertson, Sn Commercial Lead, GH Labs seconded this notion and shared her experience of the issue, “the biggest issue with maintenance for Kenyan, Nigeria and Ghana Ministries of Health, which is the three countries that we conducted a workshop in, is that there were not enough trained engineers”. In some countries there is one Biomedical Engineer per whole state or country. The problem really needs new thinking.

In addition, things like filters are a simple thing to replace nonetheless; if there is no engineer, no one with the accountability to replace filters, and no global supply chain for parts, such as medical filters, then the device will fail without preventative maintenance. The alternative is a more robust concentrator with smarter technology that does not require filters, which will cost a lot of money. The robustness of the system around the concentrator is key to its sustained usage.

Elana also said that, ‘this is not about how to make an old concentrator better, but how to make it fit into a system more effectively. People die when one part of a system fails, from cylinders not being delivered on time, or governments not paying for the cylinders on time. Cylinders will never be replaced. The system will never be all cylinders or all concentrators. I think there’s different ways we need to think about this in the future and it’s not about the concentrator being more robust but also how to work on the system around it too.’

The group shared that parts are often inexpensive to replace, so it’s not an issue of price but of supply chain, logistics, and different bottlenecks that happen during the process.

  • One argument is to build a more robust device to not need repairs and therefore pay more money upfront for that device, consequently shifting the proportion of one’s investment upfront rather than to the tail end of the device’s lifetime
  • Alternately, we need more robust maintenance systems, from the ecosystem of replacement parts, to the alarms and accountability for servicing the device. As an example, how can we get a global system for parts in place? Perhaps the way to do this is to start off with in-country assembly of devices or in-country packing, but this will take time and an attractive market to do so

Bev Bradley from UNICEF’s procurement centre also shared that, “spares and sustainability is something we think about a lot and so one thing we’re working on with concentrators is preparing kits of both filters and spare parts… so that for concentrators that UNICEF have supplied we can be part of a supply chain for the spare parts needed to keep devices working over their lifespan. Obviously, it’s great if there are local distributors and ways to access parts locally too. The challenge we faced is that every model has their own bespoke list of spare parts, so a kit that has to be created for every different make and model of concentrator is difficult to manage… If anybody is interested and wants to brainstorm with us we can keep exploring how UNICEF can play a role here.”

Business Model Needs

The challenge is that these devices are not made for environments such as those found in Africa, where dust gets into the filters at a higher rate, and humidity affects the sieve beds and compressor. As a result these components need replacing more often and in the meantime the unit must draw more electricity and work 2–5% harder while the components are compromised.

Increasing the durability of components may increase the cost of the unit in the first instance, but in terms of total cost of ownership over the unit’s lifetime, the reduction in maintenance costs through fewer component replacements could be comparable, although more research is needed to understand impacts on the total cost of ownership.

Out of the box thinking is needed to work within the constraints of a complex system. Here are further examples raised in the workshop.

  • No company can be expected to manufacture a product without making a bit of profit, which for oxygen concentrators is already a minimal margin and the competition on device cost is sensitive.
  • We need to think in terms of value to the Ministries of Health to understand the value of the cost of the product.
  • Buyers spend money on three year warranties, but the link between the end user of those devices and who will be on the ground in the country to honour the warranty still needs to be made. A lot of buyers are unaware of this need for linkages. And we need to make a clear distinction between what a warranty includes and then when we start getting into a service level agreement and what that might include.

Device feature optimisation that reduces total cost of ownership

Participants felt that these features would bring down the total cost of ownership;

  • An easy example is membrane technologies that are not as fragile as zeolite and increase the life of the sieve bed, albeit this is a longer term innovation.
  • A clear and simple mechanism to prevent overdraw would help end users. Drawing 10 litres out of a 5 litre machine makes the devices life shorter.
  • More robust compressors, given they have not been developed for the African environment. Wobble compressors seem to be the best option at the moment.
  • An indication of sieve bed degradation. Make an easier indication to the user for when it needs changing that the user can act upon.
  • A cost effective DC powered compressor would make a big difference on energy consumption. Designed for 30,000 hours instead of 20,000 hours.
  • Literature has shown it is possible to reduce the concentrator energy consumption by half on larger devices, albeit this is not yet commercialised. This would result in a significant reduction to total cost of ownership

Where do oxygen concentrators fit into the market?

There does seem to be a transition towards piped oxygen and PSA plants, but there’s still a real role for concentrators within these systems given there’s not enough supply. Portable concentrators are needed as part of a backup system, there are many points of failure in any cylinder based solution so a backup is critical to continuous oxygen supply.

One participant talked to a study in Kenya that showed how a rural patient needs to travel 100 kilometres before they get to the first oxygen point. There are even rural setting where the roads are bad and Oxygen cylinders are out of range for delivery. There is therefore a real need for concentrators in primary health care centres in small district and rural clinics.

GH Labs and GCE developed a medium pressure device that stores oxygen by connecting to a concentrator to create oxygen cylinders. At this stage, the solution fits best into the ecosystem at the district or regional hospital level, which during COVID is where most patients tend to go.

A storage device attached to a concentrator is really useful where there are issues with electricity. When you have electricity you generate oxygen and store capacity and when the electricity is off users can use the stored supply. This means the concentrator is a primary source of oxygen as opposed to a back up making medium pressure reservoirs an adaptable component of an Oxygen system.

Each country has a different strategy for delivering care and even within tertiary care the supply of oxygen is different from hospital to hospital. In India one participant explained that there is a need for oxygen concentrators in some wards but not all wards. Hospitals may have plumbed in oxygen for ICU and surgical units, but in maternal care or in paediatric ICU they use oxygen concentrators. Concentrators might also be used where there is a semi-reliable electrical supply but the roads make cylinder delivery challenging. A variety of solutions will always be needed.

The critical component in developing a concentrator is to make sure that it can sustain oxygen supply without outages or a total breakdown during power issues, which often happens. The concentrator must be robust enough to survive a variable electricity supply.

How might we increase the number of hours of access to oxygen?

GH Labs and GCE’s medium pressure system worked on addressing variability in supply by providing greater autonomy to the user. Users like Oxygen supplies with more autonomy and without the need for infrastructure and a big oxygen plant investment to do so. This hybrid system is attractive because it mixes existing devices and is similar to a cylinder system.

When multiple health workers are depleting a finite supply of cylinders it can become tricky to manage whereas an Oxygen concentrator generates Oxygen supply with more abundance. For instance, during the night there’s less pressure on electricity so the tanks can be refilled, then during the day the concentrator can be used as normal, so there’s flexibility.

Gareth Pemberton, Director of Innovation at GCE said that

“when we launched our project roughly three years ago, we couldn’t find concentrators because they were stored away somewhere awaiting maintenance. That’s why a robust concentrator that can withstand all the fluctuation in electricity is very important, because you need as many as you can to generate enough oxygen. With two or three concentrators our device creates greater access for more than one patient at a time. This is why efficient solutions are so important for insufficient electrical supplies.”


We started with a question of what feature innovation will bring down the total cost of ownership of a device, but we learnt that perhaps we were asking the wrong questions. The group’s consensus was that total cost of ownership is most affected by the local ecosystem for that device. Biomedical engineer training and maintenance and accessible spare parts will extend the life of a device beyond the typical device lifetime in the field. The ecosystem the device is maintained within is as important as making the device more robust.

In addition, it’s hard to quantify the value of new innovations to reduce maintenance and make Oxygen concentrators more robust. On the one hand, making a device more robust is likely to also increase the unit cost. On the other hand, making a device more costly if it reduces the total cost of ownership over its lifetime is worthwhile, assuming those that purchase and use the device also see this economic value. Finding the right balance of improvements that save total cost of ownership and make the device more robust in the most meaningful way is the challenge.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store