10,000 and counting

A deployment story and the engine that powers it

At Swasth, we’re learning to appreciate the small wins, even as we keep the bigger picture clearly in sight. If you look at our live deployment dashboard, you will see that we have now deployed over 10,000 oxygen concentrators (OCs) across 22 states in the country as a part of #OxygenforEveryone — the Swasth Alliance and Act Grants initiative. This is almost three times as many as at the time of our previous post, but still a fifth of where we want to reach. In our demand dashboard, you will see the significant unmet need that remains.

These are not just numbers to us. Behind each concentrator being imported, deployed, and put to use is a massive coordination effort among an equally large number of people — public, private, and non-profit organisations, our small team, a dedicated band of volunteers, generous donors and partners. And sometimes, like in the case of Goa last week, even the Indian Air Force.

Last weekend, when Goa’s positivity rate was peaking, our algorithm highlighted the need for urgent oxygen support quite clearly. We regularly track and analyse district-level epidemiological indicators including the number of active cases and number of deaths using data from Covid19India.org. Put together, this data helps us prioritise the districts that have the greatest need at the moment when we receive confirmation from the ACT Grants sourcing team that a batch of OCs is set to arrive in India. This composite picture of need then becomes the basis for allocating OCs to entities in any given district.

We allocated 323 OCs to Goa’s districts last weekend, adding to a number of OCs that had already been delivered. Typically, once we have determined the highest priority districts in the country, we determine how to distribute the OCs equitably within each district. The majority of the allocation goes to government health centres, where the need is the greatest, and the remainder is divided among NGOs. In the case of Goa, our OCs were headed to Goa Medical College and district hospital.

Once their target destination is decided, the OCs ordinarily travel there by road, as arranged by our delivery partners. The OCs allocated to Goa were already in a Delhivery fulfilment centre in Delhi, undergoing the routine inspection and customs processes. A journey by road from Delhi to Goa (in case you’re ever planning to do one, perhaps in a post-pandemic world) takes at least 36 hours to complete, which meant that the OCs would have reached Goa 5 days after allocations were made. In an ordinary situation, to go from identifying a need to deploying a solution in 5 days is fairly quick.

But in a crisis situation, fast isn’t fast enough.

The fact is that people were in dire need of oxygen already, and if there were faster ways of getting the OCs to their destination, we needed them. Bureaucrats from the Goa state government got in touch with the Indian Air Force (IAF) who airlifted these OCs to the state, getting them to their destinations in a few hours — at least one day sooner than they would have reached by road. One day can make all the difference between life and death.

We are on a learning journey. Over the past few weeks of putting this initiative together, we’ve realised just how unique and — forgive us for using the word — unprecedented an effort like this is. There is no tried-and-tested model for how to do what we’re doing and our methods and processes have grown along the way — in order to make our work more equitable, more robust, more impactful. Given how novel initiatives like ours are, the most important thing we can do is to keep improving as we learn.

For example, when we first started allocating OCs, we looked at state-level epidemiological data, but given the size and diversity of states, this was insufficient. So we moved to looking at district-level data where available.

But even those numbers don’t always tell the whole story, and in much of the country, the figures are only a fraction of the reality. We work around that by looking at testing and positivity rates, to figure out which states’ caseloads might not be representative of the ground reality because of relatively low testing. We account for this now when allocating OCs, but this is again something we had to learn once we started deploying them.

Our demand aggregation process has also evolved as the initiative has grown. In the beginning, we crowdsourced demand for OCs through a Google form that any entity, from health sub-centres and clinics to government agencies, could fill out to express their requirements. We initially shared them with our own networks of healthcare providers, public programs and non-profit grassroots organisations. As an alliance, Swasth was well-positioned to get word out to multiple quarters. From there, word spread organically. Everything about this effort is of and by the people.

While the demand pouring in through this channel was more than we could have met, we knew we were still not capturing the full country’s needs by relying on an exclusively digital channel. And so, we started working with our partners at MyGov, where district magistrates and collectors from urban centres and rural areas of various states have shared their needs for OCs.

With the situation constantly evolving, we also need to validate and verify the demand before making an allocation, which we do through regular due diligence calls by our incredible band of volunteers. This helps us make sure our allocations are not duplicative of other efforts. When we have identified a district as requiring support, but don’t have enough information to make an allocation, we work with our partner organisations that are connected to local governments and organisations to investigate who might require OCs. All of this culminates, of course, in the updated demand and deployment dashboards on our website.

These dashboards are not just meant to guide and share allocation details. They capture the urgency of our needs, act as sources for other organisations working on OC deployment, are a record of this crisis, and can be the starting point for building a more robust health infrastructure across the country.

A big thank you to our team members working tirelessly to pull this off day after day, deployment after deployment: Pijush Sinha (Swasth), Saloni Mehta (Swasth), Natasha Prakash (Swasth/ BCG), Mekin Maheshwari (ACT Grants), Ashish Dave (ACT Grants), Ravi Trivedi (ACT Grants), Siva Sekhar (ACT Grants/ Bounce) and Srinivasan S (Delhivery).

Team Swasth Alliance

Swasth is an alliance of over 150+ healthcare organisations in India, including hospitals, rural health NGOs, insurers, health tech, med-tech and others.