To vaccinate everyone… make an app with open APIs?

Good technology must factor in societal inequities and on-ground complexities

These are terrible times for India. Thousands of people are dying of COVID-19 in the second wave every day. The total toll may be well into the millions already. Experts say that the long-term solution to this crisis is to vaccinate everyone fully and effectively. Unfortunately, there is a crippling vaccine shortage in the very country that has the world’s largest vaccine manufacturing capacity. As of May-end, under 4% of adults have been fully vaccinated – significantly lower than the world average.

The issues behind vaccine procurement are out of scope for this tech-focused article. Here we’d like to focus on the use of digital technologies for administering the vaccines.

Digital divide turns into vaccine inequity

Covishield, India’s version of the Oxford/AstraZeneca coronavirus vaccine, has been available since March 2021, albeit in insufficient numbers. To manage its administration, the Government commissioned an app and website called CoWin.

On the face of it, the app was a sensible decision intended to bring order to the vaccination process. But in a country with a significant digital divide, perhaps it wasn’t a very equitable idea. At various points over the last three months, booking slots on the app was mandatory (for certain age groups), while walk-ins at health centres was not an easy option to avail. To book a slot on the CoWin app, one had to overcome the information divide, know English (the only language the app spoke for a long time), have a smartphone or laptop, a fast Internet connection and deal with SMS-based authentication. 

Furthermore, one needed to constantly monitor the app for any open slots, which often disappeared in minutes. The joke that did the rounds was how one needed to be adept at the 'fastest finger first' game. In one of the planet’s poorest countries, leading with the digital channel CoWin, has left many vulnerable groups such as the poor, elderly and infirm, behind in the race for protection against the deadly virus.

When tech has unintended consequences

If the vaccination distribution system was already less than desirable, CoWin would soon make a change that would further skew the vaccine game in favour of the tech elite – they published an open API for CoWin. Again, this was an understandable decision in theory – allowing third party apps to access slot availability data would bring about some transparency.

But transparency for whose benefit? What the developers perhaps didn’t factor in was the emergence of a cottage industry around the open API. Several third-party apps and Telegram bots monitored CoWin 24/7, and published open slots in real-time in an untold number of increasingly exotic online forums, open-access spreadsheets and websites. Although the API couldn’t actually book anyone a slot, the mere act of publishing open slots widened the digital divide.

Tech-savvy, wealthy young people in cities like Bengaluru could, in effect, 'queue jump' not just in their own cities but also in neighbouring towns. This obviously deprived the locals in those places - especially the vulnerable groups listed above. Thus, it came to be that the knowledge of Python gave people a distinct advantage in getting themselves inoculated against a life-threatening virus. 

To build socially useful technology, understand the context fully

In the Enterprise context, open data is regarded as almost universally good. It leads to the creation of silo-free ecosystems, that can potentially benefit the end-user; open banking is a good example. Unfortunately, the benefits didn’t extend to vaccine apps. We'd urge mining the open access API episode for lessons on building 'good' tech.

Our case isn’t that technology should have no role in the vaccination process. Humanity’s ability to contain the deadliest pandemic in a century depends on vaccinating nearly everyone in a matter of months. We must harness every tool there is; technology makes supply chains for retail faster, so why not for vaccines? Indeed, other countries have made equitable use of technology in vaccinating people: by using apps as one of several good options, including door-to-door drives and bookings at public health centres.

We have two thoughts for our fellow technologists designing societally important systems:

First, ask 'cui bono?' (Latin for 'who benefits?'). Every society has inequities that impact access to technology. Technologists must understand them to answer the 'cui bono?' question effectively. Don’t leave any loopholes for the digital 'haves' to exploit.

Also, collaborate better with field experts, especially in an immensely complex country where millions don’t have access to uninterrupted internet. India’s last mile capabilities are the best in the world – we successfully administered polio drops to 110 million children over a single day this January, without any apps. We believe technologists should be led by field experts. For instance, after having laid out the pros and cons of the tech landscape (the dangers of surveillance, for example), let the experts wrestle with field complexities and map out a way forward. There will always be opportunities where technology can reduce inequities. And, in the case of the current nation-wide vaccination drive – could SMS-based information campaigns help combat vaccine hesitancy among the rural population?

Finally, we’re happy that some of the public’s feedback has made its way into CoWin, and hope the lessons of equity are taken onboard in future endeavours.