It's early when Dominic Nyarko arrives at the sprawling concrete hospital in Kumasi, Ghana's second city. He soon gets to work, checking the Covid vaccine's cold chain storage hasn't been compromised overnight.
By 7.30am the crowds start trickling into the dusty car park, where four large tents have been erected to host the district's vaccine rollout. This week the priority is second shots, and before long a queue has formed.
"It's been a busy time, let's just say we haven't had much sleep," says Nyarko, a public health official overseeing the process of vaccinating roughly 2000 people a day.
But after a hectic week the hospital's already limited supply has run dry. Of some 23,000 people who received their first shot, health workers were able to offer a second to just 10,000.
This shortage is not only a local or even national problem, it's continent-wide: last week, the World Health Organisation (WHO) said Africa urgently needs 20 million doses to provide the second doses needed for full protection.
Globally, too, disparities are acute. Roughly 75 per cent of the 1.8 billion vaccine doses administered worldwide have gone to just 10 countries. Meanwhile nations including Madagascar, South Sudan and Papua New Guinea have vaccinated less than 0.01 per cent of their population.
Back in January, WHO chief Dr Tedros Adhanom Ghebreyesus said the world was "on the brink of a catastrophic moral failure, and the price of this failure will be paid with lives".
Since then little has changed. Writing in the Telegraph on Monday Tedros, alongside the heads of three other UN bodies, said a "two-track pandemic" was developing "with richer countries having access and poorer ones being left behind".
"Inequitable vaccine distribution is not only leaving untold millions of people vulnerable to the virus, it is also allowing deadly variants to emerge and ricochet back across the world," they wrote.
But are things as bad as they seem? And what can be done to vaccinate the world? The hurdles – and the solutions – can be split into three broad categories: making, buying and distributing the shots. Here, we take a look at each.
Making the vaccines
Conversations with vaccine manufacturers about "vaccinating the world" tend to start a little tetchily, with them pointing out how far we've come. And, from a standing start last January, this is undoubtedly true.
According to forecasts from analysts Airfinity, roughly 11.14b shots will have been produced by the end of 2021, which should be enough to vaccinate 75 per cent of the global population. In a normal year, vaccine makers produce roughly 5b shots for diseases including measles, flu and polio – and the new Covid jabs are being produced on top of that.
This is not like the HIV/Aids crisis when antivirals only reached sub-Saharan Africa a decade after they were first approved, says Thomas Cueni, director general of the International Federation of Pharmaceutical Manufacturers and Associations (IFPMA).
"We had the first [Covid-19] vaccines reaching Abidjan, Accra, Kigali, within less than 100 days of the first approval. These are quite different timescales – they arrived on the same day that the first vaccines reached Tokyo."
But others disagree. US President Joe Biden recently added his voice to mounting global calls for a patent or intellectual property waiver to make the "recipe" for each vaccine publicly available, allowing other companies to produce them too.
Bolivia has already signed a deal with Canada's Biolyse Pharma Corp for 15m Johnson & Johnson vaccines, if patent rights are waived. Meanwhile Bangladesh and Indonesia both told a WHO press conference on Friday that they have spare production capacity.
"Eighty to 90 per cent of the medicines for the whole world, including developed countries, are made by companies in low- and middle-income countries," said Abdul Muktadir, chairman and managing director of Incepta Pharmaceuticals in Bangladesh.
"Let us be all clear in our mind that there are untapped, high-quality production capabilities in countries like ours, and these can be utilised to increase the supply of vaccines."
However, much of the industry (plus most European governments) have concerns about the implications of any waiver around recouping costs and stifling innovation.
Even with a waiver in place, it would take between six and nine months for new factories or producers to start turning out product. And finding the trained staff to not only make but assess the quality of the vaccines would be borderline impossible.
In an unprecedented move, companies are already collaborating to help boost supply, with 280 partnership contracts covering the production of Covid-19 vaccines.
Merck, for instance, has agreed to produce vaccines for its rival Johnson & Johnson, while GSK and Novartis are making 100m and 250m doses of the CureVax jab respectively.
Then there's AstraZeneca: through technology transfer deals it has agreed manufacturing contracts for 2.9b vaccine doses with 25 firms in 15 countries – the largest of which is with global vaccine behemoth, the Serum Institute of India (SII).
But perhaps the strongest argument against waivers is this: in October Moderna, one of the producers of new mRNA vaccines, actually offered an IP waiver. No one has yet taken it up.
Instead, "the biggest obstacle is raw materials", says Dr Richard Torbett, chief executive of the Association of the British Pharmaceutical Industry. "All of the companies are saying we could produce more if we only had more glass vials, or filters, or bio bags."
Again, this is a daunting challenge – the Pfizer vaccine, for example, has 260 ingredients that come from 60 companies in 19 different countries. Many of these products are highly specialised and it will take many months, perhaps years, to ramp production of them up.
"We're very likely to see continued shortages that set back some of the vaccine producers for several months," says Rasmus Bech Hansen, chief executive of Airfinity, adding that it is becoming harder for manufacturers with new jabs to secure the needed supplies – CureVac is already facing this problem, for example.
The third challenge is perhaps harder to tackle. Vaccines are biological products and the manufacturing process does not always go smoothly.
According to Airfinity, 1.73b doses have been distributed worldwide, far short of the 4.5b initially projected by big pharma. An overambitious manufacturing target is largely to blame for the gap. AstraZeneca's row with Europe, for instance, was triggered by a lower yield at factories than hoped.
Meanwhile Russia has produced only around 42m doses – compared to 400m from AstraZeneca and Pfizer – amid difficulties producing the second dose of Sputnik V, which uses different adenoviruses in the first and second shot.
Many of the solutions proposed are long term: the European Union, for instance, has promised to invest €1b ($1.6b) in building capacity across Africa to reduce the continent's reliance on imports – but this will not come to fruition for years.
Buying the vaccines
Most commentators agree that the key to vaccinating the world is sharing what is already available. Ahead of the G7 meeting in Cornwall this month, these calls are becoming more urgent amid fears unvaccinated regions could be the next India.
The foundations for today's unbalanced rollout were laid early last year, when wealthy countries were able to splash the cash early and secure supply ahead of others. Most contracts also give wealthy nations a first call on future doses.
The Coalition for Epidemic Preparedness and Innovation (Cepi) – a co-founder of the Covax distribution scheme, alongside the WHO and Gavi, the Vaccine Alliance – also invested in a dozen candidates early, to the tune of US$1.4b ($1.9b).
But the Covax scheme – formed last April with the aim of protecting 20 per cent of the world by the end of 2021 – simply didn't have the money to secure large supply deals until last summer; the first deal for 300m doses from AstraZeneca was announced in June.
"Covax didn't have the financial resources at the very beginning of the pandemic," says Cueni. He adds that the scheme initially bet on vaccines which appeared easier to distribute, rather than the more expensive and complex mRNA jabs made by Pfizer and Moderna.
By the time Covax turned to these companies, "many governments had already signed up most of the available capacity", Cueni says.
A year later, distribution through Covax has been disappointing. According to Unicef, which coordinates shipping, 77.7m doses have been sent worldwide – just a third of the 252.5m shots the scheme predicted it would deliver by the end of June.
Part of the problem has been what is, in hindsight, an overreliance on one manufacturer: the Serum Institute of India (SII), which had been set to produce two thirds of vaccines delivered by Covax by June.
But as India's coronavirus crisis escalated the government banned exports to focus on domestic rollout, replicating what much of the wealthy world has already done.
This decision has had a disproportionate impact on lower income countries set to receive shots for free through Covax. Of these 92 countries, 60 were earmarked to receive SII shots.
"Nobody expected this surge in the pandemic in India," Pascal Soriot, chief executive of AstraZeneca, told a Devex briefing. "The good news is our vaccine is making a big difference in India. But of course, the not so good news is we are struggling in supply to Covax and the other countries."
India has said it will resume exports in October, by which point Airfinity estimates it will have vaccinated 41 per cent of its population.
In the meantime, Covax is exploring several avenues to mitigate this, including rejigging planned supply routes – so countries set to receive SII shots instead get them directly from AstraZeneca, or use Pfizer or J&J instead.
It is also exploring scaling up manufacturing elsewhere, extending the gap between doses and even – if trial data permits – mixing and matching shots.
"It's certainly given a lot of pause for thought on what the portfolio strategy for Covax needs to be going forward, to ensure there isn't the same level of exposure to any one country or one supplier," a Unicef official says.
But wealthy nations sharing doses will be the quickest way to address stark imbalances in the global rollout, says a Unicef official. And, according to Airfinity, this is backed up by the data.
By the end of the year, six countries and blocs are expected to have a combined surplus of more than 2.6b doses: the EU with 885m; US with 539m; Japan with 300m; UK with 297m; Brazil with 177m and Canada with 175m.
Together, these jurisdictions will account for 89.7 per cent of the likely global surplus of Covid-19 vaccines by the end of this year – more than 2.6b doses. "To put this surplus in perspective, with the expected production and delivery schedule, Covax is expected to receive 1.11b doses by the end of this year," Airfinity says.
Unicef analysis also suggests that most G7 countries could share vaccines immediately without substantially delaying domestic vaccination programmes.
"It's not good enough to wait until the end of the year to donate doses that we need right now," a Unicef official says. "It's perfectly possible for those countries to both vaccinate their own populations and make available a significant portion of those doses for donations."
In particular, it is critically important to vaccinate healthcare workers and the most vulnerable, they add. Unless medics had already been largely vaccinated in India, for instance, an entire generation of doctors and nurses could have been wiped out by Covid.
But in the void between expectations and reality, countries scrambling to secure shots are increasingly turning elsewhere.
China, in particular, is executing a sophisticated campaign of vaccine diplomacy and has exported 323.3m jabs worldwide to date. This is more than double the number exported by the EU (143.8m) and dwarfs the meagre 7.5m exported by the world's biggest vaccine producer, the US.
Distributing the vaccines
Supply is one thing, but actually getting shots into arms is a huge undertaking for any country.
According to a review of low- and middle-income countries' readiness to implement vaccine campaigns conducted by the World Bank, 95 per cent have developed national plans and 82 per cent have worked out which groups should be vaccinated first.
However, crucial gaps remain. Only 59 per cent have plans to train vaccinators and less than half (48 per cent) have implemented communications strategies to encourage people to take up vaccines.
While low- and middle-income countries are used to delivering childhood vaccines, so have cold chain systems in place, a mass vaccine campaign for adults is a very different beast, says Mamta Murthi, vice president for human development at the World Bank.
"This is a very different population – adults may be at work, at home, they may be unwilling to travel or not be able to come to vaccine centres," she says.
Countries such as Rwanda, Ghana and Kenya have so far run successful campaigns with the tiny amount of supply they have.
But no developing country has begun vaccinating at scale – and this is where any problems are likely to become apparent. "You only truly road-test a system when it has a big throughput," she says.
Vaccine hesitancy has also reared its head, with concerns around rare blood clots linked to the AstraZeneca and J&J vaccines hitting public confidence in Africa.
The Democratic Republic of Congo sent 1.3m unwanted doses to countries including Togo and Senegal before they expired in late June, while Malawi destroyed 20,000 unused shots last month as hesitancy hit rollout.
"There were some assumptions in the public health community that this is such a bad pandemic... that this will change people's minds if they were ever hesitant about vaccines," Professor Heidi Larson, director of the Vaccine Confidence Project, told a Devex event.
"Well, it hasn't really – in fact, the groups and the questioning around vaccines and some of the anti sentiments have actually escalated."
There are also growing concerns that the AstraZeneca and J&J vaccines may be viewed as the "cheap relation" compared to the new mRNA vaccines produced by Pfizer and Moderna.
Given the former make up the bulk of Covax's supply and are far easier to distribute in the developing world, this is a substantial hurdle.
"The AstraZeneca row has significantly impacted confidence – not just across Africa, but around the world," says Dr Ayoade Alakija, co-chair of the Africa Union Vaccine Delivery Alliance. "But there is no choice here [to pick a different vaccine]."
However, back in Kumasi, Nyarko says it is supply rather than confidence that is currently undermining his district's rollout. And with no clear picture on when more shots will arrive, he's left with few options.
"All we can do for now is pray that Ghana can secure another batch," he says. "We are praying that the UK and Europe will help us."