Ezekiel J. Emanuel is vice provost of the University of Pennsylvania and a member of Joe Biden’s public health advisory committee. Topher Spiro is the vice president for health policy at the Center for American Progress.

In the all-important search for a covid-19 vaccine, the news seems promising. Several candidates have been found that increase antibodies with tolerable side effects. Worldwide, six have reached the final stage of testing and are now being administered to thousands of subjects to assess the vaccines’ effectiveness and safety.

But once a vaccine is found, the process of packaging, distributing and administering it to achieve herd immunity also presents significant challenges. Many months will pass between proving a vaccine is effective and being able to offer injections to 300 million Americans. Given the country’s covid-19 response so far, it’s not surprising that we are already behind.

Projections are that we will need to administer two vaccine doses, one month apart. And if the vaccine’s effectiveness is temporary, which seems likely, we may have to repeat this every 12 months. What does our government have to do to make that happen? In a report for the Center for American Progress, “A Comprehensive COVID-19 Vaccine Plan,” we identify four major potential bottlenecks — and solutions.

First, “fill and finish.” Vaccines must be put into specialized glass vials, and so far, the government has only contracted for 164 million glass vials, with the timing unknown. Corning, the major U.S. manufacturer, is expanding capacity, but it is likely to be able to produce only an additional 14 million or so vials a month with current funding.

The plants necessary to put vaccine into the vials are also highly specialized; they must be 100 times more sterile than a hospital operating room. Worldwide, these fill-finish plants are collectively operating at near-capacity. It is unclear precisely how much capacity there is in the United States, but in October 2018, a government assessment concluded “operational capability has not been adequately developed.”

It takes up to five years to build one fill-finish plant from scratch, but we can expand existing plants by installing new lines faster. Pfizer is retrofitting existing facilities for about $40 million per facility, and other companies are expanding or could expand their facilities.

We recommend that the government invest in retrofitting existing facilities, at an estimated cost of $400 million, and expanding the production of glass vials, as well as building new fill-finish facilities for 100 million doses, for $1.4 billion. Critically, the Defense Production Act must be invoked to free up and coordinate the nation’s existing manufacturing capabilities.

Second, syringes and needles. Once the vaccine is made and shipped, it has to be injected, requiring 650 million to 850 million syringes and needles. These are also in short supply. The five existing manufacturers produce 663 million injection devices per year, but most are already earmarked for many other medical purposes. The government has entered into contracts with BD, the largest manufacturer, to build production lines for an additional 320 million units. But this will take 12 months — and further production capacity is needed. The government needs to quickly invest at least $70 million to build two new manufacturing lines. It should also look into alternative delivery devices.

Third, payment. Paying for vaccines is a complex system involving physicians, pharmacies, insurers, Medicare and Medicaid, with lots of Americans falling through the cracks. In the early 1990s, Congress established the Centers for Disease Control and Prevention’s Vaccines for Children program for uninsured and low-income families. In 2009, the federal government provided free H1N1 vaccines.

The same has to occur for covid-19, and it will be comparatively cheap. The government should pay a maximum of $20 per dose — well within the range of existing CDC vaccine prices — meaning it would cost less than $20 billion to vaccinate the whole country.

Fourth, delivery. Traditionally, we administer vaccines through a patchwork of physician offices, pharmacies and public health clinics, with mixed success. Only about 45 percent of adults get an annual flu shot. To quell covid-19, we need to get 70 percent immunity, which probably means about 90 percent of Americans need to be vaccinated. To reach this goal, we calculate that we will need at least 7,300 community vaccination clinics, each providing nearly 30,000 doses per month. Some could be run by community health centers, CVS, Walmart and other existing vaccination or testing sites. But many new sites would also be needed. Collectively, we estimate these clinics would cost about $10 billion.

We all hope that by early 2021, pharmaceutical companies will be manufacturing an effective covid-19 vaccine. But it will do us no good unless we can package, ship and administer it to 300 million Americans — twice. We estimate that altogether it will cost less than $45 billion — an insignificant amount for a disease that has cost trillions of dollars in economic losses. More of a challenge is the need for a well-coordinated federal government effort to do the job. We need stronger leadership to ensure all Americans can get a vaccine and we can return to normalcy by fall 2021.

Read more: