New analysis with @SalomonJA @zhulin0406 @DFisman looking at the health costs of setting aside 2nd #COVID19 doses for future use vs using all of the available supply is published in @AnnalsofIM https://www.acpjournals.org/doi/10.7326/m20-8137 1/
To guard against a future collapse in vaccine supply many initial vaccine allocation policies placed a premium on eliminating possible delays to delivering 2nd doses by setting aside vaccines reserves. 2/
The cost of this approach is to delay receipt of first doses in many people who would benefit from earlier vaccination. 3/
We compared a fixed strategy (set aside 1/2 of available doses to make sure they’re in the freezer in 3 wks when people need their second dose) to a flexible one where we start off using most of the available doses immediately, to get as many 1st doses in arms as possible. 4/
We conservatively assumed 52% efficacy with 1 dose, 95% with 2 doses and that there was waning efficacy if a person didn’t get their second dose at 3 weeks. We also allowed for different trends in incidence over the modeled 8-week time period. 5/
We found that the flexible strategy averted 23-29% more COVID-19 cases than the fixed strategy. If there was a moderate disruption in the vaccine supply, the flexible strategy was even more attractive, averting 27-32% of cases. 6/
In sensitivity analyses, only the combination of a VERY low first-dose efficacy and a collapse in vaccine supply would result in us favouring setting aside doses for future use. 7/
Under most plausible scenarios, withholding fewer doses in order to vaccinate more people as soon as possible could substantially increase the benefits of vaccines, while still allowing most people to receive their 2nd dose on schedule. 8/
To be clear, this is not about changing the timing of the 2nd dose or not giving people their 2nd dose. It’s about using the available doses to vaccinate as many as people as possible early on, so they get the protective effect of that first 1st dose earlier. 9/
Several places that had initially indicated that they would set aside doses have changed policy. I think this is the correct decision, given what we now know about the supply chain. This could have huge benefit, particularly in LTCH settings where time is off the essence. 10/