**Background:** During 2020–2021, the United States employed a multifaceted approach to control SARS-CoV-2, including non-medical interventions (NMIs), rapid vaccine development and deployment via Operation Warp Speed, and hospital therapeutics and care improvements (HTCI). Each approach carried distinct costs and benefits. The objective was to calculate the Incremental Cost Effectiveness Ratio (ICER) for three major COVID-19 policy categories: NMIs, vaccines, and HTCI, both individually and in combination.
**Methods:** The authors developed a multi-risk Susceptible-Infected-Recovered (SIR) model where infection and fatality rates varied between six regions. The model used two equations: one representing changes in infections as a function of susceptible population, infection rate (β), and recovery rate (γ), and another representing changes in the susceptible population. Vaccines modified the susceptible population equation by removing vaccinated individuals. The model was calibrated to actual death rates with β initially set to 0.05, adjusted for seasonal weather changes. The initial infection fatality rate (IFR) was 0.5%, reduced to 0.4% in summer and 0.2% in winter 2020/21. NMIs were modeled as reducing the infection rate by 17% based on peer-reviewed evidence. HTCIs reduced the IFR: from 2.95% hospitalized case fatality in March–May 2020 to 2.07% by June–August 2020 (a 43% drop), incorporating Remdesivir (28% reduction in 29-day mortality) and Remdesivir plus Baricitinib (53% reduction). Vaccines removed individuals from the susceptible population, with herd immunity assumed at 75% vaccination or infection. Costs were assessed from a societal perspective. NMI costs included $1.7 trillion in GDP loss (difference between CBO projections and actual performance, assuming 30% baseline consumer reaction), $523 billion in reduced lifetime earnings from educational shutdowns (24.2 million children losing 44.8 days of school equivalent, leading to 1.27% lifetime earnings reduction), and $110 billion in other costs (substance abuse, domestic violence, childcare). Vaccine development costs totaled $55 billion ($11.5B development/manufacturing, $11.8B purchases, $32B distribution/administration). HTCI costs totaled $8.0 billion ($1.8B therapeutic development, $6.2B therapeutic purchases). QALY losses were calculated using 7.65 QALYs lost per COVID-19 death at median age of death. Indirect QALY losses from deferred cancer care were estimated using a 10% excess mortality among 1.8 million new cancer patients, yielding 180,000 excess deaths with average QALY loss of 10.64 (1,915,200 QALYs lost).
**Key Results:** The largest cost was GDP loss from NMIs ($1.7 trillion), followed by educational shutdowns ($523 billion). Vaccine development cost $55 billion, and HTCI cost $8.0 billion. Total costs ranged from $559 billion (do nothing) to $2.45 trillion (all interventions combined). QALYs lost ranged from 5.89 million (all interventions) to 11.88 million (NMIs only). The 'do nothing' scenario resulted in 10.27 million QALYs lost. HTCI alone had the lowest cost per QALY gained versus 'do nothing' at $2,089 per QALY. Vaccines alone cost $34,777 per QALY gained. NMIs alone were dominated (higher costs and higher QALY losses than alternatives). Compared to HTCI as the base case, adding vaccines cost $58,528 per QALY gained, while adding both vaccines and NMIs cost $3.4 million per QALY gained. Sensitivity analysis showed all alternatives were most sensitive to GDP loss estimates; NMIs remained dominated throughout a ±30% range. Vaccines ranged from $119,650 per QALY to cost-saving depending on consumer reaction assumptions.
**Clinical Implications:** HTCI was the most cost-effective intervention and well justified under any standard cost-effectiveness threshold. Vaccine development, alone or in combination, fell within standard cost-effectiveness benchmarks. NMIs reduced deaths and saved QALYs but at costs per QALY well outside accepted limits ($3.4 million per QALY when combined with HTCI and vaccines). The findings support CDC guidelines recommending against lockdowns and school closures except in unusual circumstances like hospital capacity limitations. The study suggests that future pandemic preparedness should prioritize therapeutic development and clinical care improvements, as these provide the greatest health gains at the lowest cost. NMIs may be more justified when used temporarily to delay infections until therapeutics and vaccines become available, but absent future clinical developments or hospital capacity issues, efforts to delay infections carry extremely high cost per QALY gained.