After the initial rollout of vaccines to curb the global burden of COVID-19, many countries are administering boosters to address waning immunity and alleviate the severity of new and highly transmissible variants of SARS-CoV-2 (
1). In the United States, timelines of booster vaccination have changed since the emergence of the Omicron variant, reducing from an initial recommendation of 6 months after the completion of the primary series or a known infection to 2 months when bivalent booster vaccination became available (
2). To formalize the frequency of booster doses, the U.S. Food and Drug Administration (FDA) has proposed annual, single-dose vaccination for SARS-CoV-2, similar to influenza vaccination (
3), with a potential second dose for those at risk for severe outcomes, including children younger than 2 years and adults aged 50 years or older (
4).
The seasonal patterns for SARS-CoV-2 observed in various geographic regions and countries are driven by multiple factors ranging from the emergence of new variants to heterogeneity in subnational vaccination coverage, as well as other public health responses (
5–8). These patterns may change from year to year, but SARS-CoV-2 could exhibit annual cycles similar to those of seasonal influenza once it transitions to an endemic state (
5). A synchronized pattern between the 2 diseases suggests that, as with influenza, vaccination campaigns against SARS-CoV-2 before an anticipated surge would reduce the disease burden. In the long term, coadministration of vaccines for SARS-CoV-2 and seasonal influenza is expected to become a preferred strategy. However, the effectiveness of this strategy remains undetermined, with unknown timing of a surge and the possibility of semiannual COVID-19 epidemics.
Despite wide availability of messenger RNA (mRNA) bivalent vaccines since September 2022, only 17% of the U.S. population has received a booster dose (
9). A shift to annual campaigns may enhance uptake of bivalent boosters toward the uptake seen with influenza vaccination. To evaluate the effectiveness of annual campaigns, we developed an age-structured dynamic model of infectious disease transmission and parameterized it with estimates of the epidemiologic characteristics of SARS-CoV-2 and effectiveness of mRNA vaccines. The primary simulation used the FDA-proposed annual vaccination campaign with a second dose for children younger than 2 years and adults aged 50 years or older, assuming an age-specific uptake similar to that of influenza vaccination. We explored alternative scenarios where the second dose was distributed to the following 3 age groups of adults: 65 years or older, 50 to 64 years, or 18 to 49 years. We used incidence of infection, hospitalizations, deaths, and direct health care costs as measured outcomes to compare the 2-dose vaccination strategies versus an annual single-dose campaign and determine their mitigating effect over a 1-year time horizon as well as the optimal timing of a second dose.
Discussion
In early 2023, the FDA reviewed a proposal advocating a transition to annual SARS-CoV-2 vaccinations (
4). The proposal recommends a single dose for most individuals with a second dose for children younger than 2 years and adults aged 50 years or older (
4). Using an age-structured dynamic model of infectious disease transmission calibrated to replicate winter and late summer peaks in COVID-19 hospitalization, we evaluated the effectiveness of the proposed annual vaccination with an uptake similar to that of influenza vaccination. We also estimated the optimal interval between the first and second doses that minimizes the direct health care costs. Our analysis showed that transitioning to an annual campaign would moderately reduce disease burden among adults aged 50 years or older. In comparison, the FDA-proposed campaign would lead to a 15% reduction in hospitalizations and an 18% decrease in deaths compared with the annual campaign. Assuming a second peak in hospitalizations, our model found that the FDA-proposed campaign yields the greatest benefit when the second dose is administered 5 months after the first dose, with diminishing returns beyond a 6-month dosing interval. Exploring the distribution of second doses among age groups, our findings suggest that adopting the FDA-proposed campaign is the most effective strategy of those examined by our model to mitigate the burden of SARS-CoV-2 in the United States.
We found that the FDA-proposed campaign with a second dose administered 5 months after the primary dose produced the largest reduction in direct health care costs compared with the annual campaign. However, we assumed that the vaccination rate for the first and second doses was equivalent to that estimated for influenza vaccination. Scenario analyses illustrated that the optimal time interval between the 2 doses is sensitive to vaccine uptake and the waning rate of vaccine-acquired immunity. As of 9 December 2023, SARS-CoV-2 vaccine uptake was 16%, whereas influenza vaccine uptake had already reached 42% (
38–41), potentially influencing the dosing interval. As expected, a shorter dosing interval becomes crucial to counteract accelerated waning immunity. When preexisting vaccine-acquired immunity (which we initially assumed to be absent) is considered, we expect the suggested dosing interval to extend beyond that found in our baseline analysis. Preexisting vaccine-acquired immunity could delay the surge and mitigate the fall and winter disease burden, enhancing the effectiveness of the first dose over a longer time interval. Although previous studies propose a frequency of 6 to 12 months for SARS-CoV-2 booster vaccination (
42,
43), particularly among those aged 75 years or older (
44), our analysis indicates a 5-month dosing interval in the FDA-proposed campaign, with a potential range of 3 to 6 months. This timing is projected to have the largest effect on reducing direct health care costs.
Within the framework of an annual SARS-CoV-2 vaccination initiative, our analysis highlights a moderate reduction in disease burden for the adult population aged 50 years or older on transitioning to an annual campaign. The FDA-proposed campaign, targeting this vulnerable population and children younger than 2 years, emerges as the strategy with the lowest direct health care costs among alternative strategies investigated. However, the second-dose uptake in the prioritized age groups was assumed to be 100% among those receiving the first dose. In a scenario with a 33% reduction in second-dose uptake and the optimal dosing interval of 6 months, the direct health care costs averted decreased by 47%. Among alternative vaccination strategies, we found that prioritizing adults aged either 18 to 49 years or 50 to 64 years with a second dose was more effective in reducing direct health care costs than providing second doses to only adults aged 65 years or older because of differences in population-level vaccine immunity, extent of direct protection, and contact mixing across age-stratified groups. Of note, we found that the age group selected for a second dose can influence the dosing interval, likely due to the effect of these contact patterns. Thus, if some of the prioritized populations in the FDA-proposed campaign are reluctant to pursue a second dose, then offering second doses to individuals aged 18 to 49 years could compensate for this uptake reduction by providing indirect protection for older adults.
The temporal dynamics of immunity against infection and severe disease are expected to change with the evolution of SARS-CoV-2. Our model was informed by the temporal changes in individual-level immunity against infection and severe disease using estimates of the waning immune protection from prior infection or vaccination (
11–29,
32). The data from existing epidemiologic studies inherently account for the effects of and changes in both humoral and cellular immunity after infection and vaccination, as well as behavioral confounding variables that may influence the likelihood of exposure and subsequent infection. Previous studies suggest that maintaining a robust cellular immune response to SARS-CoV-2 reduces the risk for severe disease and death, with the humoral response playing a primary role in protecting against infection (
43,
45,
46). Our formulation of waning immunity implicitly captures the effect of humoral and cellular immunity through parameterization and transition between states of immunity in the model.
Our analysis of age-specific waning of vaccine-acquired immunity shows a slower waning and higher protection against severe disease among those aged 60 years or older than among younger age groups. This result likely reflects potential biases in the data used to estimate vaccine efficacy. Specifically, observations related to hospitalization were from persons who had received 2 or 3 doses of monovalent vaccine in a test-negative case–control study in England (
18). A potential bias pertains to the occurrence of “incidental” cases in hospitals, where patients admitted for reasons unrelated to COVID-19 subsequently tested positive for SARS-CoV-2, resulting in lower estimates of vaccine efficacy against severe disease (
47). The occurrence of incidental cases in the hospital has been more pronounced among persons aged 18 to 64 years than among those aged 65 or older (
47). In an effort to alleviate the effect of this bias, we did scenario analyses by considering both fast and slow rates of waning vaccine-acquired immunity.
A strength of our study is the construction of a model that uses the U.S. demographics and age-stratified contact patterns and includes naturally acquired and vaccine-induced immunity with waning, age-specific vaccination uptake, and administration of a second dose with consideration of SARS-CoV-2 seasonal patterns. However, these factors vary among countries and even states within the United States. In the scenario of semiannual epidemic peaks, uncertainties arise regarding how each of these factors and the practice of nonpharmaceutical interventions might influence the incremental benefit and optimal timing of the second dose. Moreover, the population-based model does not account for the extent of individual variation in susceptibility to infection (for example, immune compromise) beyond the age-stratified integration. We found that reducing transmission or the probability of hospitalization decreased the effectiveness of the FDA-proposed campaign. Thus, consistent with prior studies (
48), if new immune-evasive variants emerge, vaccinating the elderly population with a second dose or potentially a new variant-specific vaccine may be critical in reducing hospitalizations and direct health care costs.
Previous analyses suggest that COVID-19 epidemic peaks will likely coincide with those of other respiratory diseases, such as seasonal influenza and respiratory syncytial virus disease, in many countries (
5). Should epidemiologic trends of SARS-CoV-2 fully adopt seasonal dynamics akin to those of influenza, we anticipate that the significance of a second dose in mitigating disease burden will diminish.
In conclusion, our study shows that adopting an annual vaccination campaign with the provision of a second dose to children younger than 2 years and adults aged 50 years or older can be a suitable approach to protect individuals against SARS-CoV-2 infection and associated outcomes. Monitoring the seasonal and evolutionary patterns of SARS-CoV-2 is critically important to inform decisions on vaccination strategies and the development of new vaccines to maintain population immunity.
Comments
0 Comments