key: cord-0720729-n01wkvwb
authors: Maschio, M.; Fust, K.; Lee, A.; Van de Velde, N.; Buck, P. O.; Kohli, M.
title: Clinical and Economic Impact of Differential COVID-19 Vaccine Effectiveness in the United States
date: 2022-04-03
journal: nan
DOI: 10.1101/2022.03.31.22272957
sha: 95c732666799fb2b7b16b55d8ad71ee46247ce13
doc_id: 720729
cord_uid: n01wkvwb

Background: In the United States (US), three vaccines are currently available for primary vaccination and booster doses to prevent coronavirus disease 2019 (COVID-19), including the 2-dose messenger ribonucleic acid (mRNA) BNT162b2 (COMIRNATY(R), Pfizer Inc) and mRNA-1273 (SPIKEVAX(R), Moderna Inc) vaccines, which are preferred by the Centers for Disease Control and Prevention's (CDC) Advisory Committee on Immunization Practice (ACIP), and the adenovirus vector Ad26.COV2.S (Johnson & Johnson) vaccine. A substantial body of evidence has now been published on the real-world effectiveness and waning of the primary series and booster doses against specific SARS-CoV2- variants. The study objective was to determine the clinical and economic impact of differences in effectiveness between mRNA-1273 and BNT162b2 booster vaccinations over one year (2022) in US adults [≥]18 years. Methods: A decision analytic model was used to compare three mRNA booster market share scenarios: (1) Current Scenario, where the booster mix observed in December 2021 continues throughout 2022; (2) mRNA-1273 Scenario, where the only booster administered in 2022 is mRNA-1273, and (3) BNT162b2 Scenario, where the only booster administered in 2022 is BNT162b2. Analyses were performed from the US healthcare system perspective. Sensitivity analyses were performed to explore the impact of COVID-19 incidence in the unvaccinated population and vaccine effectiveness (VE) on model results. Results: In the Current Scenario, the model predicts 65.2 million outpatient visits, 3.4 million hospitalizations, and 636,100 deaths from COVID-19 in 2022. The mRNA-1273 Scenario reduced each of these outcomes compared to the Current Scenario. Specifically, 684,400 fewer outpatient visits, 48,700 fewer hospitalizations and 9,500 fewer deaths would be expected. Exclusive of vaccine costs, the mRNA-1273 Scenario is expected to decrease direct medical costs by $1.3 billion. Conversely, the BNT162b2 Scenario increased outcomes compared to the Current Scenario: specifically, 391,500 more outpatient visits, 34,500 more hospitalizations and 7,200 more deaths would be expected in 2022, costing an additional $946 million in direct medical costs. For both the mRNA-1273 and BNT162b2 booster scenarios, the percent change in direct treatment costs for COVID-19 is similar to the percent change in hospitalizations as the rate of hospitalizations is the driver of the overall costs. Changing the number of projected COVID-19 cases in 2022 by varying the incidence rate has a direct effect on model outcomes. Higher incidence rates leads to higher outpatient visits, hospitalizations and deaths for all scenarios. Varying VE has an inverse effect on model outcomes. All outcomes increase when VE is lower for all vaccines and decrease when VE is higher. In all cases, additional use of mRNA-1273 leads to fewer infection outcomes while additional use of BNT126b2 results to higher infection outcomes. Conclusion: As the real-world effectiveness evidence to date indicates that mRNA-1273 may be more effective at preventing COVID-19 infection and hospitalization over time than BNT-162b2, increasing the proportion of people receiving this as a booster are expected to reduce COVID-19-related outcomes and costs in 2022, regardless of COVID-19 incidence or variant.

In the United States (US), there are currently three vaccines approved for active immunization to prevent coronavirus disease and recommended for primary vaccination and booster doses by the Centers for Disease Control and Prevention's (CDC) Advisory Committee on Immunization Practice (ACIP). 1 1, 4 Although all three vaccines are approved for use in the US, ACIP advises that the use of the mRNA vaccines are preferred due to lower effectiveness and risk of serious adverse events. 5, 6 ACIP defines a booster dose as a "dose of vaccine administered to enhance or restore protection by the primary vaccination, which might have waned over time" and recommends any of the three COVID-19 vaccines, regardless of type of vaccine received in the primary series, may be used for the purposes of a booster dose. 1 ACIP recommends that all adults, ages 18 and older, receive a COVID-19 booster shot. 7 Per ACIP, people who received either of the mRNA vaccines for their primary series should receive a booster at least 5 months after the primary series is complete, while those who received the Ad26.COV2.S primary vaccine should receive a booster dose at least 2 months following primary vaccination. 7, 8 ACIP also recommends that all people receive either one of the mRNA vaccines for the booster dose, regardless of primary vaccine type received, as these are preferred to the Ad26.COV2.S vaccine. 7 Based on data provided by the CDC, as of the end of December 2021, approximately 68.7% of US adults ages 18-64 years and 87.5% of adults ≥65 years had received their primary series; of those receiving their primary series, 29.4% of those 18-64 years and 58.5% of those ≥65 years had also received a booster dose. 9 The current scenario, as it pertains to boosting, includes a mixture of primary series and booster dose types. The majority of people who received either the BNT162b2 or mRNA-1273 primary series also received either the BNT162b2 or mRNA-1273 boosters (94.7% and 93.7%, respectively, of those receiving a booster), with the remainder receiving the Ad26.COV2.S booster. 10 Of those who received the Ad26.COV2.S primary series, . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 3, 2022. ; https://doi.org/10.1101/2022.03.31.22272957 doi: medRxiv preprint 27.2% also received the Ad26.COV2.S booster, 41.8% received the mRNA-1273 booster, and 31.1% received the BNT162b2 booster. 10 There is a growing body of effectiveness data for these vaccines. [11] [12] [13] [14] [15] While test-negative case controlled studies are the preferred method of estimating vaccine effectiveness (VE) using real world data, 16 in the case of COVID-19, these data are often outdated by time of study completion as new variants emerge. Predictive models have also been developed to estimate VE over time by correlating levels of vaccine-induced antibodies against the variant of concern (neutralizing titers) to observed VE from clinical trials and real-world test-negative case-control study results. [17] [18] [19] Thus, neutralizing titer levels may assist in the development of variant specific vaccines by allowing prediction of the level of VE over time with a moderate degree of certainty before real world data are available. Real world evidence and predictive modeling have shown that protection against symptomatic and severe infection vary by vaccine type, number of doses received, and timing since vaccination (waning). In addition, initial VE and rate of waning differs based on the circulating variants. 11, [19] [20] [21] While there is uncertainty as to how the COVID-19 pandemic will evolve, the current study uses an economic model to project various scenarios for 2022 with the aim of quantifying the potential impact of the differences in effectiveness between the two available mRNA booster vaccines. Three specific booster market share scenarios were examined, including the current scenario, which represents the mixture of booster doses across the three vaccine types as described above, and the hypothetical impact of changing the mix of boosters delivered to either 100% mRNA-1273 or 100% BNT162b2. The objective of this study was to estimate the clinical and economic impact of any potential differences in effectiveness between mRNA-1273 and BNT162b2 boosters in 2022 in adults aged 18 years and over in the US.

OVERVIEW A decision analytic model was used to compare three mRNA booster market share scenarios:

(1) Current Scenario, in which the mix of boosters administered by December 2021, which are presented in Table 1 , remains constant throughout 2022; (2) mRNA-1273 Scenario, in which the only booster administered in 2022 is mRNA-1273 (i.e., mRNA-1273 booster has 100% market share starting in January), and (3) BNT162b2 Scenario, in which the only booster delivered in 2022 is BNT162b2 (i.e., BNT162b2 booster has 100% market share starting in January). The target population for the analysis consisted of adults aged 18 years and older in the US eligible . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 3, 2022. ; https://doi.org/10.1101/2022.03.31.22272957 doi: medRxiv preprint for COVID-19 vaccination per ACIP recommendations. 1 The time horizon of the analysis was 1

year (2022). Economic analyses were performed from the perspective of the US healthcare system and included direct healthcare costs only. Additional sensitivity analyses were performed to explore the impact of COVID-19 incidence in the unvaccinated population and VE on model results. In the model, the size of the US adult population and the number of patients receiving the primary series only and primary series with a booster dose were estimated, the number of infections in each group were calculated and a decision tree was used to estimate associated infection consequences and costs ( Figure 1 ). The model was developed in Microsoft Excel using Visual Basic for Applications (Microsoft Corp; Redmond, Washington).

The size of the adult US population was estimated by age group using US Census Bureau data (See Appendix 2; Table 5 ). 22 Primary series vaccine coverage was estimated using CDC data 9 from March 2021 through December 2021 cumulatively at 68.7% for adults ages 18-64 and 87.5% ≥65 years. We made the simplifying assumption that no additional primary series were administered in 2022 as our analysis focuses on the impact of changes in booster delivery and primary series vaccinations have plateaued. December 2021 estimates of booster coverage among those who had previously received the primary series were also calculated from CDC data to be 29.4% in adults ages 18-64 and 58.5% in ≥65 years. Booster coverage was assumed to increase by 15% per month, until a maximum of 95% of those who had previously received the primary series was reached.

All market share estimates are displayed in Table 1 . Values for the primary series reflect the distribution of the total number of patients receiving each vaccine as of the end of December 2021 for the 18-64 and ≥65 years age groups and was the same for all scenarios. The same data were also used to estimate the type of booster dose received. In the Current Scenario, primary series and booster market shares are assumed to be equivalent to the estimates at the model start for the duration of 2022. In the mRNA-1273 Scenario, 100% of booster doses are assumed to be mRNA-1273, and in the BNT162b2 Scenario, 100% of booster doses are assumed to be BNT162b2, regardless of primary series type received.

For each month of the analysis, the model groups people into unvaccinated, vaccinated with primary series only, and vaccinated with primary series plus booster using the coverage data described above. The risk of infection in the unvaccinated varies by month. Amongst those who . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 3, 2022. ; https://doi.org/10.1101/2022.03.31.22272957 doi: medRxiv preprint received the primary series only or the primary series plus booster, the incidence of infection is reduced according to VE, and is recalculated each month to account for waning based on time since vaccination. For those who received the primary series in 2021, we made the simplifying assumption that their series was completed in May, as this marked the month at which 50% cumulative coverage in the population was achieved. May was therefore our anchor month to calculate time since primary series completion for the VE calculations for each month. The primary series VE was estimated as an average of the waning adjusted VE weighted by the proportion of the cohort receiving each primary series. For anyone who had received a booster in 2021, we made the simplifying assumption that the booster was received in November. As people continued to receive boosters in 2022, the booster VE was estimated as a weighted average of the waning adjusted vaccine effectiveness (time since receiving their booster) and the proportion of the cohort receiving each booster. 

The infection incidence estimates amongst the unvaccinated for January to December 2022 were based on projections from the Institute for Health Metrics and Evaluation (IHME) 23 IHME provided information on the estimated number of infections amongst the unvaccinated as well as the number of people vaccinated in their simulation for 2022. The monthly rate of infections in the unvaccinated was calculated using this data. As the IHME model does not look at incidence . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 3, 2022. ; https://doi.org/10.1101/2022.03.31.22272957 doi: medRxiv preprint by age, the same monthly rate is assumed for all age groups. Based on their modeling, IHME predicts that a large portion of individuals in countries such as the US have been infected with omicron meaning that immunity levels will be very high by March 2022. As shown in Table 2 , the monthly rate of infection is projected to drop significantly after March.

Low and high scenarios also were created by increasing and decreasing the base case cumulative incidence rates by 20% for January to December 2022. The final monthly incidence rates estimate for each scenario are displayed in Table 2 . As the IHME projections count all infections including those that are asymptomatic, we reduced the estimate to 85% of the original to account for symptomatic cases only. 26 

As the characteristics of the next variant that will become dominant in the US is unknown, the effectiveness of the existing vaccines to future variants is uncertain. Thus, three hypothetical scenarios, using existing variants as proxies, are explored: The initial VE and rate of waning for each vaccine and each scenario is displayed in Table 3 .

To inform the initial VE and the rate of waning over time following primary series for all VE scenarios, VE data for all 3 vaccines from the December 22, 2021 IHME COVID-19 model update estimates for delta and omicron were used, respectively. 11 IHME pooled findings from multiple studies to calculate VE for the delta variant. They then calculated the relative reduction in VE against omicron compared to delta from two test-negative case control studies. This relative reduction was then applied to the delta VE to obtain the omicron VE. The monthly rates of waning for VE against infection and severe disease for each vaccine were approximated from the IHME graphs (timepoint 0-50 weeks). To create the graphs, IHME pooled data from 20 studies that estimated VE as a function of time, and used regression models to estimate the rate of decline.

Waning from boosters were assumed to be the same as waning from primary series.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 3, 2022. 

The age-specific proportions of symptomatic infections and patients requiring inpatient care were estimated from Reese et al., 2021 (see Appendix Table 6 ). 26 is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 3, 2022. ; https://doi.org/10.1101/2022.03.31.22272957 doi: medRxiv preprint ventilation' was assumed to correspond to ICU with 'ventilator'. The total number of patients readmitted to the hospital in each location of care was divided by the total number of patients in each location of care. Verna et al., 2021 28 did not provide data by age category; accordingly, the same percentages of patients requiring hospital re-admission within each location of care were applied to all age groups. See Appendix Table 6 for inputs used in the model.

For those patients surviving the initial hospitalization, the rates of post-discharge mortality were calculated from Chopra et al., 2021. 29 Of those discharged, 1.96% were assumed to be readmitted to the ICU and require mechanical ventilation, 2.62% admitted to the ICU without ventilation, and 3.77% admitted to the hospital without ICU requirements. 28 Additionally, the rates of post-hospital discharge mortality were assumed to be 10.40% for patients in ICU with/without ventilation, and 5.36% for those not requiring ICU admission. 29 These rates were applied to all age groups.

Vaccine costs (including administration fees) were not included in the model, as they were assumed to be equivalent for all vaccine types and therefore do not contribute to the economic impact. 

The number of outpatient visits, hospitalizations, and deaths and the medical costs associated with treating COVID-19 for each market share scenario is shown in Table 4 for the base case analysis. In the Current Scenario (mixed boosters strategy), the model predicts 65.2 million outpatient visits, 3.4 million hospitalizations, and 636,100 deaths from COVID-19 in 2022. Given the higher overall effectiveness for the mRNA-1273 booster compared to the BNT162b2 booster, . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 3, 2022. ; https://doi.org/10.1101/2022.03.31.22272957 doi: medRxiv preprint if 100% of people received this booster, each of those outcomes would be expect to decrease by at least 1% compared to the Current Scenario. Specifically, 684,400 fewer outpatient visits, 48,700 fewer hospitalizations and 9,500 fewer deaths would be expected. As all vaccine costs are assumed to be the same, this is expected to decrease direct medical costs by $1.3 billion.

Conversely, the BNT162b2 Scenario increased these outcomes compared to the Current Scenario. Specifically, 391,500 more outpatient visits, 34,500 more hospitalizations and 7,200 more deaths would be expected in 2022, increasing direct medical costs by $946 million. For both the mRNA-1273 and BNT162b2 booster scenarios, the percent change in direct treatment costs of COVID-19 is similar to the percent change in hospitalizations as the rate of hospitalizations is the driver of the overall costs.

Given that January experienced a high incidence due to the omicron wave, we also looked at the results for February to December only, excluding January 2022 cases. There are fewer COVID-19 infections and therefore only 17.9 million outpatient visits, 965,600 hospitalizations, and 182,800 deaths. However, use of 100% mRNA-1273 still leads to a reduction in these outcomes by 2.3%-2.8%, while use of 100% BNT126b2 increases them by approximately 1.4%-2.1%. Figure 2 shows the impact of changing incidence from the base case incidence to either the high or low incidence scenarios. The incidence impacts the total number of cases projected in 2022: higher incidence leads to higher outpatient visits, hospitalizations and deaths for all scenarios.

With all incidence scenarios, additional use of mRNA-1273 leads to fewer infection outcomes while additional use of BNT126b2 leads to higher infection outcomes. The difference between the two vaccines, however, is more pronounced with higher infection incidence.

The impact of varying the VE assumptions is displayed in Figure 3 . All outcomes are increased when vaccine effectiveness is lower (omicron-like VE with double waning scenario) and decreased when VE is higher (delta-like VE scenario). In all scenarios, the mRNA-1273 booster prevents more cases than the BNT162b2 booster. Overall, the trends are as expected but the analysis quantifies the expected differences between the booster scenarios.

This study found that, when compared to the booster mix of both mRNA-1273 and BNT162b2 observed in the US as of December 2021, shifting to 100% mRNA-1273 boosters decreases the expected number of outpatient visits, hospitalization, and deaths, which in turn . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 3, 2022. ; https://doi.org/10.1101/2022.03.31.22272957 doi: medRxiv preprint results in decreased direct medical costs. Conversely, shifting to boosting with 100% BNT162b2 increased resource use and medical costs. These trends remained consistent across a range of sensivitiy analyses varying the incidence of COVID-19 and the VE against different variant types.

The results are not surprising: as the predicted VE of the mRNA-1273 has been shown to be higher against infection and severe disease compared to BNT162b2 for existing variants, 11, 13, 18, 21, 31 shifting to exclusive mRNA-1273 use as a booster would result in fewer COVID-19 cases and therefore less health care resource use.

As with all predictive modeling exercises, there are limitations to the study. The incidence of COVID-19 had to be projected over the upcoming year and is highly uncertain. Still, scenarios where a high incidence rate and low incidence rate were tested did not change the overall model results, with 100% mRNA-1273 boosters resulting in less outpatient visits, hospitalizations and costs compared to the Current Scenario and 100% BNT162b2. With a higher incidence, VE is amplified as the overall denominator of cases is larger. Overall, resource use, deaths and costs are increased, and the difference in effects of booster types is also larger than the base case. An additional limitation of the incidence used is that the incidence data included in the model contains the rapid rise of the omicron wave in January 2022 and subsequent drop in February 2022. The January 2022 incidence (33%) contributed a large proportion of cases to the model. If COVID-19 becomes endemic, this high monthly incidence is not expected to occur again. 24 However, the order of benefit (100% mRNA-1273 > booster mix > 100% BNT162b2) observed in the base case is still maintained for the February to December results only.

As the final proportion of people who choose to receive boosters is yet unknown, an assumption about future booster coverage had to be made in order to conduct this analysis.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 3, 2022. ; https://doi.org/10.1101/2022.03.31.22272957 doi: medRxiv preprint Demand for booster doses was very strong in January and February of this year as people sought additional protection against the new omicron variant. It is possible that 95% coverage will not be achieved if demand for the booster declines as incidence of disease declines. The final booster coverage achieved will impact the magnitude of cost-savings predicted: as more people are vaccinated, more individuals would receive benefit from a more-effective booster and this improves the cost savings associated with the mRNA-1273 scenario compared to the BNT162b2 scenario.

Some infectious disease models track susceptibility to infection in order to calculate an overall rate of infection in a population. 32 We used a simplified model where the incidence rate amongst the unvaccinated is an input rather than a calculation in the model. In January to March, many people, vaccinated and unvaccinated are expected to be exposed to the omicron variant leading to increased levels of susceptibility. We indirectly account for that by using a lower incidence rate in the last 6 months of the year. However, one limitation is that we are not able to directly test the impact of different levels of immunity for individuals who receive a booster alone compared to those who receive a booster plus a COVID-19 infection. While a single case of myocarditis may be expensive to diagnose and treat, the rarity of the outcome means that the occurrence of myocarditis has minimal impact on the costs. On the other . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 3, 2022. ; https://doi.org/10.1101/2022.03.31.22272957 doi: medRxiv preprint hand, individuals who develop COVID-19 are at an increased risk of developing myocarditis 35 and we did not consider the costs associated with that potential complication of COVID-19 either.

Finally, total costs savings are likely to be an underestimation of true values as many parameters, which can be significant drivers of cost, have not been included in the analyses.

Productivity loss due to acute infection, or quarantine/isolation requirements due to infection/exposure have not been included in the model. Post-acute sequelae of COVID-19 (long Covid), which could impact both productivity loss as well as increase direct medical costs, has also not been incorporated into this model. It is expected that increasing booster vaccinations would decrease rates of infection and decrease these additional costs.

In conclusion, as the real-world effectiveness evidence to date indicates that mRNA-1273 may be more effective at preventing COVID-19 infection and hospitalization over time than BNT162b2, increasing the proportion of people receiving this as a booster may reduce the number of cases and costs associated with COVID-19 in 2022, regardless of COVID-19 incidence or variant.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 3, 2022 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 3, 2022. ; https://doi.org/10.1101/2022.03.31.22272957 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 3, 2022. ; https://doi.org/10.1101/2022.03.31.22272957 doi: medRxiv preprint

The Advisory Committee on Immunization Practices' Interim Recommendations for Additional Primary and Booster Doses of COVID-19 Vaccines -United States

Food & Drug Administration. Spikevax and Moderna COVID-19 Vaccine

Food & Drug Administration. Janssen COVID-19 Vaccine

Centers for Disease Control and Prevention. Johnson and Johnson's Janssen COVID-19 vaccine overview and safety

Use of the Janssen

Vaccine: Updated Interim Recommendations from the Advisory Committee on Immunization Practices -United States

Centers for Disease Control and Prevention. COVID-19 Vaccine Emergency Use Instructions (EUI) Resources. Available at

Institute for Health Metrics and Evaluation (IHME). COVID-19 model update: Omicron and waning immunity

Effectiveness of COVID-19 vaccines against the Omicron (B.1.1.529) variant of concern

Protection of Homologous and Heterologous Vaccine Boosters Against COVID-19 in Singapore

Effectiveness of mRNA-1273 against SARS-CoV-2 omicron and delta variants

Effectiveness of BNT162b2 and mRNA-1273

Second Doses and Boosters for SARS-CoV-2 infection and SARS-CoV-2 Related Hospitalizations: A Statewide Report from the Minnesota Electronic Health Record Consortium

The test-negative design for estimating influenza vaccine effectiveness

Neutralizing antibody levels are highly predictive of immune protection from symptomatic SARS-CoV-2 infection

Neutralising antibody titres as predictors of protection against SARS-CoV-2 variants and the impact of boosting: a meta-analysis

Early Results to predict Vaccine efficacy against Omicron

Vaccine effectiveness and duration of protection of Comirnaty, Vaxzevria and Spikevax against mild and severe COVID-19 in the UK

Population Division: Washington DC. Annual Estimates of the Resident Population by Single Year of Age and Sex for the United States

Institue for Health Metrics and Evaluation (IHME)

COVID-19 will continue but the end of the pandemic is near

The annual impact of seasonal influenza in the US: measuring disease burden and costs

Estimated Incidence of Coronavirus Disease 2019 (COVID-19) Illness and Hospitalization-United States

Health outcomes and economic burden of hospitalized COVID-19 patients in the United States

Factors Associated with Readmission in the US Following Hospitalization with COVID-19

Sixty-Day Outcomes Among Patients Hospitalized With COVID-19

Brookings report: estimating potential spending on COVID-19 care

Comparative Effectiveness of mRNA and Inactivated Whole Virus Vaccines against COVID-19 Infection and Severe Disease in Singapore

Dynamic transmission modeling: a report of the ISPOR-SMDM Modeling Good Research Practices Task Force--5

Myocarditis Cases Reported After mRNA-Based COVID-19 Vaccination in the US From

Safety of the BNT162b2 mRNA Covid-19 Vaccine in a Nationwide Setting

2021 National Physician Fee Schedule Relative Value File October Release