This is the largest study to investigate the number of patients, mortality rates, and deaths during the Omicron-predominant wave of COVID-19. The strengths of this study are that it provided real-world data on patients with COVID-19 since the early pandemic period using the NDB, which contains almost all inpatient and outpatient claims data in Japan, and it compared patients with COVID-19 to those with influenza, the most common epidemic infectious disease, during the Omicron-predominant wave. The mortality rate of COVID-19 decreased significantly over time, but the number of patients was substantially higher in the Omicron-predominant wave than in the earlier waves, resulting in a large increase in the number of deaths, especially among the elderly. In a comparison of patients with COVID-19 and influenza during the Omicron-predominant wave, the mortality rate was lower for patients with COVID-19 than for those with influenza among those aged ≤ 39 years, but the mortality rate was higher for COVID-19 in patients aged ≥ 40 years, especially among the elderly.

This study found that the mortality rate of COVID-19 was lower in the Omicron-predominant wave than in the earlier waves. In previous studies, patients with the COVID-19 Omicron variant had lower mortality rates than those with the Delta variant [3,4,5,6]. A population-based study in the United Kingdom (including 440,000 and 1,060,000 patients with COVID-19 Delta and Omicron variants, respectively) reported an approximately 60% reduction in mortality for the Omicron variant compared to the Delta variant [5]. In line with that study, the mortality rate of patients with COVID-19 was 40% lower during the Omicron-predominant wave than during the Delta-predominant wave in this study of approximately 28 million patients. This reduction in mortality was presumably attributable to the decreasing virulence of SARS-Cov-2 virus, as well as the widespread use of vaccines and the development of treatments, including antiviral drugs. However, the average weekly number of deaths attributable to COVID-19 was higher in the Omicron-predominant wave than in the earlier waves. Although the Omicron variant was less virulent than the Delta variant, the former variant displayed increased transmissibility [21]. In a WHO report, the number of patients was higher during the Omicron-predominant wave than during the Delta-predominant wave [22]. In addition, strict infection control measures such as physical distancing and entry restrictions were taken in Japan until the Delta-predominant wave, but these measures were gradually relaxed after this wave. Therefore, we can speculate that the number of patients was markedly higher during the Omicron-predominant wave than during the Delta-predominant wave, resulting in a substantial increase in the number of deaths despite the decreased mortality rate. The observed increases in the number of infected patients and deaths denote an increase in health care resource consumption. Consequently, even in the Omicron-predominant wave, COVID-19 continues to have a significant social impact, and therefore, sustained control measures remain needed against this disease.

In the early stages of the pandemic, older age was reported to be associated with increased mortality in patients with COVID-19 [23,24,25]. The present study demonstrated that elderly patients with COVID-19 had a higher mortality risk than younger patients, even in the Omicron-predominant wave. An inpatient-based study during the Omicron-predominant wave reported that the mortality rate of elderly patients with COVID-19 was higher than that of elderly patients with influenza [12], in line with the results of the current study including both inpatients and outpatients. The present study found that among the elderly, the number of patients with COVID-19 was considerably higher than the number of those with influenza during the Omicron-predominant wave, resulting in significantly more deaths among elderly patients with COVID-19. As previously noted, the COVID-19 mortality rate decreased during the Omicron-predominant wave, but the total number of deaths increased substantially because of the increased number of infected patients. Notably, the elderly accounted for the majority of this increase in deaths. By contrast, the COVID-19 mortality rate was lower in younger patients despite the higher number of patients. These results suggest that the prevention of COVID-19 (e.g., promotion of vaccination) and aggressive therapeutic interventions remain equally or more important than influenza control in the elderly.

Studies have reported different mortality rates for COVID-19 and influenza depending on age [7, 11, 12]. A study limited to inpatients in the Omicron-predominant wave reported a significantly higher risk of death for COVID-19 than for influenza in patients aged > 65 years, but no significant difference in mortality risk was observed among patients aged ≤ 65 years [12]. The present study included the largest population reported to date, including outpatients and inpatients, and it demonstrated that the risk of death from COVID-19 and influenza differed by age for the first time that in the real world (Fig. 3 and Table 3). Most studies comparing COVID-19 and influenza excluded children from the study population [12,13,14,15], and therefore, studies on children have been limited to small numbers of patients [26, 27]. Therefore, the data from the present study, which included children, are significant in this regard. In the present study, the mortality rate and number of deaths attributable to COVID-19 in patients aged 0–19 years were lower than those for their counterparts with influenza during the Omicron-predominant wave. These results suggest that COVID-19 and influenza differ regarding the spread of infection and outcomes between children and adults. Therefore, measures against these infections need to be adjusted by age, and it is possible that measures against influenza should take higher priority than those against COVID-19, especially in children.

This study confirmed a significant decrease in the number of influenza cases in Japan from mid-2020 to late 2022. This period coincided with the COVID-19 pandemic, during which strict public health measures, such as infection prevention protocols and social distancing, were widely implemented. These measures likely played a substantial role in suppressing the spread of influenza during this time. However, in 2023, the number of influenza cases began to rise again. This resurgence may have been influenced by the relaxation of strict public health measures. Nevertheless, the number of influenza cases in 2023 did not reach the levels observed before the COVID-19 pandemic. Several hypotheses may explain this phenomenon. First, the influenza virus strains circulating in 2023 might have been less transmissible than those prevalent before the pandemic. This reduced transmissibility could have contributed to the suppression of case numbers. Second, even after the relaxation of public health measures, preventive behaviors such as mask-wearing and hand hygiene may have been voluntarily maintained by some individuals. These behaviors could have played a role in limiting the spread of influenza to some extent. Lastly, there may have been cross-reactive immunity between COVID-19 and influenza viruses [28,29,30]. Immune responses induced by SARS-CoV-2 infection or vaccination might have provided partial protection against influenza. Further research is needed to validate these hypotheses.

This study had several limitations. First, the NDB does not record physical findings, blood test or imaging results, or the vaccination status for COVID-19 or influenza. In Japan, following the dominance of the Alpha variant in May 2021, COVID-19 vaccination rates increased, with approximately 40% of the population vaccinated at the peak of the Delta wave and around 80% by the start of the Omicron period [31]. Thus, the decline in COVID-19 mortality observed during the study period may have been influenced not only by viral mutations and advancements in treatment but also by the widespread rollout of vaccination. Second, this study did not include information on the specific causes of death. Therefore, the recorded deaths may have included not only those directly caused by COVID-19 but also those indirectly associated with COVID-19, such as deaths resulting from the exacerbation of comorbidities or other accidental factors, as well as deaths unrelated to COVID-19. Third, the definition of the predominant wave was based on screening data in Tokyo rather than the confirmed variants detected in each patient. Fourth, the association between medications and mortality was outside the scope of this study because the NDB included data on drugs covered by insurance but not those used in clinical trials or specially approved for use without insurance coverage. Finally, asymptomatic or mild cases of COVID-19 may not have sought medical care. As a result, these cases might not have been recorded in the database, potentially leading to an underestimation of the total number of patients and affecting the calculation of mortality rates.

In conclusion, although the mortality rate of COVID-19 was lower during the Omicron-predominant wave than in previous waves, the number of deaths increased substantially because of the higher number of infected patients, especially among the elderly. Although the mortality rate and number of deaths associated with the COVID-19 Omicron variant were lower than those associated with influenza among younger patients, the opposite findings were recorded among elderly patients. Thus, COVID-19 remains associated with increased mortality in the elderly and represents a significant burden to society and healthcare. It is necessary to establish preventive measures and treatments for this disease based on age categories.