The purpose of this clinical trial is to find out if the medication called baloxavir marboxil (sold under the brand name Xofluza®) can help to reduce the time needed to recover from flu when patients need an hospitalization. Patients infected by influenza and requiring a hospitalization will be approched to be included in the study.

The main questions are:

1. When someone is hospitalized with a severe influenza infection, does baloxavir help to reduce the time needed to recover?
2. Can baloxavir marboxil help to shorten the amount of time people need to stay in the hospital with severe flu?
3. Can baloxavir marboxil help to reduce the risk of life-threatening complications as well as of death due to severe flu?
4. Can baloxavir reduce duration of contagiousness?

To be able to measure the above, the investigators will compare two groups of patients: One group receiving baloxavir marboxil, the other group receiving a mock treatment called placebo.

Participants will:

• Take one single dose of baloxavir marboxil or placebo soon after hospitalization.
• Vital signs will be followed three times per day during hospital stay.
• Have a nose swab to detect the presence of influenza virus on the first and third day of trial participation.
• Answer to a short quality of life questionnaire on the phone 3 months after receiving the study treatment.

Background: Influenza virus is a major source of seasonal outbreaks and is the leading candidate for a future pandemic. Each winter thousands of people are hospitalized with infection complications. The most at risk are the very young (< 5 years), the elderly (>65 years old) and those with a weak immune system. When a person is hospitalized because of the flu, it is not always clear if the administration of a specific medicine that fights the virus (called an antiviral) will help.

Most studies looked at antivral drug benefits among people with mild flu, who didn't need to be hospitalized. They found that if the antiviral is started early (up to two days after the symptoms started), it can make the symptoms go away a little faster. But for people in the hospital, especially if it's been more than two days since their symptoms started, it is not so sure if antiviral drugs still help. Clinical research involving patients hospitalized for disease complications is few and of lesser quality. While some indicate that early treatment might be beneficial, there is no scientific agreement about treatment benefits. Therefore, recommendations and antiviral prescription varies between hospitals and physicians.

In Switzerland two antiviral drugs are authorized to treat the flu: Tamiflu® and Xofluza®.

Study aims and methods: The goal of this industry-independent trial is to measure the benefits of Xofluza® compared to placebo (mock medication) in adult patients hospitalized for the flu.

Choice of the medication: The investigators chose Xofluza® instead of Tamiflu® because it is very simple to use (only a single pill instead of twice a day Tamiflu® for 5 days), it has fewer side effects and can be safely given to patients with a wide range of chronic diseases.

Importance: This study is very important because it could lead to a Swiss and international consensus about the utility of antiviral treatment in hospitalized patients.

If beneficial, Xofluza® might be the drug of choice in a future pandemic, when access to a simple to administer and easily stored drug effective at all stages of the illness with few side effects would be of utmost importance.

However, if the trial doesn't show any benefit of Xofluza® administration, antiviral treatment prescription won't be recommended, preserving patients from unneeded medication and from its potential side effects and saving ressources.

Tailoring antiviral use will also help prevent the risk of the virus becoming resistant which might happen in case of overuse.

Patient and public involvement: During the preparation of the study, patient representatives were involved in the development of the research idea, in the discussion concerning its ethical aspects, the feasibility of recruitment, and the selection of questionnaires to measure treatment benefits. They participated in the revision of the lay summary as well. The informed consent form has been developped together with them, in our communication efforts targeting potential participants at the start as well as in the dissemination of the results at the end of the study. Patients representatives also developped together with the investigators the strategy of participants recruitment.

Inclusion Criteria:

• Age ≥ 18 years
• Participant or participant representative capable of giving signed informed consent.
• Positive reverse transcriptase-polymerase chain reaction (RT-PCR) for influenza A and/or B confirmed on arespiratory tract sample.
• Patient requiring hospitalization.
• National Early Warning Score 2 (NEWS2) of ≥4 at planned randomization

Exclusion Criteria:

• Ongoing pregnancy or breastfeeding (self-reported by the participant or diagnosed by the treating phyisician)
• Known contraindication to baloxavir or to the placebo
• Participant weighing < 40 kg
• Patients already on NAI therapy for the current influenza episode for > 24 hours at the time of randomization.
• Prior treatment with baloxavir for the current influenza epidose
• Immunosuppression defined as 1) cancer treatment with significant negative effect on the immune system; 2) immunosuppressive therapy (treatments comprising a dose of ≥20 mg/day prednisone or equivalent when administered for ≥ 2 weeks, biological therapies, steroid sparing drugs); 3) HIV infection if CD4+ T cell count < 500/µL; 4) organ or stem cell transplantation; 5) patients on the waiting list for a transplant
• Severe underlying respiratory comorbidity requiring long-term oxygenotherapy at home.
• Severe disease requiring ICU care directly at hospitalization.
• Severe hepatic insufficiency or any other severe medical condition when participation in the study puts the patient at risk according to the investigator's judgment.
• History of inclusion in this study during a previous influenza season
• Inclusion in another interventional study with an investigational drug 30 days before inclusion in the study.
• Unability to consent or patient representative unable to consent.

The investigators plan to share individual participant data that underlie the results reported in this article, after deidentification (text, tables, figures, and appendices). IPD can be shared with investigators whose proposed use of the data has been approved by an independent review committee, for, as an example, individual participant data meta-analysis. The approval of the biobank regulation committee, which includes the coordinating PI, the sites PI and the responsibles of the biobank is also required on a case by case basis.

Beginning 3 months and up to 10 years after publication.

IPD can be shared with investigators whose proposed use of the data has been approved by an independent review committee. The approval of the biobank regulation committee, which includes the coordinating PI, the sites PI and the responsibles of the biobank is also required on a case by case basis.