Rationale Many patients with coronavirus disease (COVID-19) need hospital admission for oxygen supplementation. A substantial number of patients need intensive care unit (ICU) admission for escalation of care. ICU doctors and nurses are struggling to provide the best care for patients with COVID-19. Practice of adjunctive and supportive treatments remains uncertain.

Objective To determine and compare practice of adjunctive and supportive treatments for COVID-19 in the Netherlands, and to determine their independent associations with outcome.

Hypotheses Practice of adjunctive and supportive treatments for COVID-19 varies substantially. Adjunctive and supportive treatments have an independent association with outcome in ICU patients with COVID-19.

Study design National/international, multicenter, retrospective observational study. Study population Intensive care unit (ICU) patients with COVID-19.

Methods In this study we will collect data on diverse treatments during the first 28 days in ICU, including (a) the types of oxygen support* and awake prone positioning; (b) the types of ventilatory support, (c) rescue therapies for refractory hypoxemia during invasive ventilation (prone positioning, ventilator adjustments, continuous muscle paralysis, and extracorporeal membrane oxygenation); (d) adjunctive treatments, including thromboprophylaxis and anticoagulation, antiviral and immunomodulating therapies, and (e) experimental supportive treatments. Outcomes include duration of each adjunctive treatment, duration of ventilation, incidence of tracheostomy, duration of stay in ICU and mortality until day 90.

Study endpoints A combination of adjunctive treatments, including types of oxygen support, ventilatory support and rescue therapies for refractory hypoxemia during invasive ventilation (primary), other adjunctive and supportive treatments, tracheostomy rate; duration of ventilation and ventilator-free days and alive at day 28 (VFD-28), duration of ICU and hospital stay, and ICU, hospital and 90-day mortality.

Nature and extent of the burden and risks associated with participation, benefit and group relatedness Retrospective collection of data regarding adjunctive treatments, and clinical endpoints is without risk for ICU patients.

*In a subset of patients we will collect granular data (every two hours) regarding oxygenation (FiO2, inspiratory tidal volume, air flow, respiratory rate, SpO2, PaO2, and PEEP) over the first 2 full calendar days of ICU admission. The primary endpoint of this sub-analysis will be the amount of oxygen used with different respiratory support interventions. The statistical analysis plan for the analysis of these data that were collected in two ICUs that participated in the national study, and one additional ICU in Spain is uploaded in the document section (filename Statistical Analysis Plan PROXY-COVID)

Inclusion Criteria:

• COVID-19, confirmed with PCR; and
• Admission to one of the participating ICUs, or an emergency location that serves as an ICU during the pandemic.

Exclusion Criteria:

• Age < 18 years

Patients with severe coronavirus disease 2019 (COVID-19) admitted to intensive care units (ICUs) in the Netherlands between September 2020 and January 2021 (i.e. during the second surge of COVID-19 patients).