Treatment by Disease Severity
Mild to Moderate — Outpatient Treatment
Most COVID-19 cases are mild or moderate and can be managed at home. High-risk patients with mild-to-moderate illness may be candidates for antiviral therapy.
Paxlovid (Nirmatrelvir / Ritonavir)
Status: FDA-approved for adults with mild-to-moderate COVID-19 at high risk for severe disease. Adolescents aged 12–17 (≥40 kg) are covered separately under an Emergency Use Authorization (EUA), not the full approval.
How it works: Nirmatrelvir inhibits the SARS-CoV-2 main protease (Mpro), blocking viral replication. Ritonavir boosts nirmatrelvir levels by inhibiting its metabolism.
Efficacy: Original EPIC-HR trial: ~89% reduction in hospitalization/death vs. placebo in unvaccinated high-risk adults. Real-world effectiveness in the vaccinated population is lower but still significant, particularly for older adults and the immunocompromised.
Window: Must start within 5 days of symptom onset.
Key concern: Due to significant CYP3A4 drug interactions, prescribers review all current medications before initiating Paxlovid — it is contraindicated or requires dose adjustment with statins, certain anticoagulants, immunosuppressants, and others.
Paxlovid rebound: A subset of patients (estimated 5–10%) experience symptom return 2–8 days after completing the 5-day course. Rebound is generally mild. No evidence that a second Paxlovid course improves outcomes. Research also suggests Paxlovid may reduce the risk of Long COVID, making early treatment a priority for eligible high-risk patients.
Remdesivir (Veklury) — Outpatient
Status: FDA-approved for adults and pediatric patients — including neonates weighing at least 1.5 kg — with mild-to-moderate COVID-19 at high risk for severe disease who are not hospitalized. The outpatient course is started within 7 days of symptom onset.
How it works: Nucleotide prodrug that inhibits SARS-CoV-2 RNA-dependent RNA polymerase, blocking viral replication.
Efficacy: PINETREE trial: 3-day intravenous (IV) course reduced hospitalization/death by 87% vs. placebo in high-risk, unvaccinated adults.
Limitation: Requires IV infusion — administered in an outpatient infusion center, not at home. May be preferred when Paxlovid drug interactions are prohibitive.
Molnupiravir (Lagevrio)
Status: Not FDA-approved — available only under an Emergency Use Authorization (EUA) for adults with mild-to-moderate COVID-19 at high risk for severe disease, and only when other authorized treatments are inaccessible or clinically inappropriate.
How it works: Ribonucleoside analog that introduces errors into the viral RNA, leading to "error catastrophe."
Efficacy: MOVe-OUT trial: ~30% reduction in hospitalization/death — significantly less effective than Paxlovid. Not recommended as first-line therapy.
Concern: Theoretical mutagenic risk has limited its use. Not recommended during pregnancy.
Severe — Hospitalized Patients
Dexamethasone
Recommended for: Hospitalized patients requiring supplemental oxygen or mechanical ventilation.
Efficacy: RECOVERY trial (6,425 patients): 36% reduction in 28-day mortality in ventilated patients; 18% in those requiring oxygen without ventilation. No benefit — and possible harm — in patients not requiring oxygen.
Why it works: In severe COVID-19, the immune system's over-response (cytokine storm) causes much of the lung damage. Dexamethasone suppresses this excessive inflammation.
Mechanism contrast: Antivirals target the virus directly; dexamethasone targets the dysregulated immune response that causes severe disease.
Remdesivir (Veklury) — Inpatient
Recommended for: Hospitalized adults and children with COVID-19 requiring supplemental oxygen (but not high-flow oxygen or ventilation).
Efficacy: ACTT-1 trial: Reduced median time to recovery from 15 to 10 days in patients requiring oxygen. Mortality benefit was less clear in subsequent trials.
Dosing: 5-day IV course (200 mg loading dose, then 100 mg daily).
Baricitinib (Olumiant)
Recommended for: Hospitalized adults with severe or critical COVID-19 requiring supplemental oxygen.
How it works: JAK1/JAK2 inhibitor that reduces the inflammatory signaling cascade driving cytokine storm. Often used in combination with dexamethasone.
Efficacy: COV-BARRIER trial: Significant reduction in 28-day all-cause mortality when combined with standard of care including dexamethasone.
Home Care for Mild COVID-19
For most vaccinated, otherwise-healthy adults with mild COVID-19, supportive care at home is the primary approach:
- Rest — your immune system does the work; physical exertion during acute illness slows recovery
- Hydration — fever and respiratory illness increase fluid losses; staying well hydrated supports recovery
- Fever & pain relief — the CDC notes that acetaminophen (Tylenol) or ibuprofen are both appropriate options; follow package dosing instructions
- Monitor oxygen levels — a pulse oximeter can detect silent hypoxia; readings below 92% warrant medical attention
- Isolate — protect household members, especially high-risk individuals
- High-risk individuals — the CDC identifies adults over 65, immunocompromised individuals, and those with significant underlying conditions as high-priority candidates for antiviral treatment; these individuals should contact a healthcare provider promptly after a positive test to discuss eligibility
Remdesivir vs Paxlovid vs Molnupiravir: the Trials, Side by Side
Three outpatient antivirals, three separate placebo-controlled trials, no head-to-head comparison between any of them. Here is what each actually found — in absolute terms, not just relative ones.
| Drug | Trial | Who was enrolled | Result | Absolute difference |
|---|---|---|---|---|
| Nirmatrelvir/ritonavir (Paxlovid) | EPIC-HR | 2,246 randomised; unvaccinated adults with ≥1 risk factor; treatment started within 5 days | Hospitalisation or death by day 28: 0.77% vs 7.01% in the interim analysis of those treated within 3 days | −6.32 percentage points (95% CI −9.04 to −3.59); relative risk reduction 89.1%. Final analysis of 1,379 patients: −5.81 points (−7.78 to −3.84), 88.9%. All 13 deaths were in the placebo group. |
| Remdesivir (Veklury, outpatient) | PINETREE | 562 patients dosed; symptom onset within 7 days; ≥1 risk factor; 3-day IV course | Hospitalisation or death by day 28: 2 of 279 (0.7%) vs 15 of 283 (5.3%) | Hazard ratio 0.13 (95% CI 0.03 to 0.59). No patient in either arm died by day 28. |
| Molnupiravir (Lagevrio) | MOVe-OUT | 1,433 randomised; unvaccinated adults with ≥1 risk factor; treatment started within 5 days | Hospitalisation or death by day 29: 6.8% vs 9.7% in the full randomised population | −3.0 percentage points (95% CI −5.9 to −0.1) |
The molnupiravir number that changed between analyses
MOVe-OUT is the clearest example on this page of why an interim result should be read with care. At the planned interim analysis reported in the NEJM, when half the target enrolment had been followed to day 29, molnupiravir reduced hospitalisation or death from 14.1% to 7.3% — a 6.8-percentage-point difference (95% CI −11.3 to −2.4). Those were the numbers that drove the drug's authorisation and the headlines around it.
In MOVe-OUT’s full randomised population, as published in the NEJM, the effect was less than half that size: 6.8% versus 9.7%, a difference of 3.0 percentage points with a confidence interval of −5.9 to −0.1. The upper bound of that interval is one-tenth of a percentage point from zero. The investigators also reported that in several prespecified subgroups — patients with evidence of previous SARS-CoV-2 infection, those with low baseline viral load, and those with diabetes — the point estimate favoured placebo.
This is why the drug sits third in clinical practice and why its authorisation is conditioned on other options being inaccessible or clinically inappropriate. It is also why this page reports both analyses rather than the more flattering one.
Why "remdesivir vs Paxlovid" has no trial answer
No randomised trial has compared them. As published in the NEJM, EPIC-HR and PINETREE were separate placebo-controlled studies with different sponsors, different enrolment windows, different treatment windows (5 days from onset for Paxlovid, 7 for remdesivir), and — decisively — different baseline risks. The placebo arms tell the story: 7.01% of EPIC-HR's placebo patients were hospitalised or died, against 5.3% in PINETREE. Different underlying risk means the relative reductions are not measuring the same quantity, and PINETREE's 562 patients produced a hazard-ratio interval (0.03 to 0.59) wide enough to be compatible with a very large or a fairly modest effect.
What separates them in practice is not efficacy but logistics and interactions. Paxlovid is five days of oral tablets, taken at home, and carries substantial CYP3A4 drug-drug interactions — the reason a prescriber reviews every current medication before starting it. Remdesivir is three consecutive days of intravenous infusion at an infusion centre, which is a significant practical burden but sidesteps the interaction problem entirely. That is the actual basis on which the choice is made.
One further caveat applies to all three: every one of these trials enrolled unvaccinated adults with no documented prior infection, during 2020–2021. Almost no one being treated today matches that description. The absolute benefit in a population with hybrid immunity and a much lower baseline risk of hospitalisation is smaller than these trials measured — how much smaller is estimated from observational data, not from randomised evidence of comparable quality.
Dexamethasone: the third arm that is usually left out
The RECOVERY trial, published in the NEJM, randomised 2,104 hospitalised patients to dexamethasone and 4,321 to usual care. Overall 28-day mortality was 22.9% versus 25.7% (age-adjusted rate ratio 0.83, 95% CI 0.75 to 0.93). The benefit was concentrated entirely by level of respiratory support:
- Invasive mechanical ventilation (NEJM): 29.3% vs 41.4% — rate ratio 0.64 (0.51 to 0.81). The largest absolute benefit in the trial: about 12 fewer deaths per 100 patients treated.
- Oxygen without ventilation (NEJM): 23.3% vs 26.2% — rate ratio 0.82 (0.72 to 0.94).
- No respiratory support (NEJM): 17.8% vs 14.0% — rate ratio 1.19 (0.92 to 1.55). The point estimate favours usual care, and the interval does not exclude meaningful harm.
That third line is the one omitted from most summaries, and it is the clinically important one for anyone reading this page at home with a mild infection. A steroid that saves lives in ventilated patients has not been shown to help — and may plausibly harm — people who are not short of oxygen. Dexamethasone is a hospital drug for hypoxic patients, not a treatment for COVID-19 as such.
What is not known
- Whether any outpatient antiviral reduces hospitalisation in a vaccinated, previously infected population. The randomised evidence does not cover that population, and the absolute risk it would have to move is now very low.
- Whether Paxlovid prevents Long COVID. It is being studied on the viral-persistence hypothesis — see our Long COVID page — but no trial has established it, and this page will not assert it.
- Optimal management of Paxlovid rebound. Symptom return after the 5-day course is described in the literature and generally mild; there is no randomised evidence that a second course helps.
- Whether a second-generation oral antiviral without the interaction problem will arrive. Candidates are in development; none has replaced nirmatrelvir in practice.
For where COVID-19's severity now sits relative to other infections, see our COVID-19 Risk Perspective and the hub's Virus Risk Perspective.
Sources & References
- CDC. Clinical Care of COVID-19 (treatment). cdc.gov
- U.S. Food and Drug Administration. FDA Approves First Oral Antiviral for Treatment of COVID-19 in Adults (Paxlovid). May 25, 2023. fda.gov
- U.S. Food and Drug Administration. Lagevrio (molnupiravir) — Letter of Authorization (EUA). fda.gov
- World Health Organization. Therapeutics and COVID-19: living guideline. who.int
- Hammond J, et al. "Oral Nirmatrelvir for High-Risk, Nonhospitalized Adults with Covid-19" (EPIC-HR). NEJM 2022;386:1397–1408. doi:10.1056/NEJMoa2118542
- Gottlieb RL, et al. "Early Remdesivir to Prevent Progression to Severe Covid-19 in Outpatients" (PINETREE). NEJM 2022;386:305–315. doi:10.1056/NEJMoa2116846
- Jayk Bernal A, et al. "Molnupiravir for Oral Treatment of Covid-19 in Nonhospitalized Patients" (MOVe-OUT). NEJM 2022;386:509–520. doi:10.1056/NEJMoa2116044
- RECOVERY Collaborative Group. "Dexamethasone in Hospitalized Patients with Covid-19." NEJM 2021;384:693–704. doi:10.1056/NEJMoa2021436
Continue Reading
Antivirals Compared
Paxlovid vs remdesivir vs molnupiravir side by side, and the interaction problem.
COVID Symptoms
How symptoms differ across variants and when to seek emergency care.
Read more →Long COVID
Symptoms, risk factors, and ongoing research into PASC.
Read more →Misinformation
Debunked treatments: hydroxychloroquine, ivermectin, and more.
Read more →