Publication

• Title: Ivermectin for Critically and Noncritically Ill Hospitalized Patients With COVID-19: Randomized, Embedded, Multifactorial Adaptive Platform Trial for Community-Acquired Pneumonia (REMAP-CAP)
• Acronym: REMAP-CAP COVID-19 Antiviral Therapy (Ivermectin) Domain
• Year: 2026
• Journal published in: Critical Care Medicine
• Citation: Hashmi M, Haniffa R, Jayakumar D, Beane A, Lorenzi E, Berry LR, et al; REMAP-CAP Investigators. Ivermectin for Critically and Noncritically Ill Hospitalized Patients With COVID-19: Randomized, Embedded, Multifactorial Adaptive Platform Trial for Community-Acquired Pneumonia (REMAP-CAP). Crit Care Med. 2026;54(00):XX–XX.

Context & Rationale

• Background

  • Ivermectin is an inexpensive, widely available antiparasitic agent with a long safety record in licensed indications.
  • Early pandemic interest was driven by in vitro inhibition of SARS-CoV-2 replication, but the translational relevance was uncertain because concentrations required for antiviral effects were difficult to achieve safely with standard human dosing.¹²
  • By 2021, ivermectin had become widely promoted and used in some settings despite limited and heterogeneous clinical evidence.
  • Most early clinical trials were outpatient, small, single-centre, methodologically limited, or vulnerable to bias; critically ill and hospitalised patients were under-represented.
• Research Question/Hypothesis

  • Does enteral ivermectin, compared with no ivermectin, improve organ support-free days and survival in hospitalised patients with COVID-19?
  • The investigators evaluated critically ill patients receiving ICU organ support and noncritically ill hospitalised patients separately, while allowing Bayesian dynamic borrowing between illness states when treatment effects were concordant.
• Why This Matters

  • Ivermectin was low-cost and potentially scalable, making it attractive in resource-limited settings if effective.
  • Unproven use risked diverting attention from proven COVID-19 therapies and from enrolment into high-quality randomised trials.
  • This REMAP-CAP domain specifically addressed the evidence gap in hospitalised and critically ill patients, where robust randomised evidence had been sparse.

Design & Methods

• Research Question: Among critically and noncritically ill hospitalised adults with suspected or confirmed COVID-19, does ivermectin improve respiratory and cardiovascular organ support-free days and survival compared with no ivermectin?
• Study Type: International, multicentre, randomised, open-label, response-adaptive, multifactorial Bayesian platform trial embedded within REMAP-CAP. Patients were enrolled in hospitals in Pakistan, India, and Ireland between 11 June 2021 and 9 September 2022.
• Population:
  • Adults aged ≥18 years admitted to hospital with suspected or confirmed COVID-19.
  • Critically ill state: receiving respiratory support in an ICU, defined as high-flow nasal oxygen ≥30 L/min with Fio2 ≥0.4, non-invasive ventilation, invasive mechanical ventilation, or cardiovascular organ support with vasopressors or inotropes.
  • Noncritically ill state: all other hospitalised patients, including those in ICU for reasons such as infection-control zoning rather than organ support.
  • Key exclusions: intention to receive or ongoing receipt of ivermectin; receipt of an antiviral not licensed for COVID-19; participation in another ongoing antiviral trial; confirmed MERS-CoV infection; hypersensitivity; pregnancy or breastfeeding; severe liver disease; creatinine clearance <30 mL/min unless receiving renal replacement therapy.
  • Timing exclusions: >48 hours since start of ICU organ support for critically ill patients, or >96 hours since hospital admission for noncritically ill patients.
• Intervention:
  • Ivermectin 0.2 mg/kg enterally once daily, maximum 24 mg/day.
  • Given for up to 5 days or until hospital discharge, whichever occurred first.
  • Tablets could be dispersed in water and delivered via an enteral feeding tube when required.
  • Other care, including corticosteroids, remdesivir, anticoagulation, and immunomodulators where available, was delivered according to local practice and other REMAP-CAP domain assignments.
• Comparison:
  • No ivermectin, without placebo.
  • Licensed COVID-19 antivirals such as remdesivir were permitted.
  • Other aspects of care were provided according to local standard care.
• Blinding: Open-label treatment allocation. There was no placebo, so treating clinicians knew whether ivermectin was assigned. The International Trial Steering Committee and clinical staff were not provided aggregate outcome data while the domain was ongoing; the DSMB was unblinded.
• Statistics: No conventional fixed sample-size power calculation was used. REMAP-CAP used a Bayesian adaptive design with no maximum sample size; recruitment was intended to continue until prespecified posterior statistical triggers were met. Ivermectin efficacy required ≥99% posterior probability of superiority versus control; futility required <5% probability of at least a 20% odds-ratio improvement; harm was prespecified as >90% probability of OR <1, although the harm trigger was not implemented at adaptive analyses because futility would precede it. The primary analysis was by assigned group, using a Bayesian cumulative logistic model adjusted for domain eligibility and assignments, site nested within country, age category, sex, and 2-week calendar epoch.
• Follow-Up Period: Primary outcome was respiratory and cardiovascular organ support-free days to day 21, with in-hospital death through day 90 assigned the worst category. Secondary outcomes included survival to hospital discharge and 90-day survival.

Key Results

This trial was stopped early. Enrolment closed for operational, not statistical, futility after 153 patients were randomised and 150 were analysed, because recruitment was low after external outpatient trials suggested no benefit with ivermectin. No Bayesian statistical trigger for ivermectin efficacy, futility, or harm was reached.

• The primary endpoint was almost exactly neutral in both severity strata: adjusted OR 0.94 in critically ill patients and 1.04 in noncritically ill patients.
• The prespecified remdesivir subgroup analyses did not show a consistent effect: for organ support-free days, adjusted ORs were 1.13 (95% CrI 0.44 to 2.97) without remdesivir and 0.82 (95% CrI 0.30 to 1.99) with remdesivir among critically ill patients; 1.16 (95% CrI 0.45 to 3.19) without remdesivir and 0.92 (95% CrI 0.38 to 2.36) with remdesivir among noncritically ill patients.
• The negative interpretation is driven less by a definitive null estimate than by the combination of neutral point estimates, wide uncertainty, absence of a Bayesian efficacy signal, and concordance with larger external ivermectin trials.

Internal Validity

• Randomisation and Allocation: Randomisation used a centralised algorithm with response-adaptive allocation. Initial allocation was 1:1; after a May 2022 adaptive analysis, allocation ratios changed to ivermectin:control 1.0:2.0 for critically ill patients and 1.2:1.0 for noncritically ill patients. Allocation concealment before assignment appears robust.
• Dropout or Exclusions: Of 153 randomised patients, 3 withdrew consent, leaving 150 analysed. There were no missing primary outcome data among analysed participants. This is a strength, although the overall sample was much smaller than needed for precise estimation.
• Performance/Detection Bias: The trial was open-label and had no placebo, so treatment decisions after randomisation could theoretically be influenced by clinicians’ knowledge of allocation. The primary endpoint was largely objective, especially death, but organ support-free days can still be influenced by local thresholds for escalation, weaning, ICU discharge, and ward transfer.
• Protocol Adherence: Separation of the treatment variable was excellent: 37/37 critically ill and 44/44 noncritically ill patients assigned to ivermectin received ivermectin, while 0/24 critically ill and 0/45 noncritically ill controls received ivermectin.
• Separation of the Variable of Interest: Median ivermectin duration was 5 days (IQR 4 to 5) in critically ill patients and 5 days (IQR 4.5 to 5) in noncritically ill patients. Median daily dose was 14.8 mg (IQR 14 to 17.6) in critically ill patients and 14.0 mg (IQR 13 to 16) in noncritically ill patients.
• Baseline Characteristics: Baseline characteristics were broadly similar but the small sample created imbalances. In critically ill patients, median APACHE II score was 11 in the ivermectin group versus 9 in control; vasopressor use was 1/37 (2.7%) versus 3/24 (12.5%); invasive ventilation was 2/37 (5.4%) versus 1/24 (4.2%). Noncritically ill patients had median APACHE II scores of 7 versus 6.
• Heterogeneity: Critical and noncritical illness were analysed separately with Bayesian borrowing. This was methodologically appropriate, but the clinical cohorts were small and outcome distributions differed substantially: critically ill patients had very high mortality, while noncritically ill patients frequently achieved the maximum organ support-free day score.
• Timing: Enrolment was early after hospital admission: median 0.7 days versus 1.0 days in critically ill patients and 0.5 days versus 0.6 days in noncritically ill patients. Time from ICU admission in critically ill patients was 12.9 hours versus 15.6 hours. Symptom-onset-to-treatment timing was not clearly central to the reported baseline table, limiting assessment of whether antiviral treatment occurred during peak viral replication.
• Dose: The tested dose was pragmatic and aligned with accepted ivermectin dosing for other indications, but it was not a high-dose antiviral strategy. This matters because pharmacokinetic analyses questioned whether standard doses can achieve antiviral concentrations relevant to SARS-CoV-2.²
• Adjunctive Therapy Use: Corticosteroid use was high: 33/37 (89.2%) versus 23/24 (95.8%) in critically ill patients and 43/44 (97.7%) versus 39/45 (86.7%) in noncritically ill patients. Remdesivir use was also high and somewhat greater in ivermectin arms: 29/37 (78.4%) versus 15/24 (62.5%) in critically ill patients and 40/44 (90.9%) versus 35/45 (77.8%) in noncritically ill patients. Tocilizumab or sarilumab use was low: 3/37 (8.1%) versus 2/24 (8.3%) in critically ill patients and 1/44 (2.3%) versus 2/45 (4.4%) in noncritically ill patients.
• Crossover: Clinically relevant crossover was not a problem: no controls received ivermectin and all patients allocated to ivermectin received at least one dose.
• Outcome Assessment: The primary outcome was prespecified and clinically meaningful for critical care, combining survival and organ support. Its ordinal construction appropriately assigns death the worst outcome, but interpretation in a small cohort can be dominated by mortality in critical illness and by ceiling effects in less severe illness.
• Statistical Rigor: The Bayesian modelling strategy matched REMAP-CAP’s adaptive design. The statistical analysis plan was posted before analysis. The analysis was revised before unblinding to constrain covariate estimation to Pakistan, India, and Ireland because all enrolled patients came from those countries. The major limitation is that the domain was closed before statistical stopping criteria were met.

Conclusion on Internal Validity: Internal validity is moderate. Randomisation, allocation concealment, adherence, and objective outcome ascertainment were strong, but the open-label design, very small sample, operational early closure, imprecision, and dose uncertainty substantially limit causal confidence for small or moderate treatment effects.

External Validity

• Population Representativeness: The trial enrolled real hospitalised COVID-19 patients, including critically ill patients, which was the key evidence gap. However, 96% of patients were randomised in Pakistan, with only 5 patients from India and 1 from Ireland.
• Severity Spectrum: The critically ill cohort was mostly managed with non-invasive respiratory support: 18/61 (29.5%) received high-flow nasal cannula, 40/61 (65.6%) non-invasive ventilation, and only 3/61 (4.9%) invasive mechanical ventilation at baseline. Only 4/61 (6.6%) received vasopressors at baseline. This differs from many high-income ICU cohorts dominated by invasive ventilation and vasopressor-dependent shock.
• Healthcare System Context: The trial is highly relevant to resource-limited settings where ivermectin use was common and access to expensive immunomodulators was limited. It is less directly generalisable to systems with routine IL-6 inhibitor use, widespread vaccination, later variants, or different thresholds for ICU admission and organ support.
• COVID-19 Era: Enrolment occurred from June 2021 to September 2022. Standard care, variants, vaccination status, and baseline risk have evolved since then, making extrapolation to current lower-risk hospitalised populations imperfect.
• Clinical Applicability: The result applies to treatment of hospitalised adults with COVID-19 using ivermectin 0.2 mg/kg/day for up to 5 days. It does not test prophylaxis, early outpatient use, or high-dose regimens in hospitalised critical illness.

Conclusion on External Validity: External validity is moderate for hospitalised patients in similar resource-constrained settings and limited for contemporary high-income ICU practice, heavily vaccinated populations, and patients already receiving invasive ventilation or shock-level cardiovascular support.

Strengths & Limitations

• Strengths:
  • Randomised comparison within a mature adaptive platform trial.
  • Centralised allocation and excellent protocol adherence.
  • Clinically meaningful primary endpoint incorporating both survival and organ support.
  • Separate evaluation of critically and noncritically ill hospitalised patients.
  • Very low missingness: no missing primary outcome among analysed participants.
  • Important representation of South Asian sites and resource-limited practice contexts.
• Limitations:
  • Stopped for operational futility before any statistical stopping threshold was met.
  • Very small analysed sample: 61 critically ill and 89 noncritically ill patients.
  • Open-label design without placebo.
  • Wide credible intervals compatible with both benefit and harm.
  • Dominance of one country and a small number of enrolling sites.
  • Low use of invasive ventilation, vasopressors, and IL-6 inhibitors limits extrapolation to many ICU settings.
  • Standard-dose ivermectin only; no high-dose hospitalised strategy was tested.
  • Possible ceiling effect in the noncritically ill organ support-free day endpoint.

Interpretation & Why It Matters

• Clinical Practice

  Ivermectin should not be used as routine treatment for hospitalised or critically ill COVID-19 patients on the basis of this trial. The trial found no convincing improvement in organ support-free days, hospital survival, 90-day survival, or progression to mechanical ventilation or death.
• Evidence Integration

  The findings align with higher-quality outpatient trials and systematic reviews that failed to show clinically meaningful benefit, despite early enthusiasm from laboratory and low-quality clinical evidence.³⁴
• Methodological Lesson

  The trial illustrates both the strength and vulnerability of adaptive platform trials: they can rapidly embed controversial, policy-relevant questions into care, but operational futility can leave important domains underpowered if external evidence and falling case numbers reduce recruitment.

Controversies & Other Evidence

• The ivermectin evidence base became unusually controversial because early positive interpretations relied heavily on small, heterogeneous, and methodologically fragile trials. Published critiques identified important bias concerns and possible fraudulent data within the broader ivermectin literature.³
• The 2022 Cochrane review concluded that evidence did not support ivermectin for prevention or treatment of COVID-19 and highlighted the limited quality of much of the trial base.⁴
• Major outpatient randomised trials were consistently negative: TOGETHER found no benefit from early ivermectin in outpatients; COVID-OUT found no clinically important benefit in a factorial outpatient trial; ACTIV-6 found no improvement in sustained recovery with standard ivermectin; higher-dose ACTIV-6 found no recovery benefit with up to 600 μg/kg daily for 6 days; PRINCIPLE found no clinically meaningful improvement in recovery, hospital admission, or longer-term outcomes.⁵⁶⁷⁸⁹
• A 2026 meta-analysis of 40 RCTs involving 23,243 participants found no statistically significant mortality benefit in hospitalised patients (RR 0.94; 95% CI 0.74 to 1.20) or outpatients (RR 0.88; 95% CI 0.55 to 1.42), and no significant difference in adverse events for hospitalised patients (RR 1.02; 95% CI 0.76 to 1.37) or outpatients (RR 0.96; 95% CI 0.82 to 1.13).¹⁰
• Current WHO living guidance and IDSA guidance are concordant with this interpretation: ivermectin is not recommended for COVID-19 treatment outside clinical trials.¹¹¹²
• The main trial-specific interpretive controversy is precision: REMAP-CAP did not prove equivalence or meet a statistical futility trigger. It showed that, in the enrolled hospitalised cohorts, ivermectin was unlikely to provide a clinically important benefit, especially when interpreted alongside external RCTs and meta-analyses.

Summary

• REMAP-CAP randomised 150 analysed hospitalised COVID-19 patients to ivermectin or no ivermectin: 61 critically ill and 89 noncritically ill.
• The trial closed for operational futility, not because a Bayesian statistical efficacy or futility threshold was reached.
• Primary outcome results were neutral: critically ill adjusted OR 0.94 (95% CrI 0.40 to 2.07) and noncritically ill adjusted OR 1.04 (95% CrI 0.48 to 2.34).
• Hospital survival and 90-day survival showed no convincing benefit; serious adverse events were not reported in either group.
• The trial is underpowered on its own, but its findings are consistent with a large body of subsequent and contemporaneous evidence showing no meaningful clinical benefit from ivermectin for COVID-19.

Overall Takeaway

This is not a landmark positive trial; it is a methodologically important negative platform-trial domain addressing a high-profile therapeutic controversy in hospitalised COVID-19. Its small size prevents definitive exclusion of modest effects, but the neutral estimates, absence of a Bayesian efficacy signal, and concordant external evidence support not using ivermectin for hospitalised or critically ill COVID-19 outside research.

Bibliography

• 1.Caly L, Druce JD, Catton MG, Jans DA, Wagstaff KM. The FDA-approved drug ivermectin inhibits the replication of SARS-CoV-2 in vitro. Antiviral Res. 2020;178:104787.
• 2.Schmith VD, Zhou JJ, Lohmer LRL. The approved dose of ivermectin alone is not the ideal dose for the treatment of COVID-19. Clin Pharmacol Ther. 2020;108(4):762-765.
• 3.Hill A, Mirchandani M, Pilkington V. Ivermectin for COVID-19: addressing potential bias and medical fraud. Open Forum Infect Dis. 2022;9(2):ofab645.
• 4.Popp M, Reis S, Schiesser S, Hausinger RI, Stegemann M, Metzendorf MI, et al. Ivermectin for preventing and treating COVID-19. Cochrane Database Syst Rev. 2022;6(6):CD015017.
• 5.Reis G, Silva EASM, Silva DCM, et al; TOGETHER Investigators. Effect of early treatment with ivermectin among patients with Covid-19. N Engl J Med. 2022;386(18):1721-1731.
• 6.Bramante CT, Huling JD, Tignanelli CJ, Buse JB, Liebovitz DM, Nicklas JM, et al. Randomized trial of metformin, ivermectin, and fluvoxamine for Covid-19. N Engl J Med. 2022;387(7):599-610.
• 7.Naggie S, Boulware DR, Lindsell CJ, Stewart TG, Gentile N, Collins S, et al; ACTIV-6 Study Group and Investigators. Effect of ivermectin vs placebo on time to sustained recovery in outpatients with mild to moderate COVID-19: a randomized clinical trial. JAMA. 2022;328(16):1595-1603.
• 8.Naggie S, Boulware DR, Lindsell CJ, Stewart TG, Gentile N, Collins S, et al; ACTIV-6 Study Group and Investigators. Effect of higher-dose ivermectin for 6 days vs placebo on time to sustained recovery in outpatients with COVID-19: a randomized clinical trial. JAMA. 2023;329(11):888-897.
• 9.Hayward G, Yu LM, Little P, Gbinigie O, Shanyinde M, Harris V, et al; PRINCIPLE Trial Collaborative Group. Ivermectin for COVID-19 in adults in the community (PRINCIPLE): an open, randomised, controlled, adaptive platform trial of short- and longer-term outcomes. J Infect. 2024;88(4):106130.
• 10.Wang X, Meng J, Li Y, Xiang Y, Wu Y, Xiong Y, et al. The role of ivermectin in the prevention and treatment of SARS-CoV-2 infection: a meta-analysis of randomized controlled trials. BMC Infect Dis. Published online 31 March 2026.
• 11.World Health Organization. Therapeutics and COVID-19: living guideline, August 2025. Geneva: World Health Organization; 2025.
• 12.Bhimraj A, Morgan RL, Shumaker AH, et al. Infectious Diseases Society of America guidelines on the treatment and management of patients with COVID-19. Arlington (VA): Infectious Diseases Society of America; 2025.