Assembling the evidence…
Reliability concern: This paper was retracted on 2024-12-16. See notice: 10.1016/j.ijantimicag.2024.107416.
hydroxychloroquine treatment is significantly associated with viral load reduction/disappearance in COVID-19 patients
reported association is materially weakened by excluded treated patients and non-equivalent controls
Azithromycin added to hydroxychloroquine was significantly more efficient for virus elimination.
claim rests on six clinically selected combination patients without random allocation
hydroxychloroquine is efficient in clearing viral nasopharyngeal carriage of SARS-CoV-2 in COVID-19 patients in only three to six days
causal efficacy language exceeds the open-label non-randomized comparison
we recommend that COVID-19 patients be treated with hydroxychloroquine and azithromycin
global treatment recommendation is disproportionate to sample size, design, and endpoint
Derived from the full evaluation — not a separate score.
Strengths
The Introduction identifies the early-pandemic need for treatments that could reduce symptoms, viral carriage, and transmission. This gives the study a direct practical audience despite limitations in the evidence.
↳ Introduction, p.5
The primary endpoint is day-6 virological clearance, and Tables 2 and 3 report day-by-day PCR negativity by treatment group. This makes the main empirical comparison visible to readers.
↳ Outcome, p.9; Tables 2-3, pp.20-21
Supplementary Table 1 provides individual patient rows with clinical status, treatment, azithromycin status, hydroxychloroquine concentration, and PCR results from D0 to D6. That improves transparency around the reported endpoint calculations.
↳ Supplementary Table 1, p.22
Limitations
Six hydroxychloroquine-treated patients were excluded from the presented 20-patient treated analysis after ICU transfer, death, hospital departure, or treatment cessation. These outcomes are directly relevant to interpreting treatment response and reliability.
↳ Results, p.10
Controls from centres other than Marseille were sampled every other day in most cases and imputed as PCR-positive on missing days when adjacent days were positive. This creates a non-equivalent measurement process for the primary endpoint.
↳ Table 2 note a, p.20
The Discussion recommends hydroxychloroquine and azithromycin for COVID-19 patients worldwide despite a small, open-label, non-randomized study with short follow-up. The recommendation is not proportionate to the evidence presented.
↳ Discussion, p.13
Withdrawn — shown for reference only
The article reports a large apparent difference in day-6 PCR negativity, with 70% of hydroxychloroquine-treated patients versus 12.5% of controls classified as virologically cured in Table 2. That result is difficult to interpret because the Results exclude six treated patients, including ICU transfers and one death, while Table 2 notes that many controls were sampled every other day and imputed as positive on missing days. The azithromycin claim is even less secure because the combination subgroup contains only six patients and azithromycin was added according to clinical presentation or judgment rather than random assignment. The Discussion’s recommendation that COVID-19 patients be treated with hydroxychloroquine and azithromycin is therefore much stronger than the article’s design and evidence can support.
Nabu’s assessment, alongside the field’s view.
Are you an author of this paper?
Very Limited1.8
The paper addresses an urgent early-pandemic treatment question and reports a concrete day-6 virological endpoint. The contribution is substantially limited because the central treatment and combination claims are made from a small, non-randomized comparison with excluded deteriorating treated patients and a broad recommendation in the Discussion.
we recommend that COVID-19 patients be treated with hydroxychloroquine and azithromycin
The merit-based craft score reflects a non-randomized, open-label design with centre-linked treatment allocation, excluded treated patients, differential control sampling, and no sensitivity analysis for missing or excluded outcomes. The Red reliability flag does not change the score because the uncapped Methodological Rigour score is already below the 2.0 cap.
Six hydroxychloroquine-treated patients were lost in follow-up
The article is readable and organized around a single virological endpoint, with tables showing day-by-day PCR status. Precision is limited because strong efficacy and recommendation language is used while important caveats are brief or placed in Results and table notes.
our first results are so significant and evident
The article situates hydroxychloroquine in contemporaneous in vitro and early clinical COVID-19 literature. Contextualization is limited because the discussion does not adequately frame the uncertainty created by non-randomized evidence, surrogate viral clearance, small sample size, and attrition.
Our study has some limitations including a small sample size
Lower confidence on Positioning — domain match limited.
Concerns4 of 4 checks
The main efficacy interpretation is materially affected by selective exclusion of deteriorating treated patients, non-equivalent control ascertainment, and causal treatment language beyond the non-randomized design. These issues occur in the Results, Table 2 note a, and Discussion and could change interpretation of the core findings.
Ethics approval, informed consent, and EU trial registration are declared, but code and formal data availability are absent and conflict-of-interest disclosure is not apparent in the article text. These conduct issues are transparency limitations rather than the primary driver of the Red reliability flag.
Flags: 2 declared / 5 total
Checks on statistical reporting are in development. Nothing has been assessed for this paper yet.
All 31 cited references resolved to real publications via PubMed, Crossref, or OpenAlex.
This paper was retracted on 2024-12-16. See notice: 10.1016/j.ijantimicag.2024.107416.
Sources: Retraction Watch ✓
Low2.5
The paper addresses a problem that clinicians and decision-makers were actively trying to solve during the early COVID-19 pandemic. Its traction score is high because the article explicitly frames treatment and transmission reduction as urgent practical needs, while not reaching the highest anchor because the evidence is not deployment-ready.
urgent need for an effective treatment
The translation score reflects an initial clinical observation based on short-term nasopharyngeal PCR clearance rather than clinical outcomes. The article states that clinical follow-up and side effects will be described later, leaving major validation steps incomplete.
Clinical follow-up and occurrence of side-effects will be described in a further paper
Generalizability is low because the analysed sample is small, treated patients are from Marseille, controls come from several centres, and control sampling differs across sites. These design features limit generalisation to other populations, severity levels, and clinical settings.
Controls without hydroxychloroquine treatment were recruited in Marseille, Nice, Avignon and Briançon centers
The study connects to an active early evidence stream on repurposed antivirals, chloroquine, and hydroxychloroquine. Its trajectory score is limited because it provides a preliminary, confounded signal that requires controlled validation before guiding practice.
Further studies on this combination are needed
Lower confidence on Trajectory — domain match limited.
AI-generated, human-governed. Something look off? Contact us to request a review.