A Pragmatic Trial of Glucocorticoids for Community-Acquired Pneumonia.

The Background cites multiple prior trials and meta-analyses of glucocorticoids for CAP (Dequin et al., Meduri et al., etc.), noting both evidence of benefit and remaining uncertainty. It explicitly addresses limitations of prior work: they were conducted in older patients in high-income settings, excluded common comorbidities in sub-Saharan Africa (HIV, TB), and were mostly ICU-based. The premise that these gaps justify a pragmatic trial in Kenya is logically sound.

Randomization method is reported (central list by independent statistician, sealed opaque envelopes). The unit of randomization is the individual participant. Blinding is not performed (open-label), but the rationale is implicitly given by the pragmatic design and the use of a mortality endpoint, which is objective. Power analysis is reported (85% power to detect 25% relative reduction, 2180 patients). Inclusion/exclusion criteria are pre-specified. The outlier handling sub-criterion maps to the analysis population (ITT primary, modified ITT and complete-case sensitivity); this is adequate. Replicate_distinction, controls, and independent_replication are not applicable for a human RCT as per applicability rules.

Sex is reported (46.3% women), both sexes are enrolled. Age, weight (BMI), health status (oxygen saturation, chronic illnesses, HIV status) are reported in Table 1. Demographics are thorough. Species/strain/housing are not applicable for a human trial. The reporting meets standards for an interventional human study.

The paper names the Kenyan Medical Research Institute Scientific and Ethics Review Unit (SERU 4319), the Kenya Pharmacy and Poisons Board, and the University of Oxford's Tropical Research Ethics Committee (OxTREC 4–22). Written informed consent was obtained from participants and/or their legally acceptable representative. Regulatory compliance with Kenyan and UK frameworks is implied by the named approvals.

The investigational products (dexamethasone, hydrocortisone, methylprednisolone, prednisolone, prednisone) are named with doses. Software is identified (not explicitly but the analysis uses Cox regression; no bespoke code is mentioned). Antibodies, cell lines, and mycoplasma testing are not applicable. Reagents_identified is scored against the investigational product as per applicability rules; the drugs are named with doses but not manufacturer/lot numbers. This is adequate for a pragmatic trial.

The primary test is named (Cox regression, log-rank test). Proportional hazards assumption was verified using Schoenfeld residuals. Exact p-values are reported for the primary analysis (P=0.021). Effect sizes and confidence intervals are reported for the primary outcome (HR 0.84, 95% CI 0.73-0.97). Statistical software is not identified. Data presentation includes Kaplan-Meier curves, forest plots, and tables with per-group n's. Mathematical plausibility checks on the primary outcome are not applicable due to continuous data and large N. Tests for secondary outcomes are named but CIs are noted as not adjusted for multiplicity.

A data availability statement is present in the form of trial registration numbers (PACTR and ISRCTN) and a link to the protocol at NEJM.org. However, no statement explicitly describes where individual patient data can be accessed. Repository deposit and accession numbers are not applicable for individual patient data due to privacy constraints, but no managed access procedure is described. Code sharing is not mentioned; no custom code repository is provided.

Methods are detailed enough for replication. The trial is prospectively registered (PACTR, ISRCTN). A reporting guideline is not explicitly mentioned (no CONSORT checklist reference). All pre-specified outcomes (mortality at various time points, safety) are reported; immune response data are deferred. Limitations are thoroughly discussed (heterogeneous population, open-label, no monitoring of co-interventions, oral formulations). Conclusions are proportional, noting the effect size is lower than in previous trials and recommending further investigation. Funding sources and conflict of interest are disclosed.