Clinical Cohort Data for Replicating COVID-19 Severity Score Studies

1,963 patient records contain clinical and demographic variables for a full study cohort. The dataset includes parameters required to calculate q-CSI, ISARIC-4C, SEIMC, and CALL scores, as well as 30-day mortality outcomes. It was authored by Johan Azañero-Haro and last updated on May 14, 2026.

Use Cases

• Replicating mortality prediction studies based on the included clinical scores and 30-day outcome.
• Comparing the performance of different severity scores (q-CSI, ISARIC-4C, SEIMC, CALL) using a common patient cohort.
• Conducting a complete-case analysis for model development, as missing values are explicitly maintained.

Strengths

• Contains 1,963 patient records for a full study cohort.
• Includes all parameters needed to calculate four established clinical severity scores.
• Maintains missing values to ensure transparency for complete-case analysis selection.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for large-scale modeling.
• The 1.1 MB file size suggests a small dataset, potentially limiting feature richness.

Provenance

Source

Johan Azañero-Haro via figshare

Collection Method

Likely collected as part of a clinical study cohort.

Freshness

Last updated 2026-05-14 17:27:23.