Vitamin D deficiency and SARS‑CoV‑2 infection: D-COVID study

D-COVID is a proyect to study the association between COVID-19 infection and vitamin D deficiency in patients of a terciary university hospital. To investigate the clinical evolution and prognosis of patients with COVID-19 and vitamin D deficiency, several queries qere launched into a Database containing plain text apparitions of certain terms in medical reports, as well as structured data. These apparitions were detected using NLP technology, and then saved individually in a database. The presented dataset is a bounded version of such database, containing only relevant data to the one extracted for the associated study. As for the strcture of the dataset, each row represents an apparition of a term in plain text, and the columns contain additional information.

- reportdate: date of the report where the term appears up.

- admission_days: strctured data days in hospitalization, if any.

- patient_id: anonymized patient identifier

- sex: patient sex (1=male, 2=female)

-birthdate: patient birthdate

- service: Service where the report was generated

- report_type: Type of report generated (either a discharge report, note, etc)

- record: unique identifier for the report itself

- term: term (that was read (NLP takes synonims and acronyms into account)

- exitus: medical exitus, if available in structured data, it could also be found in plan text in the previous column.