Datos solicitados: Historial clínico de pacientes tratados por COVID-19 (en formato electrónico, e incluyendo datos demográficos, comorbilidades, tratamiento, síntomas, pruebas de laboratorio)
Objetivo del proyecto: Desarrollo de algoritmos personalizados de estimación de riesgo.
Name: Agni Orfanoudaki
Institution: Massachusetts Institute of Technology, Cambridge, MA, USA
Web: Covid Analytics
Description of project:
Our objective is to utilize the totality of available Electronic Health Records in order to create and validate comprehensive, accurate, and interpretable models for mortality, intubation, and ICU admission. These models will allow physicians at the forefront of the disease to quickly identify which patients would benefit most from using the limited resources available. The data leveraged in this study will come from various sources from both US and European hospitals.
Terms of collaboration:
The requester needs little assistance besides the transfer of data. They will be happy to include up to 2-3 members of the collaborating institution in any eventual publication.
Detailed description of data:
A data set comprising records from patients admitted with COVID-19 infection across multiple hospitals will allow for an accurate and reliable prediction model for patient death and recovery/hospital discharge. This is a retrospective study. We do not have an upper bound on the number of patients we need. The higher the number, the better.
Record inclusion criteria
Records of patients meeting all of the following criteria will be included:
• All genders
• All races/ethnicities
• Age ≥12, with no upper age limit (note: while HH generally sees patients ≥18 years old, this “extended” age range allows for patients, such as pregnant minors, who might be admitted)
• Confirmed or presumed positive for COVID-19 infection
• January 1, 2020 – December 31, 2020, inclusive (at current time, an exact end date cannot be known)
Record exclusion criteria:
Records of patients meeting any of the following criteria will be excluded:
• Age <12 years old
• presumed positive for COVID-19 infection but diagnosed/confirmed COVID-19 negative after admission
• admitted after having recovered from earlier COVID-19 infection
Data Use, Collection:
All data will be extracted from the electronic medical record. Data will comprise demographic (e.g., age, gender and race/ethnicity) and clinical variables:
• blood pressure(s) and other vital signs
• cardiac status [ejection fraction (EF%), QTc and Fridericia-corrected QTc (ms), echo parameters]
• co-morbidities and past medical history (including, but not limited to, diabetes, cardiac history, lung/pulmonary history, immunosuppression therapy, cancer)
• current medications, COVID-19 treatment administered
• blood test results
• disease history (days from onset to hospitalization, symptoms)
• radiographic findings
• respiratory status [rate(s), SaO2/PaO2]
• complications and outcomes
Handling of personal/confidential information:
All the data we will receive will have been de-identified by the collaborating institutions. No attempts will be made to identify individual patients, hospitals or physicians. We will comply with the GDPR legislation as indicated by the COUHES guidelines. All data will be stored in an encrypted format at the MIT network and no local copies will be stored at the team’s personal computers. A full handling protocol is available.
Status of ethical approval:
The requester’s project is already approved by an ethical committee at MIT. The requester is willing to work with the data provider to seek further ethical approval at their institution.
Nota: En caso necesario, Crowdfight COVID-19 puede proporcionar ayuda para el procesamiento, preparación o envío de los datos. Esto incluye asesoramiento técnico, apoyo de expertos en bases de datos, o voluntarios que realicen tareas manuales (organización de datos, etc.).