2. Materials and Methods
In this study, we used the clinical information provided in the COVID-19 protocols of the largest university hospitals in Andalusia. In Andalusia, hospitals are stratified into 4 categories according to service area and number of medical and surgical specialties [6]: (1) regional university hospitals (referral hospital in the whole autonomous community with all specialties available); (2) specialized university hospitals (referral hospital in a province and with wide number of specialties available); (3) basic general hospitals (referral hospital in a region and with all basic specialties available); (4) highly-specialized hospitals (referral hospital in a local area and with all basic specialties available). We included 6 regional university hospitals and 9 specialized university hospitals: Hospital Universitario Reina Sofía (Córdoba), Hospital Universitario Juan Ramón Jiménez (Huelva), Hospital Universitario Virgen del Rocío (Sevilla), Hospital Universitario Virgen Macarena (Sevilla), Hospital Virgen de Valme (Sevilla), Complejo Hospitalario Torrecárdenas (Almería), Hospital Universitario de Jerez de la Frontera (Jerez de la Frontera), Hospital Universitario Puerta del Mar (Cádiz), Hospital de Puerto Real (Puerto Real), Hospital Universitario Virgen de las Nieves (Granada), Hospital Universitario San Cecilio (Granada), Complejo Hospitalario de Jaén (Jaén), Hospital Universitario Virgen de la Victoria (Málaga), Hospital Costa del Sol (Marbella), and Hospital Regional Universitario de Málaga (Málaga). This study was approved by the Institutional Research Ethics Committee of Málaga on 1 May 2020 (Ethical code: PI-Prot-0520)
We excluded 19 basic general hospitals and 15 highly-specialized hospitals. Normally, the treatment recommendations adopted in the reference hospital indicate how their area of influence proceeds.
Antiviral treatment, empirical antibacterial agents, adjunctive therapies, anticoagulant treatment, supportive care (oxygen, intravenous fluids, monitoring, high-flow nasal oxygen/noninvasive ventilation, and mechanical ventilation, among others), dosage, method of administration and duration of treatment, drug side effects and interactions, nonrecommended treatment or recommendation for other drugs, hospital circuits, ward organization and care planning, and discharge recommendations were collected from all clinical protocols for COVID-19.
COVID-19 protocols contained patient groups according to the patient’s clinical condition: mild upper respiratory illness with and without clinical risk factors (aged ≥60 years old and comorbidities), mild pneumonia (defined as CURB-65 Severity Score [7] ≤1 or Pneumonia Severity Index [8] I or II, basal oxygen saturation >95%, respiration rate at rest <22, unilateral opacity, D-dimer <1000 ng/mL (normal range: 220–500 ng/mL), ferritin <1000 mcg/L (normal range: 8–252 mcg/L), interleukin-6 <40 pg/mL (normal range: <4.4 pg/mL)), moderate pneumonia (defined as CURB-65 Severity Score ≥2 or Pneumonia Severity Index ≥III, basal oxygen saturation >92–95%, respiration rate at rest 22–30, bilateral opacities, D-dimer >1000 ng/mL, ferritin >1000 mcg/L, interleukin-6 <40 pg/mL), and severe pneumonia (defined as sepsis or shock with acute respiratory distress syndrome (ARDS), basal oxygen saturation <92%, respiration rate at rest >30, bilateral opacities, D-dimer >3000 ng/mL, ferritin >1000 mcg/L, interleukin-6 >40 pg/mL, elevated troponin I). Prophylaxis in suspected cases was not taken into consideration.
All protocols included were obtained from COVID-19 teams of each hospital and were the latest updates as of July 2020. Protocols are only available for healthcare providers on previous request to COVID-19 teams of each hospital. The update of protocols was routinely made by COVID-19 teams when they considered necessary.
All variables analyzed were categorical and are shown as the absolute value and percentage, respectively. Statistical analyses were performed using SPSS Statistics for Windows, version 15.0(IBM, Armonk, NY, USA).