Selection of periods studied
Our analysis focused on two periods:
Period 1 focused on calendar weeks 5 to 8 of the years 2017, 2018, 2019, 2020.
Period 1 included four classic winter weeks in France and excluded school holidays and was considered our reference period.
Period 2, centred on calendar weeks 12 to 15 of the years 2017, 2018, 2019, 2020 which included the first four weeks of lockdown decided by the French government in 2020.
Selection of the centres studied
Our 59 French hospital teleradiology partners were included for all the study periods.
In order to remain within each centre’s scope of services, only the centres who were partners since week 5 of each year were taken into account. The number of centres for each period was collected.
For each centre, we identified its geographic area and classified the regions according to Public Health France map reflecting mortality during the pandemic.
Selection of sectors studied
Our teleradiology structure has an exclusive emergency imaging activity. The ITIS teleradiology tool (Deeplink medical, Lyon, France) classifies parts of our activities into clinical sectors. Since our activity began, some of these sectors have been documented and monitored, as they serve as indicators for the support team and partner centres [10]. These included cerebrovascular accidents (CVA), cranial trauma (CTr), and body-CTs(BC) for multiple trauma and acute non-traumatic abdominal pain (ANTAP) courses.
Except for those with traumatic abdominal pain, all the patients referred in these four sectors over the selected periods were included.
Examinations classified as being pathological were based on the following criteria:
- CVAs: recent ischemic or hemorrhagic brain injury, significant supra-aortic arterial lesion (stenosis > 50%, dissection, occlusion) or intracranial vascular abnormality (stenosis, occlusion).
- BC: one or more post-traumatic lesions.
- CTr: intracranial bleeding or craniofacial fracture.
- Acute non-traumatic abdominal pain: diagnosis of acute appendicitis, sigmoiditis or renal colic.
The anatomical regions of all performed examinations can also be analysed (skull, thorax, abdomen, and pelvis): particular attention was paid to examinations of the "chest region.
Variables studied
All the data used in this study came from the teleradiological database, and all patient anonymity was guaranteed.
The total number of examinations per period was collected (as well as the average number of examinations per centre) overall and for each sector as well as for the “chest" anatomical region. The percentage of examinations in each sector was calculated, as well as the percentage of positivity (pathological results for each sector).
The age, sex, patient’s region of origin, were collected globally and in the four sectors (except for the anatomical region “chest” where this data collection was global).
The evolution rates in exam requests between time periods were compared between different subgroups by region.
In Figure 1, regions have been represented according to mortality due to COVID-19 between March 1 and June 8, 2020, and the discretization for the color classes of the regions was done according to the Jenks Index.
The Jenks Index synthesizes the homogeneity of values within each class, and the heterogeneity of classes between them. Partner centres are also shown.
Statistical analysis
Statistical analyses were carried out using SPSS version 20 software. The data were compared between years, and according to the pre- and per-containment periods.
The significance level adopted was 5%.
Our study was approved by the Ethics Committee of the College of French Radiology Academics (IRB CRM-2007-108).