To the Editor—The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged in China at the end of 2019 and spread worldwide, leading to the current pandemic. During the first pandemic wave in France, the northeastern region was one of the first and most affected areas. Thus, healthcare workers (HCWs) in the Strasbourg University Hospitals (SUH) were early and intensively exposed to SARS-CoV-2. This exposure varied between hospital departments, notably at the beginning of the pandemic. Some departments were rapidly dedicated to the care of SARS-CoV-2 patients, with earlier implementation and use of personal protective equipment (PPE). In this study, we evaluated SARS-CoV-2 seroprevalence in hospital staff according to department in SUH.

From June 22, 2020, to November 1, 2020, SARS-CoV-2 serology was offered to all hospital staff at SUH. Each participant completed a questionnaire collecting data about their occupational department and the onset of symptoms (ie, type of symptoms and delay expressed in weeks between onset and serum sampling time). Serum samples were tested using an immunochromatographic lateral flow assay (Biosynex COVID-19 BSS, Switzerland, Fribourg) detecting IgM and IgG directed against the receptor binding domain of SARS-CoV-2 spike protein (Supplementary Fig. 1 online). This assay was reported to have overall estimated sensitivity of 93% and clinical specificity of 99%. ^{Reference Velay, Gallais and Benotmane1}

Univariable and multivariable analyses were performed using R version 4.0.4 software (R Foundation for Statistical Computing, Vienna, Austria). Statistical significance was set at P < .05. Hospital departments were grouped together according to staff exposure risk for multivariable analysis (see Supplementary Material online for statistical analysis detail). This study was approved by the local ethics review committee (record CE-2021-79).

In total, 5,694 HCWs participated in this study. Participant characteristics and serological results of the cohort are presented in Supplementary Table 1 (online). The mean age was 40.4 years (SD, ±11.7). In total, 482 participants had positive serology (8.5%) and among them, 422 participants answered about the onset of symptoms, including 48% of asymptomatic subjects (Supplementary Fig. 2 and Supplementary Table 1 online). The seropositive and symptomatic subjects predominantly reported anosmia (29.2 %) and agueusia (27.2%).

The departments in the “geriatric group” showed the highest seroprevalence rate (27.7%) (Supplementary Table 2 online) and a strong association with seropositive status (OR, 5.3; 95% CI, 3.5–8.2; P < .05) (Supplementary Table 3 online and Fig. 1). Seroprevalence rates in the frontline departments were all lower than in the geriatric group: internal medicine department (which contains the infectious disease department, 15%), department of thoracic pathology (11.6%), anesthesia and reanimation department (7.9%), and emergency departments (8.1%).

Fig. 1. Odds ratio of SARS-CoV-2 seroprevalence according to a history of anosmia and/or agueusia and the type of hospital department. Data are depicted as odds ratios (black or grey symbols) and 95% confidence intervals (bars on either side), and P values are mentioned for each criterion.

However, the seroprevalence rate of 8.5% was concordant with those reported in the literature in Europe, ^{Reference Galanis, Vraka, Fragkou, Bilali and Kaitelidou2} and we found a high rate of asymptomatic infections (48%). Remarkably, working in a geriatric department was associated with positive serology. The high rate of seroconversion among staff in the geriatric department contrasted with low seroconversion rates among staff in departments on the frontline from the start of the pandemic.

Many contributing factors could explain this high seroprevalence. First, the onset of 2 clusters of contamination in geriatric departments, from the beginning of March 2020, led to contamination of caregivers. At the onset of the pandemic, no guidelines were in place for systematically wearing a surgical mask (ie, this obligation was implemented on March 23, 2020). Second, even when the wearing a mask became compulsory for caregivers and patients, implementation was laborious, especially among patients with cognitive disorders. Some of these fragile patients do not stay in their rooms but walk in the corridors, thus favoring virus circulation. Moreover, patients in geriatric departments often complain of discomfort when wearing a mask. Third, elderly COVID-19 patients were frequently admitted to geriatric departments (up to 50% of beds during the first wave in SUH). Cases of nosocomial transmission were reported probably related to the type of patient care (eg, repeated nursing and grooming). Another difficulty was the asymptomatic or atypical forms of the disease, ^{Reference Gan, Kho and Akhunbay-Fudge3} which were not well known at the beginning of the pandemic, possibly leading to underdiagnosis. The advanced age of the patients also led to prolonged shedding of the virus. ^{Reference Meyer, Meyer and Calabrese4}

Despite the implementation of protective measures for caregivers, the number of clusters remains high in geriatric units, in nursing homes, as well as in long-term care facilities, ^{Reference Martinot, Carnein, Kempf, Gantner, Gallais and Fafi-Kremer5,Reference Zollner-Schwetz, König, Krause, Pux, Laubreiter and Schippinger6} leading to substantial mortality rates. ^{Reference Meyer, Calabrese and Meyer7} The recurrence of these clusters paralleled the arrival of new variants of SARS-CoV-2. Considering the higher risk of being exposed to SARS-CoV-2 when working in a geriatric department, vaccination of staff working in these services is a priority.