Seroprevalence of anti-SARS-CoV-2 antibodies among staff at primary healthcare institutions in Prishtina

COVID-19 is a novel viral disease caused by SARS-CoV-2 originating from Wuhan, China in December 2019 [1]. This infectious disease was named COVID-19 by the World Health Organization at the beginning of last year. At the time of writing, more than 315 million cases and more than 5.5 million deaths around the world have been confirmed [2]. SARS-CoV-2 rapidly spread worldwide, which continues to pose a major challenge and is an ongoing threat for public health and healthcare systems. In several countries, the demand for medical care exceeds the available resources, requiring stakeholders to reorganize the medical landscape [3].

During pandemics, everyone can potentially be infected, but healthcare workers (HCWs) are at a particularly high risk of infection, due to direct and indirect contact with patients [4]. As SARS-CoV-2 is transmitted via airborne droplet infection and indirect contact with COVID-19 patients, all kinds of HCWs are at high risk of infection [5]. Healthcare staff face numerous challenges, including: increased workload created by outbreaks, fear of contagion for themselves and their families, working with new and frequently changing protocols and personal protective equipment (PPE), caring for patients who are very sick and quickly deteriorating, and caring for colleagues who become ill [6]. Many studies have examined seroprevalence among HCWs. For example, some studies from different countries have reported low seroprevalence of COVID-19 in HCWs who work in pediatric hospitals [7, 8], tertiary hospitals [9, 10], university or academic hospitals [11, 12] and emergency HCWs [13]. Some studies show that measurement of seroprevalence in SARS-CoV-2 in frontline HCWs can give highly varied results based on the country or region of testing, healthcare role and when testing was carried out [14‐16].

Kosovo was among the last countries in the region and Europe to be hit by the coronavirus pandemic, with the first cases confirmed on 13th March 2020 [17]. Prishtina, as a capital city and a more populated area, has recorded more cases of the virus than other cities [18]. As a result, the frontline health workforce has experienced a high workload during the pandemic, along with multiple psychosocial stressors, which may affect their mental and emotional health, leading to burnout symptoms [19].

SARS-CoV-2 infection is followed by an antibody response with IgG and IgM antibodies, and therefore serological tests can provide more information on SARS-CoV-2. Antibody tests are potentially useful for detecting previous infection when measured 15 or more days after the onset of symptoms, but the duration of elevated antibody levels is currently unknown [20].

In Kosovo, primary healthcare facilities are the first contact for most patients. Given their continuous exposure to the virus, extensive and continuous testing of primary HCWs is a necessity. Kosovo has not been an exception to this requirement, as the pandemic has posed considerable danger to HCWs, particularly primary HCWs. This is the first study to date examining seroprevalence in primary HCWs in Kosovo. Studies elsewhere in the world have reported 2.6% [21] and 9.17% [22] seropositivity of COVID-19 antibodies among primary HCWs. This study aims to determine the prevalence of positive anti-SARS-CoV-2 antibodies among primary HCWs in the municipality of Prishtina, and its association with different demographic, epidemiological factors and health behaviors.

Among HCWs, 83 (17.47%) were seropositive for SARS-CoV-2 antibodies IgG or IgM. A total of 189 HCWs (39.79%) reported that they had been previously diagnosed with COVID-19 by a nose swab PCR test. There were overall 231 HCWs who either reported previous PCR diagnosis, or tested positive for serological markers IgG or IgM, which was almost half of the study sample (48.63%, Table 1). Several factors are associated with decreased and increased odds for such outcomes.

Our result of 17.47% seroprevalence is comparable to the range of percentage reported in other similar studies [23‐26]. Interestingly, the majority of HCWs who reported having previously tested positive by PCR tests did not test positive for either antibody. Inbaraj et al. reported roughly 7 undetected infections for every PCR confirmed case [27], which is the opposite of the trend we have observed. It is possible that this represents the extensive and thorough testing of HCWs in Prishtina. Alkurt et al. reported that only 78.2% of PCR positive patients had IgG antibodies, and that IgG titres of asymptomatic PCR-positive patients were significantly lower than symptomatic patients. Notably, these patients were tested 52.8 ± 11.6 days after infection. Bendavid’s study in California also showed that the seroprevalence of subjects 2 weeks after symptoms was considerably higher than seroprevalence 2 months after symptoms [28]. It is possible, then, that the low amount of positive antibody tests in comparison with reported PCR tests is due to antibody loss, or to mild or asymptomatic cases among some HCWs. Although it would not account for the full extent of the variation found here, systematic reviews of serological tests have shown considerable variation between accuracy of serological tests, and lower accuracy for serological tests [11, 29, 30].

The study design and the general structure of the study were based on previously published studies [14, 31, 32]. The content of the questionnaire was based on existing studies published and on consultations with healthcare experts, public health experts and staff of the Main Family Medical Centre of Prishtina. The questionnaire was compiled, tested, and revised before data collection. Key healthcare experts and medical staff from the Main Family Medical Centre of Prishtina gave their input on the study design, questionnaire, and overall progress of the study. The main limitation of the study is the relatively small sample size.

Our study is the first study about COVID-19 seroprevalence in HCWs working in Kosovo. Another study by our group has examined the seroprevalence of anti-SARS-CoV-2 antibodies in municipal workers in Prishtina municipality. Out of 418 municpal workers surveyed, we found that 21.1% were seropositive for either IgG or IgM, of which 9.6% were positive for IgM and 19.4% for IgG [33]. There are some studies that previously have examined seroprevalence among HCWs in primary healthcare. In Spain, Barallat et al. found 9.17% of IgG seropositivity in HCWs [22]. Compared to results of other previous studies, in Essen, Germany 1.6% and New York, USA 33% our study stands in the middle of seropositivity among HCWs [34, 35]. HCWs from emergency departments reportedly have a lower percentage of seropositivity (5.9%) [36]. In contrast with our study, Airoldi et al. [37] found that female HCWs have higher odds of seropositivity OR 1.29, 95% CI (0.92 to 1.80), but other studies have shown, like our study, that male HCWs have a higher ORs [26, 38]. Similar to our study, studies have confirmed the increased odds of seropositivity in physicians compared to nurses and other HCWs [38]. Studies have also reported higher odds of seroprevalence for smokers, but without a significant relationship [39].

Our results show that around a fifth of HCWs in Prishtina Municipality have antibodies for COVID-19, and just under half have been infected. This means that there is potential for most HCWs to be infected, which could lead to further spread of COVID and more hospitalization, which again can lead to more risk for HCWs. Considering that newly qualified Kosovar physicians are often going outside Kosovo [40], healthcare staff resources are already limited, and the need for ample health professionals for patients to have sufficient regular care means that protective measures for health professionals are needed. This should alert our policy makers and trigger them to speed up the process vaccinating more people, particularly HCWs, and increase attempts to protect HCWs from burnout.

Our study did not identify a clear link between adherence to protective measures by family, wearing of masks by coworkers, and seroprevalence or odds of diagnosis by PCR. Systematic reviews have suggested that masks have a protective effect on healthcare workers, and for the public in general [41], but higher certainty evidence is still needed [42]. Similarly, the effect of other protective measures has been examined [4, 43], but the effect of family adherence to these measures has not been considered in detail.

In conclusion, we found that several factors increase the odds of IgG and IgM seropositivity. We also found that HCWs who respect basic infection-control measures have lower odds of being seropositive on IgG and IgM, although the difference was not significant. This could be due to collinearity, which was not detected when we ran tests. Further research could clarify the relationship between family adherence to protective measures and usage of masks by staff, and seroprevalence. Our study results could assist policy makers in Kosovo to continue to raise awareness about coronavirus and its health consequences [44]. Further testing in the future would allow observation of how seroprevalence changes over time.