Association of known SARS-CoV-2 serostatus and adherence to personal protection measures and the impact of personal protective measures on seropositivity in a population-based cross-sectional study (MuSPAD) in Germany

In Germany, the first SARS-CoV-2 case was registered in January 2020 [1]. The seroprevalence measured by antibodies against Spike antigens of SARS-CoV-2 in the unvaccinated population in Germany was estimated from July-December 2020 to be 1.3%–2.8% and increased in February-May 2021 to 4.1–13.1% [2].

During the pandemic and in the absence of vaccines or effective medication within the first 12 months, personal protection measures (PPM), such as physical distancing, use of face masks, adequate indoor ventilation, avoiding crowded indoor spaces, hand hygiene, and cleaning the environment were important protective measures [3, 4]. In Germany as of March 2020 lockdowns, curfews and contact bans were imposed, and various facilities, such as schools were closed and events canceled [5, 6]. The requirement of wearing face masks in public transport and for shopping was implemented on April 29, 2020, followed by lockdowns lasting from November 2020 until March 2021 [5, 6]. The vaccination campaign started at the end of December 2020 [7]. The first policy measures were eased at the beginning of March 2021, with regional differences in the opening of stores with strict hygiene measures, and citizens could get tested free of charge at test centers [5, 6].

Wearing masks, performing social distancing, having close contact for less than 15 min, and frequent hand washing were independently associated with a lower risk of a positive polymerase chain reaction (PCR) test after having had contact to a SARS-CoV-2 index case [8]. Reduced incidences of COVID-19 were associated with increased handwashing, face mask wearing, and physical distancing [9]. In addition, a combination of social distancing, testing for SARS-CoV-2, contact tracing, and quarantine could lower pressure on the health care systems [10].

In Germany, the number of newly reported infections with SARS-CoV-2 decreased between 15 to 75%, depending on the region, after the introduction of mandatory face masks, and the overall daily growth rate of reported infections decreased by 47% [11]. The requirement to wear masks in educational institutions reduced the number of infections both within and outside educational settings [11, 12].

Many of the studies performed were of ecological nature and assessed the association between recommendations or mandates and population-level infection indicators [11, 13]. The relationship between individual compliance with PPM and its impact on SARS-CoV-2 serostatus remains unclear in real-life settings. Additionally, how PPM are related to individual knowledge of vaccination and prior infection has not been assessed in larger studies – which could generate transferable knowledge to improve parametrization of future infectious disease models.

This is why we assessed participants’ behavior in the population-based study MuSPAD. The primary aim was to estimate the effect of a reported positive PCR test of SARS-CoV-2 (since the onset of the pandemic) or vaccination against SARS-CoV-2 on specific self-reported PPM behavior. The secondary aim was to evaluate the effect of self-reported adherence with PPM on seropositivity of SARS-CoV-2.

Among MuSPAD participants in Germany the adherence to personal protection measures was rather high. Participants who were vaccinated against SARS-CoV-2 were older and had the highest compliance with physical distancing as well as hand cleaning. Those participants without an immune response against SARS-CoV-2 showed the highest percentage of additional face mask wearing. Our data indicated that self-reported positive SARS-CoV-2 PCR test since the onset of the pandemic or vaccination against SARS-CoV-2 decreased the odds of adhering more strictly to hand cleaning and physical distancing, but increased the odds of wearing face masks on mandatory occasions. In addition, we found that among unvaccinated participants, a higher adherence to PPM reduced the odds of being seropositive for SARS-CoV-2.

Overall, a high adherence to hand cleaning (67%) and additional face mask wearing (87%) was reported by the study population. In contrast, MuSPAD participants complied more partially (63%) with physical distancing. A high acceptance of PPM was also reported from an online survey (N = 9,796) conducted in March 2020 in various countries [26]. Researchers concluded that the prevention measures implemented by the government, for example avoiding gatherings, closure of public places, and hand hygiene were accepted by 95.0%–99.7% of the respondents [26].

Previous research has shown that being female, of older age, part of a risk group, suffering from impaired physical and mental health, having a positive perception of government communications, and mostly a higher socio-economic status were characteristics positively associated with acceptance of PPM [27‐30]. In our study, we found that vaccinated participants were older and showed the highest compliance with hand cleaning and physical distancing. Further, we found that a known serostatus of antibodies against SARS-CoV-2 due to a self-reported positive PCR test or vaccination might be associated with adherence to PPM. The odds of not adhering to physical distancing were generally increased in vaccinated participants compared to those unvaccinated or not having a positive PCR test. A Study has shown that moods, such as anxiety, can enhance preventive behavior toward a SARS-CoV-2 infection [31]. Thus, a known protection with antibodies against SARS-CoV-2 could improve individuals’ moods and conversely affect compliance with PPM. Moods were not investigated in our analysis, but can be an unobserved factor, which needs to be considered in future studies.

We found that vaccinated participants and those with a known prior infection had increased odds of compliance regarding additional face mask wearing. In both exposures the odds were increased (vaccinated: OR: 1.28; self-reported positive PCR test OR: 1.13).

The extent to which PPM behavior was associated with SARS-CoV-2 seropositivity among unvaccinated MuSPAD participants was subject of the second research question. Results indicated that keeping physical distancing, hand cleaning, and wearing face masks on mandatory occasions decreased the odds of being identified as seropositive. Point estimates of the exposures of physical distancing (adhering partly OR: 0.78; adhering completely: OR: 0.84), hand cleaning (adhering partly: OR: 1.03; adhering completely: OR: 0.97), and additional face mask wearing (OR: 0.91) on seropositivity were shown to be protective. Other studies confirmed that PPM reduced the transmissions of SARS-CoV-2 [32‐34]. A systematic review of 25,697 participants showed that the transmission of SARS-CoV-2 was reduced when keeping physical distance of more than 1 m [pooled adjusted OR 0.18 (95% CI: 0.09–0.38)]. Wearing face masks in addition to appropriate physical distancing can reduce SARS-CoV-2 infection by up to 85% [OR: 0.15 (95% CI: 0.07–0.34)], with a difference between respirators and disposable surgical or similar masks [33].

This study provided a presentation of the behavioral patterns of MuSPAD participants across Germany, based on a population level, rather than focusing only on hotspots. Data collection allowed depicting PPM adherence across Germany and at different times. Another strength was the measurement of SARS-CoV-2 serostatus using multiplex immunoassay to distinguish between vaccinated, recovered, and individuals without antibodies against SARS-CoV-2. Compared with commercial assays, this test already demonstrated its improved sensitivity [16] in comparison to case identification based on (self-reported) PCR tests used in previous studies [8]. To capture relevant confounding, adjustment sets were derived from expertise, visualized with DAGs, and evaluated jointly in the research group.

Due to the sequential cross-sectional nature of the MuSPAD, the temporal sequence of the relationship between exposure and outcome is uncertain. In addition, response bias due to self-report questionnaires may result in overestimation [35]. Due to mandatory mask wearing, social desirability is an important factor that may influence our estimates [36], too. The Health and Containment Index to control for restrictions in Germany [25] was only available on a country and not on a communal level, limiting the accuracy of adjustment.

Another aspect that needs to be discussed is that the large overall sample size of our study may result in confidence intervals being very narrow, suggesting high precision in the effect estimates, resulting in more statistically significant findings. Therefore, in the interpretation of our findings we focused on precision estimated based on confidence intervals.

In addition, we would like to point out that for each outcome and exposure, a specific regression model was built to adequately control for confounding factors for each research question. Therefore, there is a possibility that some of our results are false positives due to multiple testing. In our questionnaire, compliance was measured in three categories (“No, not at all,” “Yes, partly,” “Yes, completely”). Therefore misclassification might have occurred, as adherence to PPM could not be captured accurately and thus differences between individuals could not be validly measured. In our categorization of vaccinated and infected due to serological status we did not account for those who were both, which potentially overestimated the association between personal protective measures and infections according to serological status. However for the time period assessed here this misclassification cannot be large as the proportion of those infected before vaccination was small.

A prior known infection of SARS-CoV-2 confirmed with a positive PCR test and vaccination against SARS-CoV-2 may have influenced compliance with certain PPM behaviors during the pandemic in Germany. Furthermore, similar to previous research, we confirmed that compliance to PPM can reduce the odds of SARS-CoV-2 seropositivity in unvaccinated participants.