This study is the first focusing primarily on PES presentations BIBP during the COVID-19 pandemic. The current study is also the first showing an absolute increase in PES presentations BIBP during the second wave compared to a control period one year earlier. We have shown the same for the first wave (based on a part of the current study’s sample [17]. What is more, this is the first study showing that COVID-19-period itself was a predictor for PES presentations BIBP, during the first and during the second wave compared to their control periods (Table 3).

One study from Taiwan on PES presentations during the COVID-19 pandemic did show an increase in police/emergency medical service presentations in 2021 but not in 2020, compared to pre-pandemic times [14]. Unfortunately, police and emergency medical service referrals were not reported separately which makes it difficult to compare to our study. Studies from Switzerland, Turkey and Australia cover only the first wave of the COVID-19 pandemic and report, in comparison to the current study, considerably lower rates of presentations BIBP and only minor changes to pre-pandemic times [19,20,21]. The current study’s rates are more in line with pre-pandemic studies [3]. One may assume that rates of presentations BIBP may differ between PES in more urban areas and those in more rural areas (although there is no scientific evidence for this assumption). When comparing the mentioned and the current study, however, all studies concern metropolitan areas. Furthermore, the differences in rates of presentations BIBP in PES in between different sites and countries are rather important. These differences may be due to country-bound differences such as different mental health care policies and police responsibilities on the one hand [2] and due to local-bound differences such as sociodemographic differences and differences in clinical characteristics on the other hand [6, 17]. More research is necessary to better understand the factors of influence of presentations BIBP to PES.

In the logistic regression analysis, COVID-19-associated effects were seen during the first and the second wave with almost identical odd’s ratios (1.435 and 1.458, respectively). These findings suggest that both waves similarly increased the probability of presentations BIBP to a PES in Berlin. As the entry of interaction effects (i.e. for ex. interactions between the presence of a specific psychiatric diagnosis and COVID-19-period) did not further add to improve the regression models (Table 4), we are not able to pinpoint a specific patient characteristic explaining the increase of presentations BIBP during the COVID-19 waves. This suggests that the COVID-19-period effect on presentations BIBP that we saw in our sample is rather complex and not mono-causal. Potentially, the explanatory factors may also differ between the two observed waves.

In comparison to other diagnostic groups, patients with schizophrenia spectrum/psychotic disorders are more likely to be BIBP (Table 3). Especially during the first wave, outpatient facilities were less available [14,15,16, 22, 23] with limited accessibility of many psychosocial [22] and psychotherapeutic [24] facilities. One may hypothesize that patients with chronic psychotic disorders and high need of psychosocial facilities might have suffered particularly from these constraints with exacerbation as a consequence. This view is supported by the fact that many studies show an increase in PES presentations of patients with psychotic disorders during the pandemic [13, 25,26,27,28,29].

In Tables 1 and 2 one can appreciate that inpatient admission and involuntary admissions are highly associated with patients BIBP, a finding that is highly plausible and has been reported earlier [30].

Independently of COVID-19, the following factors predicted presentations BIBP in all observation periods of our study: lower age, aggressive behavior towards others, and schizophrenia sprectrum/psychotic disorders(cf Table 3). Patients with depressive disorders were less likely to be BIBP. In the rather underpowered American studies (ca. 100 patients BIBP per study), age was not shown to be a predictor of presentations BIBP [5]. In more large-scale studies, such as Wang et al. from Taiwan (> 3000 patients BIBP), however, the group of patients between 30 and 39 years old were the most at risk of being BIBP [6]. This is in line with our findings. Aggressive behavior towards others has also been shown several times to be associated with presentations BIBP [3,4,5,6]. Psychotic disorders have earlier been reported in two small studies as potentially associated with police referrals [4, 31]. During the second wave and its control period, the presence of a substance use disorder, a suicide attempt prior to the presentation and male gender are predictors of patients BIBP. Both, the higher risk of being BIBP in patients with substance use disorder and the positive association of patients BIBP to a PES with suicide attempts prior to the presentation, are findings that were earlier reported in Taiwan but not yet in Western countries [6]. As depressive patients do rather often present with suicidal thoughts and after suicide attempts, the coincidental negative association with depressive disorders in patients BIBP might seem in the first place contradictory. However, these findings are in line with results of a meta-analysis conducted by Walker et al. in 2021 when we equalize BIBP and involuntary admission. They found that young patients with a primary diagnosis of affective disorders were significantly less at risk of involuntary admission when compared to patients perceived to be at risk of self-harm (including suicidal ideation or suicide attempts) (OR 2.05, p = 0.015) [32]. The same meta-analysis shows that young patients with substance use disorder were more likely to be admitted involuntarily (OR 1.87, p = 0.032) as well as patients who showed behavior of harm to others (e.g. aggression, violent acts) (OR 2.37, p = 0.002), which is also in line with the current study’s findings.

Male gender as predictor for presentations BIBP is a common finding in the literature [3,4,5,6, 33].

This study is the first focusing primarily on PES presentations BIBP during the COVID-19 pandemic. The current study covers a relatively long observation period with a comparably large number of assessed PES presentations. Indicators of mental health were based on clinical diagnoses rather than self-reports. In addition, we performed a detailed clinician-led review of each case, based on thorough clinical documentation.

The following limitations need to be considered: the control data is limited to the previous year only. The study is based on clinical routine data which can differ in quality and extent which may introduce bias. We cannot completely rule out the possibility of an interrater bias. However, to limit this bias we implemented the following measures: consulting all available data and scheduling regular meetings to discuss pressing questions, resolving them in consensus.

A further limitation is that we only gathered information about patients BIBP in a single-center psychiatric emergency department. Extrapolation of results should therefore be done with caution.