Impact of COVID-19 lockdown on suicide attempts

Our retrospective data analysis, which was approved by the local ethics committee (EK no: 1607/2020), focused on all patients (1) with a minimum age of 18 years, (2) who were admitted to the level 1 Trauma Centre of the Medical University of Vienna during the lockdown period established by the Austrian government (16 March–15 May 2020), and (3) for whom treatment in the trauma resuscitation room (shock room) was considered appropriate due to the injury mechanism. These patients were subdivided into trauma victims and individuals who had attempted suicide. Of the latter, demographic data including age, gender, outcome (survival/fatality), ethnic minority background, known substance abuse, known psychiatric diagnosis, previous suicide attempt, previous psychiatric treatment, abbreviated injury scale (AIS) for the regions head or neck (AIS_Head), face (AIS_Face), thorax (AIS_Thorax), abdominal or pelvic contents (AIS_Abdomen), extremities or pelvic girdle (AIS_Extremities), and external (AIS_External), as well as injury severity score (ISS) and zip code of residence district were extracted from patients’ medical charts. Of our study group, a special focus was set on polytrauma victims, defined by an ISS ≥ 16. In order to be able to form a valid control group all patients who were admitted to the level 1 Trauma Centre of the Medical University of Vienna following a suicide attempt from 16 March to 15 May in the years 2015–2019, were assessed. In general, survival referred to the period of hospitalization.

All patient records were pseudoanonymized and deidentified prior to statistical analysis that was performed using the SPSS system (IBM, Armonk, NY, USA, version 26). Normally distributed variables are presented as mean ± standard deviation, non-normally distributed variables and categorical data are characterized by median and range in square brackets, and dichotomous characteristics are displayed by absolute and relative frequencies. The Mann-Whitney U-test was applied to compare continuous variables, whereas categorical data were analysed by means of the χ²-test. In general, a p-value < 0.05 was considered significant.

During the national COVID-19 lockdown in Austria we identified a higher percentage of suicide-associated admissions to the resuscitation room of our level 1 Trauma Centre compared to the same 2‑month time period in each of the previous 5 years. These differences did not necessarily indicate that there were more suicide attempts in 2020 as the number of patients admitted due to other reasons than suicide was lower than average. Moreover, it turned out that the year 2017 was not comparable to the other years 2015–2019, necessitating exclusion of the data for 2017 from our analysis. Since the number of patients admitted to the trauma resuscitation room due to reasons other than suicide markedly decreased during the lockdown, we replaced this number by the mean number of patients admitted in 2015, 2016, 2018, and 2019 due to reasons other than suicide. Even after this adaptation a higher percentage of attempted suicides (p = 0.001) was revealed in the study group compared to the control group; however, a higher percentage of patients suffering an intentionally caused polytrauma could not be revealed (p = 0.084).

Regarding the mechanism of injury, most attempted suicides were performed by jumping from a height during the COVID-19 lockdown, whereas the majority of patients in the control group had presented after trying to kill themselves through cutting. Finally, the ratio of males and females, age and overall injury severity as well as known risk factors for suicide, such as ethnic minority status, socioeconomic background, substance abuse, pre-existing psychiatric diagnosis, previous suicide attempt and previous psychiatric treatment, did not differ significantly between the study and the control group.

As the lockdown measures took effect and people were advised to “stay at home” we noticed a reduced number of admittances to the resuscitation room of our level 1 trauma centre following a traumatic injury, which can be explained by the reduced number of accidents generally occurring at the workplace, in traffic or at outdoor activities. While the number of patients admitted to our institution due to accidental traumatic injury decreased, the number of patients admitted following an intentional traumatic injury due to a suicide attempt increased. Therefore, the percentage of patients admitted following a suicide attempt during the COVID-19 lockdown was significantly higher compared to the previous 5 years.

The reasons that lead an individual to suicidal behavior are still not fully understood, as the causes are complex and multifactorial [27]. One of the reasons for the significantly higher percentage of suicide-associated admissions to our resuscitation room during the COVID-19 lockdown might be the increased prevalence of depression and anxiety symptoms in this period, compared to previous epidemiological data in Austria [28]. Interestingly, when Pieh et al. performed the same survey they conducted in Austria on a study population in the United Kingdom, they found that the prevalence of severe depressive, anxiety and insomnia symptoms was about three times higher there than in Austria [29]. These findings lead to the hypothesis that other countries might have experienced an even higher increase in the rate of suicide attempts than our Austrian study population.

A level 1 trauma centre is a verified facility that provides the highest level of surgical care to trauma patients and has a full range of specialists and equipment available 24 h per day. The Medical University of Vienna is the only level 1 Trauma Centre in the greater Vienna region with two resuscitation rooms, where patients can be treated simultaneously. Additionally, it is known that polytrauma patients have a significantly lower mortality when care is obtained at a trauma center [8]. Therefore, it stands to reason that our institution received an influx of major trauma and polytrauma patients during this time period.

The absolute number of suicide-related polytraumas as well as the proportion of suicide-related polytraumas in comparison to all admissions was higher in our study group than in the groups referring to 2015, 2016, 2018, and 2019. Surprisingly, we detected a spike of suicide-associated polytrauma patients admitted to our institution for the 2‑month period in 2017 and further investigated the reasons for this occurrence. During the spring of 2017 another major hospital in Vienna had partially closed its trauma center due to a change in infrastructure [30]. For this reason, data on polytrauma from 2017 are not comparable and were excluded from our analysis. We further studied if a difference in suicide-associated polytrauma admittance could be detected according to the week of admission. Although no suicide-associated polytrauma was treated during the first 2 weeks of the lockdown, no significant difference could be detected; however, this might also be a result of our small sample size. Moreover, we could not find a difference in the mechanism of injury between the study and control groups.

Risk factors associated with higher suicide rates, such as age, gender, substance misuse and mental health concerns, such as depression have been identified [21] and thus at-risk populations defined. It was hypothesized that the number of suicide attempts in the senior population could increase as a result of social isolation during lockdown [26]; however, our data showed no difference in the age distribution of patients who attempted suicide during the lockdown from those in the previous 4 years. Regarding gender, we detected that more males than females attempted suicide in both study and control groups, which is in line with previously published literature [7].

Due to the small number of patients in our study group we could not observe an increase in suicidal behavior isolated for individuals with previous mental health problems, but more generally an increase in suicidality across different groups at risk. Nevertheless, people suffering from psychiatric disorders or mental health conditions tend to be more susceptible to stress than the general population and could therefore experience relapses, new manifestations or substantial worsening of symptoms, such as depression, anxiety, and posttraumatic stress. It follows that all these factors could lead to increased psychological distress and thus lower the inhibition to perform a suicide attempt. Furthermore, many outpatient clinics were closed during the lockdown period and thus regular visits could not take place and the continuance of prescriptions was not always guaranteed, increasing the burden on patients already suffering from mental health symptoms even further [31].

The COVID-19 pandemic has become an amplifier of already existing socioeconomic vulnerabilities and its economic fallout might also lead to further healthcare threats. It is known that unemployed individuals have a higher suicide risk than their employed counterparts and that during spikes of unemployment rates a spike in the number of suicides can also be detected [32‐34]. Considering the current global unemployment rates and their projected rise, an increase in suicide attempts should also be expected and prepared for by the mental health community [35, 36]. Unfortunately, we were unable to assess the current employment status of our study population and can therefore not comment if the economic instability led to the increase of suicidality in our study group. We did analyze the various zip codes to determine the socioeconomic background through the patients’ district of residence; however, no geographical cluster could be detected. We also found no association between suicide attempts during the COVID-19 lockdown and belonging to an ethnic minority, even if it has been hypothesized that ethnic minorities, especially migrant populations, such as refugees, migrant workers and asylum seekers could be exponentially more affected by the current pandemic as they are especially vulnerable to the aforementioned dynamics [37].

Many researchers discussed the possible impact of the COVID-19 lockdown on mental health and suicide attempts; however, to our knowledge this is the first study statistically confirming this hypothesis. By observing the number of suicide-associated admissions to the resuscitation room of our level 1 trauma centre, we are one of the few studies publishing actual data on how many suicide attempts were hospitalized during the lockdown period.

We are aware of the limitations of our study. By the design of our study we had only access to the number and mechanism of injury admitted to the resuscitation room of the level 1 Trauma Centre of the Medical University of Vienna and cannot estimate how many patients received care in other departments or hospitals in the greater Vienna area or Austria in general; however, the absolute number of suicidal acts performed will always be impossible to determine as many individuals do not seek treatment in a hospital or help from mental health professionals after attempting suicide.