Suicides in Germany During the COVID-19 Pandemic

Historically it is known that economic crises, epidemics, or other risk scenarios influence the suicide rates of the affected populations. An potential increase in suicide rates in connection with the COVID-19 pandemic was therefore of concern. However, an analysis of international data from the first few months of the pandemic largely disproved this assumption (1). Studies from Germany, which have so far only examined smaller regional populations, also found no noticeable increase in suicide rates; for instance, the recent work from Wollschläger et al. (2) analyzed the Rhineland-Palatinate mortality register data from the year 2020 and found only minor deviations for the age group of >70 year olds.

The present work is based on suicide statistics from three German federal states, which have a total of eleven million inhabitants, over a survey period up to and including December 2021. The aim is to update and expand the aforementioned studies with special consideration of age and gender.

Methods

Data on suicide deaths from the police crime statistics (PKS) of Rhineland-Palatinate, Saxony, and Schleswig-Holstein in the period 01/2017 to 12/2021, stratified by gender and age groups, were used. The minimum requirements for inclusion of PKS data in the study were:

• Data reported monthly and rapidly available

• Stratification by age group and gender

• Coverage of the period from 01/2017 to 12/2021

• No changes in the survey methodology during the study period.

The observed monthly suicides were plotted separately by gender against the expected number based on a 95% prediction interval from the data years 2017 to 2019 (negative binomial regression model, adjusted for linear temporal trend and calendar month [categorical]). For the period 2020 to 2021, the expected annual suicides were extrapolated based on the trends determined for previous years and updated population figures. In addition, suicide rate ratios (total or age-specific) for the periods 2020 to 2021 versus 2017 to 2019 were calculated using an unadjusted negative binomial regression model and the age group–specific proportions of the total number of suicides (compositional, according to centered log ratio transformation with linear regression) (SAS 9.4).

Results

For both men and women, the observed monthly suicide events were sporadically outside the expected monthly predictions, both at the beginning of the COVID-19 pandemic and afterwards; however, the total number of observed suicides was within the expected number (figure 1).

Figure 1.

Expected and observed number of suicides, according to gender (2017–2021)

The age distribution of suicides and the total number of suicides, broken down by gender and year, are shown in Figure 2.

Figure 2.

Number of suicides by age group and calendar year

For men, the number of total suicides fell slightly in the period 2020 to 2021 as compared to the period 2017 to 2019, but the rate ratio with a 95% confidence interval (CI) was not statistically significant (0.96; 95% CI [0, 65; 1.41]). There was a significant reduction for the age group of 81 to 90 year old men (0.77 [0.62; 0.94]) and a significant increase for the age group of ≥ 91 year old men (1.20 [1.00; 1.44]). These effects were also evident in the compositional analysis: the proportion of suicides for men in the groups of 41 to 50 year olds and 71 to 80 year olds also decreased, while the proportion of suicides of the 61 to 70 year olds increased.

For women, the number of suicides in 2021 was higher than in previous years, but no statistical significance was observed for either the rate ratio of 2021 versus 2017 to 2019 (1.13 [0.49; 2.59]) or for the age-stratified rate ratios. In the compositional analysis of 2021 versus 2017 to 2019, however, there were significant reductions for women in the groups of ≤ 20 years old and 51 to 60 years old, as well as an increased proportion of suicides among women in the age groups of 61 to 70 years old and 81 to 90 years old.

Discussion

In the total sample for the pandemic period up to and including December 2021, we did not observe any noticeable increase in the number of suicides outside of the expected interval. Viewed according to age and gender, absolute and relative increases and decreases in suicide frequency were found in different age groups, but no overarching pattern was discernible.

In particular, we could not demonstrate a general increase in the number of suicides by older people as compared to the pre-pandemic period (for instance, due to loneliness imposed by contact restrictions). Of note, the increase of suicide in the group of men ≤ 90 years old was remarkable. However, as the characteristics of being a man and of old age are both strongly associated with suicide, there is already a high risk of suicide in this group. The sample examined for the age group of children and adolescents likewise showed no increased suicide rate, similar to findings from Great Britain (3). As there are comparatively low suicide rates in this age range, changes can only be detected if there are large deviations due to the wide confidence intervals. With respect to suicidality in children and adolescents in the context of the COVID-19 pandemic, the existing literature does not provide a uniform picture: while American adolescents were found to have increased suicidal thoughts (4), a (non-representative) German sample showed a significant decrease in reported suicidal plans (5).

The present work is based on a sample with a long intra-pandemic observation period and with stratification into comparatively narrow age bands. In addition, the PKS data used have the advantage that they are very up-to-date and available in sufficient detail. A limitation is that the PKS data are to be regarded as provisional and were only available to the required extent for the present study for three federal states (about 13% of the German resident population); as a result, smaller-sized effects may not have been detected. Overall, the significance of PKS and cause of death statistics with regard to completed suicides is comparable; no systematic difference to the cause of death statistics can therefore be assumed for the PKS data used.

In summary, there was no increase in suicide rates in the German population studied during the COVID-19 pandemic. With respect to the suicide risk of individual age groups, a complex picture emerges. This underscores the complexity and multicausality of suicidal phenomena. Continued monitoring of suicide rates seems sensible, as side effects and consequences of the pandemic are to be expected in the coming months and years.