An Analysis of 398 In-Hospital Suicides

Evaluation of the Werner-Felber-Institute Database

The suicide risk among patients receiving inpatient psychiatric treatment is many times higher than that among the general population (1). The average annual suicide rate has been estimated at 148 per 100 000 admissions (1970–2005), with a fluctuation range of 51–253. Hospital suicides have a significant negative impact on fellow patients, family member as well as on members of staff. In order to effectively reduce suicides and suicide attempts, it is crucial to obtain a valid recording of their context and characteristics (e.g., with the help of databases). In this way, patterns and risk factors can be identified and evidence-based prevention approaches designed (2). The first overview analysis presented here describes the number and frequency as well as certain selected parameters of suicide cases documented by hospitals. It serves the purposes of further research.

Methods

This article analyzes suicides in the 2019–2022 period based on the hospital suicide database. Attempted suicides were also documented (N = 545). The continuous online survey is based on the questionnaire that was issued by the Working Group “Suicidality and Psychiatric Hospitals” (“Suizidalität und psychiatrisches Krankenhaus”) and which was further developed and finalized by experts in a consensus process (3). Data is entered by hospitals participating on a voluntary basis (as of late 2022, 119 hospitals) in Germany and German-speaking countries. Due to the particularly high risk following hospital discharge, post-inpatient suicides (up to 4 weeks after discharge) are also recorded. Demographic data and information on the act of suicide are recorded.

The descriptive analysis of data as well as the calculation of frequencies, mean values (M), and standard deviations (SD) were carried out using SPSS 28. Due to missing responses, the total of subgroups may differ from the overall total. Correlations between the data were calculated correlatively (Pearson or, in the case of assumption violations, Spearman’s rho correlation coefficient), using the chi-square test or—in the case of assumption violations—exact test statistics.

Results

Of 398 suicides analyzed, 256 were committed by men, 139 by women, and one by an individual self-identifying as “diverse.” For two cases, no data were available. Most suicides occurred in the 51- to 60-year age group (M = 48.0; SD = 18.1; range 13–91). The most frequent methods of suicide were hanging, jumping from heights, and jumping or lying in front of a moving object (Table). A large proportion of the suicides were carried out at a distance of more than 2 km from the hospital. Almost a quarter occurred within the hospital building. While most suicides that took place further away were by hanging or by jumping from heights (n = 35 each), suicides in the hospital building were most frequently by hanging (n = 56). The methods differ in relation to the location (Monte Carlo simulation: P_MC < 0.001).

Table

Overall survey from 119 hospitals on 398 suicides

Suicides outside the hospital building mostly took place while patients were on planned unaccompanied time away from the hospital premises (n = 55), during therapeutic leave (n = 43), or after the individual had left the ward without permission (n = 41).

The majority of patients (n = 248, 66.5%, 95% confidence interval: [61.9; 71.0]) were receiving inpatient treatment at the time of their suicide.

In the 369 entries regarding a main diagnosis, affective disorders, schizophrenia as well as schizotypal and delusional disorders are the most common diagnostic groups. A history of attempted suicide was common.

The number of admissions was recorded for 2021 and 2022 and respective suicide rates of 52.35 and 71.07 per 100 000 admissions were calculated.

Discussion

In agreement with epidemiological studies, the data show that men commit suicide more frequently compared to women. The average age of hospital suicides is approximately 10 years lower than that in the general population (4)—this is explained by the younger age structure in psychiatric hospitals and can also been seen in other studies on this group of people. The same applies to the methods and location of suicide (5), which provide hospitals with indications on how to improve security—for example, structural safety measures to prevent strangulation attempts.

In agreement with the results of other studies, our analyses suggest that a previous suicide attempt may be a predictor of a later suicide, although the lack of a comparison group must be cited as the greatest limitation of our results. We recommend that when assessing suicide risk, any history of a suicide attempt should also be recorded in a standardized manner during medical history-taking, while bearing in mind that not all suicides have a known history of attempted suicide.

The fact that the completeness of the data could not be verified represents a limitation of our study results. When designing the questionnaire, compulsion to answer was kept to a minimum, making the threshold to enter a response low. In addition, numerous measures were taken to improve data quality (e.g., feedback on the data entered in the previous year, together with a request to complete the data).

Summary

The fact that many suicides occur during therapeutic leave indirectly highlights the difficulty inherent in predicting suicide. Therefore, it is important to promote not only structural but also non-structural approaches to suicide prevention in hospitals.

The results of the main diagnoses in suicides confirm the results of other studies (5).

Centralized, comprehensive documentation and analysis of hospital suicides and suicide attempts can be an important tool for the prevention of suicide in the context of inpatient treatment. This underscores the aim of this database, namely to make an important contribution to suicide prevention.

Katharina König, Luna Grosselli, Thomas Reisch, Nadine Glasow, Manfred Wolfersdorf, Friedrich M. Wurst, Ute Lewitzka

Funding

This study was co-financed by public funds based on the budget passed by members of the Saxon State Parliament (Ministerium für Soziales und Gesellschaftlichen Zusammenhalt [Ministry of Social Affairs and Cohesion]) (100645157).

Conflict of interest statement
LG is a Member of the Board of the DGS.

FMW is head of the Working Group “Suizidalität und Psychiatrisches Krankenhaus” (Suicidality and Psychiatric Hospitals).

NG received royalties for publications relating to the topic from Logos and the Medizinisch Wissenschaftliche Verlagsgesellschaft. She is Deputy Chairperson of the Werner Felber Institute e. V.

MW is a Member of the Advisory Board of the DGPPN, from which he received travel cost reimbursement. He is Honorary Chairman of the Working Group “Suizidalität und Psychiatrisches Krankenhaus” (Suicidality and Psychiatric Hospitals) and a member of NaSPro.

UL has received consultancy and lectures fees from Janssen Cilag GmbH. She is Chairwoman of the Board of the German Society for Suicide Prevent (Deutsche Gesellschaft für Suizidprävention e. V.) and the Werner Felber Institute e. V. as well as being active for the Suicidology Department of the DGPPN. She received reimbursement of travel expenses and congress participation fees from these institutions.

KK and TR declare that no conflict of interest exists.

Manuscript received on 2 October 2023, revised version accepted on 23 February 2024.

Translated from the original German by Christine Rye.

Cite this as:
König K, Grosselli L, Reisch T, Glasow N, Wolfersdorf M, Wurst FM, Lewitzka U: An analysis of 398 in-hospital suicides—evaluation of the Werner-Felber-Institute database. Dtsch Arztebl Int 2024; 121: 473–4. DOI: 10.3238/arztebl.m2024.0042

References

1. Wolfersdorf M, Vogel R, Vogl R, et al.: Suizid im psychiatrischen Krankenhaus – Ergebnisse, Risikofaktoren, therapeutische Maßnahmen. Nervenarzt 2016; 87: 474–82

2. WHO: Suizidprävention: Eine globale Herausforderung. Genf, Schweiz: WHO 2014.

3. Glasow N, König K, Wurst F: Werner-Felber-Institut erstellt Datenbank zu Kliniksuiziden und Suizidversuchen. Suizidprophylaxe 2019; 46: 35–6.

4. Suizide in Deutschland – Statistisches Bundesamt. www.destatis.de/DE/Themen/Gesellschaft-Umwelt/Gesundheit/Todesursachen/Tabellen/suizide.html (last accessed on 16 February 2023).

5. Madsen T, Erlangsen A, Nordentoft M: Risk estimates and risk factors related to psychiatric inpatient suicide—an overview. Int J Environ Res Public Health 2017; 14: 253