6. REGISTRATION AND MONITORING OF SUICIDE AND SUICIDE ATTEMPT

Suicide deaths are recorded as a part of vital statistics in the Nordic countries. Causes of death are determined by a medical doctor, which is likely to improve the validity.

To determine the cause of death, the medical doctor may request a medico-legal examination or an autopsy. Based on a sample study of suicides in Denmark, Norway and Sweden, the percentage of autopsies, which were performed to determine the cause of death in incidents suspected to be suicide deaths, was seemingly higher in Sweden (81%) than in Norway (54%) and Denmark (32%) (Tøllefsen, et al., 2015). Still, when asking experts from the respective countries to re-evaluate the cause of death for a sample of potential suicides, accidents and undetermined deaths, these evaluated that 77%, 87%, and 92% of suicides had been recorded correctly in Sweden Norway, and Denmark, respectively. Thus, suggesting a good level of reliability in the registration of suicide deaths (Tøllefsen, et al., 2015).

Suicide deaths are recorded in the national Cause of Death register by public health authorities. Following the 10^th revision of International Classification of Diseases and related Health Problems (ICD-10), specific codes are used to classify suicide deaths (ICD-10: X60-X84, Y87.0) (World Health Organization, 2021b). In Sweden, an additional category, which often is referred to as ‘undetermined deaths’ (ICD-10: Y10-Y34, Y87.2) has been suggested to also include suicide deaths (Tøllefsen, et al., 2015, Björkenstam, et al., 2014). In relation to suicide deaths, the category of undetermined deaths is proportionally larger in Sweden than in Denmark and Norway (Tøllefsen, et al., 2015). This implies that Swedish register-based studies often tend to be based on both suicide and undetermined deaths, while Danish, Finnish and Norwegian register-based studies only consist of cases classified as suicide deaths (Erlangsen, et al., 2020b, Burrell, et al., 2018, Aaltonen, et al., 2019, Mogensen, et al., 2016b). For this reason, figures on suicide deaths presented in this report from Sweden are based on both suicide and undetermined deaths, while figures for all other countries are solely based on suicide deaths.

Consulted experts from the Nordic countries unanimously confirmed that the registration of suicide death was considered to be reliable in their country (Table 6.1). The yearly number of suicide deaths was also assessed on a regular basis. As a part of the national membership obligations, Denmark, Finland, Iceland, Norway and Sweden report the yearly number of suicide deaths to the WHO Mortality Database, thus, contributing to an international monitoring of suicide. Aaland Islands, Faeroe Islands, and Greenland do not report to the WHO Mortality Database. An international comparison conducted by the Global Burden of Disease Study revealed that the suicide rate in Greenland was the highest suicide rate in the world (Naghavi, 2019). For this reason, it would be relevant that data from Greenland are considered in reports and resources on suicide prevention published by the WHO (World Health Organization, 2014, World Health Organisation, 2021).

Table 6.1 Information regarding registration of suicide and suicide attempts.

^* Defined as an implemented procedure, which produces yearly updated statistics.

Suicide attempts may be identified through presentations to emergency departments (ED) or intensive care units and recorded in electronic hospital registers. Nevertheless, there seems to be a general consensus that suicide attempts are under-recorded in hospital registers (Morthorst, et al., 2016, Qin & Mehlum, 2020). To estimate the true figure of hospital presentations for suicide attempts, different approaches have been used in the Nordic countries (see Box 6.1 and 6.2).

Almost none of the Nordic countries conduct an annual monitoring of suicide attempts. Systematic surveillance of suicide attempts have been recommended by the WHO as essential for guiding preventive efforts and providing an option for monitoring these activities (World Health Organization, 2016).

Two measures are used to identify suicide attempts in the Danish National Patient Register; namely suicide attempt and probable suicide attempt.

A hospital recorded event is considered to be a suicide attempt if at least one of two markers is recorded: 1) reason for contact is listed as being suicide attempt; and 2) a main or a secondary diagnosis of intentional self-harm (ICD-10: X60-X84).

Probable suicide attempt is defined, in addition to the inclusion criteria listed above, as: 3) a main diagnosis of a psychiatric disorder (ICD-10: F00–F99) and a supplementary diagnoses of lesion to the lower arm (ICD-10: S51, S55, S59, S61, S65, S69) or poisoning (ICD-10: T36–T50, T52–T60) where both diagnoses are listed in the same record; or 4) poisoning; defined as a main diagnoses of poisoning with weak analgesics, narcotics, psychotropic drugs, or carbon monoxide (ICD-10: T39, T40 except T401, T42, T43, T58).

Source: Morthorst et al, 2016

Data on all hospital records for somatic treatment due to external causes were retrieved from the Norwegian Patient Registry. A sophisticated procedure was applied to exclude records of indirect contacts, planned treatments, fatal injuries and poisonings, injuries, which were clearly accidental or inflicted by others, secondary outcomes of other medical conditions, and contacts of children younger than 10 years of age and non-residents.

For the remained records, three measures were applied hierarchically to identify probable incidents of suicide attempt: 1) All treatment contacts with a comorbid diagnosis of DSH or suicide attempt (ICD-10: X6n, Y87.0); 2) Treatment contacts with a diagnosis of poisoning (ICD-10: T4n, T50–T55, T57–T60, T62, T62, T65), open wounds (ICD-10: S10, S11, S15, S17, S19, S21, S25–27, S31, S35–39, S41, S45, S50–51, S54–56, S59, S61, S64–66, S69, S71, S88, T01, T09, T11) or suffocation/drowning or burning (ICD-10: T18, T19, T27–28, T31, T68, T69, T71, T95) and a comorbid diagnosis of mental or behavioral problems (ICD-10: F0–F9); 3) Treatment contacts with poisoning (ICD-10: T4n, T50), which had not been identified in the above listed measures.  

Source: Qin et al, 2020

The COVID-19 pandemic demonstrated the usefulness of real-time data, i.e. data that is available for publication immediately after being collected. Although the registration of suicide deaths is conducted electronically and the data is transferred to the national offices administrating the Cause of Death registers, data on suicides are often first made available with 1–3 years of delay. This is unfortunate from a suicide preventive perspective. Having real-time data – or updates of the preceding years’ numbers published as early as possible in the new calendar year would allow policymakers and researcher to monitor whether the national goals for suicide prevention are being reached and whether new risk groups or suicide methods are emerging.

The Nordic register data offer unique opportunities for monitoring of suicide and suicide attempts for high-risk groups, such as individuals with mental and chronic disorders, unemployed, divorced, widowed, disability retirees. The Danish FORSAM-database is a first initiative in this direction and provides data on suicide and suicide attempt on municipal, hospital cluster, regional, and national level. It has been developed in collaboration with stakeholders within suicide prevention (www.forsam.dk). In Norway, the Norwegian Surveillance System for Suicide in Mental Health and Substance Use Services has been established to monitor all suicide deaths, which occur within one year after contact with mental health and substance misuse services (https://www.med.uio.no/klinmed/forskning/sentre/nssf/kartleggingssystemet/).