Nested case-control study in the dynamic cohort of the population of residents in Catalonia (~ 7.5 million inhabitants) in the study period 2014-2019. The study population consists of persons attended in any public service of the Catalan health system for a suicide attempt during the study period.
The present study will use data from several population-based registries:
(1) Administrative data, such as the Mortality Registry of Catalonia and the Registry of Insured Persons, which contains sociodemographic information (for example, age, sex, nationality, socioeconomic indicators) as well as information on geographical area (dor example, catchment area), migration, deaths, and healthcare information to monitor the healthcare needs of the Catalan population (for example, depression, severe mental illness, and the number of visits to health services). This information will be used to construct the nested case-control sample in the cohort;
(2) healthcare data, such as electronic registries of various health services in the public sector: primary care, accident and emergency departments, general hospitals, psychiatric hospitals and outpatient mental health clinics, that gather information on a large number of factors potentially predictive of suicidal behaviour (for example, a history of suicidal behaviour; any type of physical disease; neurodevelopmental disorders, mental disorders, personality disorders, substance abuse disorders; any type of medical procedure; and detailed information on the number and type of healthcare contacts). These diagnoses and procedures have been codified using the International Classification of Diseases 9th and 10th Editions, Clinical Modification (ICD-9-CM and ICD-10-CM); Pharmacy Registry, which contains information on drugs prescribed by the publicly-funded health system that are relevant for predicting suicidal behaviour (for example, psychotropic medication and other drugs for the treatment of somatic diseases or which have psychotropic effects); the Catalonia Suicide Risk Code (CSRC Programme) Registry, which collects information on episodes of suicidal behaviour occurring in Catalonia since 2014. For each episode of suicidal behaviour, the following variables potentially predictive of repeat suicidal behaviour are gathered: the suicide scale of the Mini-International Neuropsychiatric Interview (MINI) 5.0.01; the lethality of suicidal behaviour; the presence and type of mental disorder; other psychological factors such as hopelessness, impulsivity, or aggressivity; use of or dependence on alcohol or illegal drugs, severe physical disease, living arrangements, family or social support, social problems, stressful life events, access to lethal means, and a family history of suicide.
In the first phase of the study, cases of attempted suicide will be identified by using the CSRC Programme Registry. Subsequently, we will randomly select 30 participants without a history of suicide attempts (control participants) for each case of attempted suicide, who will be matched without replacement by age and sex, following a nested case-control design (incidence-density sampling or risk sampling.
We will study the frequency of suicidal behaviour (attempted suicide, repeat attempted suicide, and suicide) in Catalonia and a large number of factors potentially predictive of these events (>700 factors). Machine learning techniques will be used to develop algorithms for personalised prediction of the risk of suicide attempts and repeat suicide attempts for each health service. The CSRC Programme will be evaluated, analysing (1) the effectiveness (in terms of hospitalisations for attempted suicide and deaths by suicide) and the costs of the programme in comparison with routine treatment; and (2) the implementation of the programme, using coverage and quality indicators (that is, quantifying the proportion of the target population included in the programme and the degree of adherence to the standard of care in the programme).