Suicidal behaviour in children and adolescents in Sweden and some European countries : epidemiological and clinical aspects

Abstract

The starting point for this thesis is the empirical observation that in Sweden, as in many other European countries, suicide is the leading cause of death among men aged 15-34 and ranks second for young females. Attempted suicide also constitutes a major public-health problem both in Sweden and in the rest of Europe.

The purpose of the study was to analyse suicide trends in Sweden, the relationship between suicide and attempted suicide, suicide-attempt repetition patterns, recommendations for care in some European centres after attempted suicide, and the quality of care in Sweden for young people who subsequently committed suicide.

The material utilised in the study comprises 5,893 certain and 1,739 undetermined suicides among people aged 10-29 in Sweden during the period 1974-97, and also medical records of suicide attempters in the 15-24 age group. The data were collected within the framework of the VHO/EURO Multicentre Study on Suicidal Behaviour. This study was commenced in 1989 in 15 European centres (including two in Sweden) with an aggregate catchment area of approximately 5,6 million inhabitants.

Regression analysis of suicides, broken down by gender and five-year age group, shows no decrease in juvenile suicide rates in Sweden during the period. For the females' part, the prevalent suicide method changed from drug intoxication during the 1970s to violent methods in the '80s and '90s. Males consistently use more violent suicide methods.

Attempted-suicide rates in 1989-92 found by the 15 monitoring centres among people aged 15-24 (n=3,551) covary positively (Spearman rank order) with both sexes' suicide rates in the centres' areas and countries of location. For males, this correlation attained statistical significance (p<0.02).

A longitudinal follow-up (mean: 204 weeks) of adolescents aged 15-19 (n=1,264) in seven centres revealed that more than 10% repeated their suicide attempts within a year of the index attempt. Life-table analysis indicated that 24% of the individuals who had attempted suicide before the index suicide attempt tried again within a year, compared with 6.8% of those with no such history. Stepwise Cox regression analysis showed that the most significant predictor of repetition was a history of attempted suicide.

Analysis of 1,540 suicide-attempt events recorded at nine centres showed that adolescents aged 15-19 with a history of one or more suicide attempts were - with no significant gender difference -recommended care after attempting suicide to a significantly higher degree than those with no such history (OR 2.0, 95% CI 1.5-2.6). Care was, moreover, recommended significantly more often for those who used violent methods than for those who did not (OR 1.7, 95% CI 1.5-1.96). Analysis of individual centres showed large disparities in care recommendations, indicating that no uniform criteria for recommending care for young suicide attempters were in use.

A qualitative retrospective study of medical records relating to 34 psychiatric inpatients aged 15-24 before their deaths from suicide showed a striking discontinuity - especially in adult psychiatric care - in terms of contacts with doctors, therapists and other staff (with three to 30 different doctors involved during the period of care). This discontinuity may have contributed to the fatal outcome. Suicide-risk analyses were lacking, although two-thirds of the patients were known to have attempted suicide previously and 75% were in situations of acute stress. In most European countries, expert follow-up and care of juvenile suicide attempters, and also scientifically based clinical guidelines for diagnostics and treatment, are therefore urgently needed.