Deliberate Self Harm (DSH) and Suicide: gender specific trends in eight European regions

Transcription

1 Deliberate Self Harm (DSH) and Suicide: gender specific trends in eight European regions Ella Arensman, Tony Fitzgerald, Tore Bjerke, Jayne Cooper, Paul Corcoran, Diego De Leo, Onja Grad, Keith Hawton, Heidi Hjelmeland, Nav Kapur, Ivan J Perry, Ellinor Salander-Renberg, Kees van Heeringen Network for International Collaboration on Evidence in Suicide Prevention (NICE-SP) 2 nd EuroSafe Conference Paris, 9-10 th October 2008

2 Background Limited information about long term trends in DSH at international level Limited information about gender specific patterns of DSH at international level Absence of information on the relationship between DSH and suicide rates across different countries

3 Objectives 1. To examine time trends in rates of DSH and suicide by gender in eight European regions 2. To examine how rates of DSH at regional level and suicide rates nationally covary by gender

4 Participating regions in the Network for International Collaboration on Evidence in Suicide Prevention (NICE-SP) Sor-Trondelag Manchester Umea Vilnius Limerick Gent Ljubljana Cork Oxford Padua

5 Method Data collection in 8 regions in 6 European countries Patients, aged > 15 years presenting to hospital following DSH from hospital catchment area Standardised method of data collection, developed as a standard for European regions collaborating in the WHO Multicentre Study on Suicidal behaviour (Platt et al, 1992; Schmidtke et al, 1996; Schmidtke et al, 2004): Demographic variables and characteristics of the DSH episode. All regions except Manchester participated in the WHO Multicentre study between 1989 and 2000.

6 Method ctd. Start monitoring of DSH: Oxford, Gent, Sor-Trondelag, Umea, Padua: 1989 Cork, Limerick, Ljubljana: 1995 Manchester: 1998 Data on suicides for the countries involved was obtained from the national statistics offices in each country (ICD9 / ICD10).

7 Method ctd. Definition of Deliberate Self Harm, according to Platt et al. (1992). Non-fatal outcome Deliberately initiated non-habitual self-harming behaviour Varying behaviours (e.g. self cutting, overdosis) Varying intentions (e.g. wish to die, self-punishment) Excluded: Acts of self harm by individuals with learning disability.

8 Method ctd. Data analysis Annual person-based rates of DSH and suicide, agestandardized Three-year rolling averages Annual rates of change in DSH and suicide were estimated for each region/country using a negative binomial model

9 Results Average annual national person-based rates of DSH per 100,000 by gender (Total N 44,495) Region Females Males Rate Rate Manchester Oxford Gent Limerick Cork Sor-Trondelag Umea Ljubljana

10 Average annual national suicide rates per 100,000 by gender in the participating countries Country Females Males Rate Rate Slovenia Flanders Sweden Norway Ireland UK/EW

11 Deliberate Self Harm: person-based three-year rolling average rates per 100,000 for women in 8 European regions 550 Manchester Rate per 100,000 Smoothed Local Suicide Rate Gent Oxford Sor-Trondelag Umea Limerick Cork Ljubljana Year Year

12 Deliberate Self Harm: person-based three-year rolling average rates per 100,000 for men in 8 European regions Rate per 100,000 Smoothed Local Suicide Rate Gent Oxford Sor-Trondelag Umea Limerick Cork Ljubljana Manchester Year

13 DSH repetition rates 12 months prospectively by region and gender Region Men CI Women CI Repetition (%) Repetition (%) (%) (%) Oxford Manchester Cork Ljubljana Sor- Trondelag Limerick Period covered

14 Repetition of DSH by men 12 months prospectively following index DSH act in 6 European regions 20 Risk of repeat DSH (%) Oxford Manchester Cork Ljubljana Limerick Sor-Trondelag Time(months) Kaplan-Meier survival analysis

15 Repetition of DSH by women 12 months prospectively following index DSH act in 6 European regions 20 Risk of repeat DSH (%) Sor-Trondelag Oxford Manchester Cork Ljubljana Limerick Time(months) Kaplan-Meier survival analysis

16 Suicide: three-year rolling average rates per 100,000 for men ( ) and women( ) 35 Flanders Ireland Norway Sweden UK-EW 30 Rate per 100,000 SuicideRate 25 Rate per 100, Year Year Year

17 Suicide: three-year rolling average rates per 100,000 for men ( ) and women( ) S lovenia Rate per SuicideRate 100, Y e Year

18 Association between changes in local rates of DSH and national suicide rates for men (positive values represent an increase) Rate of change in DSH (%) Smoothed Local Suicide Rate Oxford Ljubljana Manchester Sor-Trondelag Umea Gent Limerick Cork Year Rate of change in suicide (%)

19 Association between changes in local rates of DSH and national suicide rates for women (positive values represent an increase) 8 Rate of change in DSH (%) Rate of Smoothed change Local in DSH Suicide (%) Rate Umea Ljubljana Manchester Oxford Sor-Trondelag Cork Limerick Gent Year Rate of change in suicide (%)

20 For males a significant association was found between the rate of change in DSH rates at regional level and suicide rates nationally (r=0.71, p<0.05), but not for females (r=0.57, p=0.14)

21 Summary Trends in DSH rates varied considerably across the different European regions and by gender DSH rates were consistently higher for women, with highest rates in Manchester, Oxford and Gent In most regions similar trends in DSH were found for female and male DSH rates Across the European regions there was greater temporal variation in both genders in the incidence of DSH compared to the temporal variation in the incidence of suicide

22 Summary ctd. Regions with high DSH rates did not consistently show a pattern of high suicide rates at national level However, in males a significant association was found between the rate of change in DSH rates at regional level and national suicide rates The extremely high suicide rates among males in Slovenia may be particularly related to high levels of alcohol abuse and access to firearms (Bilban and Skibin, 2005)

23 Conclusions v An increase in DSH may be accompanied by an increase in suicide, in particular among males v Based on the study outcomes it is recommended that prevention programmes target both DSH and suicide v Due to the use of different death registration procedures, in some countries suicide cases may have been misclassified (e.g. as undetermined or accident ) v The findings underline the need to develop a system to link DSH data with suicide mortality data

24 Acknowledgments: Participants in the Network on International Collaboration on Evidence in Suicide Prevention (NICE-SP) Heidi Hjelmeland Tore Bjerke Tony Fitzgerald Paul Corcoran Ivan Perry Jayne Cooper Nav Kapur Keith Hawton Kees van Heeringen Ellinor Salander- Renberg Danute Gailiene Onja Grad Diego De Leo

25 Contact details Dr Ella Arensman National Suicide Research Foundation 1 Perrott Avenue College Road Cork Ireland T: (0) ella.nsrf@iol.ie