Objective: Individual-level data from clinical settings lack information on people who did not seek professional help prior to suicide. We used records of the Nova Scotia Medical Examiner Service (NSMES) to compare people who had contact with a health professional prior to suicide with those who did not.
Method: We linked data from the NSMES to routine administrative data of the province.
Results: The NSMES recorded 108 suicides in Nova Scotia from January 1, 2006, to December 31, 2006; there were 90 male and 18 female suicide deaths. Mean and median age at death were 44.73 (SD 13.33) and 44 years, respectively. Patients aged 40 to 49 years made up one-third of the cases (n = 35) and this was the decade of life with the highest number of suicides. This was also the group least likely to have suicidal intent recorded in the NSMES files (χ^sup 2^ = 3.86, df = 1, P = 0.05). Otherwise, there were no significant differences between people who sought help, or disclosed intent, prior to suicide and people who did not. The samples in all cases were predominately male and single.
Conclusions: People aged 40 to 49 years were the age group with the highest absolute number of suicides, but were the least likely to have suicidal intent recorded in the NSMES files. This finding merits further investigation. Medical examiner or coroner data may provide additional information not obtained elsewhere for the surveillance of suicide.
Can J Psychiatry. 2011;56(7):436-440.
* Patients aged 40 to 49 years have the highest number of suicides but were least likely to have suicidal intent recorded in the files. This merits further research to establish whether this a true finding or reflects reporting bias.
* Although reluctance to seek help is one suggestion for higher suicide rates in men, threequarters of male decedents had health service contact in the year prior to their death.
* Medical examiner and coroner data may provide additional information not obtained elsewhere for the surveillance of suicide. Their use for research or surveillance would be enhanced if medical examiners and coroners collected the information in a more systematic way.
* We cannot exclude the possibility of type Il error given the small number of subjects.
* Information collected by medical examiners in the course of an investigation is meant to meet the operational requirements ofthat investigation. Data are thus not systematically collected and may also lack details important to researchers but of little practical importance to an investigative agency.
* Administrative data are subject to recording bias and lack indicators of disease severity or psychosocial functioning.
Key Words: health outcomes, mental health services, suicide
Suicide is the 11th leading cause of death for all ages in Canada, accounting for 1.7% of all deaths.1 There are about 88 deaths due to suicide in Nova Scotia each year and an age-specific mortality rate of 9.6 per 100 000 population.2 Because suicide often occurs in working life, the potential years of life lost are significant, coming third after cancer and heart disease in men.3 Mental illness, particularly depression, is the strongest individual risk factor.46 Most studies report a higher risk of suicide among males, often through reluctance to seek help, or the use of more lethal methods.3,5,7,8 Other risk factors include stressful life events,5 increasing age,8 previous suicide attempts,5 sexual orientation,9 rural residence,4 firearm ownership,10 social isolation,3 low socioeconomic status,4,8 chronic pain, terminal illnesses or disabilities," or being the victim or perpetrator of domestic violence. Hotels are a location for suicide, particularly among local residents.12
Most information on suicide comes either from population data, or individual-level data from clinical settings.13 There is less individual-level information on people who did not seek help prior to suicide and who thus may not appear in clinical data. …