Suicide is recognised as a public health priority and remains one of the leading causes of death worldwide. “Every 40 seconds a person dies by suicide somewhere in the world” (WHO, 2014).
In numbers, it means that over 800,000 people die by suicide each year and around 1,500 people in Sweden (Jiang et al., 2012), where the study that I’m blogging about today is focused . It could be naturally hypothesised that among all the suicide victims, some have mental health difficulties. This number is rounded up from Cavanagh et al. (2003) up to 90%, with individuals affected by bipolar disorder constituting a high-risk group and being 17-20 times more likely to attempt suicide compared to the general population.
Certainly, research aims to identify causation and potential risk factors associated with suicide. However, there is limited evidence on the identification of risk factors for specific populations, e.g. studies targeting people affected by a mental health disorder. Indeed, mental health difficulties can severely affect the individual and the impact might differ across diagnoses (Pompili et al., 2013).
Over a decade ago, Hawton et al. (2005) identified a list of risk factors for both completed suicide and suicide attempts in bipolar disorder, which included male gender, history of prior suicide attempts and expressed hopelessness. The downside of the review is that the studies included have a very small sample, while the majority of them do not have a prospective design. Thus, one cannot be sure of the temporal sequence between exposure and outcome.
Hamsson et al (2018) conducted a prospective longitudinal cohort study which allowed them to identify potential predictors for completed suicide. Recent studies based on the same cohort identified associations between risk factors and suicide attempt (Tidemalm et al., 2014), but little is known about how these factors may differ for completed suicide.
The longitudinal cohort study was based on the linkage of Swedish registers known as the Swedish National Quality Register for Bipolar Affective Disorder (BipoläR) and the Cause of Death Register. BipoläR follows people who registered between 2004 and 2013 and had a bipolar diagnosis according to DMS-IV-TR criteria, including bipolar disorder type 1, type 2, not otherwise specified, or schizoaffective disorder of bipolar type. The register allows the exploration of basic epidemiological data on the history and clinical course of patients’ diagnosis.
The outcome of the study was death by suicide and the identification was based on Cause of Death Register, which provides mortality data for almost every death occurring in Sweden (99%) between 2004 and 2014. Suicide was classified according to ICD-10 codes as either definite suicide or death by self-harm with undetermined intent.
An extended list of factors was tested, including:
- Demographics – sex, age, BMI, education
- Social factors – living alone, any psychosocial and environmental problems, violent behaviour towards others, recent criminal conviction
- Clinical characteristics – subtype of bipolar disorder, any recent affective episode, family history of affective disorder in first-degree relatives, age at onset of any psychiatric disorder, any comorbid psychiatric disorders including substance use disorder, anxiety disorder, eating disorder or personality disorder, any previous suicide attempts, recent psychiatric inpatient care, and recent involuntary hospitalisation.
The association between predictors and death by suicide was assessed through Cox models, which were adjusted for age and sex. The date of the first entry into BipoläR was set as the start of time, while the date of suicide, date of other causes of death, or 31 December 2014 was set as the end of the follow-up. Additionally, in the models assessing the connection between specific affective episodes and suicide, only participants who had at least one follow-up were included.
Out of 12,850 people with bipolar disorder, a record of 90 deaths by suicide was identified during the follow-up period until the end of 2014.
Significant associations for completed suicide after the adjustment of covariates were:
- Being male
- Living alone
- Previous suicide attempts
- Any comorbid psychiatric disorders (comorbid substance-use disorder, anxiety disorder, personality disorder)
- Recent affective episode
- Recent depressive episode
- Criminal conviction
- Psychiatric inpatient care
- Involuntary hospitalisation.
Interestingly, risk factors varied by gender. The evidence suggests that:
- Living alone, comorbid substance use disorder, involuntary commitment, and having had at least one affective episode in the previous year were significant predictors of suicide in men, but not in women.
- Conversely, criminal conviction, comorbid personality disorder, and having had at least one depressive episode in the previous year were significant predictors of suicide in women, but not in men.
- Risk factors differ significantly by gender
- Bipolar subtype was not associated with suicide
- Recent affective episodes can predict death by suicide
- Several risk factors are shared between attempted and completed suicide for people with bipolar.
Strengths and limitations
This study has methodological strengths, such as its prospective design and both large and representative sample of 12,850 people. BipoläR records data for both inpatients and outpatients with various symptoms and severity levels. Cause of Death Register captures 99% of deaths nationally, so the possibility of missing a suicide is highly unlikely. Moreover, Swedish registers are found to be valid, reliable, and suitable for research (Hollander et al., 2017). Linked registers not only provide accurate and rich data but also the opportunity to explore research questions in depth. The effects of nature and nurture were explored through the examination of basic epidemiological and demographic characteristics, family history and prior individual clinical history.
However, BipoläR doesn’t contain all patients with bipolar disorder in Sweden; the national coverage is 29% (Pålsson and Landén, 2016), which is relatively low. Thus, inclusion bias cannot be eliminated, as some population groups with certain characteristics may be more likely to choose not to participate in such registers. A comparison between this register and the national patient register confirms that there are no differences in terms of sex and age.
Furthermore, certain risk factors were not considered, including early life adversities, family history of suicide, physical comorbidity, polarity of the first episode, total number of lifetime episodes, and psychotic features in the depressive or manic episodes.
Implications for policy and research
The acknowledgment of risk factors for people with bipolar disorder can lead to an improved provision of professional mental health care and greater prevention of future suicide attempts and/or completed suicides. In line with previous studies, the new evidence can be turned into a very powerful tool for professionals who work closely with this clinical population. As depression and other psychiatric comorbidities can predict a completed suicide, professionals should reconsider adopting a holistic approach and provide a person-centred intervention that can potentially meet all patients’ needs.
Future research through mixed-methods should focus more on under-studied and relatively forgotten populations, such as prisoners and ex-convicts. Suicide rates are very high in prisons, yet we have limited knowledge regarding the aetiology behind suicidal thoughts, what might trigger suicide attempts and what may have the power to prevent a completed suicide. The examination of views among prisoners and staff could have major implications on service delivery and development.
Conflicts of interests
Hansson C, Joas E, Pålsson E, Hawton K, Runeson B, Landén M. (2018) Risk factors for suicide in bipolar disorder: a cohort study of 12,850 patients. Acta Psychiatr Scand 2018: 138: 456–463. doi:10.1111/acps.12946
World Health Organization. Preventing suicide: a global imperative; 2014.
Jiang G, Hadlaczky G, Wasserman D. Sj€alvmord i Sverige, Data: 1980-2012, (NASP, ed); 2012.
Cavanagh JT, Carson AJ, Sharpe M, Lawrie SM. Psychological autopsy studies of suicide: a systematic review. Psychol Med 2003;33:395–405.
Pompili M, Gonda X, Serafini G et al. Epidemiology of suicide in bipolar disorders: a systematic review of the literature. Bipolar Disord 2013;15:457–490.
Hawton K, Sutton L, Haw C, Sinclair J, Harriss L. Suicide and attempted suicide in bipolar disorder: a systematic review of risk factors. J Clin Psychiatry 2005;66:693–704.
Tidemalm D, Haglund A, Karanti A, Landen M, Runeson B. Attempted suicide in bipolar disorder: risk factors in a cohort of 6086 patients. PLoS One 2014;9:e94097.
Hollander A, Kirkbride J, Pitman A, Lundberg M, Lewis G, Magnusson C, Dalman, C. (2017). Are refugees at increased risk of suicide compared with non-refugee migrants and the host population. Eur J Public Health 2017;27 (suppl_3).
Pålsson E. and Landén M. (2016). The quality register BipoläR annual report 2016. https://bipolar.registercentrum.se/om-bipolar/arsrapporter/p/SkMRS-nGb; 2017.
- Photo by Julian Jemison on Unsplash
- Photo by Andrii Podilnyk on Unsplash
- Photo by Robert Bye on Unsplash