Post-traumatic stress disorder (PTSD) is a psychiatric diagnosis with a lifetime prevalence ranging from 2.1% to 5% (Koenen et al., 2017). It is characterised by a prolonged or delayed response to a traumatic event (or events) of a threatening or catastrophic nature. Individuals diagnosed with PTSD typically experience intrusive memories (“flashbacks”) or dreams of the traumatic event, distress when exposed to situations associated with their trauma, inability to recall elements of their trauma, and persistent psychological arousal (NIMH, 2019).

Previous research has clearly demonstrated that PTSD is a risk factor for subsequent suicidal ideation (Bentley et al., 2016), which leads to concerns regarding the potential association between PTSD and death by suicide. However, the evidence base examining this association remains shaky, with most studies lacking generalisability due to their use of a veteran/military sample. To date, only two population-based Danish studies have looked at this association, and both indicated significantly higher rates of suicide in those diagnosed with PTSD (Gradus et al., 2010; Gradus et al., 2015).

However, these studies were still unable to detect variations in the association by demographic characteristics or to control for psychiatric history. This highlights the need for further research, with even larger population-based samples, to gain a clearer picture. Fox and colleagues (2020) have swooped in to fill this research gap, specifically aiming to examine:

1. Whether a PTSD diagnosis predicts death by suicide
2. Whether this association differed between men and women
3. The potential impact of PTSD on death by suicide at the population level.

Methods

Fox et al. (2020) sought to answer these questions through a large, nationwide Swedish cohort. Their sample included all individuals born between 1973-1997 who were living in Sweden on or after their 14^th birthday, including immigrants to Sweden up to the end of 2011. Participants were followed from their 14^th birthday (or date of immigration to Sweden if later) until their exit from the cohort (i.e., by death, emigration, or end of follow-up).

The authors relied on Swedish register data to measure both their primary outcome and exposure. Their primary outcome (death by suicide) was measured as reported in the Cause of Death Register, while their exposure, PTSD diagnosis, was measured as reported in the National Patient Register. Both these registers utilised the International Classification of Diseases (ICD-9 and ICD-10).

To examine whether suicide mortality rates differed depending on the presence of PTSD diagnoses, they conducted Cox proportional hazard models, where they accounted for an array of sociodemographic characteristics (e.g. natal sex and neighbourhood deprivation level), parental severe mental illness, and personal psychiatric history (e.g. previous non-fatal suicide attempts). To determine the proportion of suicides in the population-at-risk and those exposed to PTSD, they also calculated the population attributable risk fractions and attributable fraction in the exposed.

Results

From their initial cohort of 3,194,141 individuals, 3,177,706 (99.5%) with complete data were included in the final analyses, adding up to 49.2 million total person-years of follow up. The authors suggested that 0.7% of their sample was diagnosed with PTSD, with PTSD diagnoses being more common in migrants than Swedish-born individuals (1.1% vs 0.6%). Of those with PTSD, 0.9% died by suicide during follow-up, with a subsequent median time of 2.4 years to suicide after PTSD diagnosis. Death by suicide in those with PTSD was also more common in women than men (64.1% vs 35.9%). Furthermore, the number of deaths by suicide per 100,000 was also greater in those with PTSD in comparison to both the population as a whole and those without PTSD.

Through their analyses, Fox et al. (2020) found that those diagnosed with PTSD were at greater risk of death by suicide (HR: 4.28), with the risk being greater in women than men (HR: 2.61 vs 1.67). When the authors took sociodemographic factors into consideration, this association slightly increased (HR: 5.36). However, accounting for parental SMI and personal psychiatric history greatly decreased it (HR: 2.15), mostly due to a history of non-fatal suicide attempts pre-PTSD diagnosis.

The authors also report that assuming causality, if no one in the Swedish population experienced PTSD, up to 1.6% (and 3.8% in women) of all suicides could be prevented. However, for those with a PTSD diagnosis, 53.7% of all suicides could be prevented if we could identify and remove all factors associated with increased suicide risk in people with PTSD.

Sensitivity analyses also demonstrated that these findings were not impacted by the use of a broader definition of PTSD (i.e., including “acute stress reactions” and “crisis reactions, not otherwise specified”), the inclusion of “death by undetermined intent”, and the exclusion of people diagnosed with PTSD in outpatient settings between 2001-2005 (7.2% of all PTSD cases).

Conclusions

• Fox et al. (2020) have provided longitudinal evidence that PTSD is associated with increased suicide rates in the population
• Individuals with PTSD are twice as likely to die by suicide, when compared to people without PTSD
• Women with PTSD are at a particularly heightened risk
• However, there is a light at the end of the tunnel, as Fox et al. (2020) report that, assuming causality, “over 50% of all suicides in people with PTSD” could be prevented.

Strengths and limitations

The study had several clear strengths, including:

• The use of a large population-based cohort with a 49.2 million person-year follow-up
• The use of Swedish Register data, which is known to be reliable for psychiatric research
• Minimal risk of bias due to the rigorous investigation of several potential sources of bias
• The fact that the study’s results replicated the findings of the two aforementioned Danish population-based cohort studies.

However, it was also limited in many ways:

• Use of Swedish Register data – while this was strength in one way, it was also a limitation in that they could only eliminate potential confounders that existed within the Register data. This meant that information that was not on the Register (e.g., psychiatric diagnoses made in primary care or drug/alcohol misuse) was overlooked
• Inability to identify when PTSD symptoms first began – it’s possible that they missed previous suicide attempts brought on by pre-diagnosis PTSD symptoms, which would influence the association found
• Inability to distinguish between traumas of different nature or between the type of PTSD (e.g., multi-event complex PTSD vs single event PTSD), as both of these factors, could have different implications for the association between PTSD and death by suicide
• Lack of clarity regarding whether asylum seekers and refugees were also included within the sample, as these two groups are at particular risk for PTSD (Kien et al., 2019; Blackmore et al., 2020)
• It is unclear whether we can generalise these findings to populations within other regions. This study, as well as the Danish population studies, were all conducted in Scandinavian countries with stereotypically “Western”/individualistic cultures. We don’t really know if similar results would be found if a study of this nature was conducted in other regions, such as Asia, Africa, or the Middle East.

Implications for practice

Where can research go from here?

Firstly, there is a need to investigate how the nature of the trauma experienced (e.g. sexual assault vs combat trauma) affects the association between PTSD and death by suicide. As previous research has shown that there are specific traumas that predict a PTSD diagnosis in men (combat experience) vs women (rape and other sexual assault) (Frans et al. ,2015), the types of traumas experienced by women may be underpinning their relatively higher risk of suicide. Examining this could help us identify why we see gender differences in this association, as well as help highlight if specific traumas are linked to a greater risk of suicide.

Second, future research should aim to examine the effectiveness of a variety of interventions for PTSD, with a specific focus on suicide-related outcomes, as research has largely focused on symptom reduction so far.

Third, albeit difficult due to the lack of Register data, it would be beneficial to attempt studies of this nature in other regions of the world.

How can we use these findings in clinical practice?

Broadly speaking, clinical guidance on the management of PTSD needs to recognise this increased risk. Similarly, suicide prevention strategies should highlight those with PTSD as a high-risk group and recommend effective treatments for PTSD. For mental health services, there is a need to recognise this elevated risk of suicide in people with PTSD and given the short median time between diagnosis and death by suicide, act quickly. Means restriction, currently the most effective suicide prevention intervention (Zalsman et al., 2016), should be high on the list of considerations, particularly in relation to prescribing medication.

These findings are also particularly concerning given the ongoing context of the COVID-19 crisis. Recent studies have demonstrated elevated rates of PTSD symptoms across an array of samples, including COVID-19 survivors (Cai et al., 2020) and frontline workers (Jonhson, Ebrahimi & Hoffart, 2020). In my work with UCL’s Trauma Group, I was involved in researching frontline workers’ experiences and views of psychosocial support. In doing so, I’ve heard harrowing stories of the reality of working on the frontlines during the COVID-19 pandemic, and how little psychological support frontline workers are receiving. Fox et al’s (2020) findings have made it even clearer to me how crucial it is that additional efforts are made to identify PTSD during and beyond these trying times. Knowing that a lot of PTSD-related suicides are preventable, it is vital that we do all we can to minimise the number of people falling through the cracks.

Statement of interests

None.

Links

Primary paper

Fox, V., Dalman, C., Dal, H., Hollander, A.C., Kirkbride, J.B., & Pitman, A. (2020). Suicide risk in people with post-traumatic stress disorder: A cohort study of 3.1 million people in Sweden. Journal of Affective Disorders, 279, 609-616.

Other references

Bentley, K.H., Franklin, J.C., Ribeiro, J.D., Kleiman, E.M., Fox, K.R., & Nock, M.K. 2016. Anxiety and its disorders as risk factors for suicidal thoughts and behaviors: a metaanalytic review. In: In Clinical Psychology Review, Vol, 43. Elsevier Inc, pp. 30–46.

Blackmore, R., Boyle, J.A., Fazel, M., Ranasinha, S., Gray, K.M., Fiztgerald, G., Misso, M., & Gibson-Helm, M. 2020.  The prevalence of mental illness in refugees and asylum seekers: A systematic review and meta-analysis. PLoS Med, 17(9): e1003337.

Cai, X., Hu, X., Ekumi, I.O., Wang, J., An, Y., Li, Z., & Yuan, B. (2020). Psychological Distress and Its Correlates Among COVID-19 Survivors During Early Convalescence Across Age Groups. The American Journal of Geriatric Psychiatry, 28(10), 1030-1039.

Frans, O., Rimmo, ¨ P.A., Åber, L., & Fredrikson, M. 2005. Trauma exposure and posttraumatic stress disorder in the general population. Acta Psychiatrica Scandinavica, 111(4), 291–290.

Gradus, J.L., Antonsen, S., Svensson, E., Lash, T.L., Resick, P.A., & Hansen, J.G., 2015. Trauma, comorbidity, and mortality following diagnoses of severe stress and adjustment disorders: a nationwide cohort study. American Journal of Epidemiology, 182(5), 451–458.

Gradus, J.L., Qin, P., Lincoln, A.K., Miller, M., Lawler, E., Sørensen, H.T., & Lash, T.L..2010. Posttraumatic stress disorder and completed suicide. American Journal of Epidemiology, 171(6), 721–727.

Johnson, S.U., Ebrahimi, O.V., & Hoffart, A. 2020. PTSD symptoms among health workers and public service providers during the COVID-19 outbreak. PLOS ONE 15(10): e0241032.

Kien, C., Sommer, I., Faustmann, A., Gibson, L., Schneider, M., Krczal, E., Jank, R., Klerings, I., Szelag, M., Kerschner, B., Brattström, P., & Fartlehner, G. 2019. Prevalence of mental disorders in young refugees and asylum seekers in European Countries: a systematic review. European Child and Adolescent Psychiatry, 28, 1295-1310.

Koenen, K.C., Ratanatharathorn, A., Ng, L., McLaughlin, K.A., Bromet, E.J., Stein, D.J., Karam, E.G., Meron Ruscio, A., Benjet, C., Scott, K., Atwoli, L., Petukhova, M., Lim, C.C.W., Aguilar-Gaxiola, S., Al-Hamzawi, A., Alonso, J., Bunting, B., Ciutan, M., De Girolamo, G., & Kessler, R.C.. 2017. Posttraumatic stress disorder in the World Mental Health Surveys. Psychological Medicine, 47(13), 2260–2274.

National Institute of Mental Health. 2019. Post-Traumatic Stress Disorder. Retrieved January 17^th, 2020, fromhttps://www.nimh.nih.gov/health/topics/post-traumatic-stress-disorder-ptsd/index.shtml

Zalsman, G., Hawton, K., Wasserman, D., van Heeringen, K., Arensman, E., Sarchiapone, M., Carli, V., Hoschl, ¨ C., Barzilay, R., Balazs, J., Purebl, G., Kahn, J.P., S´ aiz, P.A., Lipsicas, C.B., Bobes, J., Cozman, D., Hegerl, U., & Zohar, J. 2016. Suicide prevention strategies revisited: 10-year systematic review. Lancet Psychiatry, 3(7), 646–659.

Photo credits

• Photo by Gift Habeshaw on Unsplash
• Photo by Gabrielle Henderson on Unsplash
• Photo by Lily Banse on Unsplash