Response

In the Phrygian city of Gordium, Greek mythology tells of an oxcart tied to a post by way of an intricate knot with hidden ends. The oxcart belonged to the founder of the city, Gordias, and an oracle foretold that whoever could untie the knot would rule Asia. Legend has it that Alexander the Great, passing through the city during his campaigns, was presented this puzzle. Stumped by the complexities of the knot, Alexander’s unconventional solution was to slash through the fibres with a single swipe of his sword – hence the expression ‘cutting the Gordian knot’. This knot has become a byword for complex unsolvable problems or challenges requiring novel solutions. ^{Reference Lane Fox1,Reference Graves2}

The journal recently published a consecutive descriptive case series by Aggarwal and colleagues examining self-harm in 544 adolescents (ages 10–19) who died by suicide in the UK between 2016 and 2018. ^{Reference Aggarwal, Tham, Ibrahim, Turnbull, Webb and Appleby3} From this UK-wide cohort, 49% (267/544) had a history of self-harm, with about half of these (139/267) having injured themselves within three months of death. The authors concluded that detecting youth who have self-harmed should be a priority, with their detection a critical opportunity to engage them in treatment, potentially preventing suicide.

While not inaccurate, these conclusions represent only part of the story, ignoring the greater challenge of preventing suicide in the half of the cohort (51%; 277/544) without a known history of self-harm when they died. They were unknown in other ways as well, with most having no recorded diagnosis of mental illness (77%; 212/277), health service contact (60%; 167/277) or history of psychiatric medication treatment (88%; 245/277). With these subjects lacking indicators of concern, how are they to be identified? Akin to the Greek myth, preventing suicide in individuals offering no obvious clues before death is a Gordian Knot that resists unravelling.

Our published studies (2016–2024) on suicide provide further evidence supporting the presence of this Gordian knot. ^{Reference Bostwick, Pabbati, Geske and McKean4–Reference McKean, Pabbati, Geske and Bostwick6} Whereas Aggarwal et al explore a consecutive case series of youth dead by suicide, we conducted a retrospective–prospective study recruiting a cohort defined by having made first suicide attempts coming to medical attention (index attempts), regardless of initial management and – novelly – including those who died on first attempt. We identified 1490 (ages 10.1–91.8 years old) individuals over a 22-year time frame followed for 3–25 years, of which 813 were youth ages 10–24. The lethality of index attempts was striking, with almost three-quarters (71.4%; 20/28) of suicides in the youth sample occurring on index attempt, with over 40% of these index attempters having no antecedent psychiatric appointments, diagnoses or medication trials. ^{Reference McKean, Pabbati, Geske and Bostwick6} Despite using a different methodology and studying a differing population than Aggarwal and colleagues (consecutive case series, UK-wide cohort of youth ages 10–19 versus retrospective–prospective study, US-community cohort of youth ages 10–24), our findings concur with theirs, underscoring that a sizeable proportion of youth have no contact with psychiatric care prior to index attempt, as we report, or death by suicide, as Aggarwal and colleagues show.

Most suicide prevention efforts target youth with known histories of psychiatric diagnoses, self-injurious behaviours or suicidal behaviours. The findings of Aggarwal and colleagues show that half of their subjects had no history of self-harm, and only one in ten of those with no history of self-harm had had recent contact of any sort with mental health services. If current efforts are failing to reach a substantial portion of the dead, how will the mental health community first recognise and then untie this suicide prevention Gordian Knot? Aggarwal and colleagues cite numerous laudable efforts to address the human misery that manifests in suicidal ideation and behaviour, most prominently the National Institute for Health and Care Excellence (NICE) guidelines for self-harm management. They rightfully point out that ‘each encounter [with a suicidal individual] should be seen as an opportunity to intervene’, adding that ‘addressing the key social determinants of suicide in young people, such as poverty and deprivation, alcohol and drug use, interpersonal violence and childhood abuse, as well as insecurity in jobs and housing, lack of or fears about opportunity and demand for healthcare that services are struggling to meet, are also integral to suicide prevention.’ We would add that individuals should not have to be suicidal to access socially responsible and responsive mental health care.

While universal measures that restrict lethal means, increase screening and improve detection of psychiatric illness hold promise, not all those dying by suicide have recognisable risk factors. A significant proportion will inevitably elude detection. Moreover, suicide prevention does not belong to psychiatry alone: as a complex phenomenon affected by cultural, religious and economic factors as well as mental illness, it demands a multifaceted, population-level public health approach. Even then, Aggarwal’s data as well as ours suggests that it may never be possible to prevent all suicides. For some proportion of youth with undetected risk factors, we face the uncomfortable truth that the Gordian Knot will not be unravelled any time soon.