Sisters share the highest suicide attempt risk of any family pair, Swedish data show

Worldwide, roughly 700,000 people die by suicide each year. The sex differences in suicidal behavior are well documented but poorly explained: men die by suicide more often, while women attempt it approximately twice as frequently. Genetics clearly plays a role in suicide risk. But does it explain the sex difference?

A large population study of over 3 million Swedes, published in BMJ Mental Health, provides the most detailed answer yet. The short version: genetics contributes substantially to suicide attempt risk, but it does not account for the sex disparity. Something else is going on, and the data point toward shared family environment and social factors that operate differently for women than for men.

3 million people, 14 million family pairs

The researchers drew on Swedish population registers covering everyone born between 1963 and 1998, along with their parents, full siblings, and half siblings. Mental health and suicide attempts were tracked from age 10 until the end of 2019. The final cohort ranged in age from 21 to 57.

In total, 89,278 people (3% of the cohort) had made at least one suicide attempt. Over half (55%) were women. The sex difference was consistent: 3.3% of women attempted suicide compared with 2.6% of men. Psychiatric disorders were present in 76% of those who attempted suicide, and were more common among female attempters than male ones.

The familial analysis covered staggering numbers: 3.65 million mother-child pairs, 3.48 million father-child pairs, 4.99 million full sibling pairs, 908,740 maternal half-sibling pairs, and 1.16 million paternal half-sibling pairs.

The sister effect

Suicide attempts clustered within families, as expected. Relatives of people who had attempted suicide were at higher risk themselves, particularly among first-degree relatives. But the pattern was not uniform across family types.

The highest risk of any family pairing was among sisters, who showed nearly four times the risk. Mother-daughter pairs also showed elevated clustering. In general, risks were higher among same-sex relatives than among different-sex relatives, and higher among female family pairings than among male ones.

This is where the story gets complicated. If genetics alone drove the familial clustering, you would expect the pattern to be roughly symmetric between same-sex male and same-sex female pairs, since full siblings share approximately 50% of their DNA regardless of sex. The asymmetry, with stronger clustering among female relatives, suggests that shared environmental and social factors are amplifying the genetic risk specifically among women.

Heritability is real but sex-neutral

The quantitative genetics analysis estimated that 42% of suicide attempt risk is heritable. This figure did not differ significantly between the sexes. Shared environmental factors contributed a small but statistically significant additional proportion, around 4%.

There was also a moderate to strong genetic association between psychiatric disorders and suicide attempts, with the strongest link observed for substance misuse. This is consistent with previous research establishing psychiatric illness as a major risk factor for suicidal behavior.

The critical finding is the disconnect between equal heritability and unequal familial clustering. If genes contribute equally to risk in men and women, but the risk clusters more tightly among female relatives, then non-genetic factors must be doing additional work in women. These could include hormonal influences, neurobiological differences, shared social environments, modeling of coping behaviors within families, or other mechanisms the study was not designed to identify.

What the study cannot resolve

This is observational research using registry data, which means it can identify patterns but not prove causation. The researchers cannot distinguish between shared environment (growing up in the same household) and shared social influences (such as similar peer groups or cultural norms) that might operate among female relatives even outside the home.

The cohort included relatively young people born in or migrating to Sweden during childhood. Whether the same patterns hold in older populations, in other countries, or in cultures with different gender norms and family structures is unknown. The study also could not account for family-level factors like abuse, neglect, or parental mental illness beyond psychiatric diagnoses captured in the registry.

The 4% shared environment contribution may underestimate the true effect, since registry data cannot capture all the ways a family environment shapes suicide risk. Qualitative factors like family communication patterns, attitudes toward help-seeking, and exposure to a relative's suicidal behavior are invisible to this methodology.

Still, the scale of the study and the consistency of the sex-specific pattern make a strong case that the higher rate of suicide attempts among women is not primarily genetic. Understanding what non-genetic factors drive this difference is the next critical question for prevention.