Anew study involving Pompeu Fabra University (UPF), based on administrative data from South Korea, projects that as higher education and lifelong singlehood become more common, the composition of older-age suicide deaths may shift toward never-married men with tertiary education. This does not mean that tertiary-educated individuals face higher suicide risks: their suicide rates remain lower than those of less educated groups. Rather, the projected shift is driven by population composition. As tertiary education becomes the norm among younger generations, the growing size of this group could lead to a higher absolute number of suicide deaths among tertiary-educated, never-married men.
The study also shows that the recent decline in suicide rates among older adults has been uneven. While suicide rates among divorced or never-married older adults have declined overall in the past two decades, this improvement was concentrated among people with higher educational attainment. Among less educated never-married and divorced men, suicide rates remained persistently high.
The research, which was carried out by Chia Liu (School of Geography and Sustainable Development, University of St Andrews, UK), Ryohei Mogi (Department of Political and Social Sciences, Pompeu Fabra University) and D. Susie Lee (Max Planck Institute for Demographic Research in Rostock, Germany), published in BMJ Public Health, examines how changes in family structures and educational expansion in South Korea are reshaping the demographic profile of suicide among older adults. The results reveal highly notable disparities by sex, marital status and level of education, and foresee a demographic change in the population most at risk of death by suicide in old age.
"Whether (or not) the expected increase in suicides materializes depends, at least in part, on the evolution of the conventions of family formation and policy intervention"
"This study highlights the importance of examining changes in at-risk populations due to changes in the composition of the main demographic covariates of suicide –marital status and education– in a population with high suicide rates among the elderly. Whether (or not) the expected increase in suicides materializes depends, at least in part, on the evolution of the conventions of family formation and policy intervention", the authors point out.
Thus, for example, in an analysis focusing on men born between 1981 and 1985, the difference between a 30% reduction (the most successful scenario) and a 5% reduction (the least successful) in this period could mean about 75 fewer suicide deaths per 100,000 people.
Marital status and education are closely associated with suicide rates
According to the authors, singleness and divorce continue to be closely related to higher suicide rates, especially among men. This confirms that marriage might reflect certain selection mechanisms, especially among the older adults studied here (those who are currently between 60 and 69 years old), who lived through periods when marriage was nearly universal in South Korea.
Among less educated, single or divorced men, high suicide rates persisted
However, the recent decline in suicide rates among older people has been unequal. While overall suicide rates among divorced or single older adults have declined in the past two decades, such decline was only noted among those with a higher level of education. Among less educated, single or divorced men, high suicide rates persisted. In previous birth cohorts, people with a lower level of education were most at risk. In younger cohorts, the education gap in suicide rates has begun to narrow, but only among people with higher education: people with a lower level of education remain particularly vulnerable.
Policy implications
The study highlights a fundamental gap in suicide prevention policies: the need to take into account not just the rates, but also the absolute number of suicide deaths resulting from demographic change. As never-married older people with higher education account for an increasing share of South Korea's population, suicide prevention strategies targeting this group should recognize its diversity.
"Continued and targeted investment will be essential to address the social isolation associated with singleness in old age"
The researchers note that the Korean government has already invested considerably in suicide prevention, including the 2011 Act, which sets out the responsibilities of the national and local governments. "Continued and targeted investment will be essential to address the social isolation associated with singleness in old age", they assert. The results are also relevant for other East Asian countries and the rest of the world, where suicide in old age is increasing in a context of ageing populations, educational expansion and declining marriage.
"In a country where the family –and not the state– has traditionally served as a social safety net in old age, these structural changes have profound implications for the well-being of the elderly", the authors conclude.
A country with a high suicide rate in the midst of family transition
South Korea has one of the highest suicide rates among OECD countries —20.6 per 100,000 inhabitants in 2021— compared to 11.1 in Europe, 14.1 in the United States, and 14.7 in neighbouring Japan. In Korea, among people aged 70 and over, the suicide rate peaked around 2010 with more than 100 suicides per 100,000 inhabitants, a figure unparalleled in any high-income countries.
At the same time, Korean society has undergone a radical transformation in the 20th century. The proportion of unmarried women at the age of 40 has increased from 7% in the cohort of 1970 to 18% in the cohort of 1980. Total fertility has fallen from 7 children per woman in the early 1970s to below 0.7 in 2023 —the lowest ever recorded in any country. And the proportion of men with secondary or higher education has increased from 46%, among the cohort born between 1931 and 1940, to 88-93%, among the cohort born between 1951 and 1960, and secondary education is now almost universal in all marital groups.
A study designed based on civil registry and census data
The researchers used data from the Korean civil registry and census for the years 2000, 2010 and 2020 to calculate suicide rates among people aged 60 to 69, broken down by sex, marital status (single, married, divorced or widowed) and educational level, in three cohorts (those born in the 1930s, 40s and 50s).
They then projected the future number of deaths by suicide among single people born between 1961 and 1985. This approach assumes constant suicide rates or hypothetical reductions consistent with the targets of the National Suicide Prevention Plan.
Reference article:
Liu C, Mogi R, Lee DS. "Older-age suicide amid changing family structures in South Korea: a demographic analysis using administrative data." BMJ Public Health 2026;4:e003294.
DOI: 10.1136/bmjph-2025-003294
URL: https://bmjpublichealth.bmj.com/content/4/1/e003294
Funding
This research was funded by the following research grants: BA/Leverhulme SRG 2024 242178 (Chia Liu), Beatriu de Pinós BP 2024 ref. BDNS 802038 (Ryohei Mogi) and the European Research Council (ERC) BIOSFER 101071773 (D. Susie Lee). The opinions expressed are solely those of the authors.
