To our knowledge, this is the first study to examine, by sex, and at global, region, and country/nation/territory levels, and over nearly three decades, both absolute and age-relative patterns of OA suicide. Based on past studies, we expected age-relative and absolute rates of OA suicide to provide different kinds of information. Our expectation was supported.
Between 1990 and 2019 the absolute number of OA suicides increased globally, even though both OA ASSR and the OA-to-NOA ASSR-ratio decreased substantially. These global patterns may be due to population growth and to changes in the population age-structure. Because the number of older adults and their proportion in the population are growing, the number of OA suicides will likely continue to grow.2 To better understand the patterns described above, a direction for future studies is to examine changes in the leading causes of OA death by country/nation/territory.
There was considerable variation in OA absolute and age-relative suicide-rates by world-region, country/nation/territory, by sex, and over time. This variability in OA-suicide-patterns, together with the substantial decrease in older-adult absolute and age-relative suicide-rates, challenges simplistic theories of older-adult-suicide, including the Anglophone-countries-based theory that OA suicide is a relatively understandable, if not an inevitable response to aging-related adversities (e.g., increases in illnesses and disabilities).14 Between 1990 and 2019 OA age-relative suicide-rates were highest in Sub-Saharan Africa and East Asia. OA age-relative suicide-rates were generally higher in countries with a lower socioeconomic-position. These findings challenge the dominant belief that OA suicide is a problem of high-socioeconomic-position countries.
OA absolute suicide-rates were significantly lower among women than among men in all regions. At the same time, OA age-relative suicide-rates were significantly higher in women than in men in many GBD regions. Also, in Southeast Asia, East Asia, and Oceania, OA women’s age-relative suicide-rates increased monotonously while OA men’s started to drop after 2010. Furthermore, in Sub-Saharan Africa, OA women’s age-relative suicide-rates kept growing, though with large fluctuation, while OA men’s age-relative suicide-rates increased monotonously. These findings indicate that in many GBD regions, particularly in lower socioeconomic-position regions, women’s likelihood of suicide is greater in older adulthood than prior to older adulthood; and also that, in those regions, OA women’s relative suicide-risk is on the increase. There are several possible explanations for these patterns. One is that there have been improvements in the recording of female suicide. Female suicide is less likely to be recognized as such, and/or to be reported than male suicide,15 especially in communities where women’s agency is systemically-restricted by men, and where female suicide is viewed as a form of defiance of male de-jure or de-facto ownership of women.16 Another explanation is that there has been an increase in female suicide in lower-socioeconomic-position regions. A study found that the relatively-high suicide-rates of women in low- and middle-income countries, as compared to the suicide-rates of women in high-income countries, are predicted by the greater institutional-discrimination that women experience in low- and middle-income countries–including restricted access to productive and financial assets and justice, and lesser family-law rights.17 Because the effects of institutional discrimination accumulate over the lifespan, institutional discrimination often weighs heavier on OA women. In an increasingly-connected world, recent generations of women living in low socioeconomic-position countries with high levels of institutional discrimination may be more aware than earlier generations of women of the human-rights violations that they experience, with suicide being their desperate protest against the human-rights abuses, at least, in countries/nations/territories where female suicide follows the protest-script.16,17
The findings that, in many GBD regions, OA women’s age-relative suicide-rates were significantly higher than OA men’s age-relative suicide-rates challenge the dominant belief that in late-adulthood, suicide is men’s problem. This belief is supported by the absolute OA suicide-rates—with OA men having higher absolute suicide-rates. The belief that suicide is men’s problem has become dominant also because men, particularly men of European descent, have the highest suicide-rates in high-income, Anglophone countries,14 given that studies from high-income, Anglophone countries are over-represented in the scientific literature. This study’s examination of both age-relative and absolute OA suicide-rates, across countries/nations/territories, regions, and globally, provides a window on OA women’s and men’s different ways of suicide-vulnerability.
Interpretations of this study’s findings require consideration of its method’s strengths and weaknesses. A strength is that it used GBD data that are comparable across country/nation/territory and region, and over time. Another strength is that it examined absolute and age-relative OA suicide-data by sex, across different scales of location.
Limitations of this study result from GBD-2019 data-gaps and data-quality variability. Data-gaps and data-quality variations are more likely in lower-socioeconomic-position countries more than in higher-socioeconomic-position countries. For example, many Sub-Saharan countries have limited vital-registration data. Data from neighboring countries are used to impute the missing data, leading to more homogenous estimates in Sub-Saharan Africa than in other regions.9
Another issue is selective suicide-underreporting and misclassification, by country/nation/territory. While suicide-underreporting and misclassification occur in all countries, they are more common in countries/nations/territories where suicide is subject to more negative cultural and religious sanctions.18 Also, as noted earlier, underreporting and misclassification tend to be more of a problem in terms of women’s suicides,15 especially in cultures where there is a strong prohibition of women’s suicide.16 For these reasons, the fact that the GBD definition of suicide encompasses both suicide and intentional self-harm-deaths is an asset of this study.
Other limitations have to do with construct operationalization. We set age 60 as the older-adulthood threshold, across countries. This decision was necessary but problematic. One reason is that there is substantial variability in longevity, by sex and by country/nation/territory. Another reason is that being age 60-and-older has different meanings and involves different experiences, depending on sex and culture, with implications for suicide.
Using the SDI as a measure of country/nation/territory socioeconomic-position is a limitation. The SDI is a composite measure of total fertility-rate for persons > 25-years, mean education-years for persons > 15-years, and per-capita income. Given the variability in women’s and men’s education, paid work, and income, by country/nation/territory and region, due to discrimination against women that varies by location, the SDI provides different information about women’s and men’s socioeconomic position, depending on location and culture. It is also a limitation to use, in a study of late-adulthood suicide, a socioeconomic-position measure that includes fertility as one of its three indices.