Older adults have high rates of suicide (Global Burden of Disease Collaborative Network, 2018), and projections indicate subsequent cohorts of older adults will usher in even higher rates (Phillips, Reference Phillips2014). Thus, we can expect an increase in suicide deaths among older adults in coming decades. While there are promising approaches, there are no clear evidence-based interventions to prevent suicide in later life (Wallace et al., Reference Wallace, Miller, Fields, Xu and Mercado-Sierra2021).
In this issue, Ki et al. (Reference Ki, Lapierre, Gim, Hwang, Kang, Dargis, Jung, Koh and Mishara2024) propose that considering protective factors – not just risk factors – for suicide could identify novel intervention targets for prevention, representing a potential direction for prevention research. They present findings from a systematic review of psychosocial protective factors for suicide ideation, attempts, and deaths in older adults to describe the degree to which prior research has considered protective factors and to identify which protective factors, if any, are most reliably associated with reduced suicide risk.
Their systematic review included 70 quantitative studies, suggesting that protective factors have been considered in prior research, though less commonly than risk factors. They found that most studies that met their design requirements (e.g., psychological autopsy studies were excluded) examined suicide ideation, while fewer examined attempts and deaths. Their search indicated that protective factors examined in prior research were reliably associated with suicide ideation. These factors were perceived control, well-being and quality of life, life satisfaction, purpose-in-life, resilience, coping, religiosity, hope, sense of belonging, positive relationships, social support, social connectedness, and social participation. The authors note that results appeared strongest for purpose-in-life, resilience, and positive relationships, suggesting these protective factors may represent useful intervention targets.
Ki and colleagues' focus was on psychosocial protective factors, which they defined as positive psychology constructs. Their review identified several factors worthy of future study in observational studies and as intervention targets. They note that there are forms that protective factors may take. Along those lines, it is worth considering whether protective factors could usefully be defined in other ways – for example, as factors that, when in place, interrupt a trajectory toward suicide, regardless of whether they represent psychological or social strengths. As well, as the authors note, protective factors could function at various points along the trajectory toward suicide, including at the point of transition from ideation to attempt, as highlighted in “ideation to action” theories of suicide (Klonsky & May, Reference Klonsky and May2015; O’Connor, Reference O’Connor2011; Van Orden et al., Reference Van Orden, Witte, Cukrowicz, Braithwaite, Selby and Joiner2010). Considering characteristics of late-life suicide suggests several directions for exploring protective factors as those that interrupt the trajectory toward suicide ideation, attempts, and deaths in later life.
First, older adults are often seen in primary care in the weeks and months before their deaths, but often risk is not detected (Van Orden et al., Reference Van Orden, Silva, Conwell and Knight2019). This likely has many causes, including that older people are less likely to have a history of suicide attempts; older people are less likely to spontaneously disclose suicidal thoughts; and they frequently present with physical health concerns that do not indicate to a clinician that suicide risk should be assessed (Van Orden et al., Reference Van Orden, Silva, Conwell and Knight2019). Prior work on personality characteristics of older adults who die by suicide suggests that low openness to experience characterizes many older people who die by suicide (Duberstein et al., Reference Duberstein, Conwell and Caine1994), which could predispose an older person to be less likely to express their feelings making it less likely for others to detect distress and thus intervene (Duberstein, Reference Duberstein2001). In an earlier issue of IPG examining aspects of geriatric depression, Delhom et al. (Reference Delhom, Meléndez and Satorres2022) tested an intervention to increase emotional intelligence, which shares common aspects with openness to experience and is conceptualized as fostering use of adaptive coping strategies when faced with life stressors and negative mood. While not focused specifically on suicide risk, the authors note the relevance of emotional intelligence for suicide prevention, including prior studies linking low emotional intelligence to increased risk for suicidal behavior (Cha & Nock, Reference Cha and Nock2009). The intervention was provided in a group setting over 10 90-min sessions that provided psychoeducation on emotions as well as practice with emotion regulation skills and coping skills. Compared to control, older adults assigned to the emotional intelligence intervention demonstrated improvements in key dimensions of emotional intelligence – clarity (understanding of one’s emotions) and emotional repair (i.e., emotion regulation). Participants also improved on dimensions of adaptive coping assessed by self-report, including problem-solving, positive re-appraisal, and seeking social support. Most relevant to suicide prevention, participants assigned to the emotional intelligence intervention also demonstrated significant reductions in depressive symptoms and hopelessness, two characteristics of older people who die by suicide. The promise of this intervention may lie in its non-stigmatizing frame of increasing emotional intelligence to improve adaptive coping to aging-related stressors. This may be more acceptable to older adults compared to preventing depression or addressing loneliness, both of which carry stigma.
A second characteristic of suicide in later life is older adults most often die on their first attempt (Van Orden et al., Reference Van Orden, Silva, Conwell and Knight2019). Thus, any factors that might reduce the transition from ideation to attempt or capability to engage in lethal suicidal behavior could be important protective factors. Older adults are more likely to use lethal means and less likely to be discovered. As one example, in the U.S., most older people who die by suicide use a firearm (Van Orden et al., Reference Van Orden, Silva, Conwell and Knight2019); thus, an openness/willingness to use safe storage methods (e.g., locked safe, bullets stored separately) or willingness to have another person keep the firearm during times of stress are likely protective against a lethal suicide attempt. Research investigating psychosocial, social, and environmental predictors of willingness for means safety could be useful in assessing potential strengths to be capitalized on or targets for intervention to enhance protection against suicide.
A third characteristic of late-life suicide is that it occurs in the context of adult development and changes that accompany aging, including psychosocial strengths associated with aging. There is a rich gerontological and geropsychological literature documenting that the normative trajectory in later life is toward greater socio-emotional well-being, including greater attention to and experience of positive emotions, reduced reactivity to emotional stimuli, greater prioritization of emotionally meaningful goals, and use of flexible coping strategies (Charles et al., Reference Charles and Carstensen2010). Thus, preventing suicide in later life can be conceptualized as “course correcting” to return an older person to the normative developmental trajectory toward enhanced socio-emotional well-being. “Upstream suicide prevention” efforts that aim to prevent the development of suicide ideation given its greater lethality in later life are an essential component of multifaceted approaches to suicide prevention in later life. In that sense, late-life suicide prevention is in large part an effort to enhance protective factors such as those examined in the paper by Ki and colleagues (Reference Ki, Lapierre, Gim, Hwang, Kang, Dargis, Jung, Koh and Mishara2024).
A final characteristic of late-life suicide that is important to consider in the context of protective factors is social connection. Social disconnection – including social isolation, low social support, and loneliness (low belonging) – is one of the most consistent characteristics of older people who die by suicide, present in 70–90% of older adults who die by suicide (Duberstein et al., Reference Duberstein, Conwell, Conner, Eberly, Evinger and Caine2004) and listed as a key circumstance contributing to suicide by coroner/medical examiner/law enforcement reports in 39-42% of suicide deaths (Stone et al., Reference Stone, Simon, Fowler, Kegler, Yuan, Holland, Ivey-Stephenson and Crosby2018). In contrast, social connection is protective: socially connected adults have greater than a two-fold reduction in suicide risk over the long-term (20 years) (Tsai et al., Reference Tsai, Lucas, Sania, Kim and Kawachi2014, Reference Tsai, Lucas and Kawachi2015). Ki and colleagues raise the question of whether a factor that can be conceptualized regarding both its presence and absence, such as social connection, should be considered as a protective factor. For example, is social connection a protective factor and social disconnection a risk factor? Their results indicate that positive social relationships are reliably associated with lower suicide ideation. However, social isolation is strongly associated with increased risk for suicide deaths and low belonging is posited as a key cause of suicide ideation by several psychological theories of suicide (with empirical support) (Chu et al., Reference Chu, Buchman-Schmitt, Stanley, Hom, Tucker, Hagan, Rogers, Podlogar, Chiurliza, Ringer, Michaels, Patros and Joiner2017). It may be useful to categorize protective factors not regarding positive or negative valence, but as non-causal factors that, when present, can decrease risk by attenuating the strength or frequency of causal factors, akin to a moderator in statistical analysis. In this way, a protective factor could be defined by the function it serves in the etiology of suicide rather than its valence. Using this logic, one could conceptualize the broad construct of social connection as both a causal factor (e.g., unmet need to belong) and a protective factor (e.g., living with a spouse to support means safety and supportive friends who notice distress), depending on which dimension of social connection is considered.
This logic could be applied to factors associated with suicide risk in later life that have been studied in prior issues of IPG, including sleep quality. Poor sleep is reliably associated with suicide ideation, attempts, and deaths across the lifespan. To the extent that poor sleep may play a role in the causal process toward suicide risk, treating poor sleep may function as reducing a suicide risk factor. Also possible is that high-quality sleep could attenuate the impact of drivers of suicide risk, such as perceptions of being a burden and hopelessness, by providing rest and a reprieve from these states and fostering the capacity for emotion regulation. Cognitive performance is associated with poor sleep, as addressed by Zuidersma and colleagues (De Deyn et al., Reference Zuidersma, Lugtenburg, van Zelst, Reesink, De Deyn, Strijkert, Zuidema and Oude Voshaar2022) (though their findings highlight that the direction of the effect may differ across individuals). It is also interesting to consider the role of cognitive performance and function through the lens of risk/protective factors for suicide. Some dimensions of impaired cognition, such as cognitive control, may function as risk factors for the transition from ideation to attempt in later life. However, the association between dementia and suicidal behavior is not yet fully understood. Given that lethal suicide attempts require a degree of cognitive performance to enact – dying by suicide is not an easy thing to do (Van Orden et al., Reference Van Orden, Witte, Cukrowicz, Braithwaite, Selby and Joiner2010) – advanced dementia is in some sense protective against dying by suicide. The definition of protective factors used by Ki and colleagues would certainly not include dementia as a protective factor, as it is not a psychosocial strength. Rather, dementia might be considered a factor that blocks capability for suicide. Alternatively, positive psychological strengths could be considered one domain among a diverse range of factors that protect against suicide. These distinctions may be important when testing causal models or in conceptualizing suicide risk interventions as targeting causal mechanisms versus bolstering strengths to keep someone alive while causal mechanisms are targeted and treated, which in some cases, may take time.
A final consideration regarding protective factors for suicide is that they may be more acceptable intervention targets for older people, increasing the number who may be willing to engage. Given that older adults are less likely to seek out specialty mental health care, interventions that address factors other than mental health disorders and can be delivered in community settings may be more acceptable and feasible and thereby potent interventions to prevent suicide and promote well-being in later life. In this way, suicide prevention interventions for older adults should address and capitalize upon aging-associated strengths to allow those at risk to return to the normative developmental trajectory toward increased socio-emotional health and well-being in later life.
You have
Access
Older adults have high rates of suicide (Global Burden of Disease Collaborative Network, 2018), and projections indicate subsequent cohorts of older adults will usher in even higher rates (Phillips, Reference Phillips2014). Thus, we can expect an increase in suicide deaths among older adults in coming decades. While there are promising approaches, there are no clear evidence-based interventions to prevent suicide in later life (Wallace et al., Reference Wallace, Miller, Fields, Xu and Mercado-Sierra2021).
In this issue, Ki et al. (Reference Ki, Lapierre, Gim, Hwang, Kang, Dargis, Jung, Koh and Mishara2024) propose that considering protective factors – not just risk factors – for suicide could identify novel intervention targets for prevention, representing a potential direction for prevention research. They present findings from a systematic review of psychosocial protective factors for suicide ideation, attempts, and deaths in older adults to describe the degree to which prior research has considered protective factors and to identify which protective factors, if any, are most reliably associated with reduced suicide risk.
Their systematic review included 70 quantitative studies, suggesting that protective factors have been considered in prior research, though less commonly than risk factors. They found that most studies that met their design requirements (e.g., psychological autopsy studies were excluded) examined suicide ideation, while fewer examined attempts and deaths. Their search indicated that protective factors examined in prior research were reliably associated with suicide ideation. These factors were perceived control, well-being and quality of life, life satisfaction, purpose-in-life, resilience, coping, religiosity, hope, sense of belonging, positive relationships, social support, social connectedness, and social participation. The authors note that results appeared strongest for purpose-in-life, resilience, and positive relationships, suggesting these protective factors may represent useful intervention targets.
Ki and colleagues' focus was on psychosocial protective factors, which they defined as positive psychology constructs. Their review identified several factors worthy of future study in observational studies and as intervention targets. They note that there are forms that protective factors may take. Along those lines, it is worth considering whether protective factors could usefully be defined in other ways – for example, as factors that, when in place, interrupt a trajectory toward suicide, regardless of whether they represent psychological or social strengths. As well, as the authors note, protective factors could function at various points along the trajectory toward suicide, including at the point of transition from ideation to attempt, as highlighted in “ideation to action” theories of suicide (Klonsky & May, Reference Klonsky and May2015; O’Connor, Reference O’Connor2011; Van Orden et al., Reference Van Orden, Witte, Cukrowicz, Braithwaite, Selby and Joiner2010). Considering characteristics of late-life suicide suggests several directions for exploring protective factors as those that interrupt the trajectory toward suicide ideation, attempts, and deaths in later life.
First, older adults are often seen in primary care in the weeks and months before their deaths, but often risk is not detected (Van Orden et al., Reference Van Orden, Silva, Conwell and Knight2019). This likely has many causes, including that older people are less likely to have a history of suicide attempts; older people are less likely to spontaneously disclose suicidal thoughts; and they frequently present with physical health concerns that do not indicate to a clinician that suicide risk should be assessed (Van Orden et al., Reference Van Orden, Silva, Conwell and Knight2019). Prior work on personality characteristics of older adults who die by suicide suggests that low openness to experience characterizes many older people who die by suicide (Duberstein et al., Reference Duberstein, Conwell and Caine1994), which could predispose an older person to be less likely to express their feelings making it less likely for others to detect distress and thus intervene (Duberstein, Reference Duberstein2001). In an earlier issue of IPG examining aspects of geriatric depression, Delhom et al. (Reference Delhom, Meléndez and Satorres2022) tested an intervention to increase emotional intelligence, which shares common aspects with openness to experience and is conceptualized as fostering use of adaptive coping strategies when faced with life stressors and negative mood. While not focused specifically on suicide risk, the authors note the relevance of emotional intelligence for suicide prevention, including prior studies linking low emotional intelligence to increased risk for suicidal behavior (Cha & Nock, Reference Cha and Nock2009). The intervention was provided in a group setting over 10 90-min sessions that provided psychoeducation on emotions as well as practice with emotion regulation skills and coping skills. Compared to control, older adults assigned to the emotional intelligence intervention demonstrated improvements in key dimensions of emotional intelligence – clarity (understanding of one’s emotions) and emotional repair (i.e., emotion regulation). Participants also improved on dimensions of adaptive coping assessed by self-report, including problem-solving, positive re-appraisal, and seeking social support. Most relevant to suicide prevention, participants assigned to the emotional intelligence intervention also demonstrated significant reductions in depressive symptoms and hopelessness, two characteristics of older people who die by suicide. The promise of this intervention may lie in its non-stigmatizing frame of increasing emotional intelligence to improve adaptive coping to aging-related stressors. This may be more acceptable to older adults compared to preventing depression or addressing loneliness, both of which carry stigma.
A second characteristic of suicide in later life is older adults most often die on their first attempt (Van Orden et al., Reference Van Orden, Silva, Conwell and Knight2019). Thus, any factors that might reduce the transition from ideation to attempt or capability to engage in lethal suicidal behavior could be important protective factors. Older adults are more likely to use lethal means and less likely to be discovered. As one example, in the U.S., most older people who die by suicide use a firearm (Van Orden et al., Reference Van Orden, Silva, Conwell and Knight2019); thus, an openness/willingness to use safe storage methods (e.g., locked safe, bullets stored separately) or willingness to have another person keep the firearm during times of stress are likely protective against a lethal suicide attempt. Research investigating psychosocial, social, and environmental predictors of willingness for means safety could be useful in assessing potential strengths to be capitalized on or targets for intervention to enhance protection against suicide.
A third characteristic of late-life suicide is that it occurs in the context of adult development and changes that accompany aging, including psychosocial strengths associated with aging. There is a rich gerontological and geropsychological literature documenting that the normative trajectory in later life is toward greater socio-emotional well-being, including greater attention to and experience of positive emotions, reduced reactivity to emotional stimuli, greater prioritization of emotionally meaningful goals, and use of flexible coping strategies (Charles et al., Reference Charles and Carstensen2010). Thus, preventing suicide in later life can be conceptualized as “course correcting” to return an older person to the normative developmental trajectory toward enhanced socio-emotional well-being. “Upstream suicide prevention” efforts that aim to prevent the development of suicide ideation given its greater lethality in later life are an essential component of multifaceted approaches to suicide prevention in later life. In that sense, late-life suicide prevention is in large part an effort to enhance protective factors such as those examined in the paper by Ki and colleagues (Reference Ki, Lapierre, Gim, Hwang, Kang, Dargis, Jung, Koh and Mishara2024).
A final characteristic of late-life suicide that is important to consider in the context of protective factors is social connection. Social disconnection – including social isolation, low social support, and loneliness (low belonging) – is one of the most consistent characteristics of older people who die by suicide, present in 70–90% of older adults who die by suicide (Duberstein et al., Reference Duberstein, Conwell, Conner, Eberly, Evinger and Caine2004) and listed as a key circumstance contributing to suicide by coroner/medical examiner/law enforcement reports in 39-42% of suicide deaths (Stone et al., Reference Stone, Simon, Fowler, Kegler, Yuan, Holland, Ivey-Stephenson and Crosby2018). In contrast, social connection is protective: socially connected adults have greater than a two-fold reduction in suicide risk over the long-term (20 years) (Tsai et al., Reference Tsai, Lucas, Sania, Kim and Kawachi2014, Reference Tsai, Lucas and Kawachi2015). Ki and colleagues raise the question of whether a factor that can be conceptualized regarding both its presence and absence, such as social connection, should be considered as a protective factor. For example, is social connection a protective factor and social disconnection a risk factor? Their results indicate that positive social relationships are reliably associated with lower suicide ideation. However, social isolation is strongly associated with increased risk for suicide deaths and low belonging is posited as a key cause of suicide ideation by several psychological theories of suicide (with empirical support) (Chu et al., Reference Chu, Buchman-Schmitt, Stanley, Hom, Tucker, Hagan, Rogers, Podlogar, Chiurliza, Ringer, Michaels, Patros and Joiner2017). It may be useful to categorize protective factors not regarding positive or negative valence, but as non-causal factors that, when present, can decrease risk by attenuating the strength or frequency of causal factors, akin to a moderator in statistical analysis. In this way, a protective factor could be defined by the function it serves in the etiology of suicide rather than its valence. Using this logic, one could conceptualize the broad construct of social connection as both a causal factor (e.g., unmet need to belong) and a protective factor (e.g., living with a spouse to support means safety and supportive friends who notice distress), depending on which dimension of social connection is considered.
This logic could be applied to factors associated with suicide risk in later life that have been studied in prior issues of IPG, including sleep quality. Poor sleep is reliably associated with suicide ideation, attempts, and deaths across the lifespan. To the extent that poor sleep may play a role in the causal process toward suicide risk, treating poor sleep may function as reducing a suicide risk factor. Also possible is that high-quality sleep could attenuate the impact of drivers of suicide risk, such as perceptions of being a burden and hopelessness, by providing rest and a reprieve from these states and fostering the capacity for emotion regulation. Cognitive performance is associated with poor sleep, as addressed by Zuidersma and colleagues (De Deyn et al., Reference Zuidersma, Lugtenburg, van Zelst, Reesink, De Deyn, Strijkert, Zuidema and Oude Voshaar2022) (though their findings highlight that the direction of the effect may differ across individuals). It is also interesting to consider the role of cognitive performance and function through the lens of risk/protective factors for suicide. Some dimensions of impaired cognition, such as cognitive control, may function as risk factors for the transition from ideation to attempt in later life. However, the association between dementia and suicidal behavior is not yet fully understood. Given that lethal suicide attempts require a degree of cognitive performance to enact – dying by suicide is not an easy thing to do (Van Orden et al., Reference Van Orden, Witte, Cukrowicz, Braithwaite, Selby and Joiner2010) – advanced dementia is in some sense protective against dying by suicide. The definition of protective factors used by Ki and colleagues would certainly not include dementia as a protective factor, as it is not a psychosocial strength. Rather, dementia might be considered a factor that blocks capability for suicide. Alternatively, positive psychological strengths could be considered one domain among a diverse range of factors that protect against suicide. These distinctions may be important when testing causal models or in conceptualizing suicide risk interventions as targeting causal mechanisms versus bolstering strengths to keep someone alive while causal mechanisms are targeted and treated, which in some cases, may take time.
A final consideration regarding protective factors for suicide is that they may be more acceptable intervention targets for older people, increasing the number who may be willing to engage. Given that older adults are less likely to seek out specialty mental health care, interventions that address factors other than mental health disorders and can be delivered in community settings may be more acceptable and feasible and thereby potent interventions to prevent suicide and promote well-being in later life. In this way, suicide prevention interventions for older adults should address and capitalize upon aging-associated strengths to allow those at risk to return to the normative developmental trajectory toward increased socio-emotional health and well-being in later life.
Conflicts of interest
Kim Van Orden is responsible for the entire manuscript and has no conflicts of interest to disclose.
Source of funding
This paper is supported by a grant from the National Institute of Mental Health (R61MH129476, PI, Van Orden) and National Institute on Aging (R01AG082882, PI, Van Orden).