Distinct Suicidal Thought Patterns Flag Those at Highest Risk

Batya Swift Yasgur, MA, LSW

February 28, 2023

Long-term assessment of suicide risk and ideation in older adults may help identify distinct ideation patterns and predict potential future suicidal behavior, new research suggests.

Investigators studied over 300 older adults, assessing suicidal ideation and behavior for up to 14 years at least once annually. They then identified four suicidal ideation profiles.

They found that risk of suicide attempt/death was almost five times greater for chronic severe ideators and more than three times greater for highly variable ideators compared with fast-remitting ideators. In turn, fast-remitting ideators were at higher risk in comparison to low/non-ideators with no attempts or suicide.

Chronic severe ideators also showed the most severe levels of dysfunction across personality, social characteristics, and impulsivity measures, while highly variable and fast-remitting ideators displayed more specific deficits.

"We identified longitudinal ideation profiles that convey differential risk of future suicidal behavior to help clinicians recognize high suicide risk patients for preventing suicide," lead author Hanga Galfalvy, PhD, associate professor, Department of Psychiatry, Vagelos College of Physicians and Surgeons, Columbia University Irving Medical Center, New York, told Medscape Medical News.

"Clinicians should repeatedly assess suicidal ideation and ask not only about current ideation but also about the worst ideation since the last visit [because] similar levels of ideation during a single assessment can belong to very different risk profiles," said Galfalvy, also a professor of biostatistics at the Mailman School of Public Health and a co-investigator in the Conte Center for Suicide Prevention at Columbia.

The study was published online February 15 in the Journal of Clinical Psychiatry.

Vulnerable Population

"Older adults in most countries, including the US, are at the highest risk of dying of suicide out of all age groups," said Galfalvy. "A significant number of depressed older adults experience thoughts of killing themselves, but fortunately, only a few transition from suicidal thoughts to behavior."

Senior author Katalin Szanto, MD, professor of psychiatry, University of Pittsburgh School of Medicine, told Medscape Medical News that currently established clinical and psychosocial suicide risk factors have "low predictive value and provide little insight into the high suicide rate in the elderly."

These traditional risk factors "poorly distinguish between suicide ideators and suicide attempters and do not take into consideration the heterogeneity of suicidal behavior," said Szanto, principal investigator at the University of Pittsburgh's Longitudinal research Program in Late-Life Suicide, where the study was conducted. "Suicidal ideation measured at one time point — current or lifetime — may not be enough to accurately predict suicide risk," the investigators note.

The current study, a collaboration between investigators from the Longitudinal Research Program in Late-Life Suicide and the Conte Center for Suicide Prevention, investigates "profiles of suicidal thoughts and behavior in patients with late-life depression over a longer period of time," Galfalvy said.

The researchers used latent profile analysis (LPA) in a cohort of adults with nonpsychotic unipolar depression (aged 50 – 93 years, n = 337, mean [AD] age, 65.12 [8.75] years) to "identify distinct ideation profiles and their clinical correlates" and to "test the profiles' association with the risk of suicidal behavior before and during follow-up."

LPA is "a data-driven method of grouping individuals into subgroups, based on quantitative characteristics," Galfalvy explained.

The LPA yielded four profiles of ideation:

Ideator subgroup Percentage of total sample
Low/non-ideators 22.8%
Chronic severe 27.6%
Highly variable 18.7%
Fast-remitting 30.9%


At baseline, the researchers assessed the presence or absence of suicidal behavior history and the number and lethality of attempts. They prospectively assessed suicidal ideation and attempts at least once annually thereafter over a period ranging from 3 months to 14 years (median, 3 [IQR, 1.6 – 4] years).

At baseline and at follow-ups, they assessed ideation severity.

  • Current ideation (within past week)

  • Worst lifetime ideation

  • Current ideation

  • Worst levels of ideation since past visit


They also assessed depression severity, impulsivity, and personality measures, as well as perception of social support, social problem solving, cognitive performance, and physical comorbidities.

Personalized Prevention

Of the original cohort, 92 patients died during the follow-up period, with 13 dying of suicide (or suspected suicide).

Over half (60%) of the chronic severe as well as the highly variable groups and almost half (48%) of the fast-remitting group had a history of past suicide attempt — all significantly higher than the low-non-ideators (0%).

Despite comparable current ideation severity at baseline, the risk of suicide attempt/death was greater for chronic severe ideators vs fast-remitting ideators, but not greater than for highly variable ideators. On the other hand, highly variable ideators were at greater risk, compared to fast-remitting ideators.

Ideator category Age-adjusted HR (95% CI) P value
Chronic severe vs fast-remitting 5.75 (2.25 – 1.57) < .001
Chronic severe vs highly variable 2.02 (.91 – 4.49) = .083
Highly variable vs fast-remitting 3.21 (1/03 – 10.1) = .045
Fast-remitting vs low/non-ideators No events of interest in low-ideators < .001


Cognitive factors "did not significantly discriminate between the ideation profiles, although…lower global cognitive performance predicted suicidal behavior during follow-up," the authors note.

This finding "aligns with prior studies indicating that late-life suicidal behavior but not ideation may be related to cognition...and instead, ideation and cognition may act as independent risk factors for suicidal behavior," they add.

"Patients in the fluctuating ideator group generally had moderate or high levels of worst suicidal ideation between visits, but not when asked about current ideation levels at the time of the follow-up assessment," Galfalvy noted. "For them, the time frame of the question made a difference as to the level of ideation reported."

The study "identified several clinical differences among these subgroups which could lead to more personalized suicide prevention efforts and further research into the heterogeneity of suicidal behavior," she suggested.

New Insight

Commenting for Medscape Medical News, Ari Cuperfain, MD, Temerty Faculty of Medicine, University of Toronto, Ontario, Canada, said the study "adds to the nuanced understanding of how changes in suicidal ideation over time can lead to suicidal actions and behavior."

The study "sheds light on the notion of how older adults who die by suicide can demonstrate a greater degree of premeditated intent relative to younger cohorts, with chronic severe ideators portending the highest risk for suicide in this sample," added Cuperfain, who was not involved with the current research.

"Overall, the paper highlights the importance of both screening for current levels of suicidal ideation in addition to the evolution of suicidal ideation in developing a risk assessment and in finding interventions to reduce this risk when it is most prominent," he stated.

The research was supported by the National Institutes of Health. The authors and Cuperfain have disclosed no relevant financial relationships.

J Clin Psychiatry. Published online February 15, 2023. Abstract

Batya Swift Yasgur, MA, LSW is a freelance writer with a counseling practice in Teaneck, NJ. She is a regular contributor to numerous medical publications, including Medscape and WebMD, and is the author of several consumer-oriented health books as well as Behind the Burqa: Our Lives in Afghanistan and How We Escaped to Freedom(the memoir of two brave Afghan sisters who told her their story).

For more Medscape Psychiatry news, join us on Facebook and Twitter.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.