Association of Frailty and Suicide in Adults 65 Years and Older

Abstract Nearly 10,000 adults aged 65 years and older die by suicide in US annually. Although prior studies have linked individual diagnostic factors to late-life suicide risk, to our knowledge none have examined how accumulated health burden affects suicide risk. Such a metric could be studied utilizing a frailty index (FI). Our primary study objective was to determine the relationship of FI to risk of suicide. We examined a longitudinal cohort of 2,858,876 veterans 65 years and older from fiscal year 2012-2013 (baseline) through 12/31/2017, linking the VA’s suicide and mortality databases with medical record data. FI was defined by 31 variables, including morbidity, function, cognition, mood, sensory loss, chronic pain, and failure to thrive. We used Fine-Gray proportional hazards regression to examine time to suicide attempt (fatal and non-fatal). Our sample’s average age was 75 (SD 8), 88% White, 9% Black, and 98% male. Thirty-seven percent of veterans were non-frail, 30% were pre-frail, 17% mildly frail, 9% moderately frail, and 7% severely frail. Over the course of the study, 9,043 veterans had a documented suicide attempt with >60% fatal. After adjusting for race, gender, region, substance use disorder, and PTSD, risk of suicide attempt increased across frailty categories: Hazard ratios increased from 1.37 (95%CI: 1.30-1.45) for pre-frail individuals to 1.57 (1.43-1.72) for severely frail individuals. We found similar results after further adjustment for the Charlson Comorbidity Index, suggesting cumulative deficit FI may be a strong prognostic marker for risk of suicide in adults over 65; informing late-life suicide prevention efforts.

self-efficacy in discussing EOL care with family members and whether such an association differs by acculturation levels among older Chinese Americans.Data were collected from 207 Chinese Americans aged 65-102 in two metropolitan areas in 2017.Ordinary least squares regression was conducted to examine the association between family conflict, acculturation, and self-efficacy in discussing EOL care with family.Family conflict was negatively associated with older adults' self-efficacy in discussing EOL care with family.More specifically, the negative association between family conflict and self-efficacy in discussing EOL care with family members was more pronounced for those with higher levels of acculturation.Findings highlighted differential effects of family conflict on self-efficacy of EOL care plan discussion for older adults with different acculturation levels.Those with higher acculturation may be more independent in their EOL care planning and aware of the possible negative effects of family conflict in their EOL care planning discussions.Acculturation needs to be considered by geriatric health providers to develop family-centered interventions in improving end-of-life care planning for this population.Widowhood is associated with decreased emotional well-being, particularly increased depression.Religiosity may help improve mental health among widowed individuals.However, longitudinal studies exploring the role of religiosity on emotional well-being among widowed older adults is lacking, as are studies which examine this relationship using different dimensions of religiosity.This study analyzed data from the 2006-2016 waves of the nationally representative Health and Retirement Study (HRS).Trajectories of depression among older adults >50 years (N=5,486) were examined to explore patterns of depression among those entering widowhood and the potential impact of religiosity on depressive symptoms during widowhood.Ordinary least squares (OLS) regression analysis was used to examine the association between widowhood and depression as well as the role of religiosity as a moderator of this association.Older adults experienced an increase in depressive symptomology after the onset of widowhood, and although the levels of depressive symptomology decrease post-widowhood, they do not return to their pre-widowhood levels.Additionally, high religious service attendance and higher intrinsic religiosity were both associated with lower depressive symptomology.High religious service attendance moderated the relationship between widowhood and depression.The relationship between high religious service attendance and depression was stronger among widowed older adults living alone.This study highlights the long-term effects of widowhood on depressive symptomology among older adults.The findings also suggest that higher religious service attendance can lessen the effects of widowhood on depressive symptoms, especially for those living alone.These findings may inform intervention development around increased screening and treatment for depression.

8-YEAR CHANGES IN FRAILTY IN ADULTS: LINKS TO COGNITIVE AND PHYSICAL FUNCTION AND MORTALITY
Felicia Simpson, 1 Jamie Justice, 2 Judy Bahnson, 3 Joni Evans, 3 Kathleen Hayden, 3 Stephen Krtichevsky, 2 Karen Johnson, 4  Deficit accumulation frailty indices are being evaluated as clinical markers of biological aging.In this context, it is to be expected that changes over time in such indices should be predictive of downstream changes in cognition, physical function, and mortality.We derived a frailty index (FI) based on deficit accumulation in 38 functional, behavioral, and clinical characteristics and examined associations between 8-year changes in FI and subsequent standardized measures of cognitive and physical function and mortality collected over years 8-18.We drew data from the Look AHEAD clinical trial of a multidomain intensive lifestyle intervention (ILI) in 3841 adults, aged 45-76 years at baseline with overweight/obesity and type 2 diabetes mellitus.Greater FI increases tended to occur among individuals who were older, non-Hispanic White, heavier, and who had greater baseline multimorbidity.Greater increases in FI were associated with subsequently worse levels of composite cognitive function and 400m walk speed (all p<0.001).Additionally, compared with the lowest tertile of FI change, hazard ratios [95% confidence intervals] for 10-year mortality for the middle and highest tertiles of FI change were 1. 28 [1.03.1.58]and 1 .56[1.24,1.96],respectively.While assignment to ILI was associated with smaller 8-year increases in FI, this did not translate overall to better cognitive functioning compared to the Diabetes Support and Education control condition across years 8-18.Increase in FI over 8 years predicts subsequent reduced function and greater mortality.However, whether interventions generally targeting FI reduce risks for downstream outcomes remains to be seen.Nearly 10,000 adults aged 65 years and older die by suicide in US annually.Although prior studies have linked individual diagnostic factors to late-life suicide risk, to our knowledge none have examined how accumulated health burden affects suicide risk.Such a metric could be studied utilizing a frailty index (FI).Our primary study objective was to determine the relationship of FI to risk of suicide.We examined a longitudinal cohort of 2,858,876 veterans 65 years and older from fiscal year 2012-2013 (baseline) through 12/31/2017, linking the VA's suicide and mortality databases with medical record data.FI was defined by 31 variables, including morbidity, function, cognition, mood, sensory loss, chronic pain, and failure to thrive.We used Fine-Gray proportional hazards regression to examine time to suicide attempt (fatal and non-fatal).Our sample's average age was 75 (SD 8), 88% White, 9% Black, and 98% male.Thirty-seven percent of veterans were non-frail, 30% were pre-frail, 17% mildly frail, 9% moderately frail, and 7% severely frail.Over the course of the study, 9,043 veterans had a documented suicide attempt with >60% fatal.After adjusting for race, gender, region, substance use disorder, and PTSD, risk of suicide attempt increased across frailty categories: Hazard ratios increased from 1.37 (95%CI: 1.30-1.45)for pre-frail individuals to 1.57 (1.43-1.72)for severely frail individuals.We found similar results after further adjustment for the Charlson Comorbidity Index, suggesting cumulative deficit FI may be a strong prognostic marker for risk of suicide in adults over 65; informing late-life suicide prevention efforts.

FRAILTY AND MORTALITY IN A COMMUNITY-DWELLING RELATIVELY HEALTHY OLDER POPULATION
A R M Saifuddin Ekram, 1 Joanne Ryan, 2 Sara Espinoza, 3 Michael Ernst, 4 Anne Murray, 5 and Robyn Woods, 2 1. Monash University, Monash University, Melbourne,  Victoria, Australia, 2. Monash University, Melbourne,  Victoria, Australia, 3. University of Texas Health Science  Center San Antonio, San Antonio, Texas, United States,  4. University of Iowa, Iowa City, Iowa, United States,  5. Hennepin HealthCare Research Institute, Hennepin  HealthCare, Minneapolis, Minnesota, United States  This study examined factors associated with frailty and studied the association between frailty status and mortality in healthy community-dwelling older persons.Participants included 19,114 individuals from the "ASPirin in Reducing Events in the Elderly" (ASPREE) trial.Frailty was defined using modified Fried phenotype comprising exhaustion, body mass index, grip strength, gait speed and physical activity.A deficit accumulation frailty index (FI) using 66 items was also developed.Correlates of frailty were examined using multinomial logistic regression.The association between frailty status at baseline and mortality was analyzed using Cox regression.At baseline, 39.0% (95% CI: 38.3, 39.7) of participants were prefrail, and 2.2% (95% CI: 2.0, 2.4) were frail according to Fried phenotype, while 40.6% (95% CI: 40.0, 41.3) of participants were pre-frail and 8.1% (95% CI: 7.7, 8.5) were frail according to FI. Older age, female sex, lower education, African-American and Hispanic ethno-racial status, smoking, alcohol use, comorbidities, and polypharmacy were associated with frailty status.Pre-frailty increased risk of all-cause mortality significantly (Fried HR: 1.48; 95% CI: 1.28, 1.71; FI HR: 1.54; 95% CI: 1.31, 1.81); and the risk was even higher for frailty (Fried HR: 2.24; 95% CI: 1.67, 3.00; FI HR: 2.34; 95% CI: 1.83, 2.99) after adjustment for covariates.Cardiovascular disease (CVD) and non-CVD-related mortality showed similar trends.These results highlight a considerable burden of pre-frailty among a large group of community-dwelling, initially healthy older adults.Both Fried phenotype and deficit accumulation FI similarly predicted all-cause, CVD and non-CVD-related mortality in relatively healthy older adults.

IMPACT OF PHYSICAL AND SOCIAL FRAILTY ON THE UTILIZATION OF NURSING CARE SERVICES IN VERY OLD ADULTS Jaroslava Zimmermann, University of Cologne, Cologne, Nordrhein-Westfalen, Germany
Frailty, characterized by increased vulnerability to external stressors, has been found to increase the risk of healthcare utilization and nursing home admission.As the age group of 80 years or older remains frequently underrepresented in previous research, this study examined the impact of physical and social frailty on the utilization of nursing care services in very old population of North Rhine-Westphalia.Using data from a representative cross-sectional study, 1,577 communitydwelling and institutionalized individuals aged ≥80 years were included.Physical frailty was defined according to Fried's criteria (exhaustion, weight loss, low handgrip strength, low physical activity).Social frailty was measured with selfreported loneliness, social isolation, and time spent with others.The use of outpatient care services, day care, informal and inpatient care were considered.Multinomial regression was applied to investigate the impact of physical and social frailty on the use of outpatient and inpatient care services, controlling for relevant sociodemographic and health related characteristics.Compared to very old adults who did not use any care services, no association was found between frailty and the use of outpatient or informal care.Comparing nonusers of care services with institutionalized individuals, nursing home residents were less likely to experience physical frailty and pre-frailty, but were more likely to be socially isolated and to feel lonely.These findings suggest that physical frailty might have been successfully prevented in the context of institutional inpatient care.However, early identification and intervention focused on social inclusion of the institutionalized very old individuals are needed to reverse social frailty.

AN EVALUATION OF SOCIAL BRIDGING AND BONDING MECHANISMS IN THE ASSOCIATION OF SOCIAL NETWORKS AND COGNITIVE FUNCTION
Siyun Peng, 1 Brea Perry, 1 Adam Roth, 1 Max Coleman, 1 and Hope Sheean, 2 1. Indiana University, Bloomington, Indiana, United States, 2. Indiana University, Indianapolis, Indiana, United States Background and Objectives: Substantial evidence links social connectedness prospectively to cognitive aging Frances Hawes, 1 Jane Tavares, 2 Corina Ronneberg, 3 and Edward Miller, 4 1.University of Wisconsin-Eau Claire, eau claire, Wisconsin, United States, 2. LeadingAge LTSS Center @UMass Boston, Boston, Massachusetts, United States, 3. University of Illinois Chicago, Chicago, Illinois, United States, 4. University of Massachusetts Boston, Boston, Massachusetts, United States and Mark Espeland, 3 1.Department of Mathematics, Winston-Salem State University, Winston-Salem, North Carolina, United States, 2. Wake Forest School of Medicine, Wake Forest School of Medicine, North Carolina, United States, 3. Wake Forest School of Medicine, Winston-Salem, North Carolina, United States, 4. University of Tennessee Health Science Center, Memphis, Tennessee, United States Randall Kuffel, 1 Ruth Morin, 2 Kenneth Covinsky, 3 John Boscardin, 3 Yixia Li, 4 and Amy Byers, 3 1.San Francisco Veterans Affairs Health Care System, San Francisco, California, UnitedStates, 2. John D. Dingell  VA Medical Center, Detroit, Michigan, United States, 3.  University of California, San Francisco, San Francisco,  California, United States, 4. San Francisco VA Health Care  System, San Francisco, California, United States