Predictors of suicidal ideation in older people: a decision tree analysis

Constructing prediction models and analyzing factors in suicidal ideation using machine learning, focusing on the older population

Suicide among the older population is a significant public health concern in South Korea. As the older individuals have long considered suicide before committing suicide trials, it is important to analyze the suicidal ideation that precedes the suicide attempt for intervention. In this study, six machine learning algorithms were employed to construct a predictive model for suicidal thinking and identify key variables. A traditional logistic regression analysis was supplementarily conducted to test the robustness of the results of machine learning. All analyses were conducted using a hierarchical approach to compare the model fit of each model in both machine learning and logistic regression. Three models were established for analysis. In Model 1, socioeconomic, residential, and health behavioral factors were incorporated. Model 2 expanded upon Model 1 by integrating physical health status, and Model 3 further incorporated mental health conditions. The results indicated that the gradient boosting algorithm outperformed the other machine learning techniques. Furthermore, the household income quintile was the most important feature in Model 1, followed by subjective health status, oral health, and exercise ability in Model 2, and anxiety and depression in Model 3. These results correspond to those of the hierarchical logistic regression. Notably, economic and residential vulnerabilities are significant factors in the mental health of the older population with higher instances of suicidal thoughts. This hierarchical approach could reveal the potential target population for suicide interventions.

Using recursive partitioning to predict presence and severity of suicidal ideation amongst college students

OBJECTIVE

Predicting the presence and severity of suicidal ideation in college students is important, as deaths by suicide amongst young adults have increased in the past 20 years.

PARTICIPANTS

We recruited college students (N = 5494) from ten universities across eight states.

METHOD

Participants answered three questionnaires related to lifetime and past month suicidal ideation, and an indicator of suicidal ideation in a DSM-5 symptom measure. We used recursive partitioning to predict the presence, absence, and severity, of suicidal ideation.

RESULTS

Recursive partitioning models varied in their accuracy and performance. The best-performing model consisted of predictors and outcomes measured by the DSM-5 Level 1 Cross-Cutting Symptom Measure. Sexual orientation was also an important predictor in most models.

CONCLUSIONS

A single measure of DSM-5 symptom severity may help universities understand suicide severity to promote targeted interventions. Though further work is needed, as similar scaling amongst predictors could have influenced the model.

Exploring risk factors and their differences on suicidal ideation and suicide attempts among depressed adolescents based on decision tree model

BACKGROUND

Suicide has been recognized as a major global public health issue. Depressed adolescents are more prone to experiencing it. We explore risk factors and their differences on suicidal ideation and suicide attempts to further enhance our understanding of suicidal behavior.

METHODS

2343 depressed adolescents aged 12-18 from 9 provinces/cities in China participated in this cross-sectional study. We utilized decision tree model, incorporating 32 factors encompassing participants' suicidal behavior. The feature importance of each factor was measured using Gini coefficients.

RESULTS

The decision tree model demonstrated a good fit with high accuracy (SI = 0.86, SA = 0.85 and F-Score (SI = 0.85, SA = 0.83). The predictive importance of each factor varied between groups with suicidal ideation and with suicide attempts. The most significant risk factor in both groups was depression (SI = 16.7 %, SA = 19.8 %). However, factors such as academic stress (SI = 7.2 %, SA = 1.6 %), hopelessness (SI = 9.1 %, SA = 5.0 %), and age (SI = 7.1 %, SA = 3.2 %) were more closely associated with suicidal ideation than suicide attempts. Factors related to the schooling status (SI = 3.5 %, SA = 10.1 %), total years of education (SI = 2.6 %, SA = 8.6 %), and loneliness (SI = 2.3 %, SA = 7.4 %) were relatively more important in the suicide attempt stage compared to suicidal ideation.

LIMITATIONS

The cross-sectional design limited the ability to capture changes in suicidal behavior among depressed adolescents over time. Possible bias may exist in the measurement of suicidal ideation.

CONCLUSION

The relative importance of each risk factor for suicidal ideation and attempted suicide varies. These findings provide further empirical evidence for understanding suicide behavior. Targeted treatment measures should be taken for different stages of suicide in clinical interventions.

Inequities in Mental Health Care Facing Racialized Immigrant Older Adults With Mental Disorders Despite Universal Coverage: A Population-Based Study in Canada

OBJECTIVES

Contemporary immigration scholarship has typically treated immigrants with diverse racial backgrounds as a monolithic population. Knowledge gaps remain in understanding how racial and nativity inequities in mental health care intersect and unfold in midlife and old age. This study aims to examine the joint impact of race, migration, and old age in shaping mental health treatment.

METHODS

Pooled data were obtained from the Canadian Community Health Survey (2015-2018) and restricted to respondents (aged ≥45 years) with mood or anxiety disorders (n = 9,099). Multivariable logistic regression was performed to estimate associations between race-migration nexus and past-year mental health consultations (MHC). Classification and regression tree (CART) analysis was applied to identify intersecting determinants of MHC.

RESULTS

Compared to Canadian-born Whites, racialized immigrants had greater mental health needs: poor/fair self-rated mental health (odds ratio [OR] = 2.23, 99% confidence interval [CI]: 1.67-2.99), perceived life stressful (OR = 1.49, 99% CI: 1.14-1.95), psychiatric comorbidity (OR = 1.42, 99% CI: 1.06-1.89), and unmet needs for care (OR = 2.02, 99% CI: 1.36-3.02); in sharp contrast, they were less likely to access mental health services across most indicators: overall past-year MHC (OR = 0.54, 99% CI: 0.41-0.71) and consultations with family doctors (OR = 0.67, 99% CI: 0.50-0.89), psychologists (OR = 0.54, 99% CI: 0.33-0.87), and social workers (OR = 0.37, 99% CI: 0.21-0.65), with the exception of psychiatrist visits (p = .324). The CART algorithm identifies three groups at risk of MHC service underuse: racialized immigrants aged ≥55 years, immigrants without high school diplomas, and linguistic minorities who were home renters.

DISCUSSION

To safeguard health care equity for medically underserved communities in Canada, multisectoral efforts need to guarantee culturally responsive mental health care, multilingual services, and affordable housing for racialized immigrant older adults with mental disorders.

NSSI questionnaires revisited: A data mining approach to shorten the NSSI questionnaires

BACKGROUND AND OBJECTIVE

Non-suicidal self-injury (NSSI) is a psychological disorder that the sufferer consciously damages their body tissues, often too severe that requires intensive care medicine. As some individuals hide their NSSI behaviors, other people can only identify them if they catch them while injuring, or via dedicated questionnaires. However, questionnaires are long and tedious to answer, thus the answers might be inconsistent. Hence, in this study for the first time, we abstracted a larger questionnaire (of 662 items in total) to own only 22 items (questions) via data mining techniques. Then, we trained several machine learning algorithms to classify individuals based on their answers into two classes.

METHODS

Data from 277 previously-questioned participants is used in several data mining methods to select features (questions) that highly represent NSSI, then 245 different people were asked to participate in an online test to validate those features via machine learning methods.

RESULTS

The highest accuracy and F1 score of the selected features-via the Genetics algorithm-are 80.0% and 74.8% respectively for a Random Forest algorithm. Cronbach's alpha of the online test (validation on the selected features) is 0.82. Moreover, results suggest that an MLP can classify participants into two classes of NSSI Positive and NSSI Negative with 83.6% accuracy and 83.7% F1-score based on the answers to only 22 questions.

CONCLUSION

While previously psychologists used many combined questionnaires to see whether someone is involved in NSSI, via various data mining methods, the present study showed that only 22 questions are enough to predict if someone is involved or not. Then different machine learning algorithms were utilized to classify participants based on their NSSI behaviors, among which, an MLP with 10 hidden layers had the best performance.

Factors Related to Suicidal Ideation and Prediction of High-Risk Groups among Youngest-Old Adults in South Korea

(1) Background: The suicide of older adults shows different factors between the youngest-old adults and the old-old adults. This study aimed to identify factors predicting suicidal ideation among youngest-old adults (ages 65 to 74 years) and predict high-risk groups’ characteristics. (2) Methods: The subjects of this study were 970 youngest-old adults who participated in the Korean National Health and Nutrition Examination Survey (KNHANES VIII Year 1, 2019). Logistic regression analysis identified factors related to suicidal ideation, and decision tree analysis identified combined characteristics among high-risk groups. Data were analyzed using SPSS 27.0. (3) Results: Suicidal ideation became more common among those with relatively lower income levels (OR = 1.48, 95% CI = 1.04−2.12), those whom had experienced depression (OR = 9.28, 95% CI = 4.57−18.84), those with relatively higher stress levels (OR = 2.42, 95% CI = 1.11−5.28), and those reporting a relatively worse perceived health (OR = 1.88, 95% CI = 1.23−3.11). Complex characteristics that combined depression, low personal income level, and low perceived health predicted a high risk of suicidal ideation (64.6%, p < 0.05). (4) Conclusions: The findings indicate that this high-risk group should be prioritized when developing suicide prevention strategies.

Screening for suicide risk among older adults: assessing preliminary psychometric properties of the Brief Geriatric Suicide Ideation Scale (BGSIS) and the GSIS-Screen

OBJECTIVES

To initially assess psychometric properties of two abbreviated versions of the Geriatric Suicide Ideation Scale (GSIS): a 10-item Brief Geriatric Suicide Ideation Scale (BGSIS), and a 5-item Geriatric Suicide Ideation Scale-Screen (GSIS-Screen).

METHODS

A series of psychometric analyses was conducted, assessing the internal consistency, test-retest reliability, construct and predictive validity of the abbreviated GSIS scales. This was done by selecting-out GSIS items from a combined dataset of studies on suicide ideation in older adults: 1) The GSIS scale development study (n = 107); 2) A clinical trial of Interpersonal Psychotherapy (IPT) modified for suicidal older adults(n = 25); 3) A longitudinal study of risk and resiliency to suicide ideation in community-residing older adults (n = 173).

RESULTS

Overall findings demonstrated strong internal consistency, test-retest reliability, concurrent and predictive validity for the BGSIS and GSIS-Screen with older adults across community, clinical, and residential settings.

CONCLUSION

Study findings support the use of the abbreviated GSIS scales when conducting research on suicide risk identification among older adults. Future research is recommended testing these scales prospectively in public health, residential, and clinical settings, in research and healthcare delivery contexts.

The Development of a Suicidal Ideation Predictive Model for Community-Dwelling Elderly Aged >55 Years

PURPOSE

Suicide is an important health and social concern worldwide. Both suicidal ideation and suicide rates are higher in the elderly population than in other age groups; thus, more careful attention and targeted interventions are required. Therefore, we have developed a model to predict suicidal ideation in the community-dwelling elderly aged of >55 years.

PATIENTS AND METHODS

A random forest algorithm was applied to those who participated in the Korea Welfare Panel. We used a total of 26 variables as potential predictors. To resolve the imbalance in the dataset resulting from the low frequency of suicidal ideation, training was performed by applying the synthetic minority oversampling technique. The performance index was calculated by applying the predictive model to the test set, which was not included in the training process.

RESULTS

A total of 6410 elderly Korean aged of >55 (mean, 71.48; standard deviation, 9.56) years were included in the analysis, of which 2.7% had suicidal ideation. The results for predicting suicidal ideation using the 26 chosen variables showed an AUC of 0.879, accuracy of 0.871, sensitivity of 0.750, and specificity of 0.874. The most significant variable in the predictive model was the severity of depression, followed by life satisfaction and self-esteem factors. Basic demographic variables such as age and gender demonstrated a relatively small effect.

CONCLUSION

Machine learning can be used to create algorithms for predicting suicidal ideation in community-dwelling elderly. However, there are limitations to predicting future suicidal ideation. A predictive model that includes both biological and cognitive indicators should be created in the future.

Suicide in obstructive lung, cardiovascular and oncological disease

Healthcare institutions face increasing demands stemming from the burden of noncommunicable diseases. The personal, social, financial and societal impact of these diseases are well-documented. However, the mental health concerns and trajectories of patients afflicted by chronic medical diseases have been under-recognized and are under-resourced. Despite that chronic diseases are associated with substantially increased risk of suicide, the medical world has largely failed to properly address suicide in the medically ill. Considering their high prevalence and mortality rate, this review article will highlight the mental health burden and suicide risk in obstructive lung, cardiovascular (including stroke) and oncological disease, in light of relevant data and conceptual models of suicide. Finally, general evidence-based suicide intervention strategies and potential selective adaptation of these strategies to the chronic medically ill patient populations and medical settings will be reviewed.

A scoping review on the use of machine learning in research on social determinants of health: Trends and research prospects

Background

Machine learning (ML) has spread rapidly from computer science to several disciplines. Given the predictive capacity of ML, it offers new opportunities for health, behavioral, and social scientists. However, it remains unclear how and to what extent ML is being used in studies of social determinants of health (SDH).

Methods

Using four search engines, we conducted a scoping review of studies that used ML to study SDH (published before May 1, 2020). Two independent reviewers analyzed the relevant studies. For each study, we identified the research questions, Results, data, and algorithms. We synthesized our findings in a narrative report.

Results

Of the initial 8097 hits, we identified 82 relevant studies. The number of publications has risen during the past decade. More than half of the studies (n = 46) used US data. About 80% (n = 66) utilized surveys, and 70% (n = 57) employed ML for common prediction tasks. Although the number of studies in ML and SDH is growing rapidly, only a few studies used ML to improve causal inference, curate data, or identify social bias in predictions (i.e., algorithmic fairness).

Conclusions

While ML equips researchers with new ways to measure health outcomes and their determinants from non-conventional sources such as text, audio, and image data, most studies still rely on traditional surveys. Although there are no guarantees that ML will lead to better social epidemiological research, the potential for innovation in SDH research is evident as a result of harnessing the predictive power of ML for causality, data curation, or algorithmic fairness.

Predicting Suicidal Behavior Without Asking About Suicidal Ideation: Machine Learning and the Role of Borderline Personality Disorder Criteria

OBJECTIVE

Identifying predictors contributing to suicide risk could help prevent suicides via targeted interventions. However, using only known risk factors may not yield accurate enough results. Furthermore, risk models typically rely on suicidal ideation, even though people often withhold this information.

METHOD

This study examined the contribution of various predictors to the accuracy of six machine learning models for identifying suicidal behavior in a prison population (n = 353), including borderline personality disorder (BPD) and antisocial personality disorder (APD) criteria, and compared how excluding data about suicidal ideation affects accuracy.

RESULTS

Results revealed that gradient tree boosting accurately identified individuals with suicidal behavior, even without relying on questions about suicidal ideation (AUC = 0.875, F1 = 0.846). Furthermore, the model maintained this accuracy with only 29 predictors. Meeting five or more diagnostic criteria of BPD was an important risk factor for suicidal behavior. APD criteria, in the presence of other predictors, did not substantially improve accuracy. Additionally, it may be possible to implement a decision tree model to assess individuals at risk of suicide, without focusing upon suicidal ideation.

CONCLUSIONS

These findings highlight that modern classification algorithms do not necessarily require information about suicidal ideation for modeling suicide and self-harm behavior.

Early Trauma and Relationships among Recent Stress, Depressive Symptoms, Anxiety Symptoms, and Suicidal Ideation in Korean Women

BACKGROUND

Evidence continues to accumulate that the presence or absence of early trauma (ET) implies unique characteristics in the relationships between suicidal ideation and its risk factors. We examined the relationships among recent stress, depressive symptoms, anxiety symptoms, and suicidal ideation in Korean suicidal women with or without such a history.

METHODS

Using data on suicidal adult females, 217 victims and 134 non-victims of ET, from the Korean Cohort for the Model Predicting a Suicide and Suicide-related Behavior, we performed structural equation modeling to investigate the contribution of recent stress, depressive symptoms, and anxiety symptoms on suicidal ideation within each group according to the presence or absence of a history of ET.

RESULTS

Structural equation modeling with anxiety and depressive symptoms as potential mediators showed a good fit. Recent stress had a direct effect on both depressive symptoms and anxiety symptoms in both groups. Only anxiety symptoms for victims of ET (standardized regression weight, 0.281; P = 0.005) and depressive symptoms for non-victims of ET (standardized regression weight, 0.326; P = 0.003) were full mediators that increased suicidal ideation. Thus, stress contributed to suicidal ideation by increasing the level of anxiety and depressive symptoms for victims and non-victims, respectively.

CONCLUSION

Tailored strategies to reduce suicidal ideation should be implemented according to group type, victims or non-victims of ET. Beyond educating suicidal women in stress-management techniques, it would be effective to decrease anxiety symptoms for those with a history of ET and decrease depressive symptoms for those without such a history.

Physical multimorbidity and lifetime suicidal ideation and plans among rural older adults: the mediating role of psychological distress

BACKGROUND

Previous studies have revealed that single physical chronic condition was associated with suicidal ideation/plans, but few studies have examined the relationship between multimorbidity and suicidal ideation/plans, and no studies have explored the underlying potential mechanism on this relationship in China. This study aimed to explore association between physical multimorbidity and suicidal ideation as well as plans, and further examine the mediating role of psychological distress (PD) on this relationship.

METHODS

This study was based on the data from a survey about the health service of rural elderly household in Shandong, China. A total of 3242 adults aged 60 years and older were included in this study. PD was measured by Kessler Psychological Distress Scale (K10). Ordinal and binary logistic regression analyses were employed to explore the association between physical multimorbidity, PD and suicide ideation/plans. Bootstrapping analysis was further used to examine the mediation effect of PD on the association of multimorbidity and suicidal ideations/plans.

RESULTS

The prevalence of multimorbidity, lifetime suicidal ideation, and suicidal plan in rural older adults was 35.2, 10.6 and 2.2%, respectively. Older adults living in rural areas with two or more chronic physical conditions experienced significantly higher risk of suicidal ideation and suicidal plans. The association between multimorbidity and suicidal ideations/plans was partially mediated by PD, of which, the mediating effect of PD accounted for 31.7 and 25.5% of the total effect, respectively.

CONCLUSION

This study demonstrated the associations between physical multimorbidity and suicidal ideation/plans, and the mediating role of PD on this relationship among Chinese rural elderly. Healthcare providers in rural community should provide regular surveillance for the mental health status among the rural elderly with multimorbidity, and carry out various effective intervention measures to improve the mental health status, so as to reduce the risk of suicide.

Effects of Health-Related Quality of Life on Suicidal Ideation and Depression among Older Korean Adults: A Cross-Sectional Study

OBJECTIVE

The aim of this study was to investigate the prevalence of depression and suicidal ideation among older Korean adults and the predictors of suicidal ideation.

METHODS

We analyzed data from 5,604 Koreans aged 65 or older collected by the Korea National Health and Nutrition Examination Survey in 2012-2016, an annual national, cross-sectional, epidemiologic study of the Korean population. A chi-square test was used to assess differences in socio-demographic characteristics by suicidal ideation. Psychological variables, self-rated physical health, and the five dimensions of the Korean version of the EQ-5D (EuroQoL) were compared between four subgroups categorized according to the presence/absence of suicidal ideation and depression. Multiple logistic regression was used to estimate the predictive power of health-related quality of life variables, as well as physical and psychological factors regarding suicidal ideation.

RESULTS

Overall, 11.5% of the sample reported suicidal ideation, and 20.0% of the suicidal ideation group were diagnosed with depression. Respondents with suicidal ideation had significantly lower levels of health-related quality of life than those without suicidal ideation. EQ-5D self-control (OR=1.42), anxiety/depression (OR=2.32), depression (OR=2.07), and self-rated physical health (OR=1.36) were found to significantly predict suicidal ideation.

CONCLUSION

Suicidal ideation without depression was prevalent among older Koreans. Self-control, anxiety/depression of EQ-5D, selfrated physical health, and depression predicted suicidal ideation. Interventions to screen older Koreans for early detection of suicide risk and prevention strategies should consider evaluating each EQ-5D dimension.

Factors Associated With Suicidal Ideation in Older Adults From Three Cities in Colombia, 2016

INTRODUCTION

Suicidal ideation refers to thoughts that range from a vague idea of committing suicide to a specific suicide plan.

OBJECTIVE

To explore factors such as demographic, social, family, abuse, risk of depression, habits and health conditions, which influence suicidal ideation in the elderly people in the cities of Medellín, Barranquilla, and Pasto (Colombia), with the intention to identify those associated factors that can be used in public health programs focused on this population.

METHODS

Cross-sectional analytical study was conducted using a secondary source, demographic, social, clinical variables, social support, discrimination, abuse, happiness, depression, functional capacity, and as a dependent variable were asked the question: "Have you ever thought about committing suicide?" A descriptive, bivariate and multivariate analysis was performed.

RESULTS

The median age was ≤ 69 [interquartile range, 11] years, and 58.2% were women. The prevalence of suicidal ideation was 6.4%, and of these, 28.7% had made plans to end their lives, and 66.7% had tried at least once. A statistical association was found with informal employment, cigarette consumption, alcohol and psychoactive substances, risk of depression, having a disability, dissatisfaction with their quality of life, with their health, with their economic situation, as well as feeling unhappy, bad treatment and bad relationships among family members, poor social support, sexual and economic abuse, and finally, discrimination.

CONCLUSIONS

Suicidal ideation in older adults in three cities of Colombia is explained by the sexual and economic abuse that this population is suffering, as well as bad personal relationships between the members of the family of the older adult. The risk of depression increases the probability of having thoughts against one's life.

A Cross-Sectional Study of the Prevalence of and Risk Factors for Suicidal Ideation Among the Elderly in Nursing Homes in Hunan Province, China

Our study aims to explore the risk factors for suicidal ideation and their interaction among the elderly in nursing homes in Hunan province, China. A cross-sectional study was conducted among the elderly in nursing homes in Hunan Province. Twenty-four nursing homes were selected by multistage cluster random sampling, and 817 elderly residents were investigated using a set of structured questionnaires. The main outcome measures included general information, suicidal ideation, depression symptoms, social support, activities of daily living (ADL), stressful life events, and sleep quality. Multivariate binary logistic regression was performed to explore the risk factors for suicidal ideation among the elderly in nursing homes, and additive interaction was used to analyze the interaction between risk factors. The prevalence of suicidal ideation among the elderly in nursing homes in Hunan province was 17.9% (95% confidence interval(CI): 15.2%, 20.6%). Living in a rural area (odds ratio(OR)=1.88, 95% CI: 1.03, 3.44), infrequent visits from relatives (OR=2.61, 95% CI: 1.42, 4.78), history of chronic disease (OR=2.34, 95% CI: 1.09, 5.01), depression symptoms (OR=8.11, 95% CI: 4.52, 14.54), lower social support (OR=3.85, 95% CI: 1.94, 7.61), and ADL disability status (OR=4.38, 95% CI: 2.10, 9.14) increased the risk of suicidal ideation. Additive interactions were detected between depression symptoms and ADL status, with a relative excess risk due to interaction (RERI) of 8.73 (95% CI: 2.04, 15.43), and between depression symptoms and social support, with an RERI of 5.98 (95% CI: 0.86, 11.10). The prevalence of suicidal ideation among the elderly in nursing homes is relatively high. Both physical conditions and psychosocial factors were associated with suicidal ideation among the elderly. These findings have significant implications for the prediction and prevention of suicidal behaviors.

Suicide risk around the world: a meta-analysis of longitudinal studies

PURPOSE

Suicidal thoughts and behaviors (STBs) have been a persistent problem worldwide. Identifying risk factors for STBs across distinct areas of the world may help predict who or where requires the greatest attention. However, risk factors for STBs are infrequently explored cross-nationally. The present study examined whether psychopathology prospectively predicts STBs across different areas of the world, and whether certain country-level factors moderate the degree of risk conferred.

METHODS

We conducted a meta-analysis of 71 longitudinal studies from 30 different countries that featured psychopathology-related variables predicting STB outcomes. Meta-regression was used to evaluate whether the following country-level factors modified risk: geographic region, income level, and degree of mental health structural stigma.

RESULTS

Over 90% of studies had been conducted in North America and Europe. When assessed by country income level, it was found that only one longitudinal study on psychopathology and STB was conducted outside of a high-income country. Moreover, less than 10% of studies were conducted in high structural stigma contexts. Meta-regression findings revealed that the variation in risk effect sizes across studies was not explained by models including country-level factors.

CONCLUSIONS

Our findings show critical underrepresentation of low- and middle-income countries, which account for a large proportion of global suicide deaths. This reveals a need to broaden the scope of longitudinal research on STB risk, such that countries across more regions, income levels, and degrees of structural stigma are fully accounted for. Such lines of research will improve generalizability of findings, and more precisely inform prevention efforts worldwide.

Individual and regional factors associated with suicidal ideation among Korean elderly: a multilevel analysis of the Korea Community Health Survey

OBJECTIVES

This study aimed to identify the individual and regional characteristics that influence suicidal ideation among the Korean elderly population.

METHODS

Using data collected from the 2013 Korea Community Health Survey, a multilevel analysis was performed to establish an understanding of individual behavioral patterns and regional influences on suicidal ideation.

RESULTS

Among the 77,407 individuals sampled, 11,236 (14.5%) elderly people over 60 years of age experienced suicidal ideation. Among individual factors, age, frequency of communication with friends, religious activity, social activity, leisure activity, trust in neighbors, subjective stress level, depressive symptoms, and subjective health status were significantly associated with suicidal ideation. The results showed that the lower the regional deprivation level, the higher the suicidal ideation odds ratio. In terms of regional size, the most significant effects were found in rural areas.

CONCLUSIONS

This study suggested that suicidal ideation in the elderly is associated with community factors, such as the regional deprivation index, as well as personal factors.

Chronic obstructive pulmonary disease as a risk factor for suicide: A systematic review and meta-analysis

BACKGROUND

Patients living with chronic obstructive pulmonary disease (COPD) commonly present several limitations in their daily activities, high depression rates, and low quality of life, which makes this population a risk group for suicide. This study aims to systematically assess the literature on the association between CPOD and the likelihood of suicide.

METHODS

The protocol was registered in PROSPERO (CRD42018096618). The Latin-American and Caribbean Health Sciences Literature (LILACS), PubMed, SciELO, Scopus, LIVIVO, Web of Science, and PsychNET databases were used as primary study sources. OpenThesis and OpenGrey were used to partially capture the "grey literature". A manual search was also performed through a systematized analysis of the references of eligible articles. The risk of bias among the studies included was assessed with the Joanna Briggs Institute Critical Appraisal Tools for Systematic Reviews. A random effects meta-analysis was performed to estimate the variation in odds ratio (OR) and 95% confidence intervals (95% CI).

RESULTS

The search provided 4762 results, from which only seven met the eligibility criteria and were ultimately included in the qualitative assessment of the review. The studies were published from 2002 to 2015. All studies presented low risk of bias. The total sample included 1390 suicide cases of COPD patients. The meta-analysis, which was based on five eligible case control studies, found that people with history of COPD are more likely to commit suicide (OR = 1.90; 95% CI = 1.27-2.48; p = 0.002).

CONCLUSION

COPD patients are 1.9 times more likely to commit suicide than people without COPD.

The use of machine learning in the study of suicidal and non-suicidal self-injurious thoughts and behaviors: A systematic review

BACKGROUND

Machine learning techniques offer promise to improve suicide risk prediction. In the current systematic review, we aimed to review the existing literature on the application of machine learning techniques to predict self-injurious thoughts and behaviors (SITBs).

METHOD

We systematically searched PsycINFO, PsycARTICLES, ERIC, CINAHL, and MEDLINE for articles published through February 2018.

RESULTS

Thirty-five articles met criteria to be included in the review. Included articles were reviewed by outcome: suicide death, suicide attempt, suicide plan, suicidal ideation, suicide risk, and non-suicidal self-injury. We observed three general aims in the use of SITB-focused machine learning analyses: (1) improving prediction accuracy, (2) identifying important model indicators (i.e., variable selection) and indicator interactions, and (3) modeling underlying subgroups. For studies with the aim of boosting predictive accuracy, we observed greater prediction accuracy of SITBs than in previous studies using traditional statistical methods. Studies using machine learning for variable selection purposes have both replicated findings of well-known SITB risk factors and identified novel variables that may augment model performance. Finally, some of these studies have allowed for subgroup identification, which in turn has helped to inform clinical cutoffs.

LIMITATIONS

Limitations of the current review include relatively low paper sample size, inconsistent reporting procedures resulting in an inability to compare model accuracy across studies, and lack of model validation on external samples.

CONCLUSIONS

We concluded that leveraging machine learning techniques to further predictive accuracy and identify novel indicators will aid in the prediction and prevention of suicide.

Characteristics of the "young-old" and "old-old" community-dwelling suicidal Ideators: A longitudinal 6-month follow-up study

BACKGROUND

Despite heterogeneity of older people in suicidal behavior, research identifying characteristics by age groups is scarce. We examined baseline features of older community-dwelling suicidal ideators by dichotomized age groups and the 6-month trajectory of their suicidal ideation along with its related psychopathology. Predictors of suicidal ideation within each group were investigated.

METHODS

Older community-dwelling suicidal ideators enrolled in the Korean Cohort for the Model Predicting a Suicide and Suicide-related Behavior study were subdivided into the "young-old (65-74 years)" and "old-old (≥75 years)" group. Baseline, 1-, and 6-month assessments were compared. Within each group, multiple regression analysis using rating scales (Patient Health Questionnaire-9, Beck Anxiety Inventory, Alcohol Use Disorders Identification Test, Stress Questionnaire for Korean National Health and Nutrition Examination Survey-Short Form, and Social Relationships Scale) was conducted to identify predictors of suicidal ideation measured with the intensity subscale of the Columbia-Suicide Severity Rating Scale. Two-way repeated-measures analysis of variance (RM-ANOVA) was used to compare changes in suicidal ideation, depression, anxiety between age groups over time, and one-way RM-ANOVA to examine changes within each age group.

RESULTS

Among 29 "young-old" and 53 "old-old" ideators, the latter were less likely to be receiving psychiatric treatment (odds ratio [OR] = 4.065) and make suicide attempts (OR = 2.874), whereas the former revealed greater levels of anxiety and stress. Baseline depression and stress in the "young-old" group and the "old-old" group, respectively, predicted the intensity of suicidal ideation at both baseline and 1-month assessments. No significant age group x time interactions on suicidal ideation and depression were found. However, within each age group, both suicidal ideation and depression significantly decreased only during the first month with no further improvement.

CONCLUSION

We speculate cautiously that more attention may need to be paid to the "old-old" ideators in the evaluation of psychiatric issues and for referral to psychiatrists. To decrease suicidal ideation, tailored approaches involving proactive, timely management of depression in the "young-old" and interventions focusing on stress reduction in the "old-old," would be helpful.

Prescription opioid and benzodiazepine misuse is associated with suicidal ideation in older adults

OBJECTIVES

Suicide in older adults is a major public health issue. Past research across the US adult population has linked prescription medication misuse with suicidal ideation. No work has evaluated associations between prescription opioid or benzodiazepine misuse and suicidal ideation in older adults, and this work aimed to address that gap.

METHODS/DESIGN

Data were from adults 50 years and older participating in the 2015 to 2016 National Survey on Drug Use and Health (n = 17 608). Design-based logistic regression evaluated links between any past-year prescription opioid or benzodiazepine use without misuse or prescription misuse and past-year suicidal ideation, after controlling for sociodemographic, physical health, mental health, and substance use correlates associated with suicidal ideation.

RESULTS

After controlling for all correlates, past-year use without misuse of prescription opioids or benzodiazepines was not associated with past-year suicidal ideation in older adults. In contrast, past-year opioid misuse (AOR = 1.84, 95% CI = 1.07-3.19) and benzodiazepine misuse (AOR = 2.00, 95% CI = 1.01-3.94) were significantly associated with past-year suicidal ideation, even after controlling for all covariates. While 2.2% of US older adults not engaged in either opioid or benzodiazepine misuse reported past-year suicidal ideation, 25.4% of those who misused both medication classes endorsed such suicidality (AOR = 4.73, 95% CI = 2.07-10.79).

CONCLUSIONS

Both past-year prescription opioid and benzodiazepine misuse are associated with past-year suicidal ideation in US older adults. Clinicians encountering older adult patients at-risk for or engaged in prescription medication misuse also should screen for suicidality.

Characteristics of a suicide attempt predict who makes another attempt after hospital discharge: A decision-tree investigation

The year following discharge from psychiatric hospitalization is a high-risk period for suicidal behavior, particularly among patients initially hospitalized after a suicide attempt. Demographic and clinical correlates have been identified; however, characteristics of the initial attempt may provide insight into risk for subsequent attempts as well. This investigation examined whether individual or a combination of suicide attempt characteristics predicted future attempts. Two hundred and eighteen psychiatric inpatients from the MacArthur Violence Risk Assessment Study with a recent suicide attempt were administered items from the Suicide Intent Scale and followed one year after discharge. Sixty-nine (31.65%) made a subsequent attempt. Data were analyzed by a stepwise logistic regression, followed by an iterative receiver operator curve (IROC) analysis, a recursive partitioning classification tree. The cross-validated IROC, but not logistic regression, predicted subsequent suicide attempts. Furthermore, the IROC found that participants who made definite plans and underwent extensive preparation were at highest risk for subsequent attempts. These findings suggest that suicide attempt characteristics preceding psychiatric hospitalization can help identify patients at elevated risk for another attempt post-discharge.

Suicidal Ideation among the Chinese Elderly and Its Correlates: A Comparison between the Rural and Urban Populations

Background: As China is going through a profound aging process, the mental health of the elderly is becoming an issue. As in many other societies, the elderly in China is a population at high risk of suicide; Methods: Data for the study were taken from the Sample Survey of the Aged Population in Urban/Rural China (SSAPUR) accomplished in 2010 by the China Ministry of Civil Affairs. The valid sample for this study was composed of 18,683 individuals, including 9416 urban residents and 9267 rural residents both aged 60 or more years; Results: Logistic regression analyses showed that household income and expenditure, the number of children, chronic diseases, disability of daily living, depression, the frequency of visiting neighbors and having friends or relatives who can help or not had remarkable effects on the suicidal ideation among urban and rural old people. Gender, education, political affiliation, marital status and self-rated health status did not work on the dependent variable. However, some risk factors for suicidal ideation among the Chinese elderly were different between rural and urban regions; Conclusions: We should take different measures when facing the different groups of the elderly.

Functional disability and suicidal behavior in middle-aged and older adults: A systematic critical review

BACKGROUND

Middle-aged and older adults have elevated rates of suicide around the globe, but there is a paucity of knowledge about risk factors for suicide in these age groups. One possible risk factor may be functional disability, which is more common at later ages.

METHODS

The current systematic critical review examined findings regarding the associations between functional disability and suicidal behavior (suicidal ideation, suicide attempts, and death by suicide) in middle-aged and older adults (i.e. age 50 and older).

RESULTS

Forty-five studies were found that examined these associations. The majority of studies supported a significant association between functional disability and suicidal ideation. In addition, findings to date strongly suggest that depression serves as a mediator of the association between functional disability and suicidal ideation, though most studies did not directly test for mediation.

LIMITATIONS

Firm conclusions regarding suicide attempts and death by suicide, as well as mediation, cannot be drawn due to a relative lack of research in these areas.

CONCLUSIONS

The association between functional disability and suicidal behavior suggests an important area for prevention and intervention among middle-aged and older adults, but additional research is necessary to clarify the specifics of these associations and examine appropriate intervention strategies. Important future directions for research in this area include the direct comparison of associations of risk factors with different types of suicidal behavior, greater use of longitudinal data with multiple time points, and further examination of potential mediators and moderators of the association between functional disability and suicidal behavior.

Death wishes among older people assessed for home support and long-term aged residential care

OBJECTIVE

Death wishes in older people are common and may progress to suicidal ideation and attempts. This study used routinely collected data from the interRAI Home Care assessment to examine the prevalence and clinical predictors of death wishes in older New Zealanders assessed for home support and long-term aged residential care.

METHODS

Data were collected from 35 734 people aged over 65 during 2012-2014. Chi-squared analyses were used to determine significant relationships between the presence of death wishes and demographic factors, health and functional status, and emotional and psychosocial well-being. A three-step hierarchical logistic regression model was used to determine the predictive variables of death wishes, and odds ratios were calculated.

RESULTS

Death wishes were present in 9.5% of the sample. The following factors were significantly associated with death wishes: physical health (poor self-reported health, recurrent falls, severe fatigue and inadequate pain control), psychological factors (depression, major stressors and anxiety), social factors (loneliness and decline in social activities) and impaired cognition. Depression (odds ratio = 2.54, 95% confidence interval = 2.29-2.81), loneliness (odds ratio = 2.40, 95% confidence interval = 2.20-2.63) and poor self-reported health (odds ratio = 2.34, 95% confidence interval = 1.78-3.07) had the greatest odds ratios in the full model.

CONCLUSIONS

Clinically significant depression alone cannot fully account for the development of death wishes in the elderly, and several factors are independently associated with death wishes. This knowledge can help clinicians caring for older persons to identify people who are most at risk of developing death wishes. Copyright © 2016 John Wiley & Sons, Ltd.

Predictors of suicidal ideation in Korean American older adults: analysis of the Memory and Aging Study of Koreans (MASK)

OBJECTIVE

Our aim is to investigate the prevalence and predictors of suicidal ideation among Korean American older adults and assess the self-rated mental health of Korean American older adults with suicidal ideation with or without depressive syndrome.

METHODS

The Memory and Aging Study of Koreans is a cross-sectional, epidemiologic study of a community-representative sample of Korean American older adults (N = 1116) residing in the Baltimore-Washington area. Participants were interviewed using the Korean version of the Patient Health Questionnaire (PHQ-9K). In addition, demographic information, self-rated mental health, and self-rated physical health status were obtained.

RESULTS

In this study, 14.7% of Korean American older adults reported suicidal ideation. Predictors of suicidal ideation included living alone, major or minor depressive syndrome (diagnosed by the PHQ-9K), shorter duration of residency in the USA, and poorer self-rated mental health status. Of those who reported suicidal ideation, 64% did not have minor or major depressive syndrome. However, their self-rated mental health was as poor as that of those with major or minor depressive syndrome but without suicidal ideation.

CONCLUSION

Suicidal ideation without depressive syndromes was common among Korean American older adults. For this group of elders with poor self-rated mental health, future studies should look to improving early detection of suicide risks and developing feasible suicide prevention interventions. Copyright © 2016 John Wiley & Sons, Ltd.

Profiling Patient Characteristics Associated With the Intensity of Nurse Care Coordination

The purpose of this study was to identify characteristics of chronically ill community-dwelling older adults that differentiate the intensity of care provided by nurse care coordinators. We performed data mining on electronic health records, nurses' activity logs, and health status measures from 784 care episodes provided to 196 older adults. An inductively created decision tree identified nine groups from a combination of the six participant characteristics including medication regimen complexity, cognition, physical and mental health, hospital admission, and physical functioning. Overall there was a 5-hr difference in the intensity (or contact hours) per quarter of nurse care coordinators between individuals in the highest versus lowest intensity groups. The highest intensity group presented higher medication complexity and lower mental/physical health status. With caseloads of 30 to 35 participants, nurse care coordinators were able to provide care based on participant needs that were not influenced by regulatory payment requirements.

Suicidal ideation reported by adults with Type 1 or Type 2 diabetes: results from Diabetes MILES-Australia

AIMS

To examine the prevalence and correlates of suicidal ideation (SI) in a community-based sample of adults with Type 1 or Type 2 diabetes.

METHODS

Participants were 3338 adults aged 18-70 years with Type 1 diabetes (n = 1376) or Type 2 diabetes (non-insulin: n = 1238; insulin: n = 724) from a national survey administered to a random sample registered with the National Diabetes Services Scheme. Depression and SI were assessed using the Patient Health Questionnaire, and diabetes-specific distress with the Problem Areas In Diabetes scale. Separate logistic regression analyses by diabetes type/treatment were used to determine relative contribution to SI.

RESULTS

Overall, we observed a SI rate of 14% in our sample. Participants with Type 2 diabetes using insulin reported more frequent depressive symptoms, and were more likely to report recent SI (19%) compared with those with either Type 1 diabetes or Type 2 diabetes not using insulin (14 and 12%, respectively). After controlling for depression, there was little difference in the prevalence of SI between diabetes types/treatments, but higher diabetes-specific distress significantly increased the odds of SI.

CONCLUSIONS

As SI is a significant risk factor for a suicide attempt, the findings have implications for healthcare professionals, pointing to the importance of adequate screening and action plans for appropriate follow-up of those reporting depression. Our findings are also indicative of the psychological toll of diabetes more generally, and the need to integrate physical and mental healthcare for people with diabetes.

Depression and Suicidal Ideation During Two Psychosocial Treatments in Older Adults with Major Depression and Dementia

BACKGROUND

Depression is prevalent in dementia and contributes to poor outcomes for patients and their families. Antidepressants have limited efficacy in older adults with major depression and dementia, and psychosocial interventions are under-investigated.

OBJECTIVE

To examine the course, predictors and moderators of depression and suicidal ideation during 12 weeks of home-delivered Problem Adaptation Therapy (PATH) versus Supportive Therapy for Cognitively Impaired Older Adults (ST-CI) in 39 older adults with major depression and dementia.

METHODS

Thirty-nine older adults with major depression, mild or moderate dementia, and disability participated in a randomized controlled trial that compared the efficacy of PATH versus ST-CI. Depression and suicidal ideation were assessed with Cornell Scale for Depression in Dementia Total Score and Suicide Item.

RESULTS

PATH participants had significantly greater reduction in depression than ST-CI participants over 12 weeks of treatment. PATH participants with high social support had the greatest reduction in depression. Both treatments had comparable reduction in suicidal ideation.

CONCLUSION

PATH is more effective in reducing depression in older adults with major depression and dementia compared to ST-CI. These results are clinically significant as antidepressants have limited efficacy in this population. Home-delivered psychosocial treatments may reduce suicidal ideation in this population.

Why are suicidal thoughts less prevalent in older age groups? Age differences in the correlates of suicidal thoughts in the English Adult Psychiatric Morbidity Survey 2007

BACKGROUND

Suicidal ideation is more strongly associated with suicidal intent in later life, so risk factors may also differ by age. We investigated whether the relationship between suicidal ideation and established correlates varied by age in a representative population.

METHODS

We used data from the 2007 Adult Psychiatric Morbidity Survey of England to assess the relationship between age and suicidal thoughts across 20-year age bands, using logistic regression, adjusted for survey weights. We used mediation analyses to assess the extent to which other factors mediate the relationship between suicidal thoughts and age.

RESULTS

Reports of previous-year suicidal thoughts decreased with age. This was partly explained by (1) lower rates of reported child abuse (in those aged 75+), of depression, and of anxiety symptoms (in those aged 55+), factors all strongly associated with suicidal thoughts, and (2) higher rates of protective factors in people aged 35+, specifically homeownership and cohabitation. Rates of phobias, irritability and compulsions also decreased with age, and the association of these symptoms with suicidal thoughts was particularly strong in the youngest (16-34) age group. People who reported experiencing childhood abuse in all age groups reported more suicidal thoughts, suggesting abuse has lifelong negative effects on suicidal ideation.

LIMITATIONS

The response rate was 57%. Older people may be less likely to recall childhood abuse.

CONCLUSIONS

Sexual and physical abuse in childhood are associated with suicidal ideas throughout the lifespan, so screening for suicidal ideas in younger and older people should be routine and vigorous, and cover experiences in early life: management may require appropriate psychological interventions.