Evaluation of Electronic Health Record-Based Suicide Risk Prediction Models on Contemporary Data

Background  Suicide risk prediction models have been developed by using information from patients'' electronic health records (EHR), but the time elapsed between model development and health system implementation is often substantial. Temporal changes in health systems and EHR coding practices necessitate the evaluation of such models in more contemporary data. Objectives  A set of published suicide risk prediction models developed by using EHR data from 2009 to 2015 across seven health systems reported c-statistics of 0.85 for suicide attempt and 0.83 to 0.86 for suicide death. Our objective was to evaluate these models'' performance with contemporary data (2014–2017) from these systems. Methods  We evaluated performance using mental health visits (6,832,439 to mental health specialty providers and 3,987,078 to general medical providers) from 2014 to 2017 made by 1,799,765 patients aged 13+ across the health systems. No visits in our evaluation were used in the previous model development. Outcomes were suicide attempt (health system records) and suicide death (state death certificates) within 90 days following a visit. We assessed calibration and computed c-statistics with 95% confidence intervals (CI) and cut-point specific estimates of sensitivity, specificity, and positive/negative predictive value. Results  Models were well calibrated; 46% of suicide attempts and 35% of suicide deaths in the mental health specialty sample were preceded by a visit (within 90 days) with a risk score in the top 5%. In the general medical sample, 53% of attempts and 35% of deaths were preceded by such a visit. Among these two samples, respectively, c-statistics were 0.862 (95% CI: 0.860–0.864) and 0.864 (95% CI: 0.860–0.869) for suicide attempt, and 0.806 (95% CI: 0.790–0.822) and 0.804 (95% CI: 0.782–0.829) for suicide death. Conclusion  Performance of the risk prediction models in this contemporary sample was similar to historical estimates for suicide attempt but modestly lower for suicide death. These published models can inform clinical practice and patient care today.     

Veterans' reports of pain and associations with ratings of health, health-risk behaviors, affective distress, and use of the healthcare system

The improved management of pain among veterans seeking care in Veterans Health Administration (VHA) facilities has been established as a priority. This study documents the high prevalence of reports of pain among a convenience sample of 685 veterans seeking care in a VHA primary care setting. Also reported are associations of pain complaints with self-rated health, an index of emotional distress, health-risk behaviors such as tobacco and alcohol use, health-related concerns about diet and weight, and perceptions of the availability of social support. The relationship between the presence of pain and use of outpatient and inpatient medical and mental health services is also examined. Nearly 50% of the sample reported that they experience pain regularly and that they were concerned about this problem at the time of the index visit to their primary care provider. Persons acknowledging the presence of pain, relative to those not reporting pain, were younger, reported worsening health over the past year, had greater emotional distress, used tobacco, had diet and/or weight concerns, and were found to use more outpatient medical, but not inpatient medical or mental health services. Results support the goals of the VHA National Pain Management Strategy designed to reduce unnecessary pain and suffering among veterans receiving care in VHA facilities.     

Smoking among veterans with multiple sclerosis: prevalence correlates, quit attempts, and unmet need for services

Turner AP, Kivlahan DR, Kazis LE, Haselkorn JK. Smoking among veterans with multiple sclerosis: prevalence, correlates, quit attempts, and unmet need for services.

Objective

To describe the prevalence and correlates of smoking as well as quit attempts and unmet need for smoking cessation services in a national sample of veterans with multiple sclerosis (MS).

Design

Cross-sectional cohort study linking computerized medical record information to mailed survey data from 1999.

Setting

Veterans Health Administration (VHA).

Participants

Sixty-four percent (2994/4685) of veterans with MS who received services in VHA and also returned survey questionnaires, as well as a 20% random subsample (n=569) who completed a more extensive assessment of smoking.

Interventions

Not applicable.

Main Outcome Measures

Items assessing smoking, quit attempts, and unmet need for smoking services.

Results

Among all survey respondents with MS, 28.5% (95% confidence interval [CI], 26.9-30.2) endorsed current smoking. Of extended survey respondents, 54.5% (95% CI, 46.6-62.1) reported a quit attempt in the past year, and 59.0% (95% CI, 51.1-66.4) reported not getting needed services for smoking in the past year. In fully adjusted logistic regression, smoking was associated with younger age, lower levels of education, being unmarried, higher levels of physical pain, and poorer mental health. A quit attempt was associated with higher levels of education and greater pain intensity.

Conclusions

Smoking among veterans with MS is common, with rates similar to those for other veterans. There is substantial need for cessation services. Cessation interventions should address correlates of smoking including pain, poorer mental health, and social isolation.     

Traumatic Brain Injury and Receipt of Prescription Opioid Therapy for Chronic Pain in Iraq and Afghanistan Veterans: Do Clinical Practice Guidelines Matter?

Clinical practice guidelines admonish against prescribing opioids for individuals with chronic pain and traumatic brain injury (TBI) because of increased risk for adverse outcomes, yet no studies have described opioid prescribing patterns in these higher-risk patients. Between October 2007 and March 2015, 53,124 Iraq and Afghanistan veterans with chronic pain not prescribed opioids in the previous year were followed for 1 year after completing a Comprehensive TBI Evaluation within the Department of Veterans Affairs health care facilities. Veterans reporting the most severe TBI sequelae (eg, loss of consciousness?>30 minutes) were significantly more likely to receive short-term and long-term opioid therapy than those with less severe or no TBI sequelae (P values?<?.001). In analyses adjusted for sociodemographic characteristics, military service, pain disability, and previous nonopioid treatment modalities, veterans with moderate to severe TBI had a significantly increased risk of receiving opioid therapy. Veterans with moderate to severe TBI and comorbid post-traumatic stress disorder and depression had an even greater risk of initiating long-term opioid therapy in the year after the Comprehensive TBI Evaluation (adjusted relative risk?=?3.57 [95% confidence interval?=?2.85–4.47]). Higher-risk patients with chronic pain and TBI with mental health comorbidities may benefit from improved access to behavioral health and nonpharmacological therapies for chronic pain.

Pilot evaluation of horticultural therapy in improving overall wellness in veterans with history of suicidality

ObjectivesNovel approaches to mental health and suicide prevention are lacking. Converging evidence has shown the effectiveness of horticultural therapy (HT) in improving mental health symptoms, but whether it would reduce suicide risk and contributing risk factors is unknown.DesignUsing a cohort model, HT was delivered 3.5 h over four weekly, sessions administered by a registered horticultural therapist to veterans with history of suicide ideation or attempt who felt isolated and experienced ongoing environmental stressors with interest in learning new coping strategies.SettingHT delivery occurred in an urban garden, through a community partnership between the VA (Veterans Administration) and the New York Botanical Garden. Guided by principles of biophilia, participating veterans took part in nature walks, self-reflection and journaling, and planting activities.Outcome measuresStress, mood, pain, and social isolation levels were measured weekly pre-post HT sessions using thermometer scales, with concordant validity to validated clinical instruments.ResultsOf the 20 men and women with a history of suicide attempts/ideation, HT demonstrated immediate improvements after each session across all symptom domains in magnitude of reduction in stress, pain, mood, and loneliness. The effect sizes were in medium to large range (Cohen’s d>.5). Additionally, a single HT session showed a sustained effect over subsequent 2-to-4 weeks as observed by the significantly decreased pre-session thermometer scores in subsequent weeks. Reductions in mood symptoms correlated with decline in suicidal ideation (r_s = 0.63).ConclusionHT intervention maybe a promising therapeutic modality for improving overall wellness in suicide prevention in at-risk veteran populations.     

Migraine headaches and suicide attempt

(Headache 2012;52:723‐731) Background.— Previous cross‐sectional studies reported an increased risk of suicide attempt in persons with migraine headache, which was sustained when psychiatric comorbidity was statistically controlled. Objective.— To estimate the risk of suicide attempt in persons with migraine vs controls with no history of severe headache, using prospective data and validated diagnostic assessment. To examine the specificity of the migraine‐suicide attempt risk by comparing it to the risk associated with non‐migraine headache of comparable severity and disability. Methods.— A cohort of persons with migraine (n = 496), non‐migraine severe headaches (n = 151), and controls with no history of severe headache (n = 539) was randomly selected from the general community, assessed in 1997 and reassessed 2 years later. Results.— Persons with migraine had an increased risk of suicide attempt during the 2‐year follow‐up period, compared with controls. Odds ratio, adjusted for sex, psychiatric disorder, and previous history of suicide attempt at baseline was 4.43 (95% confidence interval [CI] 1.93, 10.2). Persons with non‐migraine headache of comparable intensity and disability also had an increased risk of suicide attempt, compared to controls: odds ratio, adjusted for the same covariates, was 6.20 (95% CI 2.40, 16.0). The difference between the 2 estimates was not significant. In the entire sample, headache severity at baseline predicted suicide attempt: a difference of 1 standard deviation (SD) in pain score increased the risk of suicide attempt by 79%, adjusting for sex and psychiatric disorders. Conclusions.— The results suggest the possibility that pain severity might account in part for the increased risk of suicide attempt associated with migraine.     

Screening for Military Service History and Risk for Suicide in a Civilian Internal Medicine Clinic

More than half of United States military veterans receive medical care outside the Veterans Health Administration. Civilian providers do not routinely screen for military service history or increased risk for suicide in veterans. This quality improvement project used the model for improvement to implement screening for military service history and suicide risk using the “VA/DoD Clinical Practice Guideline for the Assessment and Management of Patients at Risk for Suicide” in a civilian clinic. Four hundred ninety-three of 845 patients were evaluated; 13% (n = 64) were identified as veterans. Of the veterans identified, 73.4% (n = 47) required additional screening for suicide risk, and 7.7% (n = 3) were identified as at risk for suicide.     

Chronic Pain,TBI, and PTSD in Military Veterans: A Link to Suicidal Ideation and Violent Impulses?

The polytrauma clinical triad refers to the co-occurrence of chronic pain, traumatic brain injury (TBI), and posttraumatic stress disorder (PTSD). Despite research implicating dyadic relationships between these conditions and adverse outcomes, scant research has examined the polytrauma clinical triad's relation to suicide or violence. The present cross-sectional study was designed to examine whether this complex clinical presentation increases risk of suicidal ideation and violent impulses after accounting for other established risk factors. Veterans who served in the military since September 11, 2001 (N?=?667) who reported chronic pain completed an interview and self-report battery. Bivariate analyses showed that suicidal ideation and violent impulses both correlated with PTSD, TBI+PTSD, pain intensity and interference, drug abuse, and major depressive disorder (MDD). Multiple regression analyses showed that: 1) race, chronic pain with PTSD, alcohol abuse, and MDD significantly predicted suicidal ideation, 2) pain interference, chronic pain with TBI, chronic pain with PTSD, chronic pain with TBI+PTSD, drug abuse, and MDD significantly predicted violent impulses, and 3) pain interference was a more critical predictor of suicidal and violent ideation than pain intensity. Implications for risk assessment and treatment are discussed.

Suicidal Ideation and Perceived Burdensomeness in Patients with Chronic Pain

There is a clear relationship between suicide risk and chronic pain conditions. However, the exact nature of this link has been poorly understood, with risk attribution often limited to comorbid depression. Perceived burdensomeness has already been confirmed as a risk factor for suicidal ideation (SI) and suicide attempt in the general population. Self‐perceived burden, studied among medically and terminally ill medical populations, has begun to receive a great deal of attention as a suicide risk factor. However, this risk has not been considered in an outpatient chronic pain population, a group likely to experience perceived burdensomeness as a particular problem. Guidelines recommend routine suicide risk screening in medical settings, but many questionnaires are time‐consuming and do not allow for the assessment of the presence of newly identified risk constructs, such as perceived burdensomeness. This retrospective study examined the relationship between depression, perceived burdensomeness, and SI in a patient sample seeking behavioral treatment for chronic pain management. A logistic regression model was developed, with preliminary results indicating perceived burdensomeness was the sole predictor of SI, even in the presence of other well‐established risk factors such as age, gender, depressive symptoms, and pain severity. Findings highlight the potential utility of a single‐item screening question in routine clinical care as an incrementally superior predictor of SI in a chronic pain population.     

Recruitment of women veterans into suicide prevention research: Improving response rates with enhanced recruitment materials and multiple survey modalities

Women veterans (WV) are the fastest growing veteran subpopulation and recognized to be at increased risk for suicide compared with civilians. Improving engagement (e.g., response rates) of WV in survey research is critical to ensuring valid and generalizable findings, which can inform suicide prevention programs tailored for this population. Many factors are known to influence response rates, yet little is known about ways to optimize survey response rates among WV. Three recruitment cohorts (Waves 1 [W1], 2a [W2a], and 2b [W2b]) of WV were invited to participate in an online survey for a national, mixed-methods study examining suicide risk among WV using reproductive health care services paid for or provided by the Veterans Health Administration. To examine the effects of enhanced recruitment efforts, standard recruitment materials were mailed to all three cohorts, with the additions of: a study flyer aiming to build trust between participants and researchers (W2a, W2b) and a paper survey (W2b). Characteristics of responders and non-responders were compared by wave and across survey modalities. Response rates were significantly higher for groups receiving enhanced (W2a = 17.1%; W2b = 24.6%) versus standard (W1 = 12.2%) recruitment materials. WV residing in rural areas were significantly more likely to respond by paper (37.1%) than online (19.8%). Non-respondents were disproportionately racial and ethnic minorities. Disclosure of sensitive information (e.g., military sexual trauma) did not differ by survey modality. Findings suggest that enhanced recruitment materials improve survey response rates among WV, an important consideration for future research with this population.     

Involving patients who attempt suicide in suicide prevention: a focus groups study

The aim of this study is to gain insight into the individual experiences of patients who attempt suicide in order to better understand the reasons for and emotions behind a suicide attempt, thus also gaining insight, through the patients' own input, into the risk and protective factors which might influence possible repeat attempts and the attitude towards the assistance they receive. Two focus groups were conducted involving 17 participants, all hospitalized at the time of research for attempting suicide. The patients proved themselves competent, even expert in indicating reasons for, risk factors of and prevention strategies for suicide. The main findings suggest that the relational factor represents a key point both as a trigger for the suicide attempt and for promoting the communication of the intent or for preventing a repeat suicide attempt, as interpersonal relationships and an empathic environment were, in essence, what was perceived as therapeutic and protective and enabled the expression of thoughts and self-understanding. Accordingly psychotherapy, non-specific relationship 'monitoring' after discharge and tutored self-help groups have been suggested. Feasibility and implementing methods as well as the role of the nurse for such interventions were discussed.     

Rationale and study design of a trial of mindfulness-based cognitive therapy for preventing suicidal behavior (MBCT-S) in military veterans

BackgroundAlthough suicide ranks 10th as a cause of death in the United States, and 1st among active military personnel, there are surprisingly few evidence-based therapies addressing suicidality, and development of new treatments is limited. This paper describes a clinical trial testing a novel therapy for reducing suicide risk in military veterans. The intervention, Mindfulness-Based Cognitive Therapy for Preventing Suicide Behavior (MBCT-S), is a 10-week group intervention adapted from an existing treatment for depression (Mindfulness-Based Cognitive Therapy – MBCT). MBCT-S incorporates the Safety Planning Intervention, which is currently implemented throughout the Veterans Health Administration (VHA) for veterans at high suicide risk.MethodsMBCT-S is being tested in a VHA setting using an intention-to-treat, two-group randomized trial design in which 164 high suicide risk veterans are randomized to either VHA Treatment As Usual (TAU; n = 82) or TAU + MBCT-S (n = 82). Our primary outcome measure, suicide-related event, defined to include suicide preparatory behaviors, self-harm behavior with suicidal or indeterminate intent, suicide-related hospitalizations and Emergency Department (ED) visits, will be measured through five assessments administered by blinded assessors between baseline and 12 months post-baseline. We will measure suicide attempts and suicide deaths as a secondary outcome, because of their anticipated low incidence during the study period. Secondary outcomes also include severity of suicidal ideation, hopelessness and depression.SignificanceThis study has the potential to significantly enhance the quality and efficiency of VHA care for veterans at suicide risk and to substantially improve the quality of life for veterans and their families.     

Sense of coherence and suicidality in suicide attempters: a prospective study

The usefulness of the Sense of Coherence (SOC) scale in the nursing setting is well-established, and an association between SOC and suicidality has been suggested. The aim was to test whether low SOC at index attempt is an independent predictor of suicidality at 2-month follow-up and of risk for repeat attempt. The study, which had a prospective cross-sectional design, included patients admitted to hospital after a suicide attempt. They were interviewed by means of Structured Clinical Interview for DSM-IV. Participants (n=155) completed the SOC scale and the Comprehensive Psychopathological Self-rating Scale for Affective Syndromes. Suicidality was rated with the Suicide Assessment Scale. Instruments were employed again at follow-up. Non-fatal/fatal repetition within 3 years was determined by review of hospital records. Low SOC at baseline predicted high suicidality at follow-up. The association remained after adjustment for major depression and affective symptom burden. Repeat attempts were made by 54 persons. Low baseline SOC was associated with repeat attempt, but the association did not remain after adjustment for major depression and symptom burden. Low SOC ratings could be a marker of risk for high suicidality in the aftermath of a suicide attempt. The SOC scale could be incorporated in nursing assessments of suicide attempters.     

Suicide After Stroke in the United States Veteran Health Administration Population

ObjectiveTo evaluate risk for suicide among veterans with a history of stroke, seeking care within the Veterans Health Administration (VHA), we analyzed existing clinical data.DesignThis retrospective cohort study was approved and performed in accordance with the local Institutional Review Board. Veterans were identified via the VHA's Corporate Data Warehouse. Initial eligibility criteria included confirmed veteran status and at least 90 days of VHA utilization between fiscal years 2001-2015. Cox proportional hazards models were used to assess the association between history of stroke and suicide. Among those veterans who died by suicide, the association between history of stroke and method of suicide was also investigated.SettingVHA.ParticipantsVeterans with at least 90 days of VHA utilization between fiscal years 2001-2015 (N=1,647,671). Data from these 1,647,671 veterans were analyzed (1,405,762 without stroke and 241,909 with stroke).InterventionsNot applicable.Main Outcome MeasuresSuicide and method of suicide.ResultsThe fully adjusted model, which controlled for age, sex, mental health diagnoses, mild traumatic brain injury, and modified Charlson/Deyo Index (stroke-related diagnoses excluded), demonstrated a hazard ratio of 1.13 (95% confidence interval, 1.02-1.25; P=.02). The majority of suicides in both cohorts was by firearm, and a significantly larger proportion of suicides occurred by firearm in the group with stroke than the cohort without (81.2% vs 76.6%).ConclusionsFindings suggest that veterans with a history of stroke are at increased risk for suicide, specifically by firearm, compared with veterans without a history of stroke. Increased efforts are needed to address the mental health needs and lethal means safety of veterans with a history of stroke, with the goal of improving function and decreasing negative psychiatric outcomes, such as suicide.     

Unrecognized suicidal ideation in ED patients: are we missing an opportunity?

Objective

To determine whether patients who disclosed suicidal ideation during a health-risk survey had their mental health symptoms documented by physicians and were given mental health referrals and to evaluate how many of these patients subsequently attempted suicide.

Methods

As part of a larger survey, patients responded to questions on a computer kiosk about general health risk behaviors and mental health symptoms. Fifteen months after initiating the survey, we reviewed medical records on those patients who had disclosed suicidal ideation. A standardized abstraction sheet was used to collect data regarding the ED diagnosis at the time of enrollment, physician documentation of suicidal ideation, and referral to psychiatric services, as well as subsequent ED and clinic visits and suicide attempts.

Results

Of the 165 patients who disclosed suicidal ideation on the computer survey, 118 charts (72%) were available. During the index ED visit, only 25% of patients had suicidal ideation or other mental health issues noted on the chart. The majority (76%) of patients were discharged home, 10% were transferred to psychiatric services, and 3% were admitted for medical reasons. Although 72 patients had no future visits to the ED or other hospital-affiliated clinics, 39% of patients had at least one subsequent visit to the ED, and 17% had at least one visit to the psychiatric services. Four patients attempted suicide after their initial index visit to the ED.

Conclusion

Suicidal ideation was self-disclosed frequently by waiting room patients in our urban ED, and patients who disclosed suicidal ideation did not always receive referrals for mental health services.     

Prevalence and Impact of Pain in Multiple Sclerosis: Physical and Psychologic Contributors

Hirsh AT, Turner AP, Ehde DM, Haselkorn JK. Prevalence and impact of pain in multiple sclerosis: physical and psychologic contributors.

Objective

To characterize the prevalence and impact of pain in veterans with multiple sclerosis (MS) and to assess their association with demographic, biologic, and psychologic variables.

Design

Cross-sectional cohort study linking computerized medical record information to mailed survey data.

Setting

Veterans Health Administration (VHA).

Participants

Sixty-four percent (2994/4685) of veterans with MS who received services in VHA and also returned survey questionnaires.

Interventions

Not applicable.

Main Outcome Measures

Items assessing pain intensity, pain interference, and physical and mental health functioning.

Results

Ninety-two percent of participants reported bodily pain within the prior 4 weeks, with 69% of the total sample indicating pain of moderate or higher intensity. Eighty-five percent indicated that pain caused functional interference during the past 4 weeks, with 71% of the total sample reporting pain-related interference that was moderate or greater. No significant sex or race differences emerged for the pain indices. A significant but modest relationship between increasing age and pain interference emerged (r=.05, P<.01); however, age was not significantly related to pain intensity. Multivariate regression analyses identified pain intensity (β=.73), physical health functioning (β=-.07), and mental health functioning (β=-.13) variables as significant, unique contributors to the prediction of pain interference. The interaction of pain intensity and physical functioning was also significant but of minimal effect size (β=-.03).

Conclusions

Pain is highly prevalent and causes substantial interference in the lives of veterans with MS. The functional impact of pain in veterans with MS is influenced by pain intensity, physical health, and emotional functioning. Clinical practice should take each of these domains into consideration and reflect a biopsychosocial conceptualization.     

Rural Veterans’ Utilization of Non‒Veterans Administration Community Health Care Services

Rural veterans are obtaining non‒Veterans Administration health care at an increasing rate, but information on what community services are accessed is lacking. A secondary analysis of surveys obtained from 6 previous studies conducted in acute care areas was scrutinized for military history and utilization patterns. Depending on age, multiple health care agencies were used. Acute care was relied upon frequently, most often due to pain. The Veterans Administration was infrequently visited. Clinician electronic documentation of military history, inquiry into use of other health care services, and involvement in research to monitor outcomes are suggested to facilitate identification of health needs specific to this population.