Automated Society of Fetal Urology (SFU) grading of hydronephrosis on ultrasound imaging using a convolutional neural network.

INTRODUCTION: Grading of hydronephrosis severity on postnatal renal ultrasound guides management decisions in antenatal hydronephrosis (ANH). Multiple systems exist to help standardize hydronephrosis grading, yet poor inter-observer reliability persists. Machine learning methods may provide tools to improve the efficiency and accuracy of hydronephrosis grading. OBJECTIVE: To develop an automated convolutional neural network (CNN) model to classify hydronephrosis on renal ultrasound imaging according to the Society of Fetal Urology (SFU) system as potential clinical adjunct. STUDY DESIGN: A cross-sectional, single-institution cohort of postnatal renal ultrasounds with radiologist SFU grading from pediatric patients with and without hydronephrosis of stable severity was obtained. Imaging labels were used to automatedly select sagittal and transverse grey-scale renal images from all available studies from each patient. A VGG16 pre-trained ImageNet CNN model analyzed these preprocessed images. Three-fold stratified cross-validation was used to build and evaluate the model that was used to classify renal ultrasounds on a per patient basis into five classes based on the SFU system (normal, SFU I, SFU II, SFU III, or SFU IV). These predictions were compared to radiologist grading. Confusion matrices evaluated model performance. Gradient class activation mapping demonstrated imaging features driving model predictions. RESULTS: We identified 710 patients with 4659 postnatal renal ultrasound series. Per radiologist grading, 183 were normal, 157 were SFU I, 132 were SFU II, 100 were SFU III, and 138 were SFU IV. The machine learning model predicted hydronephrosis grade with 82.0% (95% CI: 75-83%) overall accuracy and classified 97.6% (95% CI: 95-98%) of the patients correctly or within one grade of the radiologist grade. The model classified 92.3% (95% CI: 86-95%) normal, 73.2% (95% CI: 69-76%) SFU I, 73.5% (95% CI: 67-75%) SFU II, 79.0% (95% CI: 73-82%) SFU III, and 88.4% (95% CI: 85-92%) SFU IV patients accurately. Gradient class activation mapping demonstrated that the ultrasound appearance of the renal collecting system drove the model's predictions. DISCUSSION: The CNN-based model classified hydronephrosis on renal ultrasounds automatically and accurately based on the expected imaging features in the SFU system. Compared to prior studies, the model functioned more automatically with greater accuracy. Limitations include the retrospective, relatively small cohort, and averaging across multiple imaging studies per patient. CONCLUSIONS: An automated CNN-based system classified hydronephrosis on renal ultrasounds according to the SFU system with promising accuracy based on appropriate imaging features. These findings suggest a possible adjunctive role for machine learning systems in the grading of ANH.

Deep Learning of Videourodynamics to Classify Bladder Dysfunction Severity in Patients With Spina Bifida.

PURPOSE: Urologists rely heavily on videourodynamics to identify patients with neurogenic bladders who are at risk of upper tract injury, but their interpretation has high interobserver variability. Our objective was to develop deep learning models of videourodynamics studies to categorize severity of bladder dysfunction. MATERIALS AND METHODS: We performed a cross-sectional study of patients aged 2 months to 28 years with spina bifida who underwent videourodynamics at a single institution between 2019 and 2021. The outcome was degree of bladder dysfunction, defined as none/mild, moderate, and severe, defined by a panel of 5 expert reviewers. Reviewers considered factors that increase the risk of upper tract injury, such as poor compliance, elevated detrusor leak point pressure, and detrusor sphincter dyssynergia, in determining bladder dysfunction severity. We built 4 models to predict severity of bladder dysfunction: (1) a random forest clinical model using prospectively collected clinical data from videourodynamics studies, (2) a deep learning convolutional neural network of raw data from the volume-pressure recordings, (3) a deep learning imaging model of fluoroscopic images, (4) an ensemble model averaging the risk probabilities of the volume-pressure and fluoroscopic models. RESULTS: Among 306 videourodynamics studies, the accuracy and weighted kappa of the ensemble model classification of bladder dysfunction when at least 75% expected bladder capacity was reached were 70% (95% CI 66%,76%) and 0.54 (moderate agreement), respectively. The performance of the clinical model built from data extracted by pediatric urologists was the poorest with an accuracy of 61% (55%, 66%) and a weighted kappa of 0.37. CONCLUSIONS: Our models built from urodynamic pressure-volume tracings and fluoroscopic images were able to automatically classify bladder dysfunction with moderately high accuracy.

Sural nerve: imaging anatomy and pathology.

High resolution ultrasound (US) and magnetic resonance (MR) neurography are both imaging modalities that are commonly used for assessing peripheral nerves including the sural nerve (SN). The SN is a cutaneous sensory nerve which innervates the lateral ankle and foot to the base of the fifth metatarsal. It is formed by contributing nerves from the tibial and common peroneal nerves with six patterns and multiple subtypes described in literature. In addition to the SN being a cutaneous sensory nerve, the superficial location enables the nerve to be easily biopsied and harvested for a nerve graft, as well as increasing the susceptibility to traumatic injury. As with any peripheral nerves, pathologies such as peripheral nerve sheath tumors and neuropathies can also affect the SN. By utilizing a high frequency probe in US and high-resolution MR neurography, the SN can be easily identified even with the multiple variations given the standard distal course. US and MRI are also useful in determining pathology of the SN given the specific image findings that are seen with peripheral nerves. In this review, we evaluate the normal imaging anatomy of the SN and discuss common pathologies identified on imaging.

Shared correlates of prescription drug misuse and severe suicide ideation among clinical patients at risk for suicide.

OBJECTIVE: Unintentional drug overdose and suicide have emerged as public health problems. Prescription drug misuse can elevate risk of overdose. Severe suicidal ideation increases risk of suicide. We identified shared correlates of both risk factors to inform cross-cutting prevention efforts. METHODS: We conducted a cross-sectional study using the Military Suicide Research Consortium's Common Data Elements survey; 2012-2017 baseline data collected from 10 research sites were analyzed. The sample included 3962 clinical patients at risk of suicide. Factors examined in relation to the outcomes, prescription drug misuse and severe suicidal ideation, included demographic characteristics and symptoms of: hopelessness; anxiety; post-traumatic stress disorder; alcohol use; other substance use; prior head/neck injury; insomnia; and belongingness. Poisson regression models with robust estimates provided adjusted prevalence ratios (aPRs) and 97.5% confidence intervals (CIs). RESULTS: Medium and high (vs. low) levels of insomnia were positively associated with prescription drug misuse (aPRs p < 0.025). Medium (vs. low) level of insomnia was positively associated with severe suicidal ideation (aPR: 1.09; CI: 1.01-1.18). Medium and high (vs. low) levels of perceived belongingness were inversely associated with both outcomes (aPRs p < 0.025). CONCLUSIONS: Research should evaluate whether addressing sleep problems and improving belongingness can reduce prescription drug misuse and suicidal ideation simultaneously.

Correlates of Intimate Partner Homicide among Male Suicide Decedents with Known Intimate Partner Problems.

OBJECTIVE: Intimate partner (IP) problems are risk factors for suicide among men. However, there is little understanding of why some male suicide decedents who had such problems killed their partners before death (i.e., "IP homicide-suicide"), while most of these decedents did not. To inform prevention efforts, this study identified correlates of IP homicide among male suicide decedents with known IP problems. METHODS: We examined IP homicide correlates among male suicide decedents aged 18+ years who had known IP problems using 2003-2015 National Violent Death Reporting System data. Prevalence odds ratios and 95% confidence intervals were estimated for demographic, incident, and circumstance variables. IP homicide-suicide narratives were examined to identify additional prevention opportunities. RESULTS: An estimated 1,504 (5.0%) of 30,259 male suicide decedents who had IP problems killed their partner. IP homicide-suicide perpetration was positively correlated with suicide by firearm and precipitating civil legal problems but negatively correlated with mental health/substance abuse treatment. An estimated 33.7% of IP homicide-suicides occurred during a breakup; 21.9% of IP homicide-suicide perpetrators had domestic violence histories. CONCLUSIONS: Connections between the criminal justice and mental health systems as well as stronger enforcement of laws prohibiting firearm possession among domestic violence offenders may prevent IP homicide-suicides.

Ten-Year Trend and Correlates of Reported Posttraumatic Stress Disorder among Young Male Veteran Suicide Decedents-Results from the National Violent Death Reporting System, 16 U.S. States, 2005-2014.

OBJECTIVE: This study examined trends and correlates of reported post-traumatic stress disorder (PTSD) among young male Veteran suicide decedents, using data from the National Violent Death Reporting System from 2005-2014 on 1,362 male U.S. Veteran suicide decedents aged 18-34 years. METHODS: Prevalence of reported PTSD (i.e., diagnosis/symptoms) was determined by mental health diagnostic fields and narratives and examined by year. Demographic, incident, and precipitating circumstance characteristics correlated with reported PTSD were identified. RESULTS: One-hundred ninety-eight (15%) decedents had PTSD evidence. A 30-fold increase in reported PTSD prevalence occurred among decedents aged 25-34 years; however, no increase was observed among younger decedents. Reported PTSD was associated with past deployments (odds ratio (OR): 14.5, 95% confidence interval (95% CI): 9.0-23.4); depression (OR: 1.8, 95% CI: 1.2-2.6); and divorce (OR: 1.7, 95% CI: 1.0-2.7). Recent crisis (OR: 0.6, 95% CI: 0.3-0.9) was inversely associated with reported PTSD. CONCLUSIONS: Reported PTSD prevalence substantially increased among Veteran suicide decedents aged 25-34 years suggesting it is beginning to play a larger role in suicide for this group. Few correlated suicide risk factors were found, suggesting that if symptoms of PTSD are present, heightened vigilance by providers for suicide risk might be warranted, irrespective of evidence of other risk factors.

Circumstances preceding suicide in U.S. soldiers: A qualitative analysis of narrative data.

To gain a better understanding of military suicide, we examined suicide narratives for 135 Soldiers extracted from two large-scale surveillance systems: the Department of Defense Suicide Event Report (DoDSER) and the Centers for Disease Control and Prevention's (CDC) National Violent Death Reporting System (NVDRS). Using coroner/medical examiner and law enforcement narratives captured in the NVDRS and mental health provider narrative data collected across multiple domains from the DoDSER, we examined circumstances surrounding military suicides using a qualitative content analysis approach. We identified five common proximal circumstances: (1) intimate partner relationship problems (63.0%); (2) mental health/substance abuse (51.9%); (3) military job-related (46.7%); (4) financial (17.8%); and (5) criminal/legal activity (16.3%). Evidence of premeditation was present in 37.0% of suicides. Decedents frequently struggled with multiple, high-stress problems and exhibited symptoms of coping and emotion regulation difficulties. Findings demonstrate potential points of intervention for suicide prevention strategies. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Circumstances Preceding Homicide-Suicides Involving Child Victims: A Qualitative Analysis.

Homicide-suicide incidents involving child victims can have a detrimental impact on survivors of the violence, family members and friends of the decedents, and other community members, but the rare occurrence of these acts makes using quantitative data to examine their associated antecedents challenging. Therefore, using qualitative data from the 2003-2011 National Violent Death Reporting System, we examined 175 cases of homicide-suicide involving child victims in an effort to better understand the complex situational factors of these events. Our findings indicate that 98% of homicide-suicides with child victims are perpetrated by adults (mostly parents) and propelled by the perpetrators' intimate partner problems, mental health problems, and criminal/legal problems. These events are often premeditated, and plans for the violence are sometimes disclosed prior to its occurrence. Findings provide support for several theoretical perspectives, and implications for prevention are discussed.

Cómo descifrar el suicidio y otros tipos de muerte asociados a la intoxicación por drogas: Consulta de los Centros para el Control y la Prevención de Enfermedades. Resumen de la reunión

La clasificación del tipo de muerte (es decir, por causas naturales o por accidente, suicidio, homicidio o causa indeterminada) afecta tanto la vigilancia de la mortalidad como las investigaciones, políticas y prácticas de salud pública. La determinación del tipo de muerte en el caso de defunciones debidas a la intoxicación por drogas puede ser más difícil, dada la marcada variabilidad entre los estados dentro de Estados Unidos. Los Centros para el Control y la Prevención de Enfermedades (CDC) de Estados Unidos organizaron una reunión multidisciplinaria para debatir el tema de las muertes debidas a intoxicación por drogas en relación con el suicidio y otros tipos de muerte. Los objetivos de la reunión fueron determinar los factores a nivel individual, del sistema y del lugar que afectan la clasificación del tipo de muerte y proponer posibles soluciones a los obstáculos para la clasificación. Entre las estrategias sugeridas se mencionaron: una mejor estandarización de las prácticas de investigación de la escena de muerte, toxicología y autopsia; una mayor rendición de cuentas; y la creación de guías de ayuda para los investigadores. Se necesita de la colaboración y coordinación continuas de las actividades entre los interesados directos en apoyo a los esfuerzos de prevención.

Manner of death (MOD) classification (i.e., natural, accident, suicide, homicide, or undetermined cause) affects mortality surveillance and public health research, policy, and practice. Determination of MOD in deaths caused by drug intoxication is challenging, with marked variability across states. The Centers for Disease Control and Prevention hosted a multidisciplinary meeting to discuss drug intoxication deaths as they relate to suicide and other MOD. The meeting objectives were to identify individual-level, system-level, and place-based factors affecting MOD classification and identify potential solutions to classification barriers. Suggested strategies included improved standardization in death scene investigation, toxicology, and autopsy practice; greater accountability; and creation of job aids for investigators. Continued collaboration and coordination of activities are needed among stakeholders to affect prevention efforts.

Deciphering Suicide and Other Manners of Death Associated with Drug Intoxication: A Centers for Disease Control and Prevention Consultation Meeting Summary.

Manner of death (MOD) classification (i.e., natural, accident, suicide, homicide, or undetermined cause) affects mortality surveillance and public health research, policy, and practice. Determination of MOD in deaths caused by drug intoxication is challenging, with marked variability across states. The Centers for Disease Control and Prevention hosted a multidisciplinary meeting to discuss drug intoxication deaths as they relate to suicide and other MOD. The meeting objectives were to identify individual-level, system-level, and place-based factors affecting MOD classification and identify potential solutions to classification barriers. Suggested strategies included improved standardization in death scene investigation, toxicology, and autopsy practice; greater accountability; and creation of job aids for investigators. Continued collaboration and coordination of activities are needed among stakeholders to affect prevention efforts.

Examining the Prevalence and Predictors of Injury from Adolescent Dating Violence.

Medical needs of youth who experience dating violence are not well understood because of limited past research examining the prevalence and predictors of injuries and medical help seeking. To address these gaps, the current study described the prevalence and predictors of injuries from dating violence from grades 8 through 12 in a large sample of youth. Results indicate that one third to one half of youth who experienced any physical and/or sexual dating violence also sustained an injury. Prevalence of injury was highest in the 8th grade and was significantly higher for females than for males across grades 8 through 11. Youth who experienced greater amounts of violent victimization in their relationships (physical, sexual, and psychological) were at highest risk for injury. Results also suggest that victims at highest risk for injury are girls, white youth, those experiencing multiple types of violence, and those who also engage in perpetration. Given the high prevalence of injury among youth who report dating violence, healthcare professionals may be in a unique position to screen and counsel youth about dating violence. Because the highest prevalence of injury occurred before high school, prevention programs should start early and selected prevention may be used for youth at highest risk for injury.

Antecedents of Suicide among Youth Aged 11-15: A Multistate Mixed Methods Analysis.

Suicide is the second leading cause of death for youth aged 11 to 15, taking over 5,500 lives from 2003 to 2014. Suicide among this age group is linked to risk factors such as mental health problems, family history of suicidal behavior, biological factors, family problems, and peer victimization and bullying. However, few studies have examined the frequency with which such problems occur among youth suicide decedents or the context in which decedents experience these risk factors and the complex interplay of risk that results in a decedent's decision to take his/her own life. Data from a random sample of 482 youth (ages 11-15) suicide cases captured in the National Violent Death Reporting System from 2003 to 2014 were analyzed. The sample had fewer girls than boys (31 vs. 69 %) and comprised primarily White youth (79 %), but also African Americans (13 %), Asians (4 %), and youth of other races (4 %). Narrative data from coroner/medical examiner and law enforcement investigative reports were coded and analyzed to identify common behavioral patterns that preceded suicide. Emergent themes were quantified and examined using content and constant comparative analysis. Themes regarding antecedents across multiple levels of the social ecology emerged. Relationship problems, particularly with parents, were the most common suicide antecedent. Also, a pattern demonstrating a consistent progression toward suicidal behavior emerged from the data. Narratives indicated that youth were commonly exposed to one or more problems, often resulting in feelings of loneliness and burdensomeness, which progressed toward thoughts and sometimes plans for or attempts at suicide. Continued exposure to negative experiences and thoughts/plans about suicide, and/or self-injurious acts resulted in an acquired capacity to self-harm, eventually leading to suicide. These findings provide support for theories of suicidal behavior and highlight the importance of multi-level, comprehensive interventions that address individual cognitions and build social connectedness and support, as well as prevention strategies that increase awareness of the warning signs and symptoms of suicide, particularly among family members of at-risk youth.

Suicide Among Military Personnel and Veterans Aged 18-35 Years by County-16 States.

INTRODUCTION: Suicide among military personnel and young Veterans remains a health concern. This study examined stateside distribution of suicides by U.S. county to help focus prevention efforts. METHODS: Using 2005-2012 National Violent Death Reporting System data from 16 states (963 counties, or county-equivalent entities), this study mapped the county-level distribution of suicides among current military and Veteran decedents aged 18-35 years. This study also compared incident circumstances of death between decedents in high-density counties (i.e., counties with the highest proportion of deaths) versus those in medium/low-density counties to better understand the precipitators of suicide in counties most affected. Last, this study identified potential military and Veteran Health Administration intervention sites. All analyses were conducted in 2015. RESULTS: Within the National Violent Death Reporting System participating states, an estimated 262 (33%) current military suicides occurred in just ten (1.0%) counties, and 391 (33%) Veteran suicides occurred in 33 (3.4%) counties. Mental health and intimate partner problems were common precipitating circumstances, and some circumstances differed between cases in high- versus those in medium/low-density counties. Multiple potential intervention sites were identified in high-density counties. CONCLUSIONS: These findings suggest that military and Veteran suicides are concentrated in a small number of counties. Increased efforts at these locales might be beneficial.

Suicidal ideation among young Afghanistan/Iraq War Veterans and civilians: Individual, social, and environmental risk factors and perception of unmet mental healthcare needs, United States, 2013.

Suicidal Ideation among Afghanistan/Iraq War Veterans remains a health concern. As young Veterans adjust to civilian life, new risk factors might emerge and manifest differently in this group versus those in the general population. We explored these differences. With 2013 National Survey on Drug Use and Health data, we examined differences in risk of past-year suicidal ideation between Veterans of the Afghanistan/Iraq War periods aged 18-34 years (N=328) and age-comparable civilians (N=23,222). We compared groups based on individual and socio-environmental risk factors as well as perceptions of unmet mental healthcare needs. We report adjusted rate ratios (aRRs); interaction terms tested for between-group differences. PY suicidal ideation rates for Veterans and civilians did not differ (52 versus 59 per 1,000, p=0.60) and both groups shared many risk factors. However, drug problems and perceived unmet mental health care needs were vastly stronger risk factors among Veterans versus civilians (interaction terms indicated that the aRRs were 3.8-8.0 times higher for Veterans versus civilians). Other differences were discovered as well. Past-year suicidal ideation rates did not differ by Veteran status among young adults. However, different risk factors per group were detected, which can inform Veteran suicide prevention efforts.

Characteristics of Youth With Combined Histories of Violent Behavior, Suicidal Ideation or Behavior, and Gun-Carrying.

BACKGROUND: Youth reporting combined histories of nonfatal violence, suicidal ideation/behavior, and gun-carrying (VSG) are at risk for perpetrating fatal interpersonal violence and self-harm. AIMS: We characterized these youth to inform prevention efforts. METHOD: We analyzed 2004 data from 3,931 seventh-, ninth-, and 11-12th-grade youth and compared VSG youth (n = 66) with non-gun carrying youth who either had no histories of violence or suicidal thoughts/behavior (n = 1,839), histories of violence (n = 884), histories of suicidal thoughts/behaviors (n = 552), or both (n = 590). We compared groups based on demographic factors, risk factors (i.e., friends who engage in delinquency, peer-violence victimization, depressive symptoms, illicit substance use), and protective factors (i.e., school connectedness, parental care and supervision). Regression models identified factors associated with VSG youth. RESULTS: Illicit substance use and having friends who engage in delinquency were more common among VSG youth in all comparisons; almost all VSG youth had high levels of these factors. Depressive symptoms were positively associated with VSG youth versus youth without either violent or suicide-related histories and youth with violent histories alone. School connectedness and parental supervision were negatively associated with VSG youth in most comparisons. CONCLUSION: Family-focused and school-based interventions that increase connectedness while reducing delinquency and substance use might prevent these violent tendencies.

A Detailed Exploration Into the Association of Prescribed Opioid Dosage and Overdose Deaths Among Patients With Chronic Pain.

BACKGROUND: High opioid dosage has been associated with overdose, and clinical guidelines have cautioned against escalating dosages above 100 morphine-equivalent mg (MEM) based on the potential harm and the absence of evidence of benefit from high dosages. However, this 100 MEM threshold was chosen somewhat arbitrarily. OBJECTIVE: To examine the association of prescribed opioid dosage as a continuous measure in relation to risk of unintentional opioid overdose to identify the range of dosages associated with risk of overdose at a detailed level. METHODS: In this nested case-control study with risk-set sampling of controls, cases (opioid overdose decedents) and controls were identified from a population of patients of the Veterans Health Administration who were prescribed opioids and who have a chronic pain diagnosis. Unintentional fatal opioid analgesic overdose was measured from National Death Index records and prescribed opioid dosage from pharmacy records. RESULTS: The average prescribed opioid dosage was higher (P<0.001) for cases (mean=98.1 MEM, SD=112.7; median=60, interquartile range, 30-120), than controls (mean=47.7 MEM, SD=65.2; median=25, interquartile range, 15-45). In a ROC analysis, dosage was a moderately good "predictor" of opioid overdose death, indicating that, on average, overdose cases had a prescribed opioid dosage higher than 71% of controls. CONCLUSIONS: A clear cut-point in opioid dosage to distinguish between overdose cases and controls was not found. However, lowering the recommended dosage threshold below the 100 MEM used in many recent guidelines would affect proportionately few patients not at risk for overdose while potentially benefitting many of those at risk for overdose.

Vital Signs: Demographic and Substance Use Trends Among Heroin Users - United States, 2002-2013.

BACKGROUND: Heroin use and overdose deaths have increased significantly in the United States. Assessing trends in heroin use among demographic and particular substance-using groups can inform prevention efforts. METHODS: FDA and CDC analyzed data from the National Survey on Drug Use and Health and National Vital Statistics System reported during 2002-2013. Trends in heroin use among demographic and substance using groups were compared for 2002-2004, 2005-2007, 2008-2010, and 2011-2013. A multivariable logistic regression model was used to identify characteristics associated with heroin abuse or dependence. RESULTS: Annual average rates of past-year heroin use increased from 1.6 per 1,000 persons aged ≥ 12 years in 2002-2004 to 2.6 per 1,000 in 2011-2013. Rates of heroin abuse or dependence were strongly positively correlated with rates of heroin-related overdose deaths over time. For the combined data years 2011-2013, the odds of past-year heroin abuse or dependence were highest among those with past-year cocaine or opioid pain reliever abuse or dependence. CONCLUSIONS: Heroin use has increased significantly across most demographic groups. The increase in heroin abuse or dependence parallels the increase in heroin-related overdose deaths. Heroin use is occurring in the context of broader poly-substance use. IMPLICATIONS FOR PUBLIC HEALTH PRACTICE: Further implementation of a comprehensive response that targets the wider range of demographic groups using heroin and addresses the key risk factors for heroin abuse and dependence is needed. Specific response needs include reducing inappropriate prescribing and use of opioids through early identification of persons demonstrating problematic use, stronger prescription drug monitoring programs, and other clinical measures; improving access to, and insurance coverage for, evidence-based substance abuse treatment, including medication-assisted treatment for opioid use disorders; and expanding overdose recognition and response training and access to naloxone to treat opioid pain reliever and heroin overdoses.

Precipitating circumstances of suicide among active duty U.S. Army personnel versus U.S. civilians, 2005-2010.

To help understand suicide among soldiers, we compared suicide events between active duty U.S. Army versus civilian decedents to identify differences and inform military prevention efforts. We linked 141 Army suicide records from 2005 to 2010 to National Violent Death Reporting System (NVDRS) data. We described the decedents' military background and compared their precipitators of death captured in NVDRS to those of demographically matched civilian suicide decedents. Both groups commonly had mental health and intimate partner precipitating circumstances, but soldier decedents less commonly disclosed suicide intent.