Association Between Adverse Social Determinants of Health and Suicide Death.

OBJECTIVE: The aim of this study was to identify adverse social determinants of health (SDoH) International Statistical Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) code prevalence among individuals who died by suicide and to examine associations between documented adverse SDoH and suicide. RESEARCH DESIGN: A case-control study using linked medical record, insurance claim, and mortality data from 2000 to 2015 obtained from 9 Mental Health Research Network-affiliated health systems. We included 3330 individuals who died by suicide and 333,000 randomly selected controls matched on index year and health system location. All individuals in the study (cases and controls) had at least 10 months of enrollment before the study index date. The index date for the study for each case and their matched controls was the suicide date for that given case. RESULTS: Adverse SDoH documentation was low; only 6.6% of cases had ≥1 documented adverse SDoH in the year before suicide. Any documented SDoH and several specific adverse SDoH categories were more frequent among cases than controls. Any documented adverse SDoH was associated with higher suicide odds [adjusted odds ratio (aOR)=2.76; 95% CI: 2.38-3.20], as was family alcoholism/drug addiction (aOR=18.23; 95% CI: 8.54-38.92), being an abuse victim/perpetrator (aOR=2.53; 95% CI: 1.99-3.21), other primary support group problems (aOR=1.91; 95% CI: 1.32-2.75), employment/occupational maladjustment problems (aOR=8.83; 95% CI: 5.62-13.87), housing/economic problems (aOR: 6.41; 95% CI: 4.47-9.19), legal problems (aOR=27.30; 95% CI: 12.35-60.33), and other psychosocial problems (aOR=2.58; 95% CI: 1.98-3.36). CONCLUSIONS: Although documented SDoH prevalence was low, several adverse SDoH were associated with increased suicide odds, supporting calls to increase SDoH documentation in medical records. This will improve understanding of SDoH prevalence and assist in identification and intervention among individuals at high suicide risk.

How can social workers facilitate suicide prevention in primary care? Findings from a national survey of primary care physicians.

There are currently no national data regarding U.S. Primary Care Physicians' (PCPs') suicide screening practices. This study surveyed 302 U.S. PCPs about their current suicide screening practices to identify service gaps and intervention points for social workers. Although one-third of PCPs reported providing screening and safety planning, few were using evidence-based tools. Factors that increased the likelihood of routine screening were belief in the importance of screening (p < .01), time (p < .01), and access to co-located behavioral health (p < .01). Findings support the role of social workers in primary care and suggest areas for training and collaboration.

Predictors of substance use disorder treatment initiation and engagement among adolescents enrolled in Medicaid.

Background: Adolescents with substance use disorders (SUD) should receive timely access to treatment to improve lifelong outcomes. The Healthcare Effectiveness Data and Information Set (HEDIS) initiation and engagement in treatment (IET) performance measure was intended to promote quality improvement for patients with SUD. Yet, few studies have assessed predictors of measure performance among adolescents or other engagement in mental health services, which is critical to understanding disparities in treatment quality or opportunities for targeted improvement strategies. The present study reports the rates and predictors of IET among adolescents with SUD, as well as receipt of any mental health services. Methods: The sample included adolescents enrolled in Medicaid in 14 states who had a qualifying diagnosis for SUD (2009-2013) and met HEDIS IET performance measure eligibility criteria. Three outcomes were assessed, including initiation of SUD treatment within 14 days of qualifying diagnosis, engagement in SUD treatment (2 or more encounters) within 30 days of initiation, and receipt of any mental health services (1 or more encounters) within 30 days of initiation. Logistic regression was used to identify demographic and clinical characteristics associated with outcomes. Results: Among 20,602 adolescents who met eligibility criteria, 49.5% initiated SUD treatment, 48.5% engaged in SUD treatment, and 70% received any mental health service. Adolescents with higher levels of clinical need (e.g., medical complexity, mental health comorbidity, and multiple SUD diagnoses) had significantly higher odds of initiating, but lower odds of engaging in treatment or receiving any mental health service. Conclusions: To increase the delivery of SUD treatment, efforts should target adolescents with co-occurring mental health needs, many of whom are receiving mental health services after SUD diagnosis. Integrating addiction and mental health services could address these missed opportunities.

Mental Health Service Utilization Among Medical Students with a Perceived Need for Care.

OBJECTIVE: The authors investigated levels of perceived need for help, patterns of mental health service utilization, and barriers to care among US medical students with a focus on students who perceived a need for help but did not report service use in the past 12 months. METHODS: The authors administered an online survey to 2,868 medical students at three schools in Ohio between January and February 2020 including validated scales for psychological distress, self-stigma, and an exploration of mental health treatment. The authors used multivariable logistic regression to identify factors associated with treatment and qualitative analysis to identify common barriers to care. RESULTS: Twenty-eight percent (N = 800) of 2,868 students responded to the survey. Fifty-six percent (n = 439) of students reported a perceived need for help, while 34.6% of these respondents (n = 152) did not receive treatment. Among those with perceived need who completed the survey (n = 388), Asian students compared to non-Hispanic white students (adjusted odds ratio [aOR] = 0.45, 95% confidence interval [CI] 0.25-0.82) and those with higher self-stigma (aOR = 0.90, 95% CI 0.87-0.94) had lower odds of service use. Students told by others to seek help (aOR = 2.82, 95% CI 1.71-4.64) were the only group with higher odds of service use. The most common barriers to care were lack of time, difficulty accessing services, and stigma. CONCLUSIONS: Despite a perceived need for help, many students do not seek care and experience treatment barriers. Schools can encourage help-seeking by identifying students in need, using targeted messaging, fostering a low-stigma environment, and removing barriers.

Suicide in Older Adults in Ohio: Characteristics, Precipitants, and Mental Health Service Utilization.

The objective of the research is to examine characteristics of Ohio suicide decedents ages 65 + (N = 1273) and factors associated with behavioral health (BH) services utilization. The Ohio Violent Death Reporting System, 2012-2015, was the data source. Logistic regression analyses were used to examine the association among characteristics, suicide means, and BH service utilization. Of the study subjects, 96.0% were non-Hispanic white; 84.6%, male; and 63.0% living in urban areas. About 75.1% used firearms; 27.6% reported recent BH treatment. Those who were never married, depressed, and had a prior suicide attempt were more likely to have BH treatment within two months of death. Findings suggest a need for training of primary and BH providers to improve screening and assessment, treatment, and follow up care for older adults, especially those with histories of suicide attempts, depression, and firearm access. Suggested interventions include annual BH screenings and lethal means restriction at the individual and community levels.

Mapping suicide mortality in Ohio: A spatial epidemiological analysis of suicide clusters and area level correlates.

Previous studies have investigated spatial patterning and associations of area characteristics with suicide rates in Western and Asian countries, but few have been conducted in the United States. This ecological study aims to identify high-risk clusters of suicide in Ohio and assess area level correlates of these clusters. We estimated spatially smoothed standardized mortality ratios (SMR) using Bayesian conditional autoregressive models (CAR) for the period 2004 to 2013. Spatial and spatio-temporal scan statistics were used to detect high-risk clusters of suicide at the census tract level (N=2952). Logistic regression models were used to examine the association between area level correlates and suicide clusters. Nine statistically significant (p<0.05) high-risk spatial clusters and two space-time clusters were identified. We also identified several significant spatial clusters by method of suicide. The risk of suicide was up to 2.1 times higher in high-risk clusters than in areas outside of the clusters (relative risks ranged from 1.22 to 2.14 (p<0.01)). In the multivariate model, factors strongly associated with area suicide rates were socio-economic deprivation and lower provider densities. Efforts to reduce poverty and improve access to health and mental health medical services on the community level represent potentially important suicide prevention strategies.

Quality of care for Medicaid-enrolled youth with bipolar disorders.

This study examined conformance to clinical practice guidelines for children and adolescents with bipolar disorders and identified patient and provider factors associated with guideline concordant care. Administrative records were examined for 4,047 Medicaid covered youth aged 5-18 years with new episodes of bipolar disorder during 2006-2010. Main outcome measures included 5 claims-based quality of care measures reflecting national treatment guidelines. Measures addressed appropriate pharmacotherapy, therapeutic drug monitoring, and psychosocial treatment. The results indicated that current treatment practices for youth diagnosed with bipolar disorder typically fall short of recommended practice guidelines. Although the majority of affected youth are treated with recommended first-line pharmacotherapy, only a minority receive therapeutic drug monitoring and/or psychotherapy of recommended duration, underscoring the need for quality improvement initiatives.

Individual and contextual-level factors associated with continuity of care for adults with schizophrenia.

This retrospective cohort study examined rates of conformance to continuity of care treatment guidelines and factors associated with conformance for persons with schizophrenia. Subjects were 8,621 adult Ohio Medicaid recipients, aged 18-64, treated for schizophrenia in 2004. Information on individual-level (demographic and clinical characteristics) and contextual-level variables (county socio-demographic, economic, and health care resources) were abstracted from Medicaid claim files and the Area Resource File. Outcome measures captured four dimensions of continuity of care: (1) regularity of care; (2) transitions; (3) care coordination, and (4) treatment engagement. Multilevel modeling was used to assess the association between individual and contextual-level variables and the four continuity of care measures. The results indicated that conformance rates for continuity of care for adults with schizophrenia are below recommended guidelines and that variations in continuity of care are associated with both individual and contextual-level factors. Efforts to improve continuity of care should target high risk patient groups (racial/ethnic minorities, the dually diagnosed, and younger adults with early onset psychosis), as well as community-level risk factors (provider supply and geographic barriers of rural counties) that impede access to care.

Association Between Marijuana Laws and Suicide Among 12- to 25-Year-Olds in the United States From 2000 to 2019.

OBJECTIVE: Cannabis use is associated with suicide-related outcomes in both adolescents and adults, and may be increasing amid shifting cannabis policies. However, little is known about the impact of medical marijuana legalization (MML) and recreational marijuana legalization (RML) policies on youth suicide. Using 20 years of national data, we examined associations between MML, RML, and suicide-related mortality among US individuals aged 12 to 25 years, and assessed whether they varied based on age and sex. METHOD: Suicide deaths (N = 113,512) from the 2000-2019 National Vital Statistics System Multiple Cause of Death files for age groups 12 to 13, 14 to 16, 17 to 19, 20 to 22, and 23 to 25 years were examined in relation to time-varying cannabis law status using a staggered adoption difference-in-difference (DiD) approach with a negative binomial regression to determine associations between MML, RML, and suicide rates, controlling for individual- and state-level covariates and accounting for the varying effective dates of MML and RML by state. RESULTS: The overall unadjusted annual suicide rate was 10.93/100,000, varying from 9.76 (states without marijuana laws (ML)) to 12.78 (MML states) to 16.68 (RML states). In multivariable analysis, both MML (incidence rate ratio [IRR] = 1.10, 95% CI: 1.05-1.15) and RML (IRR = 1.16, 95% CI: 1.06-1.27) were associated with higher suicide rates among female youth compared to those in states without ML. Youth aged 14 to 16 years had higher rates of suicide in states with RML compared to states with MML (IRR = 1.14, 95% CI: 1.00-1.30) and states without ML (IRR = 1.09, 95% CI: 1.00-1.20). Findings were consistent across sensitivity analyses. CONCLUSION: MML and RML were associated with increased suicide-related mortality in female youth and 14- to- 16-year-old individuals of both sexes. Mechanisms through which cannabis policies are related to increased youth suicide warrant further study and should inform legislative reform.

Identifying Subgroups of Youth Suicide Decedents Based on Clinical Profiles of Psychiatric and Medical Diagnoses: A Latent Class Analysis.

PURPOSE: To identify risk subgroups of youth suicide decedents using demographic and clinical psychiatric and medical diagnostic profiles to inform tailored youth suicide prevention efforts. METHODS: This study linked Ohio Medicaid and death certificate data for Medicaid enrolled youth aged 8-25 years who died by suicide between January 1, 2010, and December 31, 2020 (N = 511). Latent class analysis was used to identify distinct clinical risk subgroups. RESULTS: Three latent classes were identified. Internalizing problems were common across all classes, but especially prevalent in class 1, the High Internalizing + Multiple Comorbidities group (n = 152, 30%). A prior history of suicidal behavior was confined to class 1 decedents, who were otherwise characterized by substance misuse, and multiple psychiatric and medical comorbidities. Class 2 decedents, the Internalizing + Externalizing group (n = 176, 34%), were more often younger, male, Black, and unlikely to have a history of substance misuse. Decedents in class 3, the Internalizing + Substance Misuse group (n = 183, 36%), were more often older and likely to have a history of substance misuse, but unlikely to exhibit other externalizing problems. DISCUSSION: Internalizing psychopathology is particularly common among youth who die by suicide, with comorbid externalizing psychopathology, substance misuse, and medical problems contributing to youth suicide risk. Because less than a third of youth who die by suicide have a prior history of recognized suicidal thinking or behavior, universal screening for youth suicide risk should be considered, particularly in younger children, and efforts to integrate suicide prevention in traditional health care settings should be prioritized.

Association Between Social Determinants of Health and Deliberate Self-Harm Among Youths With Psychiatric Diagnoses.

OBJECTIVE: The authors sought to examine the association between adverse social determinants of health (SDoHs) and risk for self-harm among youths. METHODS: The authors performed a retrospective longitudinal analysis of Ohio Medicaid claims data (April 1, 2016-December 31, 2018) of 244,958 youths (ages 10-17 years) with a primary psychiatric diagnosis. SDoHs were identified from ICD-10 codes and classified into 14 categories, encompassing abuse and neglect, child welfare placement, educational problems, financial problems, exposure to violence, housing instability, legal issues, disappearance or death of a family member, family disruption by separation or divorce, family alcohol or drug use, parent-child conflict, other family problems, social and environmental problems, and nonspecific psychosocial needs. Cox proportional hazards analysis was used to examine the association between SDoHs and self-harm (i.e., nonsuicidal self-injury or suicide attempt). Analyses controlled for demographic characteristics and comorbid psychiatric and general medical conditions. RESULTS: During follow-up after an index claim event, 51,796 youths (21.1%) had at least one adverse SDoH indicator, and 3,262 (1.3%) had at least one self-harm event. Abuse and neglect (hazard ratio [HR]=1.90, 99% CI=1.70-2.12), child welfare placement (HR=1.32, 99% CI=1.04-1.67), parent-child conflict (HR=1.52, 99% CI=1.23-1.87), other family problems (HR=1.25, 99% CI=1.01-1.54), and nonspecific psychosocial needs (HR=1.41, 99% CI=1.06-1.89) were associated with significantly increased hazard of self-harm. CONCLUSIONS: Adverse SDoHs were significantly associated with self-harm, even after controlling for demographic and clinical characteristics, underscoring the need for capturing SDoH information in medical records to identify youths at elevated suicide risk and to inform targeted interventions.

Child Welfare System-Level Factors Associated with All-Cause Mortality Among Children in Foster Care in the United States, 2009-2018.

Little is known about the impact of child welfare system-level factors on child mortality as an outcome within foster care. Using data from the Adoption and Foster Care Analysis and Reporting System, 2009-2018, we examined the associations between county-level sociodemographic, foster care performance, and judicial reform characteristics with all-cause mortality rates. Results of random effects negative binomial regression analyses showed that higher proportions of younger children (<1 year: IRR = 1.06, 95% CI [1.02, 1.11]; 5-9 years: IRR = 1.05, 95% CI [1.01, 1.09]); children of color (i.e., non-Hispanic Asian: IRR = 1.07, 95% CI [1.01, 1.13]; multiracial: IRR = 1.03, 95% CI [1.01, 1.04]; non-Hispanic Black: IRR = 1.02, 95% CI [1.01, 1.02]; Hispanic: IRR = 1.01, 95% CI [1.01, 1.02]); and male children (IRR = 1.10, 95% CI [1.05, 1.15]) were associated with higher mortality risks at the county level. Current class action lawsuits (IRR = 0.79, 95% CI [0.63, 0.99]) and active consent decrees (IRR = 0.77, 95% CI [0.63, 0.94]) were associated with lower mortality risks. None of the foster care performance characteristics (e.g., foster care entry, placement stability, permanency) were associated with mortality risks. These findings have implications for addressing health disparities and reforming foster care systems through programmatic and policy efforts.

Impact of Treatment Initiation and Engagement on Deliberate Self-Harm Among Individuals With First-Episode Psychosis.

OBJECTIVE: Individuals with psychosis are at increased risk for suicide, with the greatest risk being present during the first few months after diagnosis. The authors aimed to examine whether treatment initiation within 14 days of diagnosis and treatment engagement within 90 days of initiation reduce the risk for deliberate self-harm (DSH) among individuals with first-episode psychosis (FEP). METHODS: A retrospective longitudinal cohort design was adopted by using Ohio Medicaid claims for 6,349 adolescents and young adults ages 15-24 years with FEP. Logistic regression was used to examine factors associated with treatment initiation and engagement. Cox proportional hazard models were used to estimate the impact of treatment initiation and engagement on DSH. Propensity score weighting was used to control for sociodemographic and clinical covariates. RESULTS: Approximately 70% of the sample initiated treatment, 55% of whom engaged in treatment. Treatment initiation and engagement were associated with both demographic and clinical variables. Treatment initiation significantly reduced the hazard of DSH (average treatment effect in the entire population: hazard ratio [HR]=0.62, 95% CI=0.47-0.81; average treatment effect among those treated: HR=0.64, 95% CI=0.52-0.80). In contrast, treatment engagement was not significantly associated with DSH. CONCLUSIONS: These results suggest that the initial treatment contact is essential for reducing DSH among adolescents and young adults with FEP. Additionally, the finding that treatment engagement did not reduce DSH suggests that standard clinical care may not be sufficient for reducing DSH in this population. These findings highlight the need for suicide-specific interventions for individuals with FEP.

Spatiotemporal Patterns of Deaths of Despair Across the U.S., 2000-2019.

INTRODUCTION: Deaths of despair (i.e., suicide, drug/alcohol overdose, and chronic liver disease and cirrhosis) have been increasing over the past 2 decades. However, no large-scale studies have examined geographic patterns of deaths of despair in the U.S. This ecologic study identifies geographic and temporal patterns of individual and co-occurring clusters of deaths of despair. METHODS: All individuals aged ≥10 years who died in the U.S. between 2000 and 2019 and resided within the 48 contiguous states and Washington, District of Columbia were included (N=2,171,105). Causes of death were limited to deaths of despair, namely suicide, drug/alcohol overdose, and chronic liver disease and cirrhosis. Univariate and multivariate space-time scan statistics were used to identify individual and co-occurring clusters with excess risk of deaths of despair. County-level RRs account for heterogeneity within each cluster. Analyses were conducted from late 2021 to early 2022. RESULTS: Six suicide clusters, four overdose clusters, nine liver disease clusters, and three co-occurring clusters of all three types of deaths were identified. A large portion of the western U.S., southeastern U.S., and Appalachia/rust belt were contained within the co-occurring clusters. The co-occurring clusters had average county RRs ranging from 1.17 (p<0.001) in the southeastern U.S. to 4.90 (p<0.001) in the western U.S. CONCLUSIONS: Findings support identifying and targeting risk factors common to all types of deaths of despair when planning public health interventions. Resources and policies that address all deaths of despair simultaneously may be beneficial for the areas contained within the co-occurring high-risk clusters.

Youth Suicide During the First Year of the COVID-19 Pandemic.

OBJECTIVE: To identify potential differential changes in youth suicide deaths associated with the coronavirus disease (COVID-19) pandemic to better inform suicide prevention strategies. METHODS: This cross-sectional study analyzed national suicide data for US youth aged 5 to 24 years from 2015 to 2020. Annual and monthly numbers of suicides were extracted overall and by sex, age, race and ethnicity, and method. Expected suicides were modeled from the trend in monthly deaths before COVID-19 (January 1, 2015-February 29, 2020), by using interrupted time-series analyses with quasi-Poisson regression. Rate ratios (RR) and corresponding 95% confidence intervals (CI) were used to compare expected and observed suicides during the first 10 months of COVID-19 (March 1, 2020-December 31, 2020). RESULTS: Among 5568 identified youth suicides during the 2020 pandemic, 4408 (79.2%) were male, 1009 (18.1%) Hispanic, 170 (3.3%) non-Hispanic American Indian/Alaska Native, 262 (4.7%) Asian/Pacific Islander, 801 (14.4%) Black, and 3321 (59.6%) white. There was a significant increase in overall observed versus expected youth suicides during the COVID-19 pandemic (RR = 1.04, 95% CI = 1.01-1.07), equivalent to an estimated 212 excess deaths. Demographic subgroups including males (RR = 1.05, 95% CI = 1.02-1.08), youth aged 5 to 12 years (RR = 1.20, 95% CI = 1.03-1.41) and 18 to 24 years (RR =1.05, 95% CI = 1.02-1.08), non-Hispanic AI/AN youth (RR = 1.20, 95% CI = 1.03-1.39), Black youth (RR = 1.20, 95% CI = 1.12-1.29), and youth who died by firearms (RR = 1.14, 95% CI = 1.10-1.19) experienced significantly more suicides than expected. CONCLUSIONS: Suicide deaths among US youth increased during COVID-19, with substantial variation by sex, age, race and ethnicity, and suicide method. Suicide prevention strategies must be tailored to better address disparities in youth suicide risk.

Correlates of deliberate self-harm among youth with bipolar disorder.

OBJECTIVE: Youth with bipolar disorder (BD) are at high risk for deliberate self-harm (DSH) and suicide. However, research regarding factors associated with DSH, a key suicide risk factor, among youth with BD is limited. In a population-based sample of youth with BD, we therefore investigated associations between demographic, clinical, and service utilization factors and DSH incidence and compared suicide, unintentional injury, and all-cause mortality to the general population. METHOD: We analyzed a retrospective cohort of youth aged 5 to 19 years with a new BD episode between 2010 and 2017 (n = 25,244) using Ohio Medicaid claims and death certificate data. Cox proportional hazards models examined associations between different factors and DSH. Mortality rates were compared to the general population using standardized mortality ratios. RESULTS: During follow-up, 1,517 (6.0%) youth had at least one DSH event. Older index age, female sex, comorbid psychiatric/medical conditions, prior DSH/suicidal ideation, and prior ER mental healthcare were associated with increased DSH risk. Prior DSH was most strongly associated with increased DSH risk for 3 months after a new BD episode. Being non-Hispanic Black (vs. White, non-Hispanic) and prior psychiatric hospitalization were associated with decreased DSH hazard. DSH risk was highest for 3 months after a new BD episode. Suicide, unintentional injury, and all-cause mortality rates were elevated in youth with BD. LIMITATIONS: May not generalize to other states or non-Medicaid populations; claims data cannot distinguish suicidal intent of self-harm CONCLUSION: Early intervention following a new BD episode, particularly among high-risk groups, is key to prevent DSH.

Risk Factors for Deliberate Self-harm and Suicide Among Adolescents and Young Adults With First-Episode Psychosis.

Little is known about the risk factors for deliberate self-harm (DSH) and suicide among adolescents and young adults with first episode psychosis (FEP) and the longitudinal course of DSH following the initial onset of illness. This study identifies risk factors for DSH and suicide death among Medicaid-covered adolescents and young adults with FEP along with the periods of greatest risk for DSH after diagnosis. A retrospective longitudinal cohort analysis was performed using Medicaid claims data merged with death certificate data for 19 422 adolescents and young adults (aged 15-24 years) diagnosed with the onset of FEP between 2010 and 2017. DSH per 1000 person-years and standardized mortality rates for suicide were determined. Hazard ratios of DSH and suicide were estimated by Cox proportional hazard models. During follow-up, 2148 (11.1%) individuals had at least one self-harm event and 22 (0.1%) died by suicide. The hazards of DSH were significantly higher for those with a previous DSH, suicidal ideation, child abuse and neglect, comorbid medical and psychiatric diagnoses, and prior mental health care. The median follow-up time for those who had DSH was 208.0 days (SD: 526.5 days) in adolescents and 108.0 days (SD: 340.0 days) in young adults. Risk of DSH was highest in the first 3 months following FEP. Individuals with FEP are at high risk for self-harm and suicidal behavior, and recognition of who among these individuals and when following illness onset they are at greatest risk may guide more precise clinical recognition and intervention.

Factors associated with antidepressant adherence for Medicaid-enrolled children and adolescents.

BACKGROUND: Antidepressants have been shown to be efficacious for the treatment of pediatric depression. However, many youths do not receive an adequate duration of treatment, and factors associated with nonadherence in this population remain poorly understood. OBJECTIVE: To examine rates of antidepressant adherence for depressed youth and identify factors associated with adherence during the acute and continuation phases of treatment. METHODS: A retrospective cohort analysis was conducted using claims data from a state Medicaid-enrolled population of 1650 youths (aged 5-17 years) with new episodes of depression between January 1, 2005, and December 30, 2007. These patients were treated with selective serotonin reuptake inhibitors or newer antidepressants and followed for 6 months from the first prescription fill date. Adherence measures were derived from the Health Plan Employer Data and Information Set (HEDIS) quality indicators on antidepressant management (3 months of continuous treatment for the acute phase and 6 months for the continuation phase) and assessed using the medication possession ratio. Multivariate logistic regression analyses evaluated the association between demographic, clinical, medication, and treatment factors, and adherence. RESULTS: About half (49.5%) of the youths were adherent to antidepressant medication during the acute phase, and 42% of these were adherent during the continuation phase; 21% were adherent across both treatment phases. Optimal follow-up visits and adequate antidepressant dosing was associated with better adherence during both treatment phases, as was use of other psychotropic medications. Youths prescribed trazodone for sleep had higher adherence rates during the acute phase. Minority youths and adolescents had lower adherence rates during the acute phase. Youths in foster care had higher adherence rates during both treatment phases. CONCLUSIONS: Nonadherence with antidepressant medications is common among Medicaid-covered children and adolescents. Study findings underscore the need for clinicians to deliver guideline-concordant care, assess adherence, and develop interventions that improve adherence, particularly for vulnerable subgroups.

Depression and Suicide-Risk Screening Results in Pediatric Primary Care.

BACKGROUND AND OBJECTIVES: Depression is common, and suicide rates are increasing. Adolescent depression screening might miss those with unidentified suicide risk. Our primary objective in this study was to compare the magnitude of positive screen results across different approaches. METHODS: From June 2019 to October 2020, 803 mostly Medicaid-enrolled adolescents aged ≥12 years with no recent history of depression or self-harm were screened with the Patient Health Questionnaire-9 Modified for Adolescents (PHQ-9A) and the Ask Suicide-Screening Questions (ASQ) across 12 primary care practices. Two PHQ-9A screening strategies were evaluated: screening for any type of depression or other mental illness (positive on any item) or screening for major depressive disorder (MDD) (total score ≥10). RESULTS: Overall, 56.4% of patients screened positive for any type of depression, 24.7% screened positive for MDD, and 21.1% screened positive for suicide risk. Regardless of PHQ-9A screening strategy, the ASQ identified additional subjects (eg, 2.2% additional cases compared with screening for any type of depression or other mental illness and 8.3% additional cases compared with screening positive for MDD). Of those with ≥6 month follow-up, 22.9% screened positive for any type of depression (n = 205), 35.6% screened positive for MDD (n = 90), and 42.7% with a positive ASQ result (n = 75) had a depression or self-harm diagnosis or an antidepressant prescription. CONCLUSIONS: Suicide risk screening identifies cases not identified by depression screening. In this study, we underscore opportunities and challenges in primary care related to the high prevalence of depression and suicide risk. Research is needed regarding optimal screening strategies and to help clinicians manage the expected number of screening-identified adolescents.

Health Services Use by Children in the Welfare System Who Died by Suicide.

OBJECTIVES: To examine characteristics and health service use patterns of suicide decedents with a history of child welfare system involvement to inform prevention strategies and reduce suicide in this vulnerable population. METHODS: A retrospective matched case-control design (120 suicide decedents and 1200 matched controls) was implemented. Suicide decedents included youth aged 5 to 21 who died by suicide and had an open case in Ohio's Statewide Automated Child Welfare Information System between 2010 and 2017. Controls were matched to suicide decedents on sex, race, and ethnicity. Comparisons were analyzed by using conditional logistic regressions to control for matching between the suicide and control groups. RESULTS: Youth in the child welfare system who died by suicide were significantly more likely to experience out-of-home placements and be diagnosed with mental and physical health conditions compared with controls. Suicide decedents were twice as likely to access mental health services in the 1 and 6 months before death, regardless of the health care setting. A significantly higher percentage of suicide decedents used physical health services 6 months before their death or index date. Emergency department visits for both physical and mental health conditions were significantly more likely to occur among suicide decedents. CONCLUSIONS: Suicide decedents involved in the child welfare system were more likely to use both mental and physical health care services in the months before their death or index date. Findings suggest that youth involved in the child welfare system may benefit from suicide prevention strategies in health care settings.