Use of General Primary Care, Specialized Primary Care, and Other Veterans Affairs Services Among High-Risk Veterans.

Importance: Integrated health care systems increasingly focus on improving outcomes among patients at high risk for hospitalization. Examining patterns of where patients obtain care could give health care systems insight into how to develop approaches for high-risk patient care; however, such information is rarely described. Objective: To assess use of general and specialized primary care, medical specialty, and mental health services among patients at high risk of hospitalization in the Veterans Health Administration (VHA). Design, Setting, and Participants: This national, population-based, retrospective cross-sectional study included all veterans enrolled in any type of VHA primary care service as of September 30, 2015. Data analysis was performed from April 1, 2016, to January 1, 2019. Exposures: Risk of hospitalization and assignment to general vs specialized primary care. Main Outcome and Measures: High-risk veterans were defined as those who had the 5% highest risk of near-term hospitalization based on a validated risk prediction model; all others were considered low risk. Health care service use was measured by the number of encounters in general primary care, specialized primary care, medical specialty, mental health, emergency department, and add-on intensive management services (eg, telehealth and palliative care). Results: The study assessed 4 309 192 veterans (mean [SD] age, 62.6 [16.0] years; 93% male). Male veterans (93%; odds ratio [OR], 1.11; 95% CI, 1.10-1.13), unmarried veterans (63%; OR, 2.30; 95% CI, 2.32-2.35), those older than 45 years (94%; 45-65 years of age: OR, 3.49 [95% CI, 3.44-3.54]; 66-75 years of age: OR, 3.04 [95% CI, 3.00-3.09]; and >75 years of age: OR, 2.42 [95% CI, 2.38-2.46]), black veterans (23%; OR, 1.63; 95% CI, 1.61-1.64), and those with medical comorbidities (asthma or chronic obstructive pulmonary disease: 33%; OR, 4.03 [95% CI, 4.00-4.06]; schizophrenia: 4%; OR, 5.14 [95% CI, 5.05-5.22]; depression: 42%; OR, 3.10 [95% CI, 3.08-3.13]; and alcohol abuse: 20%; OR, 4.54 [95% CI, 4.50-4.59]) were more likely to be high risk (n = 351 012). Most (308 433 [88%]) high-risk veterans were assigned to general primary care; the remaining 12% (42 579 of 363 561) were assigned to specialized primary care (eg, women's health and homelessness). High-risk patients assigned to general primary care had more frequent primary care visits (mean [SD], 6.9 [6.5] per year) than those assigned to specialized primary care (mean [SD], 6.3 [7.3] per year; P < .001). They also had more medical specialty care visits (mean [SD], 4.4 [5.9] vs 3.7 [5.4] per year; P < .001) and fewer mental health visits (mean [SD], 9.0 [21.6] vs 11.3 [23.9] per year; P < .001). Use of intensive supplementary outpatient services was low overall. Conclusions and Relevance: The findings suggest that, in integrated health care systems, approaches to support high-risk patient care should be embedded within general primary care and mental health care if they are to improve outcomes for high-risk patient populations.

Characteristics and Injury Mechanisms of Veteran Primary Care Suicide Decedents with and without Diagnosed Mental Illness.

BACKGROUND: In the United States, suicide rates are increasing among nearly all age groups. Primary care is a critical setting for suicide prevention, where interventions often rely on identifying mental health conditions as indicators of elevated suicide risk. OBJECTIVE: Quantify the proportion of suicide decedents within primary care who had no antecedent mental health or substance use diagnosis. DESIGN: Retrospective cohort study. PARTICIPANTS: Veterans who received Veterans Health Administration (VHA) primary care any time from 2000 to 2014 and died by suicide before 2015 (n = 27,741). MAIN MEASURES: We categorized decedents by whether they had any mental health or substance use diagnosis (yes/no) using ICD-9 codes available from VHA records. We compared sociodemographic, clinical, and suicide mechanism characteristics between groups using chi-square, Student's T, or Wilcoxon tests. RESULTS: Forty-five percent of decedents had no mental health or substance use diagnosis. Decedents without such a diagnosis were older (68 vs. 57 years, p < 0.001), and more likely to be male (98.3% vs. 95.8%, p < 0.001), non-Hispanic White (90.6% vs. 87.9%, p < 0.001), married/partnered (50.4% vs. 36.6%, p < 0.001), and without military service-connected disability benefits (72.6% vs. 56.9%, p < 0.001). They were also more likely to die from firearm injury (78.9% vs. 60.7%, p < 0.001). There were statistically significant differences in physical health between groups, but the magnitudes of those differences were small. Decedents without a mental health or substance use diagnosis had significantly shorter durations of enrollment in VHA healthcare, less healthcare utilization in their last year of life, and had little utilization aside from primary care visits. CONCLUSIONS AND RELEVANCE: From 2000 to 2014, of nearly thirty thousand VHA primary care patients who died by suicide, almost half had no antecedent mental health or substance use diagnosis. Within VHA primary care settings, suicide risk screening for those with and without such a diagnosis is indicated.

Mortality Among Veterans with Major Mental Illnesses Seen in Primary Care: Results of a National Study of Veteran Deaths.

BACKGROUND: Premature mortality observed among the mentally ill is largely attributable to chronic illnesses. Veterans seen within Veterans Affairs (VA) have a higher prevalence of mental illness than the general population but there is limited investigation into the common causes of death of Veterans with mental illnesses. OBJECTIVE: To characterize the life expectancy of mentally ill Veterans seen in VA primary care, and to determine the most death rates of combinations of mental illnesses. DESIGN: Retrospective cohort study of decedents. SETTING/PARTICIPANTS: Veterans seen in VA primary care clinics between 2000 and 2011 were included. Records from the VA Corporate Data Warehouse (CDW) were merged with death information from the National Death Index. MAIN MEASURES: Mental illnesses were determined using ICD9 codes. Direct standardization methods were used to calculate age-adjusted gender and cause-specific death rates per 1000 deaths for patients with and without depression, anxiety, post-traumatic stress disorder (PTSD), substance use disorder (SUD), serious mental illness (SMI), and combinations of those diagnoses. KEY RESULTS: Of the 1,763,982 death records for Veterans with 1 + primary care visit, 556,489 had at least one mental illness. Heart disease and cancer were the two leading causes of death among Veterans with or without a mental illness, accounting for approximately 1 in 4 deaths. Those with SUD (n = 204,950) had the lowest mean age at time of death (64 ± 12 years). Among men, the death rates were as follows: SUD (55.9/1000); anxiety (49.1/1000); depression (45.1/1000); SMI (40.3/1000); and PTSD (26.2/1000). Among women, death rates were as follows: SUD (55.8/1000); anxiety (36.7/1000); depression (45.1/1000); SMI (32.6/1000); and PTSD (23.1/1000 deaths). Compared to men (10.8/1000) and women (8.7/1000) without a mental illness, these rates were multiple-fold higher in men and in women with a mental illness. A greater number of mental illness diagnoses was associated with higher death rates among men and women (p < 0.0001). CONCLUSIONS: Veterans with mental illnesses, particularly those with SUD, and those with multiple diagnoses, had shorter life expectancy than those without a mental illness. Future studies should examine both patient and systemic sources of disparities in providing chronic illness care to Veterans with a mental illness.

A Resampling Based Grid Search Method to Improve Reliability and Robustness of Mixture-Item Response Theory Models of Multimorbid High-Risk Patients.

There are many statistics available to the applied statistician for assessing model fit and even more methods for assessing internal and external validity. We detail a useful approach using a grid search technique that balances the internal model consistency with generalizability and can be used with models that naturally lend themselves to multiple assessment techniques. Our method relies on resampling and a simple grid search method over 3 commonly used statistics that are simple to calculate. We apply this method in a latent traits framework using a mixture Item Response Theory (MIXIRT) model of common chronic health conditions. Model fit is assessed using Akaike's Information Criteria (AIC), latent class similarity is measured with the Variance of Information (VI), and the consistency of condition complexity and prevalence across latent classes is compared using Kendall's τ rank order statistic. From two patient cohorts at high risk for hospitalization in 2014 and 2018, we generated 19 MIXIRT models (allowing 2-20 latent classes) on 21 common comorbid conditions identified via healthcare encounter diagnosis codes. We ran these models on 100 bootstrap samples of size 10% for each cohort. Among the resulting models, combined AIC and VI statistics identified 5-7 latent classes, but the rank order correlation of condition complexity revealed that only the 5 class solutions had consistent condition complexity. The 5 class solutions were combined to produce a single parsimonious MIXIRT solution that balanced clinical significance with model fit, cluster similarity, and consistency of condition complexity.

Which patients are persistently high-risk for hospitalization?

OBJECTIVES: Many healthcare systems use prediction models to estimate and manage patient-level probability of hospitalization. Patients identified as high-risk at one point in time may not, however, remain high-risk. We aimed to describe subgroups of patients with distinct longitudinal risk score patterns to inform interventions tailored to patients' needs. STUDY DESIGN: Retrospective national cohort study. METHODS: Using a previously validated prediction algorithm, we identified a cohort of 258,759 patients enrolled in the Veterans Health Administration (VHA) who were in the top 5% of risk for hospitalization within 90 days. During each of the following 24 months, patients were placed in 1 of 6 categories: death, hospitalized, no VHA care, persistently high-risk for hospitalization (≥10% probability), initially high-risk then persistently low-risk (<10% probability), and intermittently high-risk. We used multivariable logistic regression to identify characteristics predictive of being persistently high-risk through the last study month. RESULTS: After 2 years, 17.7% had died, 13.8% had remained persistently high-risk for hospitalization, 41.5% had become persistently low-risk, and 19.9% were intermittently high-risk. Predictors of being persistently high-risk included urban residence, chronic medical comorbidities, auditory and visual impairment, chronic pain, any cancer diagnosis, and social instability. CONCLUSIONS: Few patients who were high-risk for hospitalization at baseline remained so. Nonrandomized evaluations of interventions that identify patients based on a single high-risk score may spuriously appear to have positive effects. Clinical interventions may need to focus on individuals who are persistently high-risk.

Clinical Care Quality Among Veterans Health Administration Patients With Mental Illness Following Medical Home Implementation.

OBJECTIVE: This study sought to compare quality of care following medical home implementation among Veterans Health Administration (VHA) primary care patients with and without mental illness. METHODS: VHA primary care patients seen between April 2010 and March 2013 whose medical records were reviewed by the VHA External Peer Review Program were identified. The proportion of patients meeting quality indicators in each mental illness group (depression, posttraumatic stress disorder, anxiety disorder, substance use disorder, serious mental illness, and any mental illness) was compared with the proportion of patients without mental illness. Sample sizes ranged from 210,864 to 236,421. Differences of 5.0% or greater were deemed clinically important, and higher proportions indicated higher quality of care across 33 clinical indicators. RESULTS: The proportion of veterans meeting clinical quality indicators ranged from 64.7% to 99.6%. Differences of ≥5.0% between veterans with and without mental illness were detected in six of 33 indicators. A greater proportion of veterans with mental illness received influenza immunizations (age 50-64) and had documented left ventricular functioning (among veterans with chronic heart failure) compared with veterans without mental illness. A lower proportion of veterans with substance use disorders or severe mental illness received colorectal cancer screening or met indicators related to recommended medications if diagnosed as having diabetes or ischemic heart disease. CONCLUSIONS: Contrary to earlier reports of lower-quality care, patients with and without mental illness had similar preventive and chronic disease management care quality following medical home implementation.

Identifying Latent Subgroups of High-Risk Patients Using Risk Score Trajectories.

OBJECTIVE: Many healthcare systems employ population-based risk scores to prospectively identify patients at high risk of poor outcomes, but it is unclear whether single point-in-time scores adequately represent future risk. We sought to identify and characterize latent subgroups of high-risk patients based on risk score trajectories. STUDY DESIGN: Observational study of 7289 patients discharged from Veterans Health Administration (VA) hospitals during a 1-week period in November 2012 and categorized in the top 5th percentile of risk for hospitalization. METHODS: Using VA administrative data, we calculated weekly risk scores using the validated Care Assessment Needs model, reflecting the predicted probability of hospitalization. We applied the non-parametric k-means algorithm to identify latent subgroups of patients based on the trajectory of patients' hospitalization probability over a 2-year period. We then compared baseline sociodemographic characteristics, comorbidities, health service use, and social instability markers between identified latent subgroups. RESULTS: The best-fitting model identified two subgroups: moderately high and persistently high risk. The moderately high subgroup included 65% of patients and was characterized by moderate subgroup-level hospitalization probability decreasing from 0.22 to 0.10 between weeks 1 and 66, then remaining constant through the study end. The persistently high subgroup, comprising the remaining 35% of patients, had a subgroup-level probability increasing from 0.38 to 0.41 between weeks 1 and 52, and declining to 0.30 at study end. Persistently high-risk patients were older, had higher prevalence of social instability and comorbidities, and used more health services. CONCLUSIONS: On average, one third of patients initially identified as high risk stayed at very high risk over a 2-year follow-up period, while risk for the other two thirds decreased to a moderately high level. This suggests that multiple approaches may be needed to address high-risk patient needs longitudinally or intermittently.

Patient-Centered Medical Home Implementation and Improved Chronic Disease Quality: A Longitudinal Observational Study.

OBJECTIVE: To examine associations between clinics' extent of patient-centered medical home (PCMH) implementation and improvements in chronic illness care quality. DATA SOURCE: Data from 808 Veterans Health Administration (VHA) primary care clinics nationwide implementing the Patient Aligned Care Teams (PACT) PCMH initiative, begun in 2010. DESIGN: Clinic-level longitudinal observational study of clinics that received training and resources to implement PACT. Clinics varied in the extent they had PACT components in place by 2012. DATA COLLECTION: Clinical care quality measures reflecting intermediate outcomes and care processes related to coronary artery disease (CAD), diabetes, and hypertension care were collected by manual chart review at each VHA facility from 2009 to 2013. FINDINGS: In adjusted models containing 808 clinics, the 77 clinics with the most PACT components in place had significantly larger improvements in five of seven chronic disease intermediate outcome measures (e.g., BP < 160/100 in diabetes), ranging from 1.3 percent to 5.2 percent of the patient population meeting measures, and two of eight process measures (HbA1c measurement, LDL measurement in CAD) than the 69 clinics with the least PACT components. Clinics with moderate levels of PACT components showed few significantly larger improvements than the lowest PACT clinics. CONCLUSIONS: Veterans Health Administration primary care clinics with the most PCMH components in place in 2012 had greater improvements in several chronic disease quality measures in 2009-2013 than the lowest PCMH clinics.

Housing instability and mental distress among US veterans.

Evidence has suggested increased risk for homelessness and suicide among US veterans, but little is known about the associations between housing instability and psychological distress (including suicidal ideation). We examined frequent mental distress (FMD) and suicidal ideation among a probability-based sample of 1767 Nebraska veterans who participated in the 2010 Behavioral Risk Factor Surveillance Survey who had and had not experienced housing instability in the past 12 months. Veterans experiencing housing instability had increased odds of FMD and suicidal ideation.

Prevalence of gender identity disorder and suicide risk among transgender veterans utilizing veterans health administration care.

OBJECTIVES: We estimated the prevalence and incidence of gender identity disorder (GID) diagnoses among veterans in the Veterans Health Administration (VHA) health care system and examined suicide risk among veterans with a GID diagnosis. METHODS: We examined VHA electronic medical records from 2000 through 2011 for 2 official ICD-9 diagnosis codes that indicate transgender status. We generated annual period prevalence estimates and calculated incidence using the prevalence of GID at 2000 as the baseline year. We cross-referenced GID cases with available data (2009-2011) of suicide-related events among all VHA users to examine suicide risk. RESULTS: GID prevalence in the VHA is higher (22.9/100 000 persons) than are previous estimates of GID in the general US population (4.3/100 000 persons). The rate of suicide-related events among GID-diagnosed VHA veterans was more than 20 times higher than were rates for the general VHA population. CONCLUSIONS: The prevalence of GID diagnosis nearly doubled over 10 years among VHA veterans. Research is needed to examine suicide risk among transgender veterans and how their VHA utilization may be enhanced by new VA initiatives on transgender care.

Prevalence and characteristics of suicide ideation and attempts among active military and veteran participants in a national health survey.

The relationships between military service and suicide are not clear, and comparatively little is known about the characteristics and correlates of suicide ideation and attempts among those with history of military service. We used data from a national health survey to estimate the prevalence and correlates of suicidal behaviors among veterans and service members in 2 states. The prevalence of suicidal behaviors among Veterans was similar to previous estimates of ideation and attempts among adults in the US general population.