Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Aug 1.
Published in final edited form as: Drug Alcohol Depend. 2022 Jun 3;237:109521. doi: 10.1016/j.drugalcdep.2022.109521

Association between clinically recognized suicidality and subsequent initiation or continuation of medications for opioid use disorder

Madeline C Frost 1,2, Julie E Richards 2,3, John R Blosnich 4,5, Eric J Hawkins 1,6,7, Judith I Tsui 8, E Jennifer Edelman 9, Emily C Williams 1,2
PMCID: PMC9546132  NIHMSID: NIHMS1836488  PMID: 35716644

Abstract

Background:

Among individuals with opioid use disorder (OUD), medications for OUD (MOUD) may lower suicide risk. Therefore, it is important that individuals with OUD and suicidality receive MOUD. This study examined associations between clinically recognized suicidality and subsequent initiation or continuation of MOUD among patients with OUD in the national Veterans Health Administration (VA).

Methods:

Electronic health record data were extracted for outpatients with OUD who received VA care 10/1/2016–7/31/2017. Suicidality was measured using diagnostic codes for suicidal ideation/attempt and patient record flags. Analyses were conducted separately among patients without prior-year MOUD receipt to examine MOUD initiation, and with prior-year MOUD receipt to examine MOUD continuation. Poisson regression models estimated likelihood of MOUD receipt in the following year for patients with prior-year suicidality relative to those without. Models were adjusted for sociodemographic and clinical characteristics.

Results:

Among 20,085 patients with no prior-year MOUD, 12% had suicidality and 12% received MOUD in the following year. Suicidality was positively associated with MOUD initiation (adjusted incidence rate ratio [aIRR]: 1.15, 95% confidence interval [CI]: 1.04–1.28). Among 10,162 patients with prior-year MOUD, 9% had suicidality and 84% received MOUD in the following year. Suicidality was negatively associated with MOUD continuation (aIRR: 0.95, 95% CI 0.91–0.98).

Conclusions:

Among VA patients with OUD, clinically recognized suicidality may increase likelihood of MOUD initiation but decrease likelihood of continuation. Efforts to increase initiation overall and to support retention for patients with suicidality are needed.

Keywords: suicide, suicidality, opioid use disorder, buprenorphine, methadone, Veterans

1. Introduction

Opioid use disorder (OUD) and opioid overdose is an urgent public health crisis in the United States, with overdose deaths having reached a record high (Hedegaard et al., 2021). There are three approved OUD treatment medications (MOUD), namely methadone, buprenorphine, and naltrexone (Volkow et al., 2019). Opioid agonists (methadone, buprenorphine) are considered first-line treatment (Department of Veterans Affairs, 2021; Substance Abuse and Mental Health Services Administration, 2021) and reduce overdose risk (Larochelle et al., 2018; Pierce et al., 2016; Wakeman et al., 2020). Most individuals with OUD do not receive these medications despite their effectiveness (Center for Behavioral Health Statistics and Quality, 2020).

Like overdose, suicide has substantially contributed to decline in U.S. life expectancy (Woolf and Schoomaker, 2019), and suicide is considered a leading cause of death (Centers for Disease Control and Prevention, 2021). OUD and nonmedical opioid use may increase risk of suicidal ideation, suicide attempt, and suicide mortality (Ali and Dubenitz, 2021; Ashrafioun et al., 2017; Bohnert and Ilgen, 2019; Bohnert et al., 2017; Chesin et al., 2019; Wilcox et al., 2004)

MOUD treatment may lower suicide mortality risk among individuals with OUD and suicidality (Ahmadi et al., 2018; Molero et al., 2018; Santo et al., 2021; Vakkalanka, P. et al., 2021; Watts et al., 2022). Longitudinal studies have found that individuals with OUD have lower risk of suicide attempt and mortality when they are receiving MOUD (Molero et al., 2018; Santo et al., 2021; Vakkalanka, P. et al., 2021; Watts et al., 2022). Similarly, a trial evaluating different doses of buprenorphine among patients with OUD and depression observed reduced suicidal ideation after patients received buprenorphine of any dose (Ahmadi et al., 2018). It is therefore particularly important that individuals with OUD who also have suicidality receive MOUD, as this may help prevent suicide mortality in addition to treating OUD.

Given the potential for MOUD to lower suicide risk, it is important to understand whether and how suicidality influences MOUD receipt to determine if targeted efforts are needed to increase receipt. In the Behavioral Model of Health Services Use, factors influencing utilization are characterized as predisposing (i.e., pre-existing characteristics), enabling (i.e., things that impede/facilitate utilization), or need (i.e., recognition of need for services) (Andersen and Davidson, 2007). In the context of this model, suicidality may be theorized as either positively or negatively associated with MOUD receipt. For example, suicidality may decrease patients’ perceived need for OUD treatment if they feel hopelessness. It may also increase evaluated need for OUD treatment by clinicians who are aware of a patient’s suicidality, or increase receipt through enabling factors (e.g., if patients with suicidality are more likely to be connected to MOUD through increased healthcare utilization). Suicidality may impact MOUD receipt in the context of related factors—for example, having other unmet mental health needs may present barriers to receiving MOUD (Priester et al., 2016).

Existing research assessing the relationship between suicidality and subsequent MOUD receipt is limited. One unadjusted analysis found that patients in the national Veterans Health Administration (VA) with admissions for suicide/self-harm were more likely to discontinue buprenorphine (Vakkalanka, J.P. et al., 2021). Other studies examining associations between suicidality and MOUD retention have been conducted in small samples and report mixed findings (Grella et al., 1994; Håkansson et al., 2016; Lister et al., 2019; Marcovitz et al., 2016). To our knowledge, no study has assessed whether suicidality is associated with subsequent MOUD initiation.

The VA is an important setting in which to examine the relationship between suicidality and MOUD receipt, as it is the nation’s largest provider of OUD treatment and prioritizes suicide prevention and increasing MOUD provision (Department of Veterans Affairs Office of Mental Health and Suicide Prevention, 2019; Wyse et al., 2018). This study examined associations between clinically recognized suicidality and subsequent initiation or continuation of MOUD among national VA outpatients with OUD. Because suicidality could be theorized as either positively or negatively associated with MOUD receipt, we took an exploratory approach and refrained from making a directional hypothesis.

2. Materials and Methods

2.1. Data source and study sample

This study is a secondary analysis of data from a parent study examining alcohol use and related care in the VA (Williams et al., 2021). Electronic health record (EHR) data were extracted from the VA Corporate Data Warehouse (Souden, 2017) for outpatients with a documented alcohol use screen. This screen is administered annually at a routine healthcare visit to >90% of VA outpatients (Bradley et al., 2006). A visit with a documented alcohol screen was used as the index date; if a patient had multiple visits with a screen, the most recent visit was considered the index. Measures were defined using a consistent timeframe (e.g., within the 12 or 24 months prior to or following the index visit, regardless of when the visit occurred). This study included all patients within each VA facility with an index visit 10/1/2016–7/31/2017 who had an OUD diagnosis within the 12 months prior to the index visit (International Classification of Diseases, 10th Revision, Clinical Modification [ICD-10-CM] codes for abuse or dependence, excluding in remission) (Oliva et al., 2012). The study timeframe was selected to account for the VA’s conversion to ICD-10-CM on 10/1/2015 and spans the beginning of ICD-10-CM coding for variables measured in the 12 months prior to the index visit until the end of the parent study period; the range of actual index visit dates was 10/1/2016–7/31/2017.

Analyses were conducted among two mutually exclusive subsamples: 1) an initiation subsample that included patients with no MOUD receipt within the 12 months prior to the index visit (n=20,085); and 2) a continuation subsample including patients with MOUD receipt within the 12 months prior to the index visit (n=10,162). To examine continuation of specific MOUD, the second subsample was defined based on the outcome being examined (i.e., 9,964 patients who received opioid agonist medications and 319 patients who received injectable naltrexone within the 12 months prior to the index visit). Study procedures, including waivers of Health Insurance Portability and Accountability Act (HIPAA) authorization and consent, were approved by institutional review boards at the University of Washington and VA Puget Sound.

2.2. Measures

2.2.1. Primary independent variable

Clinically recognized suicidality was defined as having ≥1 ICD-10-CM code representing suicidal ideation or suicide attempt/intentional self-harm (Hedegaard et al., 2018; VA Program Evaluation and Resource Center) and/or ≥1 patient record flag (PRF) for suicide risk (Berg et al., 2018) documented within the 12 months prior to the index visit. We considered that suicidality documented through a PRF may impact MOUD receipt differently than suicidality documented through diagnostic codes, as clinicians are immediately alerted when they open a patient’s record if a PRF is in effect and they are expected to increase frequency of contact with these patients (Berg et al., 2018). We therefore secondarily defined suicidality as only ICD-10-CM codes and as only the PRF. Measure definitions that include diagnostic codes, clinic visit codes and/or prescription fills, and a figure depicting measure timeframes, are included in Appendix A.

2.2.2. Outcome variables

The outcome of interest was receipt of any MOUD within the 12 months following the index visit, defined as ≥1 visit to a VA Opioid Treatment Program (OTP) clinic, having filled ≥1 prescription for oral (sublingual/buccal) formulations of buprenorphine, and/or ≥1 administration of injectable naltrexone with an OUD diagnosis in the year prior to administration. VA EHR data do not capture OTP-dispensed MOUD (which is typically methadone) in pharmacy data, therefore an OTP clinic visit (stop code 523) is used to measure methadone receipt (Lin et al., 2021; Manhapra et al., 2020; Wyse et al., 2022). Oral naltrexone is not recommended for OUD in clinical guidelines (Department of Veterans Affairs, 2021), therefore we did not include this formulation. Opioid agonists (methadone, buprenorphine) are considered first-line treatment (Department of Veterans Affairs, 2021), therefore we secondarily examined receipt of opioid agonist medication (OTP clinic visit and/or buprenorphine prescription fill) and receipt of injectable naltrexone separately.

2.2.3. Covariates and other variables

Covariates included patient sociodemographic and clinical characteristics considered potential confounders based on previously observed associations between these characteristics and both suicidality and MOUD receipt (Ahmedani et al., 2017; Blakey et al., 2018; Denney et al., 2009; Finlay et al., 2016; Finlay et al., 2021; Ilgen et al., 2007; Kelly et al., 2018; Krawczyk et al., 2017; Manhapra et al., 2017; Michel et al., 2015; Murphy et al., 2014; Oliva et al., 2012; Schinka et al., 2016; Shiner et al., 2017; Shiner et al., 2021; Stein et al., 2012; Wyse et al., 2019; Zivin et al., 2007; Zwald et al., 2020).

Patient age, sex, race, ethnicity, marital status, VA eligibility status, rurality/urbanicity of patient residence, and U.S. Census region of the facility were measured based on EHR documentation at the time of the index visit. Race and ethnicity were included as proxies for exposure to structural and interpersonal racism (Hardeman and Karbeah, 2020), and VA eligibility status was included as a proxy for socioeconomic status (Young et al., 2003). Rurality/urbanicity was based on VA Office of Rural Health definitions (VHA Office of Rural Health, 2015). Homelessness/housing instability was defined as having ≥1 ICD-10-CM code, visit code, or screening documentation from an electronic clinical reminder related to homelessness or other housing instability within the 24 months prior to the index visit (Blosnich et al., 2020). Legal system involvement was defined as having ≥1 visit code indicating receipt of VA services provided to Veterans with legal system involvement within the 12 months prior to the index visit (Taylor et al., 2019).

Most clinical conditions were defined based on the presence of ≥1 relevant ICD-10-CM code within the 12 months prior to the index visit, including any mental health condition (depression, post-traumatic stress disorder, anxiety, other mood disorders, bipolar disorder, psychoses, and/or schizophrenia), alcohol use disorder, stimulant use disorder, cannabis use disorder, sedative use disorder, hallucinogen use disorder, and Charlson comorbidity index conditions (adjusted for as separate binary variables rather than a continuous weighted score) (Charlson et al., 1994; D’Hoore et al., 1996). Hepatitis C and HIV were defined based on documentation within the 24 months prior to the index visit; hepatitis C as having ≥1 ICD-10-CM code, and HIV as having ≥1 inpatient or ≥2 outpatient ICD-10-CM codes (Fultz et al., 2006). Chronic pain was defined as having ≥1 inpatient or ≥2 outpatient ICD-10-CM codes indicating chronic pain documented within the 24 months prior to the index visit (Carey et al., 2018). History of mental healthcare utilization was defined as having ≥1 outpatient mental healthcare visit within the 12 months prior to the index visit (e.g., mental health clinic services, Primary Care-Mental Health Integration services). The data were clustered by VA facility, defined as the facility where the index visit occurred.

2.3. Analyses

All analyses were conducted separately in the initiation and continuation subsamples. Patient characteristics were described overall and across presence of clinically recognized suicidality. Poisson regression models estimated the relative rate (incidence rate ratio) of MOUD outcomes among patients with suicidality relative to those without. Poisson regression was used instead of logistic because outcomes were not expected to be rare, and therefore odds ratios were not likely to accurately reflect relative risk and more likely to be misinterpreted (Greenland, 1995; Zou, 2004). Standard errors were calculated with the robust sandwich estimator to correct for misspecification of the Poisson variance structure (Greenland, 1995; Liang and Zeger, 1986; Zou, 2004), and clustered by VA facility (N=130) to account for correlation of patient data within the same facility. We calculated marginal predicted prevalence of outcomes across the presence of suicidality to examine the magnitude of differences. Models were fit unadjusted, then adjusted for all covariates. As being dually enrolled in Medicare may impact VA service utilization (Petersen et al., 2010), we conducted a sensitivity analysis repeating primary regression models among patients under age 65. Secondary analyses repeated regression models with suicidality-related ICD-10-CM codes as the independent variable of interest, and again with suicide risk PRF as the independent variable of interest. In post-hoc analyses, we separately examined associations between suicidality and buprenorphine continuation (defined as filled ≥1 prescription) among those with past-year buprenorphine receipt, and with methadone continuation (defined as ≥1 OTP clinic visit) among those with past-year methadone receipt. Analyses were conducted using Stata version 16 software (StataCorp, 2019).

3. Results

3.1. Sample descriptions

Overall, 30,247 patients had a documented OUD diagnosis in the year prior to their index visit, and 11% of these patients had clinically recognized suicidality in the prior year (10% with suicidality-related ICD-10-CM codes, 3% with PRFs). Sixty-six percent had no prior-year MOUD receipt (initiation subsample, N=20,085) and 34% had any prior-year MOUD receipt (continuation subsample, N=10,162; analyses examining continuation for specific MOUD included 9,964 patients who received opioid agonist medication in the prior year, and 319 patients who received injectable naltrexone in the prior year).

Patient characteristics in the initiation and continuation subsamples are presented in Tables 1a and 1b, respectively. In both subsamples, mean age was about 50 years, and those with suicidality were younger on average than those without. The samples were majority male and White; a larger proportion of patients with suicidality were female in the initiation subsample but there was no significant difference in EHR-documented sex across suicidality in the continuation subsample. In the initiation subsample, Black patients and those with other or multiple EHR-documented race made up a larger proportion of patients with suicidality compared to those without, while in the continuation subsample, White patients made up a larger proportion of patients with suicidality compared to those without. There was no significant difference in Hispanic/Latino ethnicity across suicidality in either sample. In both subsamples, those with suicidality had higher prevalence than those without of divorced or single/never married marital status, full VA coverage, urban residence, homelessness/housing instability, legal system involvement, mental health conditions, other SUDs, hepatitis C, and history of mental healthcare utilization. In the initiation subsample, those with suicidality were more likely to be in the South and had a slightly lower mean Charlson score (calculated without HIV) and higher prevalence of HIV. In the continuation subsample, those with suicidality were more likely to have chronic pain.

Table 1a.

Sociodemographic and clinical characteristics compared across presence of clinically recognized suicidality in the initiation subsamplea

No Suicidality
(N=17,687)		 Suicidality
(N=2,398)		 p-valueb Total
(N=20,085)
N (%) N (%) N (%)
Age (mean, SD, t-test) 53.3 (14.4) 48.6 (13.6) <0.001 52.7 (14.4)
Female 1,285 (7.3) 209 (8.7) 0.011 1,494 (7.4)
Race 0.013
 Black/African American 3,185 (18.0) 478 (19.9) 3,663 (18.2)
 White 13,317 (75.3) 1,759 (73.4) 15,076 (75.1)
 Other/Multiple races 511 (2.9) 85 (3.5) 596 (3.0)
 Unknown 674 (3.8) 76 (3.2) 750 (3.7)
Hispanic/Latino ethnicity 884 (5.0) 135 (5.6) 0.186 1,019 (5.1)
Marital status <0.001
 Divorced/separated 7,162 (40.5) 1,122 (46.8) 8,284 (41.2)
 Married 6,024 (34.1) 580 (24.2) 6,604 (32.9)
 Never married/single 3,814 (21.6) 617 (25.7) 4,431 (22.1)
 Widowed 632 (3.6) 73 (3.0) 705 (3.5)
 Unknown 55 (0.3) 6 (0.3) 61 (0.3)
VA eligibility statusc 0.017
 Full VA coverage 7,704 (43.6) 1,106 (46.1) 8,810 (43.9)
 Service connection <50% 3,299 (18.7) 418 (17.4) 3,717 (18.5)
 Non-service connected 6,672 (37.7) 869 (36.2) 7,541 (37.6)
 Other/unknown 12 (0.1) 5 (0.2) 17 (0.1)
Rurality/urbanicity <0.001
 Rural 5,368 (30.4) 604 (25.2) 5,972 (29.7)
 Urban 12,259 (69.3) 1,779 (74.2) 14,038 (69.9)
 Unknown 60 (0.3) 15 (0.6) 75 (0.4)
U.S. Census region <0.001
 Midwest 3,637 (20.6) 497 (20.7) 4,134 (20.6)
 Northeast 3,537 (20.0) 409 (17.1) 3,946 (19.7)
 South 6,688 (37.8) 1,050 (43.8) 7,738 (38.5)
 West 3,764 (21.3) 421 (17.6) 4,185 (20.8)
 Other 61 (0.3) 21 (0.9) 82 (0.4)
Homelessness/housing instability 5,357 (30.3) 1,343 (56.0) <0.001 6,700 (33.4)
Legal system involvement 1,278 (7.2) 281 (11.7) <0.001 1,559 (7.8)
Mental health conditiond 12,166 (68.8) 2,267 (94.5) <0.001 14,433 (71.9)
Alcohol use disorder 5,075 (28.7) 1,334 (55.6) <0.001 6,409 (31.9)
Stimulant use disorder 3,564 (20.2) 1,136 (47.4) <0.001 4,700 (23.4)
Cannabis use disorder 2,646 (15.0) 827 (34.5) <0.001 3,473 (17.3)
Sedative/hallucinogen use disorder 1,205 (6.8) 454 (18.9) <0.001 1,659 (8.3)
Charlson scoree (mean, SD, t-test) 1.0 (1.7) 0.9 (1.6) <0.001 1.0 (1.7)
HIV 208 (1.2) 49 (2.0) <0.001 257 (1.3)
Hepatitis C 3,457 (19.6) 596 (24.9) <0.001 4,053 (20.2)
Chronic pain 12,077 (68.3) 1,645 (68.6) 0.754 13,722 (68.3)
Mental healthcare utilization 12,942 (73.2) 2,331 (97.2) <0.001 15,273 (76.0)
Open in a new tab

MOUD = medications for opioid use disorder; OUD = opioid use disorder; SD = standard deviation; SUD = substance use disorder; VA = Veterans Health Administration

a

National VA outpatients 10/1/16–7/31/17 with OUD and no MOUD receipt in the past year

b

From chi-square test unless otherwise indicated

c

Fisher’s exact test (expected cell count <5)

d

Includes depression, post-traumatic stress disorder, anxiety, other mood disorders, bipolar disorder, psychoses, and/or schizophrenia

e

Not including HIV

Table 1b.

Sociodemographic and clinical characteristics compared across presence of clinically recognized suicidality in the continuation subsamplea

No Suicidality
(N=9,236)		 Suicidality
(N=926)		 p-valueb Total
(N=10,162)
N (%) N (%) N (%)
Age (mean, SD, t-test) 50.5 (14.0) 46.3 (13.6) <0.001 50.1 (14.0)
Female 611 (6.6) 70 (7.6) 0.273 681 (6.7)
Race 0.005
 Black/African American 1,869 (20.2) 142 (15.3) 2,011 (19.8)
 White 6,844 (74.1) 731 (78.9) 7,575 (74.5)
 Other/Multiple races 208 (2.3) 22 (2.4) 230 (2.3)
 Unknown 315 (3.4) 31 (3.4) 346 (3.4)
Hispanic/Latino ethnicity 579 (6.3) 60 (6.5) 0.801 639 (6.3)
Marital statusc <0.001
 Divorced/separated 3,832 (41.5) 450 (48.6) 4,282 (42.1)
 Married 2,637 (28.6) 181 (19.6) 2,818 (27.7)
 Never married/single 2,461 (26.7) 267 (28.8) 2,728 (26.9)
 Widowed 281 (3.0) 26 (2.8) 307 (3.0)
 Unknown 25 (0.3) 2 (0.2) 27 (0.3)
VA eligibility statusc 0.003
 Full VA coverage 3,670 (39.7) 428 (46.2) 4,098 (40.3)
 Service connection <50% 1,784 (19.3) 169 (18.3) 1,953 (19.2)
 Non-service connected 3,773 (40.9) 329 (35.5) 4,102 (40.4)
 Other/unknown 9 (0.1) 0 (0.0) 9 (0.1)
Rurality/urbanicityc 0.015
 Rural 1,962 (21.2) 169 (18.3) 2,131 (21.0)
 Urban 7,260 (78.6) 753 (81.3) 8,013 (78.9)
 Unknown 14 (0.2) 4 (0.4) 18 (0.2)
U.S. Census regionc 0.510
 Midwest 1,830 (19.8) 187 (20.2) 2,017 (19.9)
 Northeast 2,663 (28.8) 268 (28.9) 2,931 (28.8)
 South 2,821 (30.5) 292 (31.5) 3,113 (30.6)
 West 1,894 (20.5) 174 (18.8) 2,068 (20.4)
 Other 28 (0.3) 5 (0.5) 33 (0.3)
Homelessness/housing instability 3,329 (36.0) 664 (71.7) <0.001 3,993 (39.3)
Legal system involvement 596 (6.5) 139 (15.0) <0.001 735 (7.2)
Mental health conditiond 6,812 (73.8) 900 (97.2) <0.001 7,712 (75.9)
Alcohol use disorder 2,428 (26.3) 510 (55.1) <0.001 2,938 (28.9)
Stimulant use disorder 1,975 (21.4) 481 (51.9) <0.001 2,456 (24.2)
Cannabis use disorder 1,386 (15.0) 281 (30.4) <0.001 1,667 (16.4)
Sedative/hallucinogen use disorder 704 (7.6) 243 (26.2) <0.001 947 (9.3)
Charlson scoree (mean, SD, t-test) 0.8 (1.5) 0.8 (1.4) 0.896 0.8 (1.5)
HIV 128 (1.4) 17 (1.8) 0.271 145 (1.4)
Hepatitis C 2,848 (30.8) 337 (36.4) 0.001 3,185 (31.3)
Chronic pain 5,531 (59.9) 681 (73.5) <0.001 6,212 (61.1)
Mental healthcare utilization 8,013 (86.8) 921 (99.5) <0.001 8,934 (87.9)
Open in a new tab

MOUD = medications for opioid use disorder; OUD = opioid use disorder; SD = standard deviation; SUD = substance use disorder; VA = Veterans Health Administration

a

National Veterans Health Administration outpatients 10/1/16–7/31/17 with OUD and any MOUD receipt in the past year

b

From chi-square test unless otherwise indicated

c

Fisher’s exact test (expected cell count <5)

d

Includes depression, post-traumatic stress disorder, anxiety, other mood disorders, bipolar disorder, psychoses, and/or schizophrenia

e

Not including HIV

3.2. Association between clinically recognized suicidality and subsequent MOUD initiation

Among 20,085 patients in the initiation subsample, 12% had prior-year clinically recognized suicidality. In the following year, 12% received any MOUD (15% among those with prior-year suicidality, 12% among those without), 11% received opioid agonist medication (13% among those with prior-year suicidality, 11% among those without), and 1% received injectable naltrexone (3% among those with prior-year suicidality, 1% among those without; Table 2a).

Table 2a.

Likelihood of MOUD initiation in the following year for patients with past-year clinically recognized suicidality compared to those without in the initiation subsamplea (N=20,085)

Outcome (in following year) No Suicidality Suicidality IRR for Suicidality Relative to No Suicidality
% (95% CI) % (95% CI) IRR (95% CI) p-value
Any MOUD
 Unadjusted Model 11.8 (10.5–13.0) 14.8 (12.8–16.7) 1.25 (1.12–1.41) <0.001
 Adjusted Modelb 11.9 (10.8–12.9) 13.7 (12.1–15.4) 1.15 (1.04–1.28) 0.009
Opioid agonist medication
 Unadjusted Model 10.8 (9.6–12.0) 13.0 (11.3–14.8) 1.20 (1.07–1.35) 0.002
 Adjusted Modelb 10.9 (9.9–11.9) 12.2 (10.7–13.7) 1.12 (1.00–1.25) 0.044
Injectable naltrexone
 Unadjusted Model 1.2 (0.8–1.6) 2.6 (1.7–3.5) 2.16 (1.53–3.05) <0.001
 Adjusted Modelb 1.2 (0.9–1.6) 2.2 (1.4–3.1) 1.81 (1.30–2.52) <0.001
Open in a new tab

IRR = incidence rate ratio; MOUD = medications for opioid use disorder; OUD = opioid use disorder

a

National Veterans Health Administration outpatients 10/1/16–7/31/17 with OUD and no MOUD receipt in the past year

b

Adjusted for age, sex, race, ethnicity, marital status, VA eligibility status, urbanicity/rurality, U.S. Census region, homelessness/housing instability, legal system involvement, mental health condition, alcohol use disorder, stimulant use disorder, cannabis use disorder, sedative use disorder, hallucinogen use disorder, cancer, cerebrovascular disease, chronic obstructive pulmonary disease, congestive heart failure, dementia, diabetes, diabetes with complications, metastatic solid tumor, mild liver disease, moderate/severe liver disease, myocardial infarction, paralysis, peptic ulcer, peripheral vascular disease, renal disease, rheumatic disease, HIV, hepatitis C, chronic pain, history of mental healthcare utilization

Results from initiation subsample analyses are presented in Table 2a. In both unadjusted and adjusted models, patients with suicidality were significantly more likely than those without to initiate any MOUD, opioid agonist medications and injectable naltrexone in the following year. When the primary regression analysis was repeated among patients under age 65, there was a significant positive association between suicidality and initiation of any MOUD (adjusted incidence rate ratio [aIRR]: 1.15, 95% confidence interval [CI]: 1.03–1.29).

Secondary analyses conducted in the initiation subsample with suicidality documented via only ICD-10-CM codes or only PRFs as the independent variable of interest are presented in Appendix B. Results from models including suicidality documented via only ICD-10-CM codes mirrored main analyses (Table B1, Appendix B). In models including suicidality documented via only PRFs, associations were non-significant for all outcomes in all models (Table B2, Appendix B).

3.3. Association between clinically recognized suicidality and subsequent MOUD continuation

Among 10,162 patients in the continuation subsample (any prior-year MOUD receipt), 9% had prior-year clinically recognized suicidality. In the following year, 84% received any MOUD (75% among those with prior-year suicidality, 84% among those without). Among the subset of 9,964 patients with prior-year opioid agonist medication receipt, 84% received opioid agonist medication in the following year (75% among those with prior-year suicidality, 85% among those without). Among the subset of 319 patients with prior-year injectable naltrexone receipt, 41% received injectable naltrexone in the following year (42% among those with prior-year suicidality, 41% among those without; Table 2b).

Table 2b.

Likelihood of MOUD continuation in the following year for patients with past-year clinically recognized suicidality compared to those without in the continuation subsamplea

Outcome (in following year) No Suicidality Suicidality IRR for Suicidality Relative to No Suicidality
% (95% CI) % (95% CI) IRR (95% CI) p-value
Any MOUD
(N=10,162)
 Unadjusted Model 84.5 (81.9–87.1) 74.7 (71.3–78.2) 0.88 (0.84–0.93) <0.001
 Adjusted Modelb 84.0 (81.8–86.1) 79.5 (76.3–82.6) 0.95 (0.91–0.98) 0.006
Opioid agonist medication
(N=9,964)
 Unadjusted Model 84.8 (82.1–87.4) 74.9 (71.4–78.5) 0.88 (0.84–0.93) <0.001
 Adjusted Modelb 84.3 (82.1–86.4) 79.8 (76.4–83.1) 0.95 (0.91–0.99) 0.009
Injectable naltrexone
(N=319)
 Unadjusted Model 41.0 (34.7–47.4) 42.2 (32.6–51.9) 1.03 (0.77–1.38) 0.851
 Adjusted Modelb 39.7 (32.8–46.5) 46.3 (36.8–55.8) 1.17 (0.86–1.59) 0.328
Open in a new tab

IRR = incidence rate ratio; MOUD = medications for opioid use disorder; OUD = opioid use disorder

a

National Veterans Health Administration outpatients 10/1/16–7/31/17 with OUD and MOUD receipt in the past year; sample limited to patients who received the type of MOUD examined as the outcome of interest during past year (differs for each outcome)

b

Adjusted for age, sex, race, ethnicity, marital status, VA eligibility status, urbanicity/rurality, U.S. Census region, homelessness/housing instability, legal system involvement, mental health condition, alcohol use disorder, stimulant use disorder, cannabis use disorder, sedative use disorder, hallucinogen use disorder, cancer, cerebrovascular disease, chronic obstructive pulmonary disease, congestive heart failure, dementia, diabetes, diabetes with complications, metastatic solid tumor, mild liver disease, moderate/severe liver disease, myocardial infarction, paralysis, peptic ulcer, peripheral vascular disease, renal disease, rheumatic disease, HIV, hepatitis C, chronic pain, history of mental healthcare utilization

Results from analyses conducted in the continuation subsample are presented in Table 2b. Patients with suicidality were significantly less likely than those without to continue any MOUD and opioid agonist medications in the following year in unadjusted and adjusted models, while associations between suicidality and continuation of injectable naltrexone were non-significant in both models. When the primary regression analysis was repeated among patients under age 65, there was a significant negative association between suicidality and continuation of any MOUD (aIRR: 0.95, 95% CI: 0.92–0.99).

Secondary analyses conducted in the continuation subsample with suicidality documented via only ICD-10-CM codes or only PRFs as the independent variable of interest are presented in Appendix C. Results from models including suicidality documented via only ICD-10-CM codes mirrored main analyses (Table C1, Appendix C). In models including suicidality documented via only PRFs, associations were non-significant for all outcomes in all models, apart from suicide risk PRF being positively associated with continuing injectable naltrexone in the adjusted model (Table C2, Appendix C). In post-hoc analyses separately examining continuation of buprenorphine and methadone, suicidality was negatively associated with buprenorphine continuation and not significantly associated with methadone continuation in adjusted models (Table D1, Appendix D).

4. Discussion

In this national sample of VA outpatients with OUD, clinically recognized suicidality was associated with a modest increase in likelihood of initiating MOUD among patients without prior-year MOUD receipt (initiation subsample), though initiation was low overall. Clinically recognized suicidality was associated with decreased likelihood of continuing MOUD among patients with prior-year MOUD receipt (continuation subsample).

This is the first study to our knowledge to examine whether suicidality is associated with subsequent MOUD initiation. Findings from the initiation subsample suggest that clinically recognized suicidality does not engender additional barriers to initiating MOUD and may slightly increase likelihood of initiation. However, initiation was low overall (12% of the initiation subsample initiated any MOUD in the following year), similar to the estimated 11% prevalence of MOUD initiation among commercially insured U.S. patients with OUD (Morgan et al., 2022). Given low overall initiation, continued efforts are needed to increase MOUD initiation for all patients with untreated OUD, including increasing buprenorphine provision in mental healthcare settings—where patients with suicidality may be seen more frequently—in addition to other settings (Gordon et al., 2020).

Mechanisms underlying the positive association between clinically recognized suicidality and MOUD initiation are unclear. Secondary analyses found that suicidality measured only as ≥1 PRF was not associated with initiation, suggesting that this method of alerting clinicians to patient suicidality may not make them more likely to prescribe or link the patient to MOUD. Future research with VA clinicians could examine whether suicide risk PRFs impact their substance use treatment decisions, and if this differs across patients (e.g., if PRFs are less likely to impact treatment decisions for patients who consistently have a PRF in their chart). Future research could also aim to understand why this finding differs from prior research finding that suicide risk PRF was associated with a subsequent increase in SUD care visits (Berg et al., 2018). For example, perhaps clinicians are not aware that MOUD may lower suicide risk among patients with OUD and suicidality. Additionally, future research with clear temporality between suicidality, subsequent mental healthcare utilization, and subsequent MOUD receipt could apply mediation analysis to examine whether increased mental healthcare utilization is a pathway through which patients with suicidality become more likely to initiate MOUD. Within the context of the Behavioral Model of Health Services Use (Andersen and Davidson, 2007), this might be considered an enabling factor.

While most patients in the continuation subsample (84%) received any MOUD in the following year, findings suggest that patients with suicidality may face increased barriers to continuing treatment with opioid agonist medications, particularly buprenorphine. We were unable to determine in these data whether patients who did not receive MOUD in following year died or discontinued MOUD. Therefore, these results could be partly driven by increased suicide mortality among patients with higher severity suicidality (Louzon et al., 2016). However, our findings align with a prior unadjusted analysis that accounted for mortality during treatment and found that VA patients with admissions for suicide/self-harm were more likely to discontinue buprenorphine treatment (Vakkalanka, J.P. et al., 2021).

Mechanisms underlying the association between suicidality and MOUD discontinuation should be investigated. Again, within the Behavioral Model of Health Services Use (Andersen and Davidson, 2007), one could hypothesize that some patients with suicidality feel a sense of hopelessness that decreases perceived need for OUD treatment. Efforts are needed to support retention in MOUD for patients with suicidality, and to address suicide risk among patients receiving MOUD. These could include integrating interventions to address suicidality and comorbid mental health conditions into MOUD care, and outreach to reengage patients who discontinue MOUD.

Social factors that may increase likelihood of both suicidality and MOUD discontinuation (e.g., homelessness, legal system involvement) should also be addressed to promote MOUD retention among patients with suicidality (Krawczyk et al., 2021; O’Connor et al., 2020). Additionally, though it was not the focus of this study, descriptive analyses suggested differences in suicidality across patient characteristics (e.g., higher prevalence among patients with co-occurring SUDs) and that some differences might be modified by MOUD receipt (e.g., different patterns in suicidality across race among patients with prior-year MOUD compared to those without). Future research should examine sociodemographic and clinical differences in suicidality among patients with OUD and whether these are impacted by MOUD receipt, with adjustment for confounders. This research should consider that associations between clinically recognized suicidality and other clinical conditions (e.g., chronic pain, SUDs) may not be causal, and could be due to increased healthcare utilization resulting in increased recognition/documentation of suicidality.

4.1. Limitations

This study has limitations. Our measure of clinically recognized suicidality likely undermeasures suicidality, as some patients may not disclose suicidal ideation or suicide attempts to providers (Richards et al., 2019) or these disclosures may not be documented in the EHR. We did not have access to additional sources of suicidality documentation (e.g., routine depression screening (Bahraini et al., 2020), clinical progress notes). We were also unable to capture MOUD that may have been received outside of the VA. Estimated associations may be biased by mortality in the year following the index visit, and by unobserved confounding. These data did not allow us to examine precise differences in retention time across suicidality, or to account for prior length of treatment when examining continuation. Future studies using time-to-discontinuation data should examine the impact of suicidality on MOUD retention (with appropriate adjustment for confounding). We were also unable to assess the potential impact of dose, frequency of MOUD receipt, or longer-term history of MOUD receipt on these associations, which should be examined in future research. Although a large majority of VA outpatients receive routine alcohol screening (Bradley et al., 2006), it is possible that patients who did not (and therefore were not included in this sample) differed from those who were screened. Additionally, this study may be missing patients with undiagnosed OUD (a phenomenon which might differ across suicidality), and results may not be generalizable to Veterans with OUD who are not engaged in VA care. Finally, these findings may have limited generalizability outside of the VA. The VA patient population has more men, older patients, and White patients compared to the general population. The VA system may also differ from other healthcare settings in its documentation of suicidality and efforts to increase provision of MOUD.

5. Conclusions

This study found that among VA patients with OUD, 11% had clinically recognized suicidality, and MOUD initiation was low overall. Patients with OUD who have clinically recognized suicidality may be more likely to initiate MOUD than those without suicidality, but less likely to continue first-line treatment (opioid agonist medications). Efforts to increase MOUD initiation among all patients with untreated OUD are needed, as well as efforts to improve retention for patients with suicidality and to address suicide risk among patients receiving MOUD.

Supplementary Material

Appendices

Acknowledgements:

The authors thank Joseph E. Glass, PhD, MSW and Gwen T. Lapham, PhD, MPH, MSW from Kaiser Permanente Washington Health Research Institute for their advice on analyses.

Role of Funding Source:

This work reflects secondary analyses of data from a study supported by the National Institute on Alcohol Abuse and Alcoholism (R21 AA025973). Ms. Frost is supported by a predoctoral training award from the Veterans Affairs (VA) Puget Sound Research and Development Service. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication. The opinions expressed in this work are the authors’ and do not necessarily reflect those of the institutions, funders, the Department of Veterans Affairs, or the United States Government.

Footnotes

Conflict of Interest: No conflict declared.

Preliminary findings from this study were presented at the 2020 Addiction Health Services Research Conference.

REFERENCES

  1. Ahmadi J, Jahromi MS, Ehsaei Z, 2018. The effectiveness of different singly administered high doses of buprenorphine in reducing suicidal ideation in acutely depressed people with co-morbid opiate dependence: a randomized, double-blind, clinical trial. Trials 19(1), 462. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Ahmedani BK, Peterson EL, Hu Y, Rossom RC, Lynch F, Lu CY, Waitzfelder BE, Owen-Smith AA, Hubley S, Prabhakar D, Williams LK, Zeld N, Mutter E, Beck A, Tolsma D, Simon GE, 2017. Major Physical Health Conditions and Risk of Suicide. Am. J. Prev. Med 53(3), 308–315. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ali MM, Dubenitz J, 2021. Suicidal behavior, opioid use disorder, and behavioral health treatment: Prevalence and correlates among adults in the United States 2015–2018. J Subst Abuse Treat 130, 108413. [DOI] [PubMed] [Google Scholar]
  4. Andersen RM, Davidson PL, 2007. Improving access to care in America: Individual and contextual indicators, in: Andersen RM, Rice TH, Kominski GF (Eds.), Changing the US health care system: Key issues in health services policy and management. Jossey-Bass, San Francisco, pp. 3–31. [Google Scholar]
  5. Ashrafioun L, Bishop TM, Conner KR, Pigeon WR, 2017. Frequency of prescription opioid misuse and suicidal ideation, planning, and attempts. J. Psychiatr. Res 92, 1–7. [DOI] [PubMed] [Google Scholar]
  6. Bahraini N, Brenner LA, Barry C, Hostetter T, Keusch J, Post EP, Kessler C, Smith C, Matarazzo BB, 2020. Assessment of Rates of Suicide Risk Screening and Prevalence of Positive Screening Results Among US Veterans After Implementation of the Veterans Affairs Suicide Risk Identification Strategy. JAMA Netw Open 3(10), e2022531. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Berg JM, Malte CA, Reger MA, Hawkins EJ, 2018. Medical Records Flag for Suicide Risk: Predictors and Subsequent Use of Care Among Veterans With Substance Use Disorders. Psychiatr. Serv 69(9), 993–1000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Blakey SM, Wagner HR, Naylor J, Brancu M, Lane I, Sallee M, Kimbrel NA, Elbogen EB, 2018. Chronic Pain, TBI, and PTSD in Military Veterans: A Link to Suicidal Ideation and Violent Impulses? J Pain 19(7), 797–806. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Blosnich JR, Montgomery AE, Dichter ME, Gordon AJ, Kavalieratos D, Taylor L, Ketterer B, Bossarte RM, 2020. Social Determinants and Military Veterans’ Suicide Ideation and Attempt: a Cross-sectional Analysis of Electronic Health Record Data. J. Gen. Intern. Med 35(6), 1759–1767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Bohnert ASB, Ilgen MA, 2019. Understanding Links among Opioid Use, Overdose, and Suicide. N. Engl. J. Med 380(1), 71–79. [DOI] [PubMed] [Google Scholar]
  11. Bohnert KM, Ilgen MA, Louzon S, McCarthy JF, Katz IR, 2017. Substance use disorders and the risk of suicide mortality among men and women in the US Veterans Health Administration. Addiction 112(7), 1193–1201. [DOI] [PubMed] [Google Scholar]
  12. Bradley KA, Williams EC, Achtmeyer CE, Volpp B, Collins BJ, Kivlahan DR, 2006. Implementation of evidence-based alcohol screening in the Veterans Health Administration. The American journal of managed care 12(10), 597–606. [PubMed] [Google Scholar]
  13. Carey EP, Nolan C, Kerns RD, Ho PM, Frank JW, 2018. Association Between Facility-Level Utilization of Non-pharmacologic Chronic Pain Treatment and Subsequent Initiation of Long-Term Opioid Therapy. J. Gen. Intern. Med 33(Suppl 1), 38–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Center for Behavioral Health Statistics and Quality, 2020. 2019 National Survey on Drug Use and Health: Detailed Tables. Rockville, MD: Substance Abuse and Mental Health Services Admininstration. [Google Scholar]
  15. Centers for Disease Control and Prevention, 2021. Suicide Prevention: Fast Facts. Accessed September 3, 2021 at: https://www.cdc.gov/suicide/facts/index.html.
  16. Charlson M, Szatrowski TP, Peterson J, Gold J, 1994. Validation of a combined comorbidity index. J. Clin. Epidemiol 47(11), 1245–1251. [DOI] [PubMed] [Google Scholar]
  17. Chesin M, Interian A, Kline A, St Hill L, King A, Miller R, Latorre M, Stanley B, 2019. Past-year opioid misuse and suicide attempt are positively associated in high suicide risk veterans who endorse past- year substance use. Addict. Behav 99, 106064. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. D’Hoore W, Bouckaert A, Tilquin C, 1996. Practical considerations on the use of the Charlson comorbidity index with administrative data bases. J. Clin. Epidemiol 49(12), 1429–1433. [DOI] [PubMed] [Google Scholar]
  19. Denney JT, Rogers RG, Krueger PM, Wadsworth T, 2009. Adult Suicide Mortality in the United States: Marital Status, Family Size, Socioeconomic Status, and Differences by Sex. Soc Sci Q 90(5), 1167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Department of Veterans Affairs, 2021. VA/DoD Clinical Practice Guideline for the Management of Substance Use Disorders. Accessed January 7, 2022 from: https://www.healthquality.va.gov/guidelines/MH/sud/VADoDSUDCPG.pdf.
  21. Department of Veterans Affairs Office of Mental Health and Suicide Prevention, 2019. National Strategy for Preventing Veteran Suicide: 2018–2028. Accessed September 7, 2021: https://www.mentalhealth.va.gov/suicide_prevention/docs/Office-of-Mental-Health-and-Suicide-Prevention-National-Strategy-for-Preventing-Veterans-Suicide.pdf.
  22. Finlay AK, Harris AH, Rosenthal J, Blue-Howells J, Clark S, McGuire J, Timko C, Frayne SM, Smelson D, Oliva E, Binswanger I, 2016. Receipt of pharmacotherapy for opioid use disorder by justice-involved U.S. Veterans Health Administration patients. Drug Alcohol Depend. 160, 222–226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Finlay AK, Harris AHS, Timko C, Yu M, Smelson D, Stimmel M, Binswanger IA, 2021. Disparities in Access to Medications for Opioid Use Disorder in the Veterans Health Administration. J Addict Med 15(2), 143–149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Fultz SL, Skanderson M, Mole LA, Gandhi N, Bryant K, Crystal S, Justice AC, 2006. Development and verification of a “virtual” cohort using the National VA Health Information System. Med. Care 44(8 Suppl 2), S25–30. [DOI] [PubMed] [Google Scholar]
  25. Gordon AJ, Drexler K, Hawkins EJ, Burden J, Codell NK, Mhatre-Owens A, Dungan MT, Hagedorn H, 2020. Stepped Care for Opioid Use Disorder Train the Trainer (SCOUTT) initiative: Expanding access to medication treatment for opioid use disorder within Veterans Health Administration facilities. Subst Abus 41(3), 275–282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Greenland S, 1995. Avoiding power loss associated with categorization and ordinal scores in dose-response and trend analysis. Epidemiology 6(4), 450–454. [DOI] [PubMed] [Google Scholar]
  27. Grella CE, Anglin MD, Wugalter SE, Rawson R, Hasson A, 1994. Reasons for discharge from methadone maintenance for addicts at high risk of HIV infection or transmission. J. Psychoactive Drugs 26(2), 223–232. [DOI] [PubMed] [Google Scholar]
  28. Håkansson A, Widinghoff C, Abrahamsson T, Gedeon C, 2016. Correlates of Nine-Month Retention following Interim Buprenorphine-Naloxone Treatment in Opioid Dependence: A Pilot Study. J Addict 2016, 6487217. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Hardeman RR, Karbeah J, 2020. Examining racism in health services research: A disciplinary self-critique. Health Serv. Res 55 Suppl 2(Suppl 2), 777–780. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Hedegaard H, Miniño A, Spencer M, Warner M, 2021. Drug Overdose Deaths in the United States, 1999–2020. NCHS Data Brief No. 428 Centers for Disease Control and Prevention. Accessed January 7, 2022 from: https://www.cdc.gov/nchs/products/databriefs/db428.htm. [Google Scholar]
  31. Hedegaard H, Schoenbaum M, Claassen C, Crosby A, Holland K, Proescholdbell S, 2018. Issues in Developing a Surveillance Case Definition for Nonfatal Suicide Attempt and Intentional Self-harm Using International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) Coded Data. Natl Health Stat Report(108), 1–19. [PubMed] [Google Scholar]
  32. Ilgen MA, Harris AH, Moos RH, Tiet QQ, 2007. Predictors of a suicide attempt one year after entry into substance use disorder treatment. Alcohol. Clin. Exp. Res 31(4), 635–642. [DOI] [PubMed] [Google Scholar]
  33. Kelly MM, Reilly E, Quinones T, Desai N, Rosenheck R, 2018. Long-acting intramuscular naltrexone for opioid use disorder: Utilization and association with multi-morbidity nationally in the Veterans Health Administration. Drug Alcohol Depend. 183, 111–117. [DOI] [PubMed] [Google Scholar]
  34. Krawczyk N, Feder KA, Fingerhood MI, Saloner B, 2017. Racial and ethnic differences in opioid agonist treatment for opioid use disorder in a U.S. national sample. Drug Alcohol Depend. 178, 512–518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Krawczyk N, Williams AR, Saloner B, Cerdá M, 2021. Who stays in medication treatment for opioid use disorder? A national study of outpatient specialty treatment settings. J Subst Abuse Treat 126, 108329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Larochelle MR, Bernson D, Land T, Stopka TJ, Wang N, Xuan Z, Bagley SM, Liebschutz JM, Walley AY, 2018. Medication for Opioid Use Disorder After Nonfatal Opioid Overdose and Association With Mortality: A Cohort Study. Ann. Intern. Med 169(3), 137–145. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Liang KY, Zeger SL, 1986. Longitudinal data analysis using generalized linear models. Biometrika 73, 13–22. [Google Scholar]
  38. Lin LA, Bohnert ASB, Blow FC, Gordon AJ, Ignacio RV, Kim HM, Ilgen MA, 2021. Polysubstance use and association with opioid use disorder treatment in the US Veterans Health Administration. Addiction 116(1), 96–104. [DOI] [PubMed] [Google Scholar]
  39. Lister JJ, Brown S, Greenwald MK, Ledgerwood DM, 2019. Gender-specific predictors of methadone treatment outcomes among African Americans at an urban clinic. Subst Abus 40(2), 185–193. [DOI] [PubMed] [Google Scholar]
  40. Louzon SA, Bossarte R, McCarthy JF, Katz IR, 2016. Does Suicidal Ideation as Measured by the PHQ-9 Predict Suicide Among VA Patients? Psychiatr. Serv 67(5), 517–522. [DOI] [PubMed] [Google Scholar]
  41. Manhapra A, Petrakis I, Rosenheck R, 2017. Three-year retention in buprenorphine treatment for opioid use disorder nationally in the Veterans Health Administration. Am. J. Addict [DOI] [PubMed] [Google Scholar]
  42. Manhapra A, Stefanovics E, Rosenheck R, 2020. Initiating opioid agonist treatment for opioid use disorder nationally in the Veterans Health Administration: Who gets what? Subst Abus 41(1), 110–120. [DOI] [PubMed] [Google Scholar]
  43. Marcovitz DE, McHugh RK, Volpe J, Votaw V, Connery HS, 2016. Predictors of early dropout in outpatient buprenorphine/naloxone treatment. Am. J. Addict 25(6), 472–477. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Michel L, Lions C, Maradan G, Mora M, Marcellin F, Morel A, Spire B, Roux P, Carrieri PM, 2015. Suicidal risk among patients enrolled in methadone maintenance treatment: HCV status and implications for suicide prevention (ANRS Methaville). Compr. Psychiatry 62, 123–131. [DOI] [PubMed] [Google Scholar]
  45. Molero Y, Zetterqvist J, Binswanger IA, Hellner C, Larsson H, Fazel S, 2018. Medications for Alcohol and Opioid Use Disorders and Risk of Suicidal Behavior, Accidental Overdoses, and Crime. The American journal of psychiatry 175(10), 970–978. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Morgan JR, Quinn EK, Chaisson CE, Ciemins E, Stempniewicz N, White LF, Linas BP, Walley AY, LaRochelle MR, 2022. Variation in Initiation, Engagement, and Retention on Medications for Opioid Use Disorder Based on Health Insurance Plan Design. Med. Care 60(3), 256–263. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Murphy SM, Fishman PA, McPherson S, Dyck DG, Roll JR, 2014. Determinants of buprenorphine treatment for opioid dependence. J Subst Abuse Treat 46(3), 315–319. [DOI] [PubMed] [Google Scholar]
  48. O’Connor AM, Cousins G, Durand L, Barry J, Boland F, 2020. Retention of patients in opioid substitution treatment: A systematic review. PloS one 15(5), e0232086. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Oliva EM, Harris AH, Trafton JA, Gordon AJ, 2012. Receipt of opioid agonist treatment in the Veterans Health Administration: facility and patient factors. Drug Alcohol Depend. 122(3), 241–246. [DOI] [PubMed] [Google Scholar]
  50. Petersen LA, Byrne MM, Daw CN, Hasche J, Reis B, Pietz K, 2010. Relationship between clinical conditions and use of Veterans Affairs health care among Medicare-enrolled veterans. Health Serv. Res 45(3), 762–791. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Pierce M, Bird SM, Hickman M, Marsden J, Dunn G, Jones A, Millar T, 2016. Impact of treatment for opioid dependence on fatal drug-related poisoning: a national cohort study in England. Addiction 111(2), 298–308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Priester MA, Browne T, Iachini A, Clone S, DeHart D, Seay KD, 2016. Treatment access barriers and disparities among individuals with co-occurring mental health and substance use disorders: An integrative literature review. J Subst Abuse Treat 61, 47–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Richards JE, Hohl SD, Whiteside U, Ludman EJ, Grossman DC, Simon GE, Shortreed SM, Lee AK, Parrish R, Shea M, Caldeiro RM, Penfold RB, Williams EC, 2019. If You Listen, I Will Talk: the Experience of Being Asked About Suicidality During Routine Primary Care. J. Gen. Intern. Med 34(10), 2075–2082. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Santo T Jr., Clark B, Hickman M, Grebely J, Campbell G, Sordo L, Chen A, Tran LT, Bharat C, Padmanathan P, Cousins G, Dupouy J, Kelty E, Muga R, Nosyk B, Min J, Pavarin R, Farrell M, Degenhardt L, 2021. Association of Opioid Agonist Treatment With All-Cause Mortality and Specific Causes of Death Among People With Opioid Dependence: A Systematic Review and Meta-analysis. JAMA psychiatry 78(9), 979–993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Schinka JA, Bossarte RM, Curtiss G, Lapcevic WA, Casey RJ, 2016. Increased Mortality Among Older Veterans Admitted to VA Homelessness Programs. Psychiatr. Serv 67(4), 465–468. [DOI] [PubMed] [Google Scholar]
  56. Shiner B, Leonard Westgate C, Bernardy NC, Schnurr PP, Watts BV, 2017. Trends in Opioid Use Disorder Diagnoses and Medication Treatment Among Veterans With Posttraumatic Stress Disorder. J Dual Diagn 13(3), 201–212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Shiner B, Peltzman T, Cornelius SL, Gui J, Forehand J, Watts BV, 2021. Recent trends in the rural-urban suicide disparity among veterans using VA health care. J. Behav. Med 44(4), 492–506. [DOI] [PubMed] [Google Scholar]
  58. Souden M, 2017. Overview of VA data, information systems, national databases and research uses. Accessed July 26, 2019 from: https://www.hsrd.research.va.gov/for_researchers/cyber_seminars/archives/2376-notes.pdf.
  59. StataCorp, 2019. Stata Statistical Software: Release 16, College Station, TX: StataCorp LLC. [Google Scholar]
  60. Stein BD, Gordon AJ, Sorbero M, Dick AW, Schuster J, Farmer C, 2012. The impact of buprenorphine on treatment of opioid dependence in a Medicaid population: recent service utilization trends in the use of buprenorphine and methadone. Drug Alcohol Depend. 123(1–3), 72–78. [DOI] [PubMed] [Google Scholar]
  61. Substance Abuse and Mental Health Services Administration, 2021. Medications for Opioid Use Disorder. Treatment Improvement Protocol (TIP) Series 63 Publication No. PEP21-02-01-002. Rockville, MD: Substance Abuse and Mental Health Services Administration. Accessed January 7, 2022 from: https://store.samhsa.gov/sites/default/files/SAMHSA_Digital_Download/PEP21-02-01-002.pdf. [Google Scholar]
  62. Taylor E, Timko C, Harris AHS, Yu M, Finlay AK, 2019. Receipt of pharmacotherapy for alcohol use disorder by justice-involved women in the Veterans Health Administration. Addict Sci Clin Pract 14(1), 1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. VA Program Evaluation and Resource Center, VA Stratification Tool for Opioid Risk Management (STORM) Definitions Report. Accessed September 18, 2021 from: https://vaww.pbi.cdw.va.gov/PBIRS/Pages/ReportViewer.aspx?/RVS/OMHSP_PERC/SSRS/Production/CDS/STORM/ORM_Definitions_ClinicalDetail&VariableName=37.
  64. Vakkalanka JP, Lund BC, Ward MM, Arndt S, Field RW, Charlton M, Carnahan RM, 2021. Telehealth Utilization Is Associated with Lower Risk of Discontinuation of Buprenorphine: a Retrospective Cohort Study of US Veterans. J. Gen. Intern. Med, 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Vakkalanka P, Lund BC, Arndt S, Field W, Charlton M, Ward MM, Carnahan RM, 2021. Association Between Buprenorphine for Opioid Use Disorder and Mortality Risk. Am. J. Prev. Med 61(3), 418–427. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. VHA Office of Rural Health, 2015. Chapter 3: Veteran Population, Enrollees, and Patients. A Rural Veterans Health Care Atlas Series -- 2nd edition Accessed September 9, 2021 at https://www.ruralhealth.va.gov/docs/atlas/CHAPTER_03_Vets_Enrollees_Pts.pdf.
  67. Volkow ND, Jones EB, Einstein EB, Wargo EM, 2019. Prevention and Treatment of Opioid Misuse and Addiction: A Review. JAMA psychiatry 76(2), 208–216. [DOI] [PubMed] [Google Scholar]
  68. Wakeman SE, Larochelle MR, Ameli O, Chaisson CE, McPheeters JT, Crown WH, Azocar F, Sanghavi DM, 2020. Comparative Effectiveness of Different Treatment Pathways for Opioid Use Disorder. JAMA Netw Open 3(2), e1920622. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Watts BV, Gottlieb DJ, Riblet NB, Gui J, Shiner B, 2022. Association of Medication Treatment for Opioid Use Disorder With Suicide Mortality. Am. J. Psychiatry 179(4), 298–304. [DOI] [PubMed] [Google Scholar]
  70. Wilcox HC, Conner KR, Caine ED, 2004. Association of alcohol and drug use disorders and completed suicide: an empirical review of cohort studies. Drug Alcohol Depend. 76 Suppl, S11–19. [DOI] [PubMed] [Google Scholar]
  71. Williams EC, Frost MC, Rubinsky AD, Glass JE, Wheat CL, Edmonds AT, Chen JA, Matson TE, Fletcher OV, Lehavot K, Blosnich JR, 2021. Patterns of Alcohol Use Among Transgender Patients Receiving Care at the Veterans Health Administration: Overall and Relative to Nontransgender Patients. J Stud Alcohol Drugs 82(1), 132–141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Woolf SH, Schoomaker H, 2019. Life Expectancy and Mortality Rates in the United States, 1959–2017. JAMA 322(20), 1996–2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Wyse JJ, Gordon AJ, Dobscha SK, Morasco BJ, Tiffany E, Drexler K, Sandbrink F, Lovejoy TI, 2018. Medications for opioid use disorder in the Department of Veterans Affairs (VA) health care system: Historical perspective, lessons learned, and next steps. Subst Abus 39(2), 139–144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Wyse JJ, McGinnis KA, Edelman EJ, Gordon AJ, Manhapra A, Fiellin DA, Moore BA, Korthuis PT, Kennedy AJ, Oldfield BJ, Gaither JR, Gordon KS, Skanderson M, Barry DT, Bryant K, Crystal S, Justice AC, Kraemer KL, 2022. Twelve-Month Retention in Opioid Agonist Treatment for Opioid Use Disorder Among Patients With and Without HIV. AIDS Behav 26(3), 975–985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Wyse JJ, Robbins JL, McGinnis KA, Edelman EJ, Gordon AJ, Manhapra A, Fiellin DA, Moore BA, Korthuis PT, Gaither JR, Gordon K, Skanderson M, Barry DT, Crystal S, Justice A, Kraemer KL, 2019. Predictors of timely opioid agonist treatment initiation among veterans with and without HIV. Drug Alcohol Depend. 198, 70–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Young BA, Maynard C, Boyko EJ, 2003. Racial differences in diabetic nephropathy, cardiovascular disease, and mortality in a national population of veterans. Diabetes Care 26(8), 2392–2399. [DOI] [PubMed] [Google Scholar]
  77. Zivin K, Kim HM, McCarthy JF, Austin KL, Hoggatt KJ, Walters H, Valenstein M, 2007. Suicide mortality among individuals receiving treatment for depression in the Veterans Affairs health system: associations with patient and treatment setting characteristics. Am. J. Public Health 97(12), 2193–2198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Zou G, 2004. A modified poisson regression approach to prospective studies with binary data. Am. J. Epidemiol 159(7), 702–706. [DOI] [PubMed] [Google Scholar]
  79. Zwald ML, Holland KM, Annor FB, Kite-Powell A, Sumner SA, Bowen DA, Vivolo-Kantor AM, Stone DM, Crosby AE, 2020. Syndromic Surveillance of Suicidal Ideation and Self-Directed Violence - United States, January 2017-December 2018. MMWR. Morb. Mortal. Wkly. Rep 69(4), 103–108. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendices
NIHMS1836488-supplement-Appendices.docx (63.2KB, docx)

RESOURCES