Skip to main content
NCBI home page
VA Author Manuscripts logoLink to VA Author Manuscripts
. Author manuscript; available in PMC: 2025 May 1.
Published in final edited form as: Am J Psychiatry. 2024 May 1;181(5):434–444. doi: 10.1176/appi.ajp.20230683

Associations Between Primary Care-Delivered Alcohol-Related Brief Intervention and Subsequent Opioid-Related Outcomes

Dan V Blalock 1,2, Sophia A Berlin 3,4, Theodore Berkowitz 1, Valerie A Smith 1,5,6, Charles Wright 7, Rachel L Bachrach 8,9, Janet M Grubber 1,10
PMCID: PMC11076009  NIHMSID: NIHMS1987828  PMID: 38706328

Abstract

Objective:

The co-occurrence of unhealthy alcohol use and opioid misuse is high and associated with increased rates of overdose, emergency healthcare utilization, and death. The current study examined whether receipt of alcohol-related brief intervention (BI) is associated with reduced risk of negative downstream opioid-related outcomes.

Methods:

This retrospective cohort study included all VISN-6 Veterans Affairs (VA) patients with an Alcohol Use Disorders Identification Test-Consumption (AUDIT-C) screening (N=492,748) from 2014–2019. Logistic regression was used to examine the association between documentation of alcohol-related BI and probability of 1) new opioid prescription, 2) new opioid use disorder (OUD) diagnosis, or 3) new opioid-related hospitalization in the following year controlling for demographic and clinical covariates.

Results:

Thirteen percent (N=63,804) of Veterans had “positive” AUDIT-C screens. Of those, 72% (N=46,216) had documented alcohol-related BI. Within one year, 8.5% (N=5430) had a new opioid prescription, 1.1% (N=698) had a new OUD diagnosis, and 0.8% (N=499) had a new opioid-related hospitalization. In adjusted models, Veterans with “positive” AUDIT-C screens that did not receive alcohol-related BI had higher odds of new opioid prescriptions (AOR=1.10, 95%C.I.[1.03,1.17]) and new OUD diagnoses (AOR=1.19, 95%C.I.[1.02,1.40]), while new opioid-related hospitalizations (AOR=1.19, 95%C.I.[0.99,1.44]) were higher though not statistically significant. Removal of medications for OUD (MOUD) did not impact associations. All outcomes were significantly associated with alcohol-related BI in unadjusted models.

Conclusions:

VA’s standard alcohol-related BI is associated with subsequent lower odds of new opioid prescription and new OUD diagnosis. Results suggest a reduction in a cascade of new opioid-related outcomes from prescription through hospitalization.

INTRODUCTION

Over half of adults with opioid use disorder (OUD) have an additional comorbid substance use disorder (SUD).1 Specifically, individuals with OUD are nearly twice as likely to meet criteria for alcohol use disorder (AUD), versus those without OUD.2 Individuals who endorse binge drinking (consuming 4/5+ drinks in a single setting for women/men) are 3.5x more likely to misuse prescription opioids, and 20% of prescription opioid deaths also involve alcohol.3,4 Alcohol involvement among opioid overdoses increased from 11.5% in 2013 to 14.9% in 2017, or 1.3 additional deaths per 100,000 persons.5 People who engage in unhealthy drinking (a range of alcohol use patterns above recommended limits),6 and use prescription opioids are at greater risk of overdose and death due to depressant effects of alcohol on the respiratory and central nervous systems. The risk of harm increases with the amount of alcohol consumed, but there is no safe level of alcohol use for people using opioids.7,8

Veterans are 3x more likely to endorse unhealthy alcohol use, and much more likely to overdose on opioids (prescription or otherwise) than civilians.9,10 A 2014 Veterans Affairs (VA) report indicated that over 440,000 patients (7.7%) were currently prescribed opioids, placing them at potential risk, and 55,000 patients (1%) were diagnosed as having a current OUD.11,12 In 2017, almost 60% of VA patients with OUD had at least one additional SUD, AUD being the most common.13 Veterans with multiple SUDs show higher utilization of inpatient psychiatric treatment, emergency department services, rates of homelessness, psychiatric medication prescriptions, and more severe medical and psychiatric comorbidities.14

Screening and Brief Intervention (BI) for alcohol use is an intervention for mitigating unhealthy alcohol use with regular surveillance of alcohol consumption in primary care settings. The goal of BIs is to change not only immediate practices or thoughts about risky behavior, but to also address long-standing problems associated with unhealthy alcohol use.15 If alcohol consumption is endorsed at a level potentially harmful to a patient’s health or well-being, in-the-moment brief educational and normative feedback can be provided by the primary care team, and if needed, referral to more intensive treatment.16–18 Screening and BI is established as an efficacious population-based intervention to reduce unhealthy alcohol use.19,20 Thus, the U.S. Preventive Services Task Force recommends routine screening and BI in primary care settings given its effectiveness at curbing unhealthy alcohol use.21,22 Additionally, the VA was a leader in the implementation of alcohol screening and BI, implementing annual screenings in 2004 and BI in 2008.19 Documentation of BI after an elevated alcohol screening is a frequently used quality indicator in VA primary care practices.23

Integrated interventions consisting of screening tools and brief interventions are priorities in addressing unhealthy alcohol use among opioid-dependent populations.24 Screening and BI has preliminary evidence of efficacy in reducing unhealthy alcohol use in opioid-dependent populations, but many studies have focused on treatment delivery and provider interactions. These studies have not investigated 1) any direct effects of screening and BI for unhealthy alcohol use on comorbid opioid misuse, or 2) any potential downstream effects of reduced alcohol consumption on the impairment or lethality associated with opioid misuse. The CDC maintains the position that screening and BI can be effective for concurrent alcohol and opioid misuse. However, screening and BI methods may require modifications when delivered to patients with OUD, which may include a greater emphasis on abstaining or decreasing alcohol intake.25

In 2013, the VA developed and implemented the Opioid Safety Initiative (OSI) to promote safer opioid-related prescribing and tracking of opioid-related outcomes in the VA healthcare system.26–29 As a result of the OSI, the VA implemented several strategies to track and recommend risk mitigation activities: for example, a Stratification Tool for Opioid Risk Mitigation (STORM). STORM estimates the risk of overdose and incidence of suicide-related events for VA patients on opioids,30 improving the delivery and efficiency of clinical care. STORM incorporates AUD in its algorithms, and current CDC guidelines recommend screening for unhealthy alcohol use in individuals with opioid prescriptions. However, research has not examined whether yearly alcohol-related screening and BI in VA is associated with opioid-related outcomes.

Numerous potential mechanisms may explain why alcohol-related screening and BI may be associated with opioid-related outcomes (reviewed in the discussion below). An overarching reason for this potential association is that a “catalyst for change” emerges somewhere in a patient’s environment (including possibly the alcohol-related screening and BI itself), and spurs action across multiple domains. Some qualitative evidence exists for this from interviews of patients who have reduced unhealthy alcohol use in the context of other problems and identified an “awareness of accumulating harms” related to alcohol use that led to these changes.31 Yet while guidance already exists to capitalize on an “awareness of accumulating harms” for alcohol use in opioid using populations, there is no guidance addressing these harms in the opposite direction. This lack of guidance is especially surprising given the earlier lifetime onset of alcohol-related problems, and their impact on increasing the severity of opioid-related harms like OUD diagnosis.32 Additionally, given the establishment of alcohol-related screening and BI at the population level in VA, the directionality of alcohol-related BI leading to improved downstream opioid-related outcomes has a much greater potential population-level impact.

Therefore, the purpose of the present study was to investigate the association between primary care-delivered alcohol-related BI and opioid-related outcomes (new opioid prescriptions, new OUD diagnosis, and new opioid-related hospitalizations) in Veterans one year after a “positive” screening for unhealthy alcohol use.

METHODS

Study design and data source

We examined a retrospective cohort of 492,748 Veterans seeking healthcare within the Veterans Integrated Service Network (VISN)-6, a southeast region of the United States, who had at least one recorded AUDIT-C score between Jan 1, 2014 and Dec 31, 2019. Data were available for Jan 1, 2013 through Dec 31, 2020 for prior year covariates and prior year and subsequent year opioid-related outcomes. Existing electronic health records (EHR) data were pulled from outpatient and inpatient records in the VA’s Corporate Data Warehouse (CDW), including “health factors” populated by medical center-specific clinical reminders such as documentation of alcohol-related BI. Veterans were excluded for implausible birth or death dates, missing covariate data, implausible data linkages or lack of AUDIT-C scores during the study period (see Figure 1). Study protocols were approved by the Durham VA Health Care System Institutional Review Board. Additional information can be found in the Technical Appendix.

Figure 1.

Figure 1.

Open in a new tab

Study Flow Diagram

Notes:

a Veteran’s birthdate was earlier than calendar year 1913 or later than calendar year 2002; death date was prior to calendar year 2014; gender was an invalid value; or had personal identifier linkage mismatches.

b Individual Veterans may have multiple opioid outcomes of interest and, thus, may contribute to more than one opioid outcome analysis as the “event of interest”.

Measures

Yearly alcohol screening scores

Yearly alcohol screening scores were derived from in-clinic measures of the Alcohol Use Disorders Identification Test-Consumption (AUDIT-C) screening tool.33,34 Scores ≥ 5 are considered “positive” screens in VA and require documentation of brief intervention and/or referral to treatment to meet quality of care performance metrics.23 AUDIT-C scores denote alcohol use severity, with specified levels of no alcohol use (AUDIT-C=0), low- (AUDIT-C<5), moderate- (AUDIT-C=5–7), or high-risk drinking (AUDIT-C≥8). The index AUDIT-C was defined as the first “positive” (≥5) AUDIT-C in the 2014–2019 observation window, otherwise the highest (<5) AUDIT-C in the observation window. This measure represents the “screening” in “screening and BI.”

Documentation of alcohol-related brief intervention

Documentation of BI was obtained 0–14 days after a “positive” AUDIT-C screen. It has been used across VA nationally to meet performance measures and the measure has been defined, pulled, and quantified in prior studies.19,23,35,36 This documentation represents the “BI” in “screening and BI.” Though content of BIs vary, consistent elements recommended by VA/DoD SUD Clinical Practice Guidelines (2021) include expressing concern about one’s drinking, providing individualized feedback on how one’s drinking may be adversely impacting one’s health, advising one to either abstain or reduce one’s drinking to within recommended drinking limits, and offering a referral to SUD treatment if appropriate.37

Opioid-related Outcomes

Opioid prescriptions included all outpatient CN101 prescriptions from the VA’s formulary (sensitivity analyses removed CN101 prescriptions associated with medications for OUD (MOUD), most notably buprenorphine, buprenorphine/naloxone, and methadone). ICD-9-CM and ICD-10-CM codes were used to identify OUD diagnoses. In order to establish temporal ordering between the receipt of screening and BI window and subsequent new opioid prescriptions and OUD diagnosis, only patients without opioid prescriptions or OUD diagnoses, respectively, in the year prior to index AUDIT-C or days 0–14 after index AUDIT-C were included. Hospitalizations with a documented reason matching one of the OUD diagnosis codes were coded as an opioid-related hospitalization. Because all opioid-related hospitalizations are new events, we did not exclude patients based on prior hospitalizations.

Covariates

Covariates were defined based on data from up to two years before index and included gender marital status, rurality, VA priority group, Nosos predictive risk score (a VA-developed score using existing VA demographic, diagnosis, drug, and utilization data to predict future patient healthcare costs relative to national VA average patient cost, where a value of 1.0 indicates the veteran is expected to have costs that are the national average of VA patients))38, and relevant mental health and substance use comorbidities including the presence vs. absence of the following: high-risk alcohol use (i.e., AUDIT-C≥8), depression/mood disorders, non-PTSD anxiety disorders, PTSD, schizophrenia and bipolar disorder, and other SUDs (excluding OUD/AUD). See Table 1 and Technical Appendix for more information on covariates.

Table 1.

Demographic and Clinical Characteristics of Veterans for Whom Brief Intervention Was Required (Index AUDIT-C ≥ 5), Overall and by Presence vs. Absence of Documentation of Brief Intervention Occurrence

Total Brief Intervention Documented within 0–14 Days after Index AUDIT-C
No Yes
N (%) N (%) N (%) SMDa
Total 63804 (100.0) 17588 (100.0) 46216 (100.0)
Age (n, mean, std dev) 63804 52 (15) 17588 52 (15) 46216 52 (15) 0.02
 <35 11527 (18.1) 3236 (18.4) 8291 (17.9)
 35–49 14657 (23.0) 3995 (22.7) 10662 (23.1)
 50–64 23006 (36.1) 6297 (35.8) 16709 (36.2)
 ≥ 65 14614 (22.9) 4060 (23.1) 10554 (22.8)
Gender 0.05
 Female 4451 (7.0) 1400 (8.0) 3051 (6.6)
 Male 59353 (93.0) 16188 (92.0) 43165 (93.4)
Race 0.06
 American Indian or Alaska Native 473 (0.7) 129 (0.7) 344 (0.7)
 Asian 282 (0.4) 72 (0.4) 210 (0.5)
 Black or African American 23786 (37.3) 6937 (39.4) 16849 (36.5)
 Native Hawaiian or Other Pacific Islander 327 (0.5) 92 (0.5) 235 (0.5)
 White 36539 (57.3) 9725 (55.3) 26814 (58.0)
 Unknown 2397 (3.8) 633 (3.6) 1764 (3.8)
Ethnicity 0.03
 Hispanic or Latino 1745 (2.7) 515 (2.9) 1230 (2.7)
 Not Hispanic or Latino 61245 (96.0) 16889 (96.0) 44356 (96.0)
 Unknown 814 (1.3) 184 (1.0) 630 (1.4)
Marital Status 0.05
 Separated/Divorced/Widowed 17359 (27.2) 5004 (28.5) 12355 (26.7)
 Single/Never Married 8883 (13.9) 2456 (14.0) 6427 (13.9)
 Married 21657 (33.9) 5735 (32.6) 15922 (34.5)
 Unknown 15905 (24.9) 4393 (25.0) 11512 (24.9)
Rurality 0.10
 Rural/Highly Rural 23744 (37.2) 5947 (33.8) 17797 (38.5)
 Urban 40060 (62.8) 11641 (66.2) 28419 (61.5)
VA Priority Group b 0.04
 1–6 55548 (87.1) 15475 (88.0) 40073 (86.7)
 7–8 8256 (12.9) 2113 (12.0) 6143 (13.3)
Clinical Substance Use and Mental Health Diagnoses in Year Prior to Index AUDIT-C
High Risk Alcohol Use (AUDIT-C ≥ 8) <0.001
 No 41015 (64.3) 11309 (64.3) 29706 (64.3)
 Yes 22789 (35.7) 6279 (35.7) 16510 (35.7)
Other Substance Use Disorder 0.10
 No 57833 (90.6) 15547 (88.4) 42286 (91.5)
 Yes 5971 (9.4) 2041 (11.6) 3930 (8.5)
Depression or Mood Disorder 0.15
 No 49511 (77.6) 12809 (72.8) 36702 (79.4)
 Yes 14293 (22.4) 4779 (27.2) 9514 (20.6)
Non-PTSD Anxiety 0.10
 No 56091 (87.9) 15040 (85.5) 41051 (88.8)
 Yes 7713 (12.1) 2548 (14.5) 5165 (11.2)
PTSD 0.12
 No 51607 (80.9) 13603 (77.3) 38004 (82.2)
 Yes 12197 (19.1) 3985 (22.7) 8212 (17.8)
Schizophrenia or Bipolar Disorder 0.07
 No 61521 (96.4) 16784 (95.4) 44737 (96.8)
 Yes 2283 (3.6) 804 (4.6) 1479 (3.2)
Nosos Score c 0.18
 < 0.50 9672 (15.2) 2386 (13.6) 7286 (15.8)
 0.50 – < 1.00 19951 (31.3) 5640 (32.1) 14311 (31.0)
 1.00 – < 1.50 8084 (12.7) 2448 (13.9) 5636 (12.2)
 1.50 – < 2.00 3254 (5.1) 1091 (6.2) 2163 (4.7)
 ≥ 2.00 4690 (7.4) 1698 (9.7) 2992 (6.5)
 Unknown 18153 (28.5) 4325 (24.6) 13828 (29.9)
Open in a new tab

SMD=Standardized Mean Difference.

a

Represents the Standardized Mean Difference between each demographic and clinical category by receipt of brief intervention. Created using the ‘tableone’ package in R (version 4.3.2).

b

A grouping of Veterans based on military service history, disability rating, income level, Medicaid qualification and receipt of other VA benefits used in helping assign the priority by which Veterans receive VA healthcare and benefits (1=highest and 8=lowest priority). According to current VA guidelines, copays for inpatient care are not required for priority groups 1–6 but are typically required for priority groups 7–8.

c

Prediction of VA patient healthcare cost for upcoming fiscal year based on prior fiscal year VA demographic, diagnosis, drug and utilization data that is reported relative to the national average healthcare cost of VA patients (1 = same, > 1 = greater, and < 1= less than the national average fiscal year cost for VA patients).We have categorized the continuous score.

Statistical Analysis

Balance in patient characteristics among patients with “positive” AUDIT-C scores was evaluated for those receiving and not receiving BI using standardized mean differences (SMDs), where values <0.1 indicate no meaningful difference.39 SMDs were calculated using the ‘tableone’ package in R (version 4.3.2).40 For Veterans who were required by VA policy to have BI (i.e., after a “positive” AUDIT-C), we calculated odds ratios and 95% confidence intervals (95% C.I.) for the associations between the presence vs. absence of the provision of a BI during days 0–14 post-“positive” AUDIT-C and each of the three primary opioid-related outcomes using logistic regression adjusting for the covariates listed in the prior section. Unadjusted analyses were also conducted to understand the impact of adjustment. A sensitivity analysis was conducted, excluding CN101 MOUD medications from identification as “opioid prescriptions” during any day in the observation window. More information on the entire sample of Veterans with AUDIT-C during the study timeframe (not just those with a “positive” AUDIT-C), can be found in eTable1 and in the Technical Appendix. Statistical significance was set to α=.05. All analyses aside from SMDs were performed using SAS version 9.4.

RESULTS

Descriptive Opioid-Related Outcome Trends Across All Patients with AUDIT-C

Rates of opioid prescriptions, OUD diagnoses, and opioid-related hospitalizations appeared to increase somewhat with higher alcohol use levels (eTable1 top pane). Rates of all opioid-related outcomes also appeared somewhat higher when documented BI was present (versus absent) in patients with AUDIT-C values below the “positive” threshold (eTable1 bottom pane).

Demographic and Clinical Characteristics for Patients with AUDIT-C ≥ 5

Of the 492,748 patients in our sample, 63,804 patients (13%) had a “positive” AUDIT-C (after removal of N=201 with missing covariates). Table 1 presents patient demographic and clinical characteristics overall, and broken down by the presence or absence of documented BI. Patients’ average age was 52 years (SD=15). A majority were White (57.3%), and a large minority were Black (37.3%). Female Veterans represented 7% of the sample. SMDs suggest relatively small differences in characteristics among those who did and did not receive BI within 14 days, with most <0.1 and all <0.2.

AUDIT-C, BI, and Opioid-Related Outcomes for Patients with AUDIT-C ≥ 5

Of those 63,804 patients with “positive” AUDIT-C screens, 46,216 (72.4%) had documented BI within 14 days of index AUDIT-C, 4.8% (N=3045) had a new opioid prescription in the subsequent 15 to 365 days after “positive” AUDIT-C, 1.1% (N=698) had a new OUD diagnosis, and 0.8% (N=499) had a new opioid-related hospitalization in the subsequent 15 to 365 days after “positive” AUDIT-C.

Among all 63,804 patients, 5764 (9.03%) had only one new opioid-related outcome, and 212 (0.33%) had multiple new opioid-related outcomes in the post-BI observation window (eTable2 top pane, with similar numbers from sensitivity analysis in eTable2 bottom pane).

BI Predicting New Opioid-Related Outcomes in Patients with AUDIT-C ≥ 5

In unadjusted analyses, absence of alcohol-related BI when clinically indicated was associated with increased one-year odds of new opioid prescriptions (OR=1.15, 95%C.I.[1.08,1.23]), new OUD diagnoses (OR=1.35, 95%C.I.[1.15,1.58]), and new opioid-related hospitalizations (OR=1.41, 95%C.I.[1.17,1.70]) (Table 2).

Table 2.

Unadjusted and Adjusted Odds Ratios and 95% Confidence Intervals for Association between Having a New Opioid Prescription, OUD Diagnosis, or Hospitalization in the 15–365 Days post “Positive” AUDIT-C (≥5) Date among Veterans Receiving vs. Not Receiving (reference) Documented BI

New Opioid Prescriptions New OUD Diagnoses New Opioid-Related Hospitalizations
(N=51,473a) (N=62,095a) (N=63,804a)
Unadjusted Adjusted Unadjusted Adjusted Unadjusted Adjusted
OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI
Absence of BI when Clinically Indicated 1.15 1.08, 1.23 1.10 1.03, 1.17 1.35 1.15, 1.58 1.19 1.02,1.40 1.41 1.17,1.70 1.19 0.99,1.44
Presence of BI when Clinically Indicated 1.00 reference 1.00 reference 1.00 reference 1.00 reference 1.00 reference 1.00 reference
Covariates
 Age 35–49 1.11 1.01,1.22 0.78 0.62,0.99 0.65 0.49,0.85
50–64 1.29 1.18,1.41 0.86 0.69,1.08 0.60 0.46,0.78
≥65 0.95 0.85,1.05 0.35 0.25,0.48 0.47 0.34,0.67
<35 1.00 reference 1.00 reference 1.00 reference
 Gender Female 1.05 0.94,1.17 0.97 0.73,1.28 0.71 0.49,1.03
Male 1.00 reference 1.00 reference 1.00 reference
 Race American Indian or Alaska Native 1.05 0.76,1.45 1.06 0.46,2.41 1.76 0.81,3.84
Asian 0.80 0.50,1.27 0.97 0.31,3.06 0.84 0.20,3.48
Black or African American 0.92 0.87,0.98 0.70 0.59,0.83 0.54 0.44,0.67
Native Hawaiian or Other Pacific Islander 0.62 0.38,1.00 0.78 0.25,2.48 0.28 0.04,2.06
Unknown 0.86 0.72,1.03 0.93 0.59,1.47 0.75 0.42,1.34
White 1.00 reference 1.00 reference 1.00 reference
 Ethnicity Hispanic or Latino 0.88 0.73,1.16 0.66 0.37,1.16 0.91 0.50,1.65
Unknown 0.63 0.45,0.87 0.53 0.19,1.46 0.50 0.12,2.08
Not Hispanic or Latino 1.00 reference 1.00 reference 1.00 reference
 Marital Status Separated/Divorced/Widowed 1.18 1.09,1.27 1.60 1.30,1.97 1.70 1.32,2.21
Single/Never Married 1.02 0.93,1.13 1.60 1.25,2.04 2.00 1.50,2.66
Unknown 1.16 1.08,1.26 1.09 0.86,1.39 1.03 0.75,1.40
Married 1.00 reference 1.00 reference 1.00 reference
 Rurality Rural/Highly Rural 1.02 0.96,1.08 0.98 0.83,1.16 0.94 0.77,1.14
Urban 1.00 reference 1.00 reference 1.00 reference
 Priority Group 1–6 1.04 0.95,1.14 1.35 0.99,1.84 1.36 0.91,2.05
7–8 1.00 reference 1.00 reference 1.00 reference
Clinical Substance Use and Mental Health Diagnoses in Year Prior to Index AUDIT-C
 High Risk Alcohol Use (AUDIT-C ≥ 8) Yes 1.12 1.06,1.19 1.79 1.54,2.10 1.38 1.15,1.66
No 1.00 reference 1.00 reference 1.00 reference
 Other Substance Use Disorder Yes 1.07 0.96,1.19 2.95 2.43,3.60 4.13 3.29,5.19
No 1.00 reference 1.00 reference 1.00 reference
 Depression or Mood Disorder Yes 1.05 0.97,1.14 1.51 1.25,1.82 1.55 1.24,1.94
No 1.00 reference 1.00 reference 1.00 reference
 Non-PTSD Anxiety Yes 1.04 0.95,1.14 0.98 0.80,1.20 1.19 0.95,1.48
No 1.00 reference 1.00 reference 1.00 reference
 PTSD Yes 1.05 0.96,1.13 1.10 0.91,1.32 1.15 0.93,1.41
No 1.00 reference 1.00 reference 1.00 reference
 Schizophrenia or Bipolar Disorder Yes 0.64 0.54,0.76 1.49 1.16,1.92 1.61 1.24,2.08
No 1.00 reference 1.00 reference 1.00 reference
 NOSOS Score <0.50 0.45 0.40,0.51 0.72 0.52,1.02 0.94 0.61,1.45
0.50– <1.00 0.69 0.63,0.75 0.68 0.51,0.89 0.83 0.58,1.19
1.5– <2.00 1.34 1.17,1.54 1.52 1.10,2.09 1.67 1.10,2.53
≥2.0 1.20 1.05,1.37 1.97 1.51,2.58 3.19 2.30,4.43
Unknown 0.56 0.51,0.62 1.03 0.78,1.36 1.26 0.87,1.82
1.00 - < 1.50 1.00 reference 1.00 reference 1.00 reference
Open in a new tab
a

Sample sizes for analyses vary by outcome due to exclusion of patients with opioid outcomes of interest in the −365 to +14 day observational window around index AUDIT-C for opioid prescription and opioid use disorder.

In analyses adjusted for relevant covariates, absence of alcohol-related BI when clinically indicated was associated with increased one-year odds of new opioid prescriptions (AOR=1.10, 95%C.I.[1.03,1.17]) and new OUD diagnoses (AOR=1.19, 95%C.I.[1.02,1.40]). The association between absence of BI and new opioid-related hospitalizations was similar in magnitude, but with half the number of observations, did not reach statistical significance (AOR=1.19, 95%C.I.[0.99,1.44]) (Table 2).

Sensitivity analyses removing all MOUD medications from the observation window also found absence of alcohol-related BI when clinically indicated was associated with increased one-year odds of new opioid prescriptions (excluding MOUD) in unadjusted (OR=1.15, 95%C.I.[1.09,1.23]) and adjusted (OR=1.10, 95%C.I.[1.03,1.17]) analyses (eTable3).

DISCUSSION

The current study sought to investigate the association between screening and brief intervention (BI) for unhealthy alcohol use and one-year opioid-related outcomes, including opioid prescriptions, OUD diagnoses, and opioid-related hospitalizations. Across five years and almost half a million VA patients, absence of BI when clinically indicated by VA policy was associated with increased odds of approximately 10% or more of opioid-related outcomes over the following year, depending on the outcome. Specifically, absence of BI was associated with 10% higher odds of obtaining a new opioid prescription within one year (with or without the inclusion of MOUD prescriptions), and 19% higher odds of receiving an OUD diagnosis within one year. Absence of BI may also be associated with higher odds of having an opioid-related hospitalization within one year, but because the 95% confidence interval in the adjusted model included 1.00, more work should be conducted before such inferences are drawn. Altogether, these data present consistent evidence across a cascade of opioid-related outcomes that absence of BI after a positive alcohol screen is associated with increased opioid-related risks in the following year.

Potential Mechanisms

Mechanisms of action were not examined in the current study, but the association between BI and opioid-related harms may exist via at least four routes. First, BI may directly reduce the chances of opioid-related harms. BI is designed to mitigate unhealthy alcohol use and existing evidence suggests unhealthy alcohol use substantially increases the harms associated with opioid use.3,4 Therefore, any direct reductions of unhealthy alcohol use at a population level may reduce opioid-related harms even with similar levels of opioid use. Second, BI may indirectly reduce the chances of opioid-related harms by reducing opioid use. Higher levels of alcohol use are associated with increased incidence and severity of chronic pain,41 as well as sharply increased risk of injury,42 both of which are strong predictors of opioid prescription. The current study is the first to produce any evidence that BI is associated with lower odds of future receipt of an opioid prescription in a general patient population. This association between BI and a lower odds of future opioid prescriptions, however, is unlikely to be fully responsible for the association between BI and a lower likelihood of future opioid-related harms (as seen by the presence of those opioid-related harms without opioid prescriptions in eTable2).

Third, BI may indirectly reduce opioid-related harms by improving substance use service utilization more broadly,43 though evidence for this route is somewhat mixed,16 or even suggestive of reduced utilization of alcohol-related SUD treatment.44 Nevertheless, there remain many other substance use services outside of alcohol-related SUD treatment that may impact opioid-related outcomes. Finally, BI and opioid-related harms may both be associated with other patient attributes, behaviors, or healthcare processes. The presence of a “positive” AUDIT-C and documented BI may represent a joint success of patient attributes (motivation to seek initial care or change behaviors, fewer competing mental or physical health concerns), behaviors (success in seeking care or making behavioral changes), and healthcare processes (successful scheduling of appointments, screening, and documentation). Subsequent opioid-related outcomes, therefore may also represent continued joint success of these patient attributes, behaviors, and healthcare system processes. While in some instances this success may be driven more by patient processes leading them to seek care or make behavioral changes related to substance use at a particular time, it may also be the case that increased documented BI is an indicator of better quality care in a specific clinic or medical center.

While the temporal ordering of these concepts and adjustments in our analyses attempted to control for as many confounding factors as were accessible, we cannot rule out the potential for residual confounding. Supplemental material from this study provides novel detailed information on the prevalence of all three opioid-related outcomes in each time period in a large sample of patients with unhealthy alcohol use, which may provide important information for future studies. This material also provides detailed information on the co-occurrence (or absence of co-occurrence) of each of these opioid-related outcomes, and underscores the need for future studies to assess opioid-related outcomes in a variety of ways to gain a more complete understanding of mechanisms.

Potential Impacts

The potential clinical implications of these results are highly relevant to the treatment of polysubstance use generally. This novel connection between alcohol-related BI and reductions in non-treated opioid use increases the importance of national initiatives already underway to expand the implementation of alcohol-related screening and BI to meet population needs for alcohol use.45 As seen in the current data, not everyone who meets screening criteria for BI receives it. Barriers to receipt of BI even after a “positive” screen may include competing medical demands, lack of time during a clinical encounter, lack of knowledge and/or low self-efficacy by providers, and lack of confidence in the efficacy of BI among others.46,47 Thus evidence for the benefits of alcohol-related screening and BI (besides its effectiveness in curbing alcohol use) may increase the adoption and fidelity of this intervention more broadly.

It is important to highlight that this work is specifically looking at lowered risk for a substance (opioids) based on an intervention entirely unrelated to that substance, which has been seen in few other settings.48 This connection underscores the importance of the CDC’s recommendation for alcohol-related screening and BI for opioid using populations specifically.25

The potential clinical implications of these results are also highly relevant to population-level screening and management of alcohol use and opioid use. VA’s national, multifaceted approach to reduce opioid prescribing since 2013 with OSI, has been successful on many fronts. OSI has led to iterative strategies over time that have helped to decrease the initiation and duration of opioid prescribing, including expanded contraindications for long-term opioid therapy in the 2017 VA/DoD issued Clinical Practice Guideline for Opioid Therapy for Chronic Pain, and the requirement for interdisciplinary team review/care coordination for Veterans at very high risk for overdose/suicide in 2018.49 While effective, the majority of these strategies have focused on patients already at high risk for opioid-related harms due to known factors like chronic pain or opioid prescription history. The current results may speak to additional strategies to employ at a population level that can continue to bolster OSI’s already effective iterative and multifaceted strategy for opioid risk mitigation.

Most notably, VA’s opioid risk surveillance tool (STORM) has been widely adopted and is effective at reducing opioid-related risk at a population level.29 However, the algorithm used to stratify patients’ risk is becoming dated, and regarding alcohol in particular, uses only AUD diagnoses which are severe underestimates of true AUD prevalence in general and in high-risk populations.50 On the other hand, screening and BI from the current study uses AUDIT-C alcohol screenings, which are required to be collected annually on every VA patient. The combination of information from both the AUDIT-C screening result and any indication of follow-up BI may provide data that is more in-depth, more accurate, and more complete for incorporation into STORM and other opioid surveillance tools.

Limitations and Future Directions

The current study has several strengths, including a large multi-year cohort, medical record data on screening and BI and a cascade of opioid-related outcomes, and temporal ordering and analyses that provide some initial evidence of the directionality of effect. Nevertheless, several limitations are worth noting, and several future directions are immediately obvious to increase the scope of evidence and understanding of BI and opioid-related outcomes. First, the current study did not examine long-term opioid therapy (LTOT), which is a common measure in the cascade of opioid-related outcomes. Future research may benefit from longer follow-up timeframes to examine this outcome, typically defined as ≥90 continuous days of opioid therapy.51 Second, all data were collected from the VA’s EHR, and there may be meaningful non-VA utilization not captured. Future research incorporating Centers for Medicare and Medicaid Services data may establish a more complete picture external to VA services utilization. Third, despite an initial cohort of almost half a million patients, the strict temporal ordering, extensive covariate adjustments, and rarity of some outcomes may have limited the ability for finding meaningful effects to reach statistical significance. This can be seen by the fact that the 95% confidence interval in adjusted analyses predicting opioid-related hospitalizations contained 1.00, despite the high clinical meaningfulness of a 19% increase in the odds of a 1-year opioid-related hospitalization. Finally, the current data did not examine other interventions that may have played a mechanistic role in the associations found, including referral to SUD treatment. While BI has consistently been found to be less effective at producing successful referrals to SUD treatment,52 it is still unclear if other specialized SUD treatments occurred beyond the BI, including those outside of VA (e.g., Alcoholics Anonymous, Narcotics Anonymous). Future research capturing the full spectrum of services utilization may provide insights into what additional substance use services may influence opioid-related outcomes.

Regarding additional future directions, first extending the current investigation to a nationally representative sample across all VA medical centers would test the generalizability of results and provide greater power for rarer outcomes such as opioid-related hospitalizations. Second, additional randomized or mixed-methods study designs could be leveraged to begin testing some of the proposed mechanisms underlying the observed association between BI and opioid-related outcomes, as briefly noted above. Finally, predictive modelling of improvements to VA’s current STORM tool are already underway and may benefit from the inclusion of alcohol-related screening and BI information. Altogether, each of these future research avenues would inform future intervention development, tailoring, targeting, and implementation that maximizes benefits across both alcohol and opioid-related harms.

Conclusion

The current study demonstrated that absence of clinically indicated alcohol-related BI is associated with increased odds of future opioid-related outcomes. While additional research needs to validate and examine mechanisms behind these associations, there are immediate clinical implications for the importance of receiving alcohol-related BI and potential additional routes of opioid-related risk mitigation. Polysubstance use is a prevalent phenomenon that increases morbidity and mortality. Results suggest that even the most common and brief single substance-related interventions may be leveraged to provide benefits beyond those intended in polysubstance using populations, reducing the likelihood of deleterious outcomes on a broader scale.

Supplementary Material

supplement

Acknowledgements:

Dr. Blalock was supported by Career Development Award 19-035 (IK2HX003085-01A2) from the United States Department of Veterans Affairs Health Services Research and Development (HSR&D) Service. Dr. Bachrach was supported by Career Development Award 20-057 (K2HX003087) from the United States Department of Veterans Affairs HSR&D. This work was also supported by grant #6754-SP SUB #21 P3630024 from the Duke University Endowment, in coordination with the Duke Opioid Collaboratory, and award CIN 13-410 from the Center of Innovation to Accelerate Discovery and Practice Transformation at the Durham VA Health Care System.

Footnotes

Disclosures: Dr. Blalock receives consulting fees from Eating Recovery Center. All other authors have no disclosures.

References

  • 1.Hassan AN, Le Foll B. Polydrug use disorders in individuals with opioid use disorder. Drug Alcohol Depend. 2019;198:28–33. doi: 10.1016/j.drugalcdep.2019.01.031 [DOI] [PubMed] [Google Scholar]
  • 2.Saha TD, Grant BF, Chou SP, et al. Concurrent use of alcohol with other drugs and DSM-5 alcohol use disorder comorbid with other drug use disorders: Sociodemographic characteristics, severity, and psychopathology. Drug Alcohol Depend. 2018;187:261–269. doi: 10.1016/j.drugalcdep.2018.03.006 [DOI] [PubMed] [Google Scholar]
  • 3.Esser MB, Guy GP Jr, Zhang K, et al. Binge Drinking and Prescription Opioid Misuse in the U.S., 2012–2014. Am J Prev Med. 2019;57(2):197–208. doi: 10.1016/j.amepre.2019.02.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Jones CM, Paulozzi LJ, Mack KA. Centers for Disease Control and Prevention (CDC). Alcohol involvement in opioid pain reliever and benzodiazepine drug abuse-related emergency department visits and drug-related deaths - United States, 2010. MMWR Morb Mortal Wkly Rep. 2014;63(40):881–885. [PMC free article] [PubMed] [Google Scholar]
  • 5.Tori ME, Larochelle MR, Naimi TS. Alcohol or Benzodiazepine Co-involvement With Opioid Overdose Deaths in the United States, 1999–2017. JAMA Netw Open. 2020;3(4):e202361. doi: 10.1001/jamanetworkopen.2020.2361 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Saitz R Clinical practice. Unhealthy alcohol use. N Engl J Med. 2005;352(6):596–607. doi: 10.1056/NEJMcp042262 [DOI] [PubMed] [Google Scholar]
  • 7.FDA Drug Safety Communication: FDA warns about serious risks and death when combining opioid pain or cough medicines with benzodiazepines; requires its strongest warning. U.S. Food and Drug Administration. Updated September 20, 2017. Accessed January 1, 2023. https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-fda-warns-about-serious-risks-and-death-when-combining-opioid-pain-or [Google Scholar]
  • 8.Gudin JA, Mogali S, Jones JD, et al. Risks, management, and monitoring of combination opioid, benzodiazepines, and/or alcohol use. Postgrad Med. 2013;125(4):115–130. doi: 10.3810/pgm.2013.07.2684 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Kessler RC, Demler O, Frank RG, et al. Prevalence and treatment of mental disorders, 1990 to 2003. N Engl J Med. 2005;352(24):2515–2523. doi: 10.1056/NEJMsa043266 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Lan CW, Fiellin DA, Barry DT, et al. The epidemiology of substance use disorders in US Veterans: A systematic review and analysis of assessment methods. Am J Addict. 2016;25(1):7–24. doi: 10.1111/ajad.12319 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Department of Veterans Affairs. Healthcare inspection: VA patterns of dispensing take-home opioids and monitoring patients on opioid therapy. Department of Veterans Affairs: Washington, DC, USA. 2014. Jul 14. [Google Scholar]
  • 12.Bennett AS, Elliott L, Golub A. Veterans’ Health and Opioid Safety-Contexts, Risks, and Outreach Implications. Fed Pract. 2015;32(6):4–7. doi: 10.3109/10826084.2013.796991 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Lin LA, Bohnert A, Blow FC, et al. Polysubstance use and association with opioid use disorder treatment in the US Veterans Health Administration. Addiction. 2020;116(1), 96–104. Doi: 10.1111/add.15116 [DOI] [PubMed] [Google Scholar]
  • 14.Bhalla IP, Stefanovics EA, Rosenheck RA. Clinical Epidemiology of Single Versus Multiple Substance Use Disorders: Polysubstance Use Disorder. Med Care. 2017;55 Suppl 9 Suppl 2:S24–S32. doi: 10.1097/MLR.0000000000000731 [DOI] [PubMed] [Google Scholar]
  • 15.Screening, Brief Intervention, and Referral to Treatment (SBIRT). Substance Abuse and Mental Health Services Administration. Updated August 12, 2022. Accessed January 1, 2023. https://www.samhsa.gov/sbirt
  • 16.Derges J, Kidger J, Fox F, et al. Alcohol screening and brief interventions for adults and young people in health and community-based settings: a qualitative systematic literature review. BMC Public Health. 2017;17(1):562. doi: 10.1186/s12889-017-4476-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Glass JE, Andréasson S, Bradley KA, et al. Rethinking alcohol interventions in health care: a thematic meeting of the International Network on Brief Interventions for Alcohol & Other Drugs (INEBRIA). Addict Sci Clin Pract. 2017;12(1):14. doi: 10.1186/s13722-017-0079-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Vendetti J, Gmyrek A, Damon D, et al. Screening, brief intervention and referral to treatment (SBIRT): implementation barriers, facilitators and model migration. Addiction. 2017;112 Suppl 2:23–33. doi: 10.1111/add.13652 [DOI] [PubMed] [Google Scholar]
  • 19.Williams EC, Rubinsky AD, Chavez LJ, et al. An early evaluation of implementation of brief intervention for unhealthy alcohol use in the US Veterans Health Administration. Addiction. 2014;109(9):1472–1481. doi: 10.1111/add.12600 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Loughran TA, Scharer JL, Rodriguez L, et al. Brief alcohol interventions in U.S. medical settings: A systematic review of the implementation literature. J Subst Abuse Treat. 2021;131:108456. doi: 10.1016/j.jsat.2021.108456 [DOI] [PubMed] [Google Scholar]
  • 21.Curry SJ, Krist AH, et al. Screening and Behavioral Counseling Interventions to Reduce Unhealthy Alcohol Use in Adolescents and Adults: US Preventive Services Task Force Recommendation Statement. JAMA. 2018;320(18):1899–1909. doi: 10.1001/jama.2018.16789 [DOI] [PubMed] [Google Scholar]
  • 22.O’Connor EA, Perdue LA, Senger CA, et al. Screening and Behavioral Counseling Interventions to Reduce Unhealthy Alcohol Use in Adolescents and Adults: Updated Evidence Report and Systematic Review for the US Preventive Services Task Force. JAMA. 2018;320(18):1910–1928. doi: 10.1001/jama.2018.12086 [DOI] [PubMed] [Google Scholar]
  • 23.Lapham GT, Achtmeyer CE, Williams EC, et al. Increased documented brief alcohol interventions with a performance measure and electronic decision support. Med Care. 2012;50(2):179–187. doi: 10.1097/MLR.0b013e3181e35743 [DOI] [PubMed] [Google Scholar]
  • 24.Henihan AM, McCombe G, Klimas J, et al. Feasibility of alcohol screening among patients receiving opioid treatment in primary care. BMC Fam Pract. 2016;17(1):153. doi: 10.1186/s12875-016-0548-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Alcohol screening and brief intervention for people who consume alcohol and use opioids. Center for Disease Control and Prevention. July 11, 2018. Accessed January 1, 2023. https://www.cdc.gov/alcohol/fact-sheets/alcohol-screening.html [Google Scholar]
  • 26.Lin LA, Bohnert ASB, Kerns RD, et al. Impact of the Opioid Safety Initiative on opioid-related prescribing in veterans. Pain. 2017;158(5):833–839. doi: 10.1097/j.pain.0000000000000837 [DOI] [PubMed] [Google Scholar]
  • 27.Minegishi T, Frakt A. Reducing Long-term Opioid Use in the Veterans Health Administration. J Gen Intern Med. 2018;33(6):781–782. doi: 10.1007/s11606-018-4352-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Gellad WF, Good CB, Shulkin DJ. Addressing the Opioid Epidemic in the United States: Lessons From the Department of Veterans Affairs. JAMA Intern Med. 2017;177(5):611–612. doi: 10.1001/jamainternmed.2017.0147 [DOI] [PubMed] [Google Scholar]
  • 29.Lavasani R, Chung M, Beatty A, et al. Opioid prescribing trends in a Veterans Affairs emergency department and dental clinic before and after implementation of opioid-prescribing guidelines. Ment Health Clin. 2020;10(5):270–274. doi: 10.9740/mhc.2020.09.270 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Oliva EM, Bowe T, Tavakoli S, et al. Development and applications of the Veterans Health Administration’s Stratification Tool for Opioid Risk Mitigation (STORM) to improve opioid safety and prevent overdose and suicide. Psychol Serv. 2017;14(1):34–49. doi: 10.1037/ser0000099 [DOI] [PubMed] [Google Scholar]
  • 31.Orford J, Hodgson R, Copello A, et al. The clients’ perspective on change during treatment for an alcohol problem: qualitative analysis of follow-up interviews in the UK alcohol treatment trial. Addiction. 2006. Jan;101(1):60–8. [DOI] [PubMed] [Google Scholar]
  • 32.Schepis TS, Hakes JK. Age of initiation, psychopathology, and other substance use are associated with time to use disorder diagnosis in persons using opioids nonmedically. Substance Abuse. 2017. Oct;38(4):407–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Bradley KA, Bush KR, Epler AJ, et al. Two brief alcohol-screening tests From the Alcohol Use Disorders Identification Test (AUDIT): validation in a female Veterans Affairs patient population. Arch Intern Med. 2003;163(7):821–829. doi: 10.1001/archinte.163.7.821 [DOI] [PubMed] [Google Scholar]
  • 34.Bush K, Kivlahan DR, McDonell MB, et al. The AUDIT alcohol consumption questions (AUDIT-C): an effective brief screening test for problem drinking. Ambulatory Care Quality Improvement Project (ACQUIP). Alcohol Use Disorders Identification Test. Arch Intern Med. 1998;158(16):1789–1795. doi: 10.1001/archinte.158.16.1789 [DOI] [PubMed] [Google Scholar]
  • 35.Lapham GT, Rubinsky AD, Shortreed SM, et al. Comparison of provider-documented and patient-reported brief intervention for unhealthy alcohol use in VA outpatients. Drug Alcohol Depend. 2015;153:159–166. doi: 10.1016/j.drugalcdep.2015.05.027 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Williams EC, Lapham GT, Shortreed SM, et al. Among patients with unhealthy alcohol use, those with HIV are less likely than those without to receive evidence-based alcohol-related care: A national VA study. Drug Alcohol Depend. 2017;174:113–120. doi: 10.1016/j.drugalcdep.2017.01.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.VA DoD Clinical Practice Guideline for the Management of Substance Use Disorders. https://www.healthquality.va.gov/guidelines/MH/sud/VADoDSUDCPG.pdf. Accessed November 22, 2023.
  • 38.Wagner T, Stefos T, Moran E, et al. Risk Adjustment: Guide to the V21 and Nosos Risk Score Programs. Technical Report 30. Health Economics Resource Center, U.S. Department of Veterans Affairs. February 2016. Accessed January 1, 2023. https://www.herc.research.va.gov/include/page.asp?id=technical-report-risk-adjustment. [Google Scholar]
  • 39.Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Statistics in Medicine. 2009;28:3083–3107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Yoshida K, Bohn J. Package ‘tableone’ for R. Accessed January 17, 2024. https://cran.r-project.org/web/packages/tableone/tableone.pdf
  • 41.Zale EL, Maisto SA, Ditre JW. Interrelations between pain and alcohol: An integrative review. Clinical Psychology Review. 2015;37:57–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Chikritzhs T, Livingston M. Alcohol and the Risk of Injury. Nutrients. 2021;13(8):2777. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Paltzer J, Brown RL, Burns M, et al. Substance Use Screening, Brief Intervention, and Referral to Treatment Among Medicaid Patients in Wisconsin: Impacts on Healthcare Utilization and Costs. J Behav Health Serv Res. 2017;44(1):102–112. doi: 10.1007/s11414-016-9510-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Frost MC, Glass JE, Bradley KA, et al. Documented brief intervention associated with reduced linkage to specialty addictions treatment in a national sample of VA patients with unhealthy alcohol use with and without alcohol use disorders. Addiction. 2020;115(4):668–678. doi: 10.1111/add.14836 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Harris B. Screening, brief intervention, and referral to treatment for substance use: A practitioner’s guide: The role of public policy in advancing SBIRT implementation and dissemination. (pp. 267–283). American Psychological Association. doi: 10.1037/0000199-015 [DOI] [Google Scholar]
  • 46.Williams EC, Achtmeyer CE, Young JP, et al. Local implementation of alcohol screening and brief intervention at five Veterans Health Administration primary care clinics: perspectives of clinical and administrative staff. Journal of substance abuse treatment. 2016;60:27–35. [DOI] [PubMed] [Google Scholar]
  • 47.Knox J, Hasin DS, Larson FR, et al. Prevention, screening, and treatment for heavy drinking and alcohol use disorder. The Lancet Psychiatry. 2019;6(12):1054–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Xu KY, Presnall N, Mintz CM, et al. Association of Opioid Use Disorder Treatment With Alcohol-Related Acute Events. JAMA Netw Open. 2021;4(2):e210061. doi: 10.1001/jamanetworkopen.2021.0061 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Sandbrink F, Oliva EM, McMullen TL, et al. Opioid Prescribing and Opioid Risk Mitigation Strategies in the Veterans Health Administration. J Gen Intern Med. 2020;35(Suppl 3):927–934. doi: 10.1007/s11606-020-06258-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Williams EC, Fletcher OV, Frost MC, et al. Comparison of Substance Use Disorder Diagnosis Rates From Electronic Health Record Data With Substance Use Disorder Prevalence Rates Reported in Surveys Across Sociodemographic Groups in the Veterans Health Administration. JAMA Netw Open. 2022;5(6):e2219651. doi: 10.1001/jamanetworkopen.2022.19651 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Karmali RN, Bush C, Raman SR, Campbell CI, Skinner AC, Roberts AW. Long-term opioid therapy definitions and predictors: A systematic review. Pharmacoepidemiol Drug Saf. 2020;29(3):252–269. doi: 10.1002/pds.4929 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Glass JE, Hamilton AM, Powell BJ, et al. Specialty substance use disorder services following brief alcohol intervention: a meta-analysis of randomized controlled trials. Addiction. 2015;110(9):1404–15. [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

supplement
NIHMS1987828-supplement-supplement.docx (59.5KB, docx)

RESOURCES