Advertisement

Young Adult Substance Use and Healthcare Use Associated With Screening, Brief Intervention and Referral to Treatment in Pediatric Primary Care

      Abstract

      Purpose

      Screening, brief intervention, and referral to treatment (SBIRT) may impact future comorbidity and healthcare utilization among adolescents screening positive for substance use or mood problems.

      Methods

      In a randomized trial sample, we compared an SBIRT group to usual care for substance use, mental health, medical diagnoses, and healthcare utilization over 7 years postscreening.

      Results

      In logistic regression models adjusting for patient characteristics, the SBIRT group had lower odds of any substance (Odds Ratio[OR] = 0.80, 95% Confidence Interval [CI] = 0.66–.98), alcohol (OR = 0.69, 95% CI = 0.51–0.94), any drug (OR = 0.73, 95% CI = 0.54–0.98), marijuana (OR = 0.70, 95% CI = 0.50–0.98), and tobacco (OR = 0.83, 95% CI = 0.69–1.00) diagnoses, and lower odds of any inpatient hospitalizations (OR = 0.59, 95% CI = 0.41–0.85) compared with usual care. Negative binomial models examining number of visits among adolescents with at least one visit of that type found that those in the SBIRT group had fewer primary care (incidence rate ratio[iRR] = 0.90, p < .05) and psychiatry (iRR = 0.64, p < .01) and more addiction medicine (iRR = 1.52, p < .01) visits over 7 years compared with usual care. In posthoc analyses, we found that among Hispanic patients, those in the SBIRT group had lower odds of any substance, any drug and marijuana use disorder diagnoses compared with usual care, and among Black/African American patients, those in the SBIRT group had lower odds of alcohol use disorder diagnoses compared with usual care.

      Discussion

      Beneficial effects of adolescent SBIRT on substance use and healthcare utilization may persist into young adulthood.

      Keywords

      Implications and Contribution
      This study of the long-term outcomes among adolescent patients in an SBIRT trial in pediatric primary care found significantly fewer substance use disorders and less healthcare utilization at 7 years among those exposed to SBIRT. This suggests that SBIRT may provide significant and lasting benefits to vulnerable adolescents well into young adulthood.
      Adolescent substance use and mental health problems are a significant public health concern and exact a considerable toll on individuals, and their families and communities. They often result in serious academic [
      • DuPont R.L.
      • Caldeira K.M.
      • Dupont H.S.
      • et al.
      America’s dropout crisis: The unrecognized connection to adolescent substance use.
      ,
      • Wickersham A.
      • Sugg H.V.R.
      • Epstein S.
      • et al.
      Systematic review and meta-analysis: The association between child and adolescent depression and later educational attainment.
      ], employment [
      • Hale D.R.
      • Bevilacqua L.
      • Viner R.M.
      Adolescent health and adult education and employment: A systematic review.
      ], legal [
      • Colins O.
      • Vermeiren R.
      • Vreugdenhil C.
      • et al.
      Psychiatric disorders in detained male adolescents: A systematic literature review.
      ] consequences for young people, and high societal costs (estimated at over half a trillion dollars annually in the U. S.) []. Research shows that adolescent substance use and mental health problems are associated with adult substance use disorders, and psychiatric and chronic medical conditions [
      • Vida R.
      • Brownlie E.B.
      • Beitchman J.H.
      • et al.
      Emerging adult outcomes of adolescent psychiatric and substance use disorders.
      ]. Early intervention during adolescence is effective in helping young people avoid future problems.
      While adolescent substance use has decreased in recent decades, it has increased for certain drugs in recent years—specifically, marijuana, vaping, and opioids. The gradual decreases in alcohol use and binge drinking have leveled off, while amphetamine, cough syrup, and inhalant use among younger adolescents has increased [
      • Johnston L.D.
      • Miech R.A.
      • O’Malley P.M.
      • et al.
      Monitoring the future national survey results on drug Use: 1975-2019: Overview: Key findings on adolescent drug use.
      ]. According to a recent national survey, in the U.S., over 400,000 adolescents and 3,400,000 young adults have alcohol use disorders, and over 600,000 and 2,600,000, respectively, have a drug use disorder, with many more reporting problematic use [
      Substance Abuse and Mental Health Services Administration
      Key substance use and mental health indicators in the United States: Results from the 2018 national survey on drug Use and health.
      ,
      • Torralba E.
      Nearly half of California adolescents report mental health difficulties.
      ]. Recent research also shows increased nonfatal drug overdoses in 59 jurisdictions among 47 states for youth aged 0–14 years old [
      • Roehler D.R.
      • Olsen E.O.
      • Mustaquim D.
      • et al.
      Suspected nonfatal drug-related overdoses among youth in the US: 2016-2019.
      ].
      Equally concerning is that fewer than half of adolescents needing treatment for substance use or mental health problems receive it [
      • Merikangas K.R.
      • He J.P.
      • Burstein M.
      • et al.
      Service utilization for lifetime mental disorders in U.S. Adolescents: Results of the national comorbidity survey-adolescent supplement (NCS-A).
      ]. Specialty treatment rates are particularly low for substance use and depressive disorders, and treatment initiation rates are particularly low among youth of color, who face many logistical, financial, and historical/cultural barriers to seeking specialty treatment [
      • Alegria M.
      • Carson N.J.
      • Goncalves M.
      • et al.
      Disparities in treatment for substance use disorders and co-occurring disorders for ethnic/racial minority youth.
      ]. Primary care is less stigmatizing than specialty addiction or psychiatry treatment [
      • Radez J.
      • Reardon T.
      • Creswell C.
      • et al.
      Why do children and adolescents (not) seek and access professional help for their mental health problems? A systematic review of quantitative and qualitative studies.
      ], and most adolescents, both privately and publicly insured, have access to primary care [
      National Committee for Quality Assurance (NCQA)
      Children and adolescents’ access to primary care Practitioners (CAP).
      ], making it an opportune setting for preventing and intervening early in behavioral health problems and for laying the groundwork for substance use behaviors—healthy or problematic—in future years. Screening, brief intervention, and referral to treatment (SBIRT) are delivered to U.S. adolescents most frequently in pediatric primary care settings as a public health approach to preventing substance use and mental health symptoms [
      • Sterling S.
      • Kline-Simon A.H.
      • Weisner C.
      • et al.
      Pediatrician and behavioral clinician-delivered screening, brief intervention and referral to treatment: Substance use and depression outcomes.
      ,
      • Tanner-Smith E.E.
      • Lipsey M.W.
      Brief alcohol interventions for adolescents and young adults: A systematic review and meta-analysis.
      ]. There is a growing evidence base on the efficacy and effectiveness of SBIRT for adolescents in healthcare settings. A pilot in Brazilian pediatric primary care clinics found that those who received brief interventions were less likely to report intention to use cannabis, and reported using it less often compared to usual care [
      • D'Amico E.J.
      • Miles J.N.
      • Stern S.A.
      • et al.
      Brief motivational interviewing for teens at risk of substance use consequences: A randomized pilot study in a primary care clinic.
      ]. In a quasi-experimental study of computer-facilitated screening and pediatrician-delivered brief advice, conducted in the U.S. and Czech Republic primary care clinics, Harris et al. found less alcohol use and any substance use, at 3 and 12 months among U.S. intervention group adolescents compared to controls, and less cannabis use at 3 and 12 months among the Czech intervention arm adolescents, compared to controls [
      • Harris S.K.
      • Csemy L.
      • Sherritt L.
      • et al.
      Computer-facilitated substance use screening and brief advice for teens in primary care: An international trial.
      ]. A randomized trial examining therapist and computer-delivered BIs, conducted with cannabis-naïve adolescents in several Federally Qualified Health Centers (FQHC), found lower rates of cannabis and other drug use, lower rates of delinquent behavior, and lower alcohol use severity, compared to usual care [
      • Walton M.A.
      • Resko S.
      • Barry K.L.
      • et al.
      A randomized controlled trial testing the efficacy of a brief cannabis universal prevention program among adolescents in primary care.
      ]. A similar randomized clinical trial, with adolescents already using cannabis, found significantly less other drug use, cannabis-related consequences, and occurrences of cannabis-related driving under the influence among those receiving the interventions compared to usual care [
      • Walton M.A.
      • Bohnert K.
      • Resko S.
      • et al.
      Computer and therapist based brief interventions among cannabis-using adolescents presenting to primary care: One year outcomes.
      ]. Other studies using nonphysicians to deliver screening and brief intervention to adolescents have also shown promising results [
      • Burke P.J.
      • Da Silva J.D.
      • Vaughan B.L.
      • et al.
      Training high school counselors on the use of motivational interviewing to screen for substance abuse.
      ,
      • Gil A.G.
      • Wagner E.F.
      • Tubman J.G.
      Culturally sensitive substance abuse intervention for Hispanic and African American adolescents: Empirical examples from the alcohol treatment Targeting adolescents in need (ATTAIN) Project.
      ,
      • Grenard J.L.
      • Ames S.L.
      • Wiers R.W.
      • et al.
      Brief intervention for substance use among at-risk adolescents: A pilot study.
      ,
      • Martin G.
      • Copeland J.
      • Swift W.
      The adolescent cannabis check-up: Feasibility of a brief intervention for young cannabis users.
      ,
      • McCambridge J.
      • Strang J.
      The efficacy of single-session motivational interviewing in reducing drug consumption and perceptions of drug-related risk and harm among young people: Results from a multi-site cluster randomized trial.
      ,
      • Srisurapanont M.
      • Sombatmai S.
      • Boripuntakul T.
      Brief intervention for students with methamphetamine use disorders: A randomized controlled trial.
      ,
      • Winters K.C.
      • Leitten W.
      • Wagner E.
      • et al.
      Use of brief interventions for drug abusing teenagers within a middle and high school setting.
      ]. There are also several systematic reviews and meta-analyses, which have found that brief interventions for substance use in adolescents, in healthcare and nonhealthcare settings alike, can be effective in reducing use and related consequences [
      • Tanner-Smith E.E.
      • Lipsey M.W.
      Brief alcohol interventions for adolescents and young adults: A systematic review and meta-analysis.
      ,
      • Das J.K.
      • Salam R.A.
      • Arshad A.
      • et al.
      Interventions for adolescent substance abuse: An overview of systematic reviews.
      ,
      • Mitchell S.G.
      • Gryczynski J.
      • O'Grady K.E.
      • et al.
      SBIRT for adolescent drug and alcohol use: Current status and future directions.
      ,
      • Steinka-Fry K.T.
      • Tanner-Smith E.E.
      • Hennessy E.A.
      Effects of brief alcohol interventions on drinking and driving among youth: A systematic review and meta-analysis.
      ,
      • Tanner-Smith E.E.
      • Risser M.D.
      A meta-analysis of brief alcohol interventions for adolescents and young adults: Variability in effects across alcohol measures.
      ,
      • Tanner-Smith E.E.
      • Steinka-Fry K.T.
      • Hennessy E.A.
      • et al.
      Can brief alcohol interventions for youth also address concurrent illicit drug use? Results from a meta-analysis.
      ,
      • Carney T.
      • Myers B.
      Effectiveness of early interventions for substance-using adolescents: Findings from a systematic review and meta-analysis.
      ].
      Although these findings are encouraging, there is still much research needed on SBI/SBIRT in primary care to broaden the evidence on the basis of its efficacy and effectiveness among adolescents. Moreover, most research has focused on short-term outcomes [
      • Harris S.K.
      • Csemy L.
      • Sherritt L.
      • et al.
      Computer-facilitated substance use screening and brief advice for teens in primary care: An international trial.
      ,
      • Steele D.W.
      • Becker S.J.
      • Danko K.J.
      • et al.
      Brief behavioral interventions for substance use in adolescents: A meta-analysis.
      ], and to our knowledge, no studies have examined the effectiveness of a range of important outcomes as patients progress into young adulthood.
      This study used electronic health record (EHR) data to examine substance use, mental health, medical and healthcare utilization outcomes over 7 years among adolescents from a randomized clinical trial [
      • Parthasarathy S.
      • Kline-Simon A.H.
      • Jones A.
      • et al.
      Three-year outcomes after brief treatment of substance use and mood symptoms.
      ,
      • Sterling S.
      • Kline-Simon A.H.
      • Jones A.
      • et al.
      Health care use over 3 years after adolescent SBIRT.
      ] that compared modalities of delivering SBIRT in pediatric primary care with usual care. We hypothesized that patients exposed to SBIRT would have lower rates of substance use, mental health, and medical comorbidities, and lower rates of ED and inpatient utilization than those receiving usual care.

      Method

      Setting and design

      The study was conducted at Kaiser Permanente Northern California (KPNC), an integrated healthcare delivery system serving a diverse population of over four million members, with “carved-in” mental health and addiction medicine services available to members as covered benefits. The study sample draws from a pragmatic, randomized hybrid clinical trial at KPNC’s Oakland Pediatrics Department, which assessed three modalities for delivering SBIRT to adolescents ages 12–18 years. Clinic pediatricians were randomized to one of three trial arms, including two brief intervention (BI) arms: (a) pediatrician-only arm (pediatricians were trained to assess patients following the screening, and where indicated, deliver BIs and refer to specialty substance use or mental health treatment); (b) embedded behavioral clinician arm (pediatricians assessed patients, and if needed, referred them to an embedded behavioral clinician for further assessment, BI, and referral to treatment); and (c) a usual care arm (pediatricians had EHR screening tools but no formal SBIRT training). Following screening for any past year alcohol and drug use (Y/N) and past two-week mood symptoms, conducted as part of a comprehensive health screening instrument administered at all adolescent well visits, adolescents who endorsed alcohol or drug use or mood symptoms were eligible for a brief intervention from their pediatrician or the behavioral health clinician, depending on the intervention arm to which their pediatrician was randomized, or treatment as usual. The trial compared implementation and patient outcomes of the three modalities delivered during routine pediatric primary care to adolescents at wellness visits. Trial details, including the pediatrician and embedded clinician training protocols, screening and assessment items, and clinical workflows, have been previously described elsewhere [
      • Sterling S.
      • Kline-Simon A.H.
      • Weisner C.
      • et al.
      Pediatrician and behavioral clinician-delivered screening, brief intervention and referral to treatment: Substance use and depression outcomes.
      ,
      • Sterling S.
      • Kline-Simon A.H.
      • Jones A.
      • et al.
      Specialty addiction and psychiatry treatment initiation and engagement: Results from an SBIRT randomized trial in pediatrics.
      ,
      • Sterling S.
      • Kline-Simon A.H.
      • Satre D.D.
      • et al.
      Implementation of screening, brief intervention, and referral to treatment for adolescents in pediatric primary care: A cluster randomized trial.
      ]. Studies of outcomes at 1 and 3 years found that SBIRT was associated with reduced substance use, mental health, medical diagnoses, and mental healthcare utilization, and similar results among adolescents endorsing both substance use and mood symptoms at the initial screening [
      • Parthasarathy S.
      • Kline-Simon A.H.
      • Jones A.
      • et al.
      Three-year outcomes after brief treatment of substance use and mood symptoms.
      ,
      • Sterling S.
      • Kline-Simon A.H.
      • Jones A.
      • et al.
      Health care use over 3 years after adolescent SBIRT.
      ]. The study was approved by the IRBs of KPNC and the University of California, San Francisco.
      Although the original trial compared pediatrician-delivered and behavioral health clinician-delivered SBIRT to each other and usual care, in real-world practice, many primary care practices have adopted flexible, hybrid SBIRT staffing and workflows, using both SBIRT-trained pediatricians and nonphysician behavioral clinicians to deliver SBIRT [
      • Levy S.J.
      • Williams J.F.
      Committee on Substance USE
      Substance use screening, brief intervention, and referral to treatment.
      ]. In the trial, we found that both SBIRT modalities produced better patient outcomes than usual care. In the current study, we have the opportunity to examine the effects of access to SBIRT, regardless of the clinician providing it, on long-term outcomes, and thus, we combined patients from both SBIRT intervention arms into a single SBIRT group.

      Study sample

      This study sample consists of a subset of adolescents aged 12–18 who endorsed either substance use or mental health symptoms when screened during an adolescent wellness visit between November 1, 2011 and October 31, 2013, and/or were deemed eligible for further assessment, BI and/or referral to specialty treatment by their pediatrician (n = 1,871). The index date was the date on which the adolescent screened positive for substance use and/or mood symptoms. Additional details of the study protocol and the screening instrument are described elsewhere [
      • Sterling S.
      • Kline-Simon A.H.
      • Weisner C.
      • et al.
      Pediatrician and behavioral clinician-delivered screening, brief intervention and referral to treatment: Substance use and depression outcomes.
      ,
      • Sterling S.
      • Kline-Simon A.H.
      • Satre D.D.
      • et al.
      Implementation of screening, brief intervention, and referral to treatment for adolescents in pediatric primary care: A cluster randomized trial.
      ].

      Measures

      Treatment groups

      A dichotomous indicator was created to determine patients in the SBIRT arm (pediatrician-delivered or embedded behavioral clinician-delivered) and usual care (=1 if SBIRT group; 0 if usual care).

      Patient characteristics

      The EHR was used to extract patient sex, age at index date, race/ethnicity (White, Black/African American, Hispanic, Asian, and other/unknown), Medicaid coverage in the year prior to index (1 = any Medicaid coverage, 0 otherwise) as a proxy for socioeconomic status, and length of enrollment calculated as the number of member months in the health plan over the 7 years postindex.

      Health services utilization

      The EHR provided all outpatient and inpatient services use over 7 years postindex, which included utilization within Kaiser and utilization outside of KPNC paid for by the health plan. Indicator and count measures were created for each type of utilization (primary care, psychiatry, addiction medicine, ED, and inpatient utilization).

      Diagnoses

      We examined diagnoses recorded in the EHR during patient visits over 7 years postindex. Specifically, we looked at any behavioral health diagnoses, which included any mental health or substance use diagnoses (ICD-9 codes: 290, 293–302, 306–319; ICD-10 codes: F01–F09, F20–F99) and separate indicators for depression (ICD-9: 296.2, 296.3, 296.82, 298.0, 300.4, 301.12, 309.0, 309.1, 309.28, 311; ICD-10: F32, F33, F34.1, F43.21, F43.23); anxiety (ICD-9: 300.00, 300.02, 300.09, 300.2, 300.3, 308.3, 309.21, 309.24, 309.81, F40, F41.1, F41.3, F41.8, F41.9, F43.22, F43.0, F43.1); and substance use disorders (ICD-9: 291, 292, 303–305; ICD-10: F10–F19). Substance use disorders were disaggregated to alcohol (ICD-9: 291, 303, 305.0; ICD-10: F10), any drug (ICD-9: 292, 304, 305.2, 305.3, 305.4, 305.5, 305.6, 305.7, 305.8, 305.9; ICD-10: F11–F17, F18–F19), marijuana (ICD-9: 304.3, 305.2; ICD-10: F12), tobacco (ICD-9: 305.1; ICD-10: F17) and opioid (ICD-9: 304.0, 304.7, 305.5; ICD-10: F11) diagnoses. The Charlson comorbidity score was calculated based on diagnosis codes made in the year prior to index, and used to account for medical comorbidity and disease severity where a higher score is associated with worse outcomes (ranging from 0-6) [
      • Charlson M.E.
      • Charlson R.E.
      • Peterson J.C.
      • et al.
      The Charlson comorbidity index is adapted to predict costs of chronic disease in primary care patients.
      ].

      Statistical analysis

      We used chi-squared and t-tests to examine differences between categorical and continuous measures, respectively, and the SBIRT and usual care groups. Measures included demographic characteristics (age, sex, race/ethnicity), substance use and mental health diagnoses, medical comorbidity and primary care, psychiatry, addiction medicine, ED, and inpatient utilization. Multilevel multivariate logistic regression models with patients nested within providers were used to examine the dichotomous outcomes (e.g., any behavioral health diagnoses, any inpatient utilization), and multilevel negative binomial regression models were used to examine Charlson score and visit counts. We examined visit counts among a subsample of patients with at least one visit of the corresponding type of utilization (e.g., number of psychiatry visits was examined among patients with at least one psychiatry visit). In these models, the exponent of the coefficient for the treatment variable represents the odds ratio (OR) and incidence rate ratio (iRR) for the logistic and negative binomial regressions, respectively, for the SBIRT group relative to usual care. Models adjusted for patient age, sex, race/ethnicity, Medicaid insurance coverage in the year prior to index, length of membership in months, and an indicator for the presence of the corresponding comorbidity or utilization in the year prior to index (e.g., an indicator for inpatient utilization in the year prior to index was include in models examining inpatient utilization). We conducted posthoc analyses examining all mental health and substance use diagnosis outcome models stratifying by race/ethnicity, and separately, by sex. Models were run as described above, excluding the indicator for Medicaid insurance, which had to be excluded due to sample size issues for some of the race/ethnicity categories. All analyses were performed using SAS© 9.3 (SAS Institute Inc., Cary, NC).

      Results

      Among the 1,871 adolescents, 55.7% were female, the average age was 15.8 years, 6.4% had Medicaid insurance coverage in the year prior to the index, 23.8% were Hispanic, 33.5% were Black/African American, 11.3% were Asian, 25.2% were White, and 6.1% had other/unknown race/ethnicity. The SBIRT group had more females (59.5% vs. 48.1%, p < .001), more Black/African American (35.6% vs. 29.2%) and Hispanic (24.9% vs. 21.6%) patients and fewer White patients (22.1% vs. 31.7%; p < .001). Age and insurance type did not differ between groups. Membership retention during the 7-year study period was quite high, with the average length of enrollment being 69.7 months (SD = 23.0 months); 55% had complete membership across all 7 years. Membership did not significantly differ between groups (SBIRT group, mean (SD) = 69.1 (23.6); usual care, 71.1 (21.7); p > .05) (Table A1).

      Bivariate comparisons of substance use and mental health diagnoses and healthcare utilization by the group over 7 years

      Fewer patients in the SBIRT group had anxiety (4.1% vs. 6.5%; <.05), alcohol- (.4% vs. 1.8%; <.01), and marijuana-related (1.2% vs. 2.8%; p < .05) diagnoses in the year prior to index compared with usual care; other diagnoses and types of utilization did not differ in the year prior to index (see Table A2).
      Over the 7-year period, the SBIRT group had fewer substance use (19.0% vs. 24.0%, p < .05), alcohol- (4.8% vs. 7.8%, p < .01), and any drug use-related (10.6% vs. 13.8%, p < .05) diagnoses compared to usual care. Mental health diagnoses and average Charlson score did not differ between groups. Eight percent of the sample had visits outside of the health plan, which were covered and captured by Kaiser. The SBIRT group had a higher prevalence of psychiatry visits (47.4% vs. 39.9%, p < .01), and fewer patients in the SBIRT group had addiction medicine visits (4.9% vs. 7.8%, p < .05) or inpatient hospitalizations (13.3% vs. 18.2%, p < .01), over the 7-year period compared to usual care. Primary care and ED utilization did not differ between groups. Among those who had at least one visit of that type, the average number of psychiatry visits (mean[SD] = 10.3[17.4] SBIRT group, 18.2[30.8] usual care, p < .01) was lower in the SBIRT group compared with usual care; the number of inpatient hospitalizations, addiction medicine visits, primary care, and ED visits did not differ between groups (Table A2).

      Multivariate models

      Substance Use and mental health diagnoses

      The SBIRT group had lower odds of having any substance use, alcohol-, any drug-, marijuana-, and tobacco-related diagnoses compared with usual care. The SBIRT group also had lower Charlson Comorbidity scores at 7 years compared with usual care. Mental health diagnoses did not differ between the SBIRT and usual care groups.
      In these same logistic regression models, females had higher odds of any behavioral health, depression, and anxiety disorders and lower odds of any substance use and tobacco-related diagnoses compared with males; there were no significant differences by sex in alcohol, any drug, or marijuana-related diagnoses. Older patients had lower odds of any behavioral health, depression, and anxiety diagnoses; substance use disorders did not differ across age with the exception of tobacco use, whereas older patients had higher odds of a tobacco diagnosis. Compared with White patients, Asians had lower odds of any behavioral health, alcohol- and tobacco-related diagnoses. Black/African Americans had lower odds of having any behavioral health, depression, anxiety, and alcohol-related diagnoses, but higher odds of any substance use and marijuana-related diagnoses compared with White patients. Hispanic patients had lower odds of depression and anxiety disorders and higher odds of any substance use- and alcohol-related diagnoses, and higher Charlson scores at 7 years compared with White patients. Medicaid insurance coverage was not significant in any of the models (Table 1).
      Table 1Multilevel logistic/negative binomial regression analyses examining mental health and substance use diagnoses and medical comorbidity over 7 years postindex among all eligible patients (n = 1,871)
      Any behavioral health diagnosisDepression disorder diagnosisAnxiety disorder diagnosisSubstance use disorder diagnosisCharlson comorbidity index
      ORp-valueORp-valueORp-valueORp-valueiRRp-value
      Intercept6.57 (2.10–20.60).0013.58 (1.22–10.51).0210.83 (0.23–2.95).7680.06 (0.02–0.19)<.0010.04 (0.01–0.24)<.001
      SBIRT Group (vs. Usual Care)1.02 (0.87–1.19).8040.83 (0.61–1.12).2151.04 (0.86–1.25).7000.80 (0.66–0.98).0270.76 (0.59–0.98).032
      Female (vs. Male)1.78 (1.41–2.25)<.0011.93 (1.54–2.44)<.0012.29 (1.86–2.82)<.0010.69 (0.56–0.86).0011.15 (0.87–1.50).326
      Age0.82 (0.77–0.87)<.0010.81 (0.76–0.86).0000.86 (0.81–0.91)<.0011.05 (0.98–1.13).1931.01 (0.95–1.09).683
      Race/Ethnicity (reference: White)
       Asian0.66 (0.44–0.98).0420.82 (0.52–1.31).4090.71 (0.47–1.07).1030.61 (0.35–1.06).0820.97 (0.53–1.78).923
       Black/African American0.56 (0.42–0.75)<.0010.65 (0.46–0.93).0180.46 (0.34–0.61)<.0011.46 (1.15–1.85).0021.49 (0.98–2.26).059
       Hispanic0.80 (0.59–1.08).1460.65 (0.47–0.90).0100.71 (0.53–0.94).0181.16 (0.92–1.46).2051.64 (1.12–2.40).010
       Missing/Unknown0.51 (0.34–0.77).0010.56 (0.34–0.90).0170.68 (0.36–1.27).2261.04 (0.59–1.83).8991.27 (0.70–2.28).432
      Medicaid Insurance
      Any Medicaid coverage in the year prior to index.
      (vs. No Medicaid Insurance)
      1.20 (0.76–1.89).4300.91 (0.63–1.33).6401.35 (0.90–2.02).1440.76 (0.43–1.32).3281.43 (0.86–2.36).169
      Membership months1.02 (1.01–1.02)<.0011.01 (1.01–1.02)<.0011.02 (1.01–1.03)<.0011.01 (1.01–1.02)<.0011.01 (0.99–1.04).169
      Diagnosis in year prior
      Corresponding diagnoses/score made in the year prior to index.
      8.61 (5.70–13.02)<.0018.44 (5.38–13.24)<.00110.96 (6.25–19.24)<.0014.31 (2.36–7.90)<.0013.21 (2.56–4.02)<.001
      Alcohol disorder diagnosisAny drug disorder diagnosisMarijuana disorder diagnosisTobacco disorder diagnosis
      ORp-valueORp-valueORp-valueORp-value
      Intercept0.02 (0.00–0.13)<.0010.05 (0.01–0.37).0030.08 (0.01–0.74).0260.01 (0.00–0.03)<.001
      SBIRT Group (vs. Usual Care)0.69 (0.51–0.94).0170.73 (0.54–0.98).0360.70 (0.50–0.98).0370.83 (0.69–1.00).046
      Female (vs. Male)0.68 (0.44–1.05).0790.77 (0.56–1.05).1030.78 (0.57–1.09).1450.63 (0.48–0.83).001
      Age1.05 (0.93–1.18).4261.01 (0.90–1.13).8630.99 (0.87–1.12).8191.13 (1.06–1.21)<.001
      Race/Ethnicity (reference: White)
       Asian0.40 (0.16–0.98).0450.95 (0.43–2.10).9020.81 (0.33–2.01).6570.59 (0.36–0.97).038
       Black/African American0.52 (0.28–0.96).0381.94 (1.44–2.62)<.0011.79 (1.21–2.65).0031.26 (0.90–1.75).174
       Hispanic1.60 (1.06–2.41).0271.70 (1.06–2.71).0261.52 (0.95–2.42).0781.02 (0.79–1.32).853
       Missing/Unknown1.37 (0.54–3.43).5081.20 (0.59–2.43).6220.81 (0.40–1.64).5591.10 (0.61–2.01).748
      Medicaid Insurance
      Any Medicaid coverage in the year prior to index.
      (vs. No Medicaid Insurance)
      1.32 (0.65–2.67).4410.55 (0.23–1.33).1840.58 (0.24–1.40).2270.84 (0.44–1.59).584
      Membership months1.01 (1.00–1.02).0081.01 (1.01–1.01)<.0011.01 (1.00–1.01).0101.01 (1.01–1.02)<.001
      Diagnosis in year prior
      Corresponding diagnoses/score made in the year prior to index.
      7.79 (2.29–6.42).0017.32 (3.58–14.98)<.0016.94 (3.72–12.98)<.0014.20 (0.64–27.68).136
      Note: iRR = incidence rate ratio; OR = Odds Ratio.
      a Any Medicaid coverage in the year prior to index.
      b Corresponding diagnoses/score made in the year prior to index.

      Healthcare utilization - any versus None

      In logistic regression models examining any utilization for each type (e.g., hospitalizations, psychiatry, addiction medicine, primary care, ED) over 7 years, the SBIRT group had lower odds of any hospitalizations compared with usual care; no differences were found between SBIRT groups in the odds of psychiatry, addiction medicine or ED visits after adjusting for patient characteristics; since almost all patients (n = 1,867) had at least one primary care visit over 7 years, logistic regression was deemed unnecessary.
      In the same logistic regression models, patient characteristic findings included females having higher odds of psychiatric utilization and hospitalizations and lower odds of addiction medicine utilization compared with males. Older patients had lower odds of psychiatric and ED utilization, and patients with any Medicaid coverage in the year prior to index had higher odds of an ED visit compared to those without any Medicaid coverage during that time period. Asians had lower odds of psychiatry and ED utilization compared with White patients. Black/African American and Hispanic patients had lower odds of psychiatry utilization and higher odds of ED and inpatient utilization compared with White patients (Table 2).
      Table 2Multilevel logistic regression analyses examining utilization
      99.8% patients had at least one primary care visit by 7 years postindex, model would not converge.
      over 7 years postindex among all eligible patients (n = 1,871)
      Any psychiatry visitsAny addiction medicine visitsAny ED visitsAny inpatient hospitalizations
      ORp-valueORp-valueORp-valueORp-value
      Intercept16.70 (2.60–107.4).0030.08 (.02–.32)<.0010.82 (0.27–2.48).7240.05 (0.01–0.24)<.001
      SBIRT Group (vs. Usual Care)0.71 (0.13–4.02).7030.87 (0.64–1.19).3801.04 (0.85–1.27).7250.59 (0.41–0.85).005
      Female (vs. Male)1.43 (1.20–1.69)<.0010.47 (0.37–0.59)<.0011.01 (0.84–1.22).8981.89 (1.41–2.52)<.001
      Age0.82 (0.77–0.87)<.0010.95 (0.87–1.05).3210.91 (0.86–0.96).0011.01 (0.91–1.12).864
      Race/Ethnicity (reference: White)
       Asian0.56 (0.39–0.80).0020.52 (0.24–1.13).0970.55 (0.40–0.77)<.0010.81 (0.50–1.33).407
       Black/African American0.65 (0.47–0.89).0070.68 (0.46–1.01).0561.84 (1.39–2.43)<.0011.44 (1.15–1.80).002
       Hispanic0.69 (0.50–0.95).0231.50 (1.05–2.14).0251.42 (1.08–1.87).0121.56 (1.06–2.31).025
       Missing/Unknown0.61 (0.42–0.89).0110.34 (0.11–1.03).0561.02 (0.60–1.71).9541.26 (0.71–2.25).426
      Medicaid Insurance
      Any Medicaid coverage in the year prior to index.
      (vs. No Medicaid Insurance)
      0.85 (0.57–1.28).4430.38 (0.13–1.13).0833.04 (1.81–5.08)<.0011.58 (0.96–2.61).071
      Membership months1.01 (1.01–1.02)<.0011.02 (1.01–1.02)<.0011.03 (1.02–1.03)<.0011.01 (1.01–1.02)<.001
      Diagnosis in year prior
      Corresponding diagnoses/score made in the year prior to index.
      1.27 (1.18–1.37)<.0011.07 (1.01–1.14).0262.30 (1.84–2.86)<.0013.34 (1.50–7.44).003
      Note: OR = odds ratio.
      a Any Medicaid coverage in the year prior to index.
      b Corresponding diagnoses/score made in the year prior to index.
      c 99.8% patients had at least one primary care visit by 7 years postindex, model would not converge.

      Healthcare utilization - visit counts

      Visit counts were examined among patients with at least one visit for the corresponding type of utilization over the 7-year period (n = 1,867 primary care model; n = 841 psychiatry model; n = 110 addiction medicine model; n = 1,142 ED model; n = 279 inpatient model). Negative binomial models examining the number of visits over 7 years found that among those with at least one visit of that type, adolescents in the SBIRT group had fewer primary care, psychiatry visits, and more addiction medicine visits compared with usual care; no differences were found in the number of emergency department visits or hospitalizations.
      Patient characteristic findings from the same models found that females had more primary care and ED visits. Older patients had fewer primary care and psychiatry visits; the number of addiction medicine, ED, and inpatient visits did not differ. Those with Medicaid coverage in the year prior to the index had more primary care visits and fewer addiction medicine visits. Asian patients had fewer primary care and addiction medicine visits compared with White patients. Black/African American patients had fewer psychiatry and addiction medicines and more ED visits compared with White patients. Hispanic patients had fewer psychiatry visits and more ED visits compared with White patients (Table 3).
      Table 3Multilevel multinomial regression examining number of visits over 7 years postindex among patients with at least one visit of the same type
      Primary care visits (n = 1,867)Psychiatry visits (n = 841)Addiction medicine visits (n = 110)ED visits (n = 1,142)Inpatient hospitalizations (n = 279)
      iRRp-valueiRRp-valueiRRp-valueiRRp-valueiRRp-value
      Intercept6.10 (3.80–9.78)<.00185.93 (25.11–294.1)<.0013.72 (0.13–110.4).4471.75 (0.87–3.52).1191.33 (0.27–6.58).726
      SBIRT Group (vs. Usual Care)0.90 (0.82–0.99).0250.64 (0.51–0.79)<.0011.52 (1.14–2.03).0040.93 (0.79–1.09).3760.89 (0.70–1.13).341
      Female (vs. Male)1.53 (1.40–1.68)<.0011.04 (0.81–1.34).7381.51 (0.88–2.58).1321.42 (1.25–1.61)<.0010.95 (0.75–1.20).653
      Age0.96 (0.94–0.98)<.0010.86 (0.81–0.91)<.0011.11 (0.94–1.31).2160.99 (0.96–1.04).8021.01 (0.93–1.10).765
      Race/Ethnicity (reference: White)
       Asian0.89 (0.80–1.00).0421.17 (0.77–1.78).4720.25 (0.08–0.83).0240.97 (0.78–1.21).7681.00 (0.71–1.41).985
       Black/African American0.95 (0.87–1.04).2470.71 (0.54–0.93).0130.35 (0.17–0.69).0031.62 (1.37–1.93)<.0011.20 (0.84–1.72).325
       Hispanic0.97 (0.89–1.06).4920.67 (0.46–0.96).0290.72 (0.37–1.41).3321.51 (1.23–1.85)<.0011.39 (0.99–1.95).057
       Missing/Unknown0.93 (0.80–1.08).3460.82 (0.54–1.24).3480.71 (0.25–2.00).5191.20 (0.92–1.56).1721.37 (0.91–2.05).130
      Medicaid Insurance
      Any Medicaid coverage in the year prior to index.
      (vs. No Medicaid Insurance)
      1.26 (1.08–1.46).0040.89 (0.54–1.47).6530.04 (0.01–0.09)<.0011.30 (0.97–1.74).0781.22 (0.61–2.42).576
      Membership months1.02 (1.01–1.02)<.0011.01 (1.00–1.01)<.0011.00 (.98–1.01).7161.00 (1.00–1.01).0261.00 (0.99–1.01).822
      Diagnosis in year prior
      Corresponding diagnoses/score made in the year prior to index.
      1.10 (1.08–1.11)<.0011.04 (1.03–1.05)<.0011.01 (1.01–1.02)<.0011.33 (1.23–1.43)<.0011.67 (0.93–3.02).087
      Note: iRR = incidence rate ratio.
      a Any Medicaid coverage in the year prior to index.
      b Corresponding diagnoses/score made in the year prior to index.

      Posthoc and sensitivity analyses

      Posthoc analyses examined all mental health and substance use diagnosis outcomes stratified by race/ethnicity, and separately, by sex. Among Asian patients, those in the SBIRT group had lower odds of any mental health, depression, and anxiety diagnoses compared with usual care. Among Hispanic patients, the SBIRT group had higher odds of an anxiety diagnosis and lower odds of any substance, other drugs, and marijuana use disorder diagnoses compared with usual care, and among Black/African American patients, the SBIRT group had lower odds of alcohol use disorder diagnoses. Among White patients, those in the SBIRT group had higher odds of any mental health diagnosis compared with usual care (See Table A3). There were no differences in the diagnosis outcomes between SBIRT groups when examining models stratified by sex (not shown).
      A sensitivity analysis was conducted, reexamining all models only among the subset of patients with continuous membership across the 7-year time period (n = 1,033, n = 671 SBIRT group, n = 362 usual care). The findings in these models mirrored what was found among the full sample, with few exceptions. Among the sample of patients with continuous membership, no differences were found between the SBIRT and usual care groups in the odds of tobacco-related and alcohol-related diagnoses, and among those with complete membership who had at least one visit of that type, there were no longer differences in the number of primary care or addiction medicine visits between groups. An additional sensitivity analysis was conducted to examine differences between the usual care group and only those patients in the SBIRT group who actually received a brief intervention in the original trial (21.3% of patients in the SBIRT group received a brief intervention). When examining models among only SBIRT group patients who received a BI (n = 267) and all eligible usual care patients (n = 616), there were no longer significant differences between groups in the odds of a substance use, alcohol, or any drug or tobacco diagnoses over 7 years. The SBIRT group that received a BI had higher odds of addiction medicine utilization (OR = 1.46, 95% CI = 1.24–1.70) compared with usual care, whereas there were no significant differences in addiction medicine utilization between groups among the full sample. Additionally, there were no significant differences in hospitalizations between the SBIRT Group who received a BI and usual care, but in models examining the full sample, the SBIRT group had lower odds of hospitalizations. Among patients with at least one visit of that type, there were no significant differences between groups in the number of primary care visits; however, the SBIRT group who received a BI had fewer inpatient hospitalizations than the usual care group (iRR = 0.77, 95% CI = 0.59, 1.00); the groups did not differ when examining the full sample (See Table A4).

      Discussion

      The field of adolescent SBIRT research is relatively new and has heretofore examined more proximal outcomes - usually up to 12 months postintervention; our 7-year outcome data address a gap in the evidence base. In this study of the long-term outcomes of adolescents participating in an SBIRT trial in pediatric primary care, we examined outcomes important to adolescent development: subsequent substance use and other mental health disorders and medical comorbidity. We also examined how access to SBIRT affected potentially costly healthcare outcomes, which is particularly salient to health policymakers. We found lasting effects on the development of future disorders and on healthcare utilization well into young adulthood. To our knowledge, this is the first study of SBIRT for adolescents to examine these critical outcomes longitudinally into young adulthood (patients ranged from 19 to 25 years by study’s end), and it expands the growing literature on the benefits of SBIRT for adolescents.
      Young adults in the SBIRT group during their adolescence were less likely to develop any substance use disorder, including a diagnosis of alcohol, marijuana, or tobacco use disorder, over the seven years after original screening. Moreover, they were less likely to experience inpatient hospitalizations, and those with a psychiatry and/or an addiction medicine visit had fewer psychiatry visits but more addiction medicine visits. These findings suggest that the referral to treatment or “RT” component of SBIRT may help facilitate specialty addiction treatment engagement for those in need of it. This is among very few studies to demonstrate the positive impact of adolescent SBIRT on specialty treatment initiation.
      Research examining drinking patterns among adolescents transitioning to adulthood suggests that many adolescents “mature out” of these problems. On the other hand, it is well-known that health behavior patterns established in adolescence, both good and risky, often persist into young adulthood [
      • Hoyt L.T.
      • Chase-Lansdale P.L.
      • McDade T.W.
      • et al.
      Positive youth, healthy adults: Does positive well-being in adolescence predict better perceived health and fewer risky health behaviors in young adulthood?.
      ,
      • Wiium N.
      • Breivik K.
      • Wold B.
      Growth trajectories of health behaviors from adolescence through young adulthood.
      ], and risky behaviors, if not addressed, can significantly affect future health outcomes [
      • Mertens J.R.
      • Flisher A.J.
      • Fleming M.F.
      • et al.
      Medical conditions of adolescents in alcohol and drug treatment: Comparison with matched controls.
      ,
      • Mertens J.R.
      • Weisner C.
      • Ray G.T.
      • et al.
      Hazardous drinkers and drug users in HMO primary care: Prevalence, medical conditions, and costs.
      ,
      • Sterling S.A.
      • Palzes V.A.
      • Lu Y.
      • et al.
      Associations between medical conditions and alcohol consumption levels in an adult primary care population.
      ]. As most adult substance use disorders originate in adolescence, this is the ideal time to intervene to prevent adolescent use entirely or prevent mild problems from progressing. SBIRT in pediatric primary care is especially important in the U.S. where adult health insurance coverage is not guaranteed; however, most adolescents are covered with access to primary care. The U.S. Preventive Services Task Force recently decided against recommending screening for unhealthy drug use among adolescents. Our study shows the benefits of SBIRT among adolescents who screen positive for substance use. We found benefits to the SBIRT group even after 7 years, which makes a strong case for making adolescent screening and subsequent BI, when needed, an important vital sign metric during well-check visits in this group.
      Recent research suggests that half of children and adolescents with behavioral health problems do not receive treatment for them [
      • Whitney D.G.
      • Peterson M.D.
      US national and state-level prevalence of mental health disorders and disparities of mental health care use in children.
      ]; pediatric primary care-based SBIRT can increase access to appropriate treatment interventions and may be advantageous for reducing racial and ethnic disparities in care among adolescents of color, and with Medicaid Managed Care enrollees in healthcare systems with “carved-out” behavioral healthcare benefits [
      • Busch A.B.
      • Frank R.G.
      • Lehman A.F.
      The effect of a managed behavioral health carve-out on quality of care for medicaid patients diagnosed as having schizophrenia.
      ]. The disparities in substance use treatment-seeking and initiation [
      • Pinedo M.
      A current re-examination of racial/ethnic disparities in the use of substance abuse treatment: Do disparities persist?.
      ] are well documented, and these youth are already less likely to receive behavioral health services [
      • Alegria M.
      • Carson N.J.
      • Goncalves M.
      • et al.
      Disparities in treatment for substance use disorders and co-occurring disorders for ethnic/racial minority youth.
      ]. Healthcare disparities are also exacerbated over time because these patients are more likely to lose health insurance coverage as they age into adulthood [
      • Artiga S.
      • Orgera K.
      Changes in health coverage by race and ethnicity since the ACA, 2010-2018.
      ,
      • Novak P.
      • Williams-Parry K.F.
      • Chen J.
      Racial and ethnic disparities among the remaining uninsured young adults with behavioral health disorders after the ACA expansion of dependent coverage.
      ,
      • Sohn H.
      Racial and ethnic disparities in health insurance coverage: Dynamics of gaining and losing coverage over the life-course.
      ]. The adult SBIRT literature provides some evidence that primary care-based SBIRT may be especially beneficial for historically under-served populations [
      • Sahker E.
      • Jones D.
      • Lancianese D.A.
      • et al.
      Racial/ethnic differences in alcohol and drug use outcomes following Screening, Brief Intervention, and Referral to Treatment (SBIRT) in federally qualified health centers.
      ], because the setting may be consistent with patients’ preferences for nonspecialty care mental health services [
      • Murry V.M.
      • Heflinger C.A.
      • Suiter S.V.
      • et al.
      Examining perceptions about mental health care and help-seeking among rural African American families of adolescents.
      ]. Findings from posthoc analyses examining outcomes stratified by race/ethnicity suggest that SBIRT can be effective, especially for Hispanic and Black/African American adolescents. Providing SBIRT in pediatric primary care, which most children and adolescents have access to, may help to improve access to groups historically under-served in specialty behavioral health treatment and reduce care disparities.

      Limitations

      The study has several limitations. It was conducted in an integrated healthcare system with an insured population, and our findings may not be generalizable to public health systems or uninsured populations. However, the health system membership is representative of the Northern California population, including a sizeable Medicaid population. Because this was an intent-to-treat analysis, all eligible patients were included in the study, regardless of whether they received a BI or had continued membership over 7 years. However, membership was adjusted for using a count of member months over the 7-year period, and we also report the results of sensitivity analyses examining: (a) only those with continuous membership and (b) including only those in the SBIRT group who received a brief intervention. While the recommended workflow was such that pediatricians with patients who endorsed alcohol or drug use or mood symptoms at screening were encouraged to either deliver a BI or refer to the behavioral health clinician, as we found out, only a minority of those patients actually received a BI. Although we could not ascertain the severity of those who did or did not receive a BI due to the pragmatic nature of the trial, we can assume with some confidence that those who did were likely more severe than those who did not (which seems to be reflected in the fact that more of those in the SBIRT group who received a BI also went on to have higher odds of specialty treatment utilization). This was a pragmatic trial, and we relied on diagnoses documented during regular clinical care rather than self-reported use or symptoms, and thus may be missing lower severity substance use and mental health symptoms. However, examining use disorder diagnoses associated with more severe substance use problems expands the range of developmentally salient outcomes examined in the literature, which more typically examines self-reported use. Because we used the health system’s EHR and administrative databases to examine services utilization, we only captured services provided within, or paid for the health plan, and thus, we may not have captured all out-of-system utilization, particularly informal services such as self-help groups. However, there is nothing to suggest that there would be a difference in outside utilization between the two groups. The SBIRT group had more girls than boys at baseline, and girls were more likely to report mental health symptoms and diagnoses, which may have dampened the effects of access to SBIRT on subsequent mental health diagnoses in that arm. While we had promising findings regarding SBIRT’s impact on healthcare utilization, additional cost-effectiveness and cost-benefit studies are needed to assess the magnitude of the economic impact of SBIRT on youth and young adult outcomes.

      Conclusions

      Our findings add to the growing evidence base on the beneficial effects of providing SBIRT for adolescents in pediatric primary care and extend the existing literature by demonstrating clinically significant effects on the development of substance use disorders and healthcare utilization into young adulthood. Like many other integrated behavioral health approaches [
      • Asarnow J.R.
      • Rozenman M.
      • Wiblin J.
      • et al.
      Integrated medical-behavioral care compared with usual primary care for child and adolescent behavioral health: A meta-analysis.
      ,
      • Kolko D.J.
      The effectiveness of integrated care on pediatric behavioral health: Outcomes and opportunities.
      ,
      • Richardson L.P.
      • McCarty C.A.
      • Radovic A.
      • et al.
      Research in the integration of behavioral health for adolescents and young adults in primary care settings: A systematic review.
      ], SBIRT may provide significant and lasting benefits to vulnerable adolescents as they mature into young adulthood. This is particularly compelling because young adults are an age group that often “falls through the cracks” of the healthcare system as they transition from pediatric care to adult medicine. They are less likely to have a primary care physician than adolescents or older adults and more likely to use the ED for their healthcare, and these disparities are especially the case for vulnerable racial, ethnic, unemployed, and under-insured sub-populations of young adults [
      • Kirzinger W.K.
      • Cohen R.A.
      • Gindi R.M.
      Health care access and utilization among young adults aged 19–25: Early release of estimates from the National Health Interview Survey, January–September 2011.
      ]. Providing an intervention during their adolescence that can lay a strong preventative foundation could have a significant impact on the trajectory of their adult health and wellbeing. It is important that future studies examine the longitudinal effects of SBIRT for youth across a broad range of not only substance use outcomes but also medical and mental health conditions and other outcomes essential to youth health and wellbeing.

      Acknowledgments

      We thank Agatha Hinman for her editorial assistance with the manuscript. We thank the KPNC Adolescent Medicine Specialists Committee, the KPNC Adolescent Addiction Medicine Coordinating Committee, and Thekla Brumder Ross, PsyD, Derek Satre, PhD, and Jennifer Mertens, PhD, for their guidance. We also thank Anna Wong, PhD, David Bacchus, MD, Patricia Castaneda-Davis, MD, and all the physicians, medical assistants, nurses, receptionists, managers, and especially the patients and parents of KPNC’s Oakland Pediatrics clinic for their participation in the activities related to this study. All contributing authors have been included above.

      Funding Sources

      This work was supported by the Conrad N. Hilton Foundation (18454), and the National Institute on Alcohol Abuse and Alcoholism (grant R01 AA016204 ).

      References

        • DuPont R.L.
        • Caldeira K.M.
        • Dupont H.S.
        • et al.
        America’s dropout crisis: The unrecognized connection to adolescent substance use.
        (Available at:)
        • Wickersham A.
        • Sugg H.V.R.
        • Epstein S.
        • et al.
        Systematic review and meta-analysis: The association between child and adolescent depression and later educational attainment.
        J Am Acad Child Adolesc Psychiatry. 2021; 60: 105-118
        • Hale D.R.
        • Bevilacqua L.
        • Viner R.M.
        Adolescent health and adult education and employment: A systematic review.
        Pediatrics. 2015; 136: 128-140
        • Colins O.
        • Vermeiren R.
        • Vreugdenhil C.
        • et al.
        Psychiatric disorders in detained male adolescents: A systematic literature review.
        Can J Psychiatry. 2010; 55: 255-263
        • National Institute on Drug Abuse
        Costs of substance abuse.
        (Available at:)
        • Vida R.
        • Brownlie E.B.
        • Beitchman J.H.
        • et al.
        Emerging adult outcomes of adolescent psychiatric and substance use disorders.
        Addict Behav. 2009; 34: 800-805
        • Johnston L.D.
        • Miech R.A.
        • O’Malley P.M.
        • et al.
        Monitoring the future national survey results on drug Use: 1975-2019: Overview: Key findings on adolescent drug use.
        (Available at:)
        • Substance Abuse and Mental Health Services Administration
        Key substance use and mental health indicators in the United States: Results from the 2018 national survey on drug Use and health.
        (Available at:)
        • Torralba E.
        Nearly half of California adolescents report mental health difficulties.
        (Available at:)
        • Roehler D.R.
        • Olsen E.O.
        • Mustaquim D.
        • et al.
        Suspected nonfatal drug-related overdoses among youth in the US: 2016-2019.
        Pediatrics. 2021; 147 (e2020003491)
        • Merikangas K.R.
        • He J.P.
        • Burstein M.
        • et al.
        Service utilization for lifetime mental disorders in U.S. Adolescents: Results of the national comorbidity survey-adolescent supplement (NCS-A).
        J Am Acad Child Adolesc Psychiatry. 2011; 50: 32-45
        • Alegria M.
        • Carson N.J.
        • Goncalves M.
        • et al.
        Disparities in treatment for substance use disorders and co-occurring disorders for ethnic/racial minority youth.
        J Am Acad Child Adolesc Psychiatry. 2011; 50: 22-31
        • Radez J.
        • Reardon T.
        • Creswell C.
        • et al.
        Why do children and adolescents (not) seek and access professional help for their mental health problems? A systematic review of quantitative and qualitative studies.
        Eur Child Adolesc Psychiatry. 2021; 30: 183-211
        • National Committee for Quality Assurance (NCQA)
        Children and adolescents’ access to primary care Practitioners (CAP).
        (Available at:)
        • Sterling S.
        • Kline-Simon A.H.
        • Weisner C.
        • et al.
        Pediatrician and behavioral clinician-delivered screening, brief intervention and referral to treatment: Substance use and depression outcomes.
        J Adolesc Health. 2018; 62: 390-396
        • Tanner-Smith E.E.
        • Lipsey M.W.
        Brief alcohol interventions for adolescents and young adults: A systematic review and meta-analysis.
        J Subst Abuse Treat. 2015; 51: 1-18
        • D'Amico E.J.
        • Miles J.N.
        • Stern S.A.
        • et al.
        Brief motivational interviewing for teens at risk of substance use consequences: A randomized pilot study in a primary care clinic.
        J Subst Abuse Treat. 2008; 35: 53-61
        • Harris S.K.
        • Csemy L.
        • Sherritt L.
        • et al.
        Computer-facilitated substance use screening and brief advice for teens in primary care: An international trial.
        Pediatrics. 2012; 129: 1072-1082
        • Walton M.A.
        • Resko S.
        • Barry K.L.
        • et al.
        A randomized controlled trial testing the efficacy of a brief cannabis universal prevention program among adolescents in primary care.
        Addiction. 2014; 109: 786-797
        • Walton M.A.
        • Bohnert K.
        • Resko S.
        • et al.
        Computer and therapist based brief interventions among cannabis-using adolescents presenting to primary care: One year outcomes.
        Drug Alcohol Depend. 2013; 132: 646-653
        • Burke P.J.
        • Da Silva J.D.
        • Vaughan B.L.
        • et al.
        Training high school counselors on the use of motivational interviewing to screen for substance abuse.
        Subst Abus. 2005; 26: 31-34
        • Gil A.G.
        • Wagner E.F.
        • Tubman J.G.
        Culturally sensitive substance abuse intervention for Hispanic and African American adolescents: Empirical examples from the alcohol treatment Targeting adolescents in need (ATTAIN) Project.
        Addiction. 2004; 99: 140-150
        • Grenard J.L.
        • Ames S.L.
        • Wiers R.W.
        • et al.
        Brief intervention for substance use among at-risk adolescents: A pilot study.
        J Adolesc Health. 2007; 40: 188-191
        • Martin G.
        • Copeland J.
        • Swift W.
        The adolescent cannabis check-up: Feasibility of a brief intervention for young cannabis users.
        J Subst Abuse Treat. 2005; 29: 207-213
        • McCambridge J.
        • Strang J.
        The efficacy of single-session motivational interviewing in reducing drug consumption and perceptions of drug-related risk and harm among young people: Results from a multi-site cluster randomized trial.
        Addiction. 2004; 99: 39-52
        • Srisurapanont M.
        • Sombatmai S.
        • Boripuntakul T.
        Brief intervention for students with methamphetamine use disorders: A randomized controlled trial.
        Am J Addict. 2007; 16: 111-116
        • Winters K.C.
        • Leitten W.
        • Wagner E.
        • et al.
        Use of brief interventions for drug abusing teenagers within a middle and high school setting.
        J Sch Health. 2007; 77: 196-206
        • Das J.K.
        • Salam R.A.
        • Arshad A.
        • et al.
        Interventions for adolescent substance abuse: An overview of systematic reviews.
        J Adolesc Health. 2016; 59: S61-S75
        • Mitchell S.G.
        • Gryczynski J.
        • O'Grady K.E.
        • et al.
        SBIRT for adolescent drug and alcohol use: Current status and future directions.
        J Subst Abuse Treat. 2013; 44: 463-472
        • Steinka-Fry K.T.
        • Tanner-Smith E.E.
        • Hennessy E.A.
        Effects of brief alcohol interventions on drinking and driving among youth: A systematic review and meta-analysis.
        J Addict Prev. 2015; 3: 11
        • Tanner-Smith E.E.
        • Risser M.D.
        A meta-analysis of brief alcohol interventions for adolescents and young adults: Variability in effects across alcohol measures.
        Am J Drug Alcohol Abuse. 2016; 42: 140-151
        • Tanner-Smith E.E.
        • Steinka-Fry K.T.
        • Hennessy E.A.
        • et al.
        Can brief alcohol interventions for youth also address concurrent illicit drug use? Results from a meta-analysis.
        J Youth Adolesc. 2015; 44: 1011-1023
        • Carney T.
        • Myers B.
        Effectiveness of early interventions for substance-using adolescents: Findings from a systematic review and meta-analysis.
        Subst Abuse Treat Prev Policy. 2012; 7: 25
        • Steele D.W.
        • Becker S.J.
        • Danko K.J.
        • et al.
        Brief behavioral interventions for substance use in adolescents: A meta-analysis.
        Pediatrics. 2020; 146: e20200351
        • Parthasarathy S.
        • Kline-Simon A.H.
        • Jones A.
        • et al.
        Three-year outcomes after brief treatment of substance use and mood symptoms.
        Pediatrics. 2021; 147 (e2020009191)
        • Sterling S.
        • Kline-Simon A.H.
        • Jones A.
        • et al.
        Health care use over 3 years after adolescent SBIRT.
        Pediatrics. 2019; 143: e20182803
        • Sterling S.
        • Kline-Simon A.H.
        • Jones A.
        • et al.
        Specialty addiction and psychiatry treatment initiation and engagement: Results from an SBIRT randomized trial in pediatrics.
        J Subst Abuse Treat. 2017; 82: 48-54
        • Sterling S.
        • Kline-Simon A.H.
        • Satre D.D.
        • et al.
        Implementation of screening, brief intervention, and referral to treatment for adolescents in pediatric primary care: A cluster randomized trial.
        JAMA Pediatr. 2015; 169: e153145
        • Levy S.J.
        • Williams J.F.
        • Committee on Substance USE
        Substance use screening, brief intervention, and referral to treatment.
        Pediatrics. 2016; 138: e20161211
        • Charlson M.E.
        • Charlson R.E.
        • Peterson J.C.
        • et al.
        The Charlson comorbidity index is adapted to predict costs of chronic disease in primary care patients.
        J Clin Epidemiol. 2008; 61: 1234-1240
        • Hoyt L.T.
        • Chase-Lansdale P.L.
        • McDade T.W.
        • et al.
        Positive youth, healthy adults: Does positive well-being in adolescence predict better perceived health and fewer risky health behaviors in young adulthood?.
        J Adolesc Health. 2012; 50: 66-73
        • Wiium N.
        • Breivik K.
        • Wold B.
        Growth trajectories of health behaviors from adolescence through young adulthood.
        Int J Environ Res Public Health. 2015; 12: 13711-13729
        • Mertens J.R.
        • Flisher A.J.
        • Fleming M.F.
        • et al.
        Medical conditions of adolescents in alcohol and drug treatment: Comparison with matched controls.
        J Adolesc Health. 2007; 40: 173-179
        • Mertens J.R.
        • Weisner C.
        • Ray G.T.
        • et al.
        Hazardous drinkers and drug users in HMO primary care: Prevalence, medical conditions, and costs.
        Alcohol Clin Exp Res. 2005; 29: 989-998
        • Sterling S.A.
        • Palzes V.A.
        • Lu Y.
        • et al.
        Associations between medical conditions and alcohol consumption levels in an adult primary care population.
        JAMA Netw Open. 2020; 3: e204687
        • Whitney D.G.
        • Peterson M.D.
        US national and state-level prevalence of mental health disorders and disparities of mental health care use in children.
        JAMA Pediatr. 2019; 173: 389-391
        • Busch A.B.
        • Frank R.G.
        • Lehman A.F.
        The effect of a managed behavioral health carve-out on quality of care for medicaid patients diagnosed as having schizophrenia.
        Arch Gen Psychiatry. 2004; 61: 442-448
        • Pinedo M.
        A current re-examination of racial/ethnic disparities in the use of substance abuse treatment: Do disparities persist?.
        Drug Alcohol Depend. 2019; 202: 162-167
        • Artiga S.
        • Orgera K.
        Changes in health coverage by race and ethnicity since the ACA, 2010-2018.
        (Available at:)
        • Novak P.
        • Williams-Parry K.F.
        • Chen J.
        Racial and ethnic disparities among the remaining uninsured young adults with behavioral health disorders after the ACA expansion of dependent coverage.
        J Racial Ethnic Health Disparities. 2017; 4: 607-614
        • Sohn H.
        Racial and ethnic disparities in health insurance coverage: Dynamics of gaining and losing coverage over the life-course.
        Popul Res Policy Rev. 2017; 36: 181-201
        • Sahker E.
        • Jones D.
        • Lancianese D.A.
        • et al.
        Racial/ethnic differences in alcohol and drug use outcomes following Screening, Brief Intervention, and Referral to Treatment (SBIRT) in federally qualified health centers.
        J Racial Ethn Health Disparities. 2019; 6: 1192-1199
        • Murry V.M.
        • Heflinger C.A.
        • Suiter S.V.
        • et al.
        Examining perceptions about mental health care and help-seeking among rural African American families of adolescents.
        J Youth Adolesc. 2011; 40: 1118-1131
        • Asarnow J.R.
        • Rozenman M.
        • Wiblin J.
        • et al.
        Integrated medical-behavioral care compared with usual primary care for child and adolescent behavioral health: A meta-analysis.
        JAMA Pediatr. 2015; 169: 929-937
        • Kolko D.J.
        The effectiveness of integrated care on pediatric behavioral health: Outcomes and opportunities.
        JAMA Pediatr. 2015; 169: 894-896
        • Richardson L.P.
        • McCarty C.A.
        • Radovic A.
        • et al.
        Research in the integration of behavioral health for adolescents and young adults in primary care settings: A systematic review.
        J Adolesc Health. 2017; 60: 261-269
        • Kirzinger W.K.
        • Cohen R.A.
        • Gindi R.M.
        Health care access and utilization among young adults aged 19–25: Early release of estimates from the National Health Interview Survey, January–September 2011.
        (Available at:)