Receipt of Addiction Treatment After Opioid Overdose in Medicaid-Enrolled Youths (original) (raw)

Figure. Kaplan-Meier Survival Curve of Time Free From Recurrent Overdose in 3606 Youths After Incident Overdose, by Type of Opioid Involved in Incident Overdose

Table 1. Baseline Characteristics of Overall Sample of 4 039 216 Youths Aged 13 to 22 Years Stratified by Opioid Overdose, January 1, 2009, to September 30, 2015

Table 2. Baseline Characteristics of 3791 Youths Aged 13 to 22 Years With Opioid Overdose by Type of Opioid Overdose, January 1, 2009, to September 30, 2015

Table 3. Multivariable Cox Proportional Hazards Model Showing Sociodemographic and Clinical Characteristics Associated With Recurrent Overdose Among 3606 Youths With Incident Opioid Overdose

Table 4. Sociodemographic and Clinical Characteristics of 3606 Youths With Opioid Overdose by Receipt of Timely Addiction Treatment Within 30 Days of Overdose, January 1, 2009, to September 30, 2015

January 6, 2020

Receipt of Addiction Treatment After Opioid Overdose Among Medicaid-Enrolled Adolescents and Young Adults

JAMA Pediatr. 2020;174(3):e195183. doi:10.1001/jamapediatrics.2019.5183

Key Points

Question What are the characteristics of youths (adolescents and young adults) who experience nonfatal opioid overdose with heroin or other opioid, and do these youths receive timely evidence-based treatment?

Findings In this cohort study of 4 039 216 Medicaid-enrolled youths 13 to 22 years of age, among 3606 individuals who experienced opioid-related overdose and had continuous enrollment for at least 30 days after overdose, less than one-third received timely addiction treatment after overdose, and only 1 in 54 youths received pharmacotherapy with buprenorphine, naltrexone, or methadone.

Meaning Interventions are needed to link youths to treatment after overdose, with priority placed on improving access to evidence-based pharmacotherapy.

Importance Nonfatal opioid overdose may be a critical touch point when youths who have never received a diagnosis of opioid use disorder can be engaged in treatment. However, the extent to which youths (adolescents and young adults) receive timely evidence-based treatment following opioid overdose is unknown.

Objective To identify characteristics of youths who experience nonfatal overdose with heroin or other opioids and to assess the percentage of youths receiving timely evidence-based treatment.

Design, Setting, and Participants This retrospective cohort study used the 2009-2015 Truven–IBM Watson Health MarketScan Medicaid claims database from 16 deidentified states representing all US census regions. Data from 4 039 216 Medicaid-enrolled youths aged 13 to 22 years were included and were analyzed from April 20, 2018, to March 21, 2019.

Exposures Nonfatal incident and recurrent opioid overdoses involving heroin or other opioids.

Main Outcomes and Measures Receipt of timely addiction treatment (defined as a claim for behavioral health services, for buprenorphine, methadone, or naltrexone prescription or administration, or for both behavioral health services and pharmacotherapy within 30 days of incident overdose). Sociodemographic and clinical characteristics associated with receipt of timely treatment as well as with incident and recurrent overdoses were also identified.

Results Among 3791 youths with nonfatal opioid overdose, 2234 (58.9%) were female, and 2491 (65.7%) were non-Hispanic white. The median age was 18 years (interquartile range, 16-20 years). The crude incident opioid overdose rate was 44.1 per 100 000 person-years. Of these 3791 youths, 1001 (26.4%) experienced a heroin overdose; the 2790 (73.6%) remaining youths experienced an overdose involving other opioids. The risk of recurrent overdose among youths with incident heroin involvement was significantly higher than that among youths with other opioid overdose (adjusted hazard ratio, 2.62; 95% CI, 2.14-3.22), and youths with incident heroin overdose experienced recurrent overdose at a crude rate of 20 700 per 100 000 person-years. Of 3606 youths with opioid-related overdose and continuous enrollment for at least 30 days after overdose, 2483 (68.9%) received no addiction treatment within 30 days after incident opioid overdose, whereas only 1056 youths (29.3%) received behavioral health services alone, and 67 youths (1.9%) received pharmacotherapy. Youths with heroin overdose were significantly less likely than youths with other opioid overdose to receive any treatment after their overdose (adjusted odds ratio, 0.64; 95% CI, 0.49-0.83).

Conclusions and Relevance After opioid overdose, less than one-third of youths received timely addiction treatment, and only 1 in 54 youths received recommended evidence-based pharmacotherapy. Interventions are urgently needed to link youths to treatment after overdose, with priority placed on improving access to pharmacotherapy.

Rates of opioid use and opioid overdose in the United States have been rapidly rising, including among adolescents and young adults,1-4 the age groups in which the majority of substance use begins.5,6 Adolescents and young adults (youths) between the ages of 15 and 24 years comprised 4027 fatal opioid overdoses in 2016,7 and the opioid overdose mortality rate for children and adolescents under age 20 has tripled in the last 2 decades.4 Rates of nonfatal opioid overdose have similarly escalated, resulting in 7410 hospitalizations and 28 207 emergency department visits for youths aged 15 to 24 years during 2015.8 Whether these overdoses involve heroin, prescription opioids, or synthetic opioids, such as fentanyl, varies by age and other sociodemographic characteristics and continues to evolve.8-10 The high risk of recurrent overdose after nonfatal overdose among adults has been described11-14 although the rate, characteristics, and type of opioid (ie, heroin or other opioids) associated with recurrent overdose among youths remain unknown.

Nonfatal opioid overdose has been identified as a potential touch point, when individuals who had not previously received a diagnosis of opioid use disorder (OUD) can be drawn into treatment.15,16 Evidence-based guidelines recommend that youths with OUD receive treatment that includes pharmacotherapy.17 Unfortunately, youths in treatment for OUD receive pharmacotherapy at only one-tenth the rate of adults.18 Other data suggest that only 23% of Medicaid-enrolled youths receive pharmacotherapy within 3 months of OUD diagnosis,19 and 27% of commercially insured youths received pharmacotherapy within 6 months of receiving an OUD diagnosis.5 Although research regarding adult access to addiction treatment is growing and has shown that timely receipt of pharmacotherapy following overdose is critical in reducing subsequent mortality,16 very little is known about health care use following opioid overdose in youths.

This study aimed to fill this knowledge gap to inform the development of secondary prevention strategies, practice guidelines, and quality measures for the delivery of youth-specific OUD treatment. Using a 16-state sample of Medicaid-enrolled youths, we sought to determine (1) the characteristics of youths who experience nonfatal opioid overdose, comparing youths with heroin vs other opioid overdose; and (2) the percentage and characteristics of youths who receive recommended treatment within 30 days following opioid overdose.

This retrospective cohort study was conducted using the 2009-2015 Truven–IBM Watson Health MarketScan Medicaid Database, encompassing 16 deidentified states representing all US census regions. Youths aged 13 to 22 years with at least 6 months of continuous enrollment were included. All inpatient, outpatient, emergency department, behavioral health service, and retail prescription drug claims between January 1, 2009, and December 31, 2015, were included. The Boston University School of Medicine Institutional Review Board granted this study exemption from formal review because this study used exclusively retrospective deidentified administrative records; thus, informed patient consent was also waived because it was not possible.

The study sample of all youths with incident opioid overdose occurring between January 1, 2009, and September 30, 2015, was created by identifying youths who had received a primary or secondary diagnosis of opioid poisoning using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes E850.0, E850.1, E850.2, 965.00, 965.01, 965.02, and 965.09 on emergency department and inpatient claims (eTable 1 in the Supplement).1,16,20-23 On the basis of ICD-9-CM codes, youths were further stratified into those with an overdose involving heroin (alone or in combination with other opioids) and those with overdose involving any other opioid.

Youths remained under observation as long as their coverage was active and were censored from the cohort if they disenrolled from their coverage.19 Those previously in the cohort were not allowed to reenter at a later date. If youths had more than 1 opioid overdose during the study, the first overdose was selected as the incident opioid overdose. Recurrent opioid overdose was defined as a subsequent diagnosis of opioid poisoning occurring any time after hospital discharge from incident overdose.

Study Variable Construction

All study covariates were chosen based on their shown association with OUD or recurrent opioid overdose.5,16,19 Sociodemographic covariates included age, sex, and race/ethnicity. Clinical covariates (diagnosed during the 3 months preceding through 1 month following the incident overdose) included pregnancy, acute pain condition, chronic pain condition, depression, anxiety disorder, attention-deficit/hyperactivity disorder, OUD, alcohol use disorder, and other substance use disorders (eTable 1 in the Supplement).19,24-27 To evaluate the overlap between overdose and intentional self-harm, claims including diagnosis codes for suicidality and self-harm occurring in the 3 months preceding through 1 month after overdose were examined. Additional clinical covariates (during the 3 months preceding overdose) included receipt of an opioid prescription or prior addiction treatment (behavioral health services, pharmacotherapy, or both). The location of care for the overdose (emergency department or inpatient setting) was added as a covariate. We additionally adjusted for calendar year because states entering or exiting the study sample did so on January 1 and December 31, respectively. Thus, adjusting for calendar year approximated adjustment for the mix of states in the sample in any given year although this approach was unable to fully adjust for within-state correlation.

Limiting the sample to youths with at least 30 days of enrollment after incident overdose, timely receipt of addiction treatment was defined as a claim for behavioral health services, pharmacotherapy (buprenorphine, naltrexone, or methadone), or both within 30 days of incident overdose.28,29 Dispensing of buprenorphine and oral and injectable naltrexone were identified using National Drug Codes in pharmacy claims (eTable 2 in the Supplement).5,19,30 In-office injectable naltrexone administration was identified using the Healthcare Common Procedure Coding System (HCPCS) code J2315 (“naltrexone, depot form”), and methadone receipt was identified via the HCPCS code H0020 (“methadone administration and/or service”).19,31,32 Behavioral health services were identified across the spectrum of settings using Current Procedural Terminology and HCPCS codes (eTable 3 in the Supplement).33,34

The overdose rate was calculated per 100 000 person-years at risk.3 The χ2 test was used to compare sociodemographic and clinical characteristics between youths with overdose vs the overall sample of youths without overdose, and between those with heroin vs other opioid overdose.

Among youths with incident overdose, the crude recurrent overdose rate was calculated. Using time since incident overdose as the time scale, the Kaplan-Meier method and log-rank test were used to compare the cumulative incidence rates of recurrent overdose among youths with overdose involving heroin vs those involving only other opioids. A multivariable Cox proportional hazards model (including all clinical and sociodemographic covariates, which were selected a priori) was used to calculate hazard ratios for recurrent overdose. Because the outcome of interest, recurrent overdose, may be associated with loss to follow-up (ie, if an individual experienced a fatal overdose and thus was removed from the insurance database), inverse probability of censoring weights35 was used to create an additional proportional hazards model using estimated weights that adjusted for emigrative selection bias.

Youths with incident opioid overdose were stratified into mutually exclusive categories of having received no timely addiction treatment, behavioral health services only, or pharmacotherapy (alone or in combination with behavioral health services) within 30 days of overdose. The χ2 test was used to compare receipt of each treatment category according to sociodemographic and clinical characteristics. Multivariable logistic regression was performed (including all sociodemographic and clinical covariates, which were selected a priori) to provide adjusted odds ratios for receipt of treatment.

Analyses were conducted from April 20, 2018, to March 21, 2019 (with the most recent data available at the time of analyses), using SAS, version 9.4 (SAS Institute Inc). All statistical tests were 2 sided and considered statistically significant at P < .05.

Characteristics Associated With Overdose

There were 4 039 216 youths aged 13 to 22 years with at least 6 months of continuous Medicaid enrollment. In total, 3791 youths (0.1%) experienced a nonfatal opioid overdose between January 1, 2009, and September 30, 2015, resulting in an incidence rate of 44.1 per 100 000 person-years. Table 1 provides the characteristics of the sample population. The median age of youths who experienced overdose was 18 years (interquartile range, 16-20 years). Of these 3791 youths with overdose, 2234 (58.9%) were female, among whom 470 (21.0%) were pregnant. Most youths, 2491 (65.7%), were non-Hispanic white. Of 3791 youths with overdose, 725 (19.1%) received a diagnosis of OUD. Characteristics associated with overdose were female sex, pregnancy, non-Hispanic white race/ethnicity, comorbid acute or chronic pain conditions, depression, self-harm, anxiety, attention-deficit/hyperactivity disorder, OUD, alcohol use disorder, and other substance use disorders.

Table 2 gives characteristics of the youths according to the opioid involved in their incident overdose. Of 3791 youths who experienced an overdose, 1001 overdoses (26.4%) involved heroin and the remainder (2790 [73.6%]) involved opioids other than heroin. Heroin overdoses were more common in the later calendar years (2014-2015) and were more often treated in the emergency department (803 [80.2%]) than inpatient setting (198 [19.8%]). Youths with heroin overdose were more likely than youths with other opioid overdose to be 18 years or older, to be non-Hispanic white race/ethnicity (768 [76.7%] vs 1723 [61.8%]), to have previously received addiction treatment, or to have received a diagnosis of OUD (497 [49.7%] vs 228 [8.2%]), alcohol use disorder (252 [25.2%] vs 461 [16.5%]), or other substance use disorders (704 [70.3%] vs 1040 [37.3%]) (all P < .001). Youths with opioid overdose were more likely than those with heroin overdose to have pain conditions, depression (1889 [67.7%] vs 440 [44.0%]), self-harm (1137 [40.8%] vs 176 [17.6%]), anxiety (1425 [51.1%] vs 387 [38.7%]), attention-deficit/hyperactivity disorder (619 [22.2%] vs 152 [15.2%]), or previously received an opioid prescription (733 [26.3%] vs 116 [11.6%]) (all P < .001).

The sample was then limited to 3606 youths with continuous enrollment for at least 30 days after overdose. Youths with heroin overdose had a more than 7-fold higher unadjusted recurrent overdose crude rate (heroin, 20 770 per 100 000 person-years vs other opioids, 2855 per 100 000 person-years; P < .001), which is 471 times as a high as the general sample overdose rate of 44 per 100 000 person-years. The Figure shows survival time free from recurrent overdose by type of opioid involved in incident overdose (log-rank test, P < .001). Among youths experiencing incident heroin overdose, the cumulative incidence rates of recurrent overdose were 4.1% by 30 days, 6.2% by 60 days, and 8.1% by 90 days. Among youths experiencing incident overdose involving other opioids, the cumulative incidence rates of recurrent overdose were 1.2% by 30 days, 1.8% by 60 days, and 2.0% by 90 days.

Table 3 gives the characteristics associated with recurrent overdose. Youths with incident heroin overdose had 2.62 (95% CI, 2.14-3.22) times greater risk of recurrent overdose than youths with incident overdose involving other opioids, controlling for all other covariates. Male sex (adjusted hazard ratio [AHA], 1.44; 95% CI, 1.22-1.70), self-harm (AHA, 1.26; 95% CI, 1.05-1.51), diagnosis of OUD (AHA, 1.65; 95% CI, 1.36-2.01) or other substance use disorders (AHA, 1.86; 95% CI, 1.54-2.24), and treatment in the emergency department (AHA, 1.24; 95% CI, 1.03-1.48) were also associated with elevated risk of recurrent overdose. Conversely, prior receipt of behavioral health services was associated with lower risk of recurrent overdose (AHA, 0.65; 95% CI, 0.54-0.79).

Timely Addiction Treatment

Table 4 provides the characteristics associated with receipt of treatment in 3606 youths with continuous enrollment for at least 30 days after overdose. Of these, 2483 youths (68.9%) received no addiction treatment within 30 days after incident opioid overdose, 1056 (29.3%) received only behavioral health services, and 67 (1.9%) received pharmacotherapy (alone or in combination with behavioral health services). Lower unadjusted rates of treatment with pharmacotherapy were found for younger adolescents (13-15 years of age, 0.5%; 16-17 years of age, 0.8%; 18-20 years of age, 1.8%; 21-22 years of age, 4.2%) and for racial/ethnic minorities (Hispanic, 0%; non-Hispanic black, 0.2%; non-Hispanic white, 2.3%).

In the adjusted multivariable model, younger youths were more likely than older youths to receive any timely addiction treatment, driven by higher rates of behavioral health services. Youths with anxiety, depression, or self-harm were more likely to receive behavioral health services but not pharmacotherapy. Youths who had received a diagnosis of OUD were 9.03 (95% CI, 3.95-20.7) times more likely to receive pharmacotherapy and 1.74 (95% CI, 1.34-2.25) times more likely to receive behavioral health services than youths who had not received a diagnosis of OUD. Youths with heroin overdose or with emergency department encounters were less likely to receive behavioral health services than youths with other opioid overdose or inpatient admission, but were no more or less likely to receive pharmacotherapy. Youths with heroin overdose were significantly less likely than youths with other opioid overdose to receive any treatment after their overdose (adjusted odds ratio, 0.64; 95% CI, 0.49-0.83). Youths who were engaged in treatment prior to overdose were more likely to continue the same treatment after overdose than individuals not previously in treatment although no more or less likely to receive another modality of treatment after overdose.

In this study of more than 4 million Medicaid-enrolled youths aged 13 to 22 years, nonfatal opioid overdose occurred in 3791 youths at a rate of 44.1 per 100 000 person-years. Approximately one-quarter of overdoses involved heroin. Youths with incident heroin overdose had 2.6 times higher risk of recurrent overdose than youths with incident overdose involving only other opioids. By 3 months, 8.1% of youths with incident heroin overdose and 2.0% of youths with incident other opioid overdose experienced recurrent overdose. Only 1 in 54 youths with overdose received timely treatment with pharmacotherapy, and less than one-third received behavioral health services, leaving more than two-thirds of youths who experienced overdose with no addiction treatment.

This study contributed several key findings. First, it complements the high risk of opioid overdose among adolescents described in the Centers for Disease Control and Prevention Annual Surveillance Report.8 In that report, 15- to 19-year-olds experienced opioid poisoning hospitalizations and emergency department visits in 2015 at a rate of 9.1 and 26.5 per 100 000 person years, respectively. Youths aged 20 to 24 years experienced higher rates: 24.2 hospitalizations and 99.7 emergency department visits per 100 000 person years. However, that report8 did not delineate between incident vs recurrent overdose, whereas the present study showed a rate of subsequent opioid overdose among youths with heroin overdose 471 times as high as the general sample of Medicaid-enrolled youth.

Second, this study characterizes factors associated with opioid overdose in Medicaid-enrolled youth, and identifies crucial differences between youths who experience heroin vs other opioid overdose. Youths with heroin overdose tended to be older and have other co-occurring substance use disorders, whereas youths with other opioid overdose tended to be younger and have a high burden of mood disorders. These other opioids were likely to be largely prescription opioids given the time frame of the study because synthetic opioids, such as fentanyl, were only beginning to increase in prevalence toward the end of the study period, in 2014.9,36,37 Given that only 26.3% of youths with other opioid overdose in the present study had been prescribed an opioid, it is likely that youths acquired prescription opioids from friends, family members, and other sources.38 Owing to recent contamination of the heroin supply with fentanyl, heroin use carries a high risk of overdose.37,39,40 Even still, prescription opioid overdose continues to remain problematic among youth, accounting for approximately one-third of fatal opioid overdoses among 15- to 19-year-olds in 2016.4 This highlights the importance of early recognition and treatment of opioid use and OUD.

Third, the present study showed a large unmet need for addiction treatment of youths following overdose. The poor follow-up of Medicaid-enrolled youths observed in this study is greater than that described for other mental health conditions.41-43 Youths are less likely to receive recommended treatment after overdose than adults. A 2016 study using commercial claims found that 16.7% of adults received pharmacotherapy within 30 days of opioid overdose and 43.3% received behavioral health services.28 To date, only 2 studies that included adolescents have been published regarding treatment trajectories after nonfatal opioid overdose. One study did not delineate outcomes for adolescents32; the other study examined a cohort of 195 Massachusetts youths, 8% of whom received pharmacotherapy within a year of overdose.44 Because timely receipt of pharmacotherapy for adults is associated with decreased mortality, addressing the treatment gap for youths is imperative.16 Future research should evaluate the population-level effects of timely addiction treatment in reducing recurrent overdose among youths and the effectiveness of programs designed to link youths to care after overdose.

There are numerous reasons for the large treatment gap observed in the present study. Foremost is a lack of pediatric clinician familiarity in treating opioid overdose and OUD. Less than 10% of youths with nonheroin opioid overdose were diagnosed as having OUD, and even among youths with heroin overdose, less than half received a diagnosis of OUD. For youths who are identified as having OUD and needing treatment, data suggest that clinicians struggle to connect youths to effective treatment amid a dearth of addiction treatment facilities that accept youths and, even more rarely, that offer pharmacotherapy.19,45,46 In general, Medicaid has generous coverage of pharmacotherapy for OUD, with little heterogeneity between states regarding buprenorphine coverage.47 However, many addiction treatment facilities do not accept Medicaid, and hurdles such as prior authorizations, lifetime limits, and requirements for concurrent behavioral health services may further limit access to pharmacotherapy.47,48 Policy changes should be aimed at making it more feasible for clinicians to treat youths with opioid overdose according to evidenced-based guidelines through payer reforms and clinician education and by increasing the number of youth-serving clinicians and facilities that prescribe pharmacotherapy for OUD. Emergency department programs that help place individuals with overdose immediately into care (including pharmacotherapy) are effective49 and have begun to be mandated through policy.50 Policymakers, public health officials, and clinicians should ensure that the needs of youths are specifically addressed in these programs. Researchers might also acknowledge the limitations of using OUD diagnoses in claims data, which likely represent a vast underreporting of the true OUD incidence.

There were several limitations to this study. First, because this study used claims data, the data reflected only the extent to which opioid overdoses and clinical covariates were accurately coded. Because there is no ICD-9-CM code for synthetic opioid overdose and because drug testing for fentanyl was uncommon during the study period, it is likely that some overdoses classified as being attributable to heroin were actually attributable to fentanyl in the last 2 years of the study period, as the prevalence of fentanyl exposure among youths has risen since late 2014.4,9,36,37 Second, some overdoses may have occurred before the study period or prior to youths’ enrollment in Medicaid. Third, because the data were deidentified with respect to state, adjustment for within-state correlation was not possible although adjustment for calendar year approximated adjustment for the mix of states in the sample in any given year. Fourth, there were small sample sizes of youths from racial/ethnic minorities; thus, the study was not powered to make statistical inferences about the roles of race and ethnicity in access to treatment. Fifth, Medicaid-enrolled youths may have accessed addiction treatment in a private setting, which would not be captured in the present data. However, this might be expected to be rare because most youths received Medicaid eligibility based on low-income status.

In this large study of Medicaid-enrolled youths with opioid overdose, youths experienced incident and recurrent opioid overdose at high rates, and incident heroin overdose was associated with recurrent overdose. After incident overdose, less than one-third of youths received any timely addiction treatment, and only 1 in 54 youths received recommended evidence-based pharmacotherapy. Interventions are urgently needed to link youths to treatment after overdose, with priority placed on improving access to recommended pharmacotherapy.

Article Information

Accepted for Publication: October 16, 2019.

Corresponding Author: Rachel H. Alinsky, MD, MPH, Division of General Pediatrics and Adolescent Medicine, Johns Hopkins School of Medicine, 200 N Wolfe St, Baltimore, MD 21287 (ralinsk1@jhmi.edu).

Published Online: January 6, 2020. doi:10.1001/jamapediatrics.2019.5183

Author Contributions: Mr Rodean had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Alinsky, Zima, Rodean, Matson, Adger, Hadland.

Acquisition, analysis, or interpretation of data: Alinsky, Zima, Rodean, Matson, Larochelle, Bagley, Hadland.

Drafting of the manuscript: Alinsky.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Rodean.

Administrative, technical, or material support: Zima, Adger, Hadland.

Supervision: Zima, Matson, Adger, Hadland.

Conflict of Interest Disclosures: Dr Alinsky reported receiving a grant from the Eunice Kennedy Shriver National Institute of Child Health and Human Development. Dr Zima reported receiving support from the California Behavioral Health Center of Excellence. Dr Bagley reported receiving support from the National Institute on Drug Abuse and the Department of Medicine at Boston University School of Medicine. Dr Matson reported receiving grants from the National Institutes of Health during the conduct of the study. Dr Larochelle reported receiving support from Boston University School of Medicine Department of Medicine during the conduct of the study, grants from the National Institute on Drug Abuse, and research funding from OptumLabs outside the submitted work. Dr Hadland reported receiving support from the National Institute on Drug Abuse, the Thrasher Research Fund, and the Academic Pediatric Association. No other disclosures were reported.

Funding/Support: Dr Alinsky was supported by the T32 HD052459 grant from the Eunice Kennedy Shriver National Institute of Child Health and Human Development. Dr Zima was supported by the California Behavioral Health Center of Excellence. Dr Bagley was supported by the K23 DA044324 award from the National Institute on Drug Abuse, and a Career Investment Award from the Department of Medicine at Boston University School of Medicine. Dr Matson was supported by the K01 DA035387 award from the National Institute on Drug Abuse. Dr Larochelle was supported by the K23 DA042168 award from the National Institute on Drug Abuse, and a Career Investment Award from the Department of Medicine at Boston University School of Medicine. Dr Hadland was supported by the K23 DA045085 award and the L40 DA042434 award from the National Institute on Drug Abuse, an Early Career award from the Thrasher Research Fund, and a Young Investigator Award from the Academic Pediatric Association.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: Maria Trent, MD, MPH, Johns Hopkins School of Medicine, provided mentorship and guidance for this study. The Society for Adolescent Health & Medicine Mentoring Forum connected the lead authors of this study. No one was financially compensated for the stated contribution.

Unick GJ, Rosenblum D, Mars S, Ciccarone D. Intertwined epidemics: national demographic trends in hospitalizations for heroin- and opioid-related overdoses, 1993-2009. PLoS One. 2013;8(2):e54496. doi:10.1371/journal.pone.0054496 PubMed Google Scholar

Morizio KM, Baum RA, Dugan A, Martin JE, Bailey AM. Characterization and management of patients with heroin versus nonheroin opioid overdoses: experience at an academic medical center. Pharmacotherapy. 2017;37(7):781-790. doi:10.1002/phar.1902 PubMed Google Scholar Crossref

Ray BR, Lowder EM, Kivisto AJ, Phalen P, Gil H. EMS naloxone administration as non-fatal opioid overdose surveillance: 6-year outcomes in Marion County, Indiana. Addiction. 2018;113(12):2271-2279. doi:10.1111/add.14426 PubMed Google Scholar Crossref

Larochelle M, Bernson D, Land T, et al. Touch points prior to opioid overdose death. Paper presented at: Association for Multi-Disciplinary Education and Research in Substance Use and Addiction (AMERSA), 42nd Annual National Conference; November 9, 2018; San Francisco, California.

Larochelle MR, Bernson D, Land T, et al. Medication for opioid use disorder after nonfatal opioid overdose and association with mortality: a cohort study. Ann Intern Med. 2018;169(3):137-145. doi:10.7326/M17-3107 PubMed Google Scholar Crossref

Reardon JM, Harmon KJ, Schult GC, Staton CA, Waller AE. Use of diagnosis codes for detection of clinically significant opioid poisoning in the emergency department: a retrospective analysis of a surveillance case definition. BMC Emerg Med. 2016;16:11. doi:10.1186/s12873-016-0075-4 PubMed Google Scholar Crossref

Maeng DD, Han JJ, Fitzpatrick MH, Boscarino JA. Patterns of health care utilization and cost before and after opioid overdose: findings from 10-year longitudinal health plan claims data. Subst Abuse Rehabil. 2017;8:57-67. doi:10.2147/SAR.S135884 PubMed Google Scholar Crossref

Ali MM, Mutter R. Patients Who Are Privately Insured Receive Limited Follow-Up Services After Opioid-Related Hospitalizations: The CBHSQ Report. Rockville, MD: Substance Abuse and Mental Health Services Administration; 2013-2016. PubMed

Stein BD, Gordon AJ, Sorbero M, Dick AW, Schuster J, Farmer C. The impact of buprenorphine on treatment of opioid dependence in a Medicaid population: recent service utilization trends in the use of buprenorphine and methadone. Drug Alcohol Depend. 2012;123(1-3):72-78. doi:10.1016/j.drugalcdep.2011.10.016 PubMed Google Scholar Crossref

Mohlman MK, Tanzman B, Finison K, Pinette M, Jones C. Impact of medication-assisted treatment for opioid addiction on Medicaid expenditures and health services utilization rates in Vermont. J Subst Abuse Treat. 2016;67:9-14. doi:10.1016/j.jsat.2016.05.002 PubMed Google Scholar Crossref

American Society of Addiction Medicine. In: Mee-Lee D, ed. The ASAM Criteria: Treatment Criteria for Addictive, Substance-Related, and Co-Occurring Conditions. 3rd ed. Chevy Chase, MD: American Society of Addiction Medicine; 2013.

Curtin SC, Tejada-Vera B, Warmer M. Drug overdose deaths among adolescents aged 15–19 in the United States: 1999–2015. NCHS Data Brief. 2017;(282):1-8.PubMed Google Scholar

Marino L, Wissow LS, Davis M, Abrams MT, Dixon LB, Slade EP. Predictors of outpatient mental health clinic follow-up after hospitalization among Medicaid-enrolled young adults. Early Interv Psychiatry. 2016;10(6):468-475. doi:10.1111/eip.12206 PubMed Google Scholar Crossref

Substance Abuse and Mental Health Services Administration, Health Services Administration. Medicaid coverage and financing of medications to treat alcohol and opioid use disorders. https://store.samhsa.gov/system/files/sma14-4854.pdf. Published 2014. Accessed November 12, 2019.