Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Adolesc Health. Author manuscript; available in PMC 2013 June 1.
Published in final edited form as:
PMCID: PMC3368381

Mental health disorders and chronic opioid use among adolescents and young adults with chronic pain

Laura P. Richardson, MD, MPH,1,6 Joan E. Russo, PhD,2 Wayne Katon, MD,2 Carolyn A. McCarty, PhD,1,6 Andrea DeVries, PhD,5 Mark J. Edlund, MD, PhD,3 Bradley C. Martin, PharmD, PhD,4 and Mark Sullivan, MD, PhD2



The purpose of this study was to examine the association between mental health disorders and subsequent risk for chronic opioid use among adolescents and young adults presenting with common chronic pain complaints (back pain, neck pain, headache and arthritis/joint pain).


Using claims data from January 1, 2001 to June 30, 2008, we conducted a longitudinal analysis of opioid use patterns among 13–24 year-olds presenting with a new episode of chronic pain. Chronic opioid use was defined as receiving >90 days of opioids within a 6-month period with no gap in use of >30 days in the 18 months following the fist qualifying pain diagnosis. Mental health disorders were identified from claims in the 6 months prior to the first qualifying pain diagnosis.


59,077 youth met criteria for a new episode of chronic pain. Among these youth, 321 (0.5%) met criteria for chronic opioid use and 16,172 (27.4%) had some opioid use. After controlling for demographic and clinical factors, youth with pre-existing mental health diagnoses had a 2.4-fold increased risk of subsequently receiving chronic opioids versus no opioids (OR: 2.36, 95% CI = 1.73–3.23) and a 1.8-fold increased likelihood of receiving chronic opioids versus some opioids (OR: 1.83, 95% CI= 1.34–2.50).


Mental health disorders are associated with increased risk for chronic opioid use among adolescents and emerging young adults. Further study is warranted to examine risks and benefits of chronic opioid use in this population.

Chronic non-cancer pain conditions (CNCP), such as back pain and headache, increase in prevalence during puberty. In one study, 28% of all adolescents (ages 11–17 years) and 50–60% of post-pubertal adolescents reported at least one pain condition that lasted a whole day or more or that occurred several times in the prior 3 months.1 Thirty-percent of post-pubertal females and 16% of post-pubertal males reported two or more pain conditions. Studies suggest that there is considerable diversity in opioid prescribing practices for this age group.2

Chronic use of opioids for CNCP has increased among adults in the last two decades.3,4 Among adults, it is estimated that 3% of the US general population without cancer receives opioids more than 30 days per year and that 90–95% of this “chronic opioid treatment” is prescribed for CNCP.5 Data regarding chronic use of opioids among youth and young adults is limited. In one prior study using data from 2005, we found that approximately 21% of commercially-insured youth with back pain, neck pain, headache, or joint pain had received an opioid prescription while 2.4% had received 30 days or more of opiates and 0.8% received 60 days or more.6

Although the balance of benefits versus risks from chronic opioid therapy in non-cancer pain is controversial, recent studies suggest that risk of overdose increases with increased dosages of chronic opioid therapy7 and that the misuse of opioid medications is more common among younger adults, as well as those with back pain and multiple pain complaints and who have a history of substance abuse.8 A 5-year community follow-up study showed that patients with chronic pain and anxiety/depressive disorders or substance abuse were approximately 2-fold and 3-fold more likely, respectively, to use opioids and that these patients are also more likely to remain on opioid medications than those without one of these disorders.9 Mental health disorders have also been shown to be associated with increased risk for opioid abuse and dependence among patients with chronic pain.10 Studies in the community suggest that there may be a bidirectional association such that mental health disorders both predict and are predicted by non-medical prescription and opioid use.1113 These findings have led to the postulation that “adverse selection” occurs in the prescription of opioid medications, meaning that patients at high risk for abuse and dependence are more likely to receive chronic opioid therapy.14 This is the opposite of the “careful selection” of patients for chronic therapy recommended in treatment guidelines.15

The onset of depression and substance use peak during the adolescent and young adult period 16 but chronic opioid use is relatively rare in this age group.6 Understanding how mental health disorders in adolescence and young adulthood are associated with subsequent risk for opioid use might allow for targeted interventions to reduce the risk for future chronic opioid use. The purpose of the current study is to examine if adolescents and young adults with pre-existing mental health disorders are more likely to be prescribed an opioid after presenting with chronic pain, and to determine if adolescents and young adults with mental health disorders are also more likely to subsequently become chronic users of opioids.


This study uses data from HealthCore Integrated Research Database (HIRDsm) which contains administrative claims and health plan eligibility data from 12 large commercial health plans representing the West, Midwest, and Southeast regions of the United States. The dataset contains information from January 1, 2001 to June 30, 2008. The claims files describe all transactions between providers and the health plans including paid claims for physicians and other clinicians, hospital, and outpatient prescriptions. All research procedures were approved by the University of Washington IRB and an IRB used by HealthCore.

Inclusion Criteria

Inclusion criteria for the study were being 13–24 years of age and having one of the most common chronic non-cancer pain (CNCP) conditions for which long-term opioids are used: back pain, neck pain, headache, and arthritis pain.17 The following criteria were used to identify new episodes of CNCP using HealthCore claims data:

  1. Two or more claims containing primary or secondary diagnoses of the same type of CNCP (back pain, neck pain, headache, arthritis) that occurred at least 1 month apart but were not separated by more than 1 year occurring between July 1, 2001 and December 31, 2006.
  2. No CNCP diagnoses of the same type and no use of opioids in the 6 months prior to the first qualifying pain diagnosis
  3. 6 months of continuous eligibility prior to and 18 months of continuous eligibility after the first qualifying diagnosis
  4. youth were excluded if they had HIV or a cancer diagnosis (other than nonmelanoma skin cancer)

Predictor Variables

Main Predictor Variables

Mental health, substance use disorder, and information on number of pain conditions were collected from the 6 months prior to the index pain condition. The “index pain episode” was defined as the first episode of back pain, neck pain, headache, or arthritis pain that resulted in inclusion in the data set.

Number of Pain Conditions

Variables were generated for the total number of pain conditions within 15 categories of pain occurring in the six months prior to the index pain episode. By study design, participants could only have pain in 14 of the 15 categories since youth were excluded if they had the index pain condition in the 6-months prior to the first qualifying diagnosis. Pain condition categories examined include: headache, back pain, neck pain, osteoarthritis, rheumatoid arthritis, chest pain, kidney/gallstone pain, gynecologic pain, temporomandibular disorder, extremity pain, abdominal pain, neuropathy, fibromyalgia, fracture/contusion/injury, and “other pain” (ICD-9 codes for each category are provided in Table 1). Number of pain categories was included in analyses as a continuous variable.

Table 1
ICD-9 Codes Used to Define Presence of Pain Conditions

Mental Health and Substance Use Disorders

Mental health and substance use disorders were identified using claims data and the Mental Health Substance Abuse Classification Software developed by the Agency for Healthcare Research and Quality. This software assigns variables that identify mental health and substance use-related diagnoses using the diagnostic coding of ICD-9-CM to generate general categories of mental health conditions. For analyses, youth were categorized as having 1 or more mental health disorder types versus none. Substance use disorder types were examined separately from mental health disorders and were similarly classified as 0 and 1 or more.

Main Outcome - Opioid Use

Chronic opioid use was defined as receiving >90 days of opioid use within a 6-month period with no gap in use of >30 days. All youth who had any opioid prescription during the 18 months following the index diagnosis but did not meet criteria for chronic use were included in the “non-chronic use” category. Youth in the “no use” category had no opioid prescriptions during the 18 months following index diagnosis.


Demographics and Community Level Variables

Age and gender came from eligibility files. Individual race and income information was not available. As a proxy, Geocodes were used to obtain the following census tract level variables: median community household income, percent of community residents who were White, and percent of community residents with some college education. These were dichotomized as being above or below median Census values. “Higher income communities” were defined as those with a median household income above the 2007 median household income of $50,233. “More White resident communities” were defined as having more White residents than the national percent of residents identifying as White in 2000 census data (>75.1%). “Higher education level communities” were defined as having a higher percentage of residents with any college than the national average in 2008 (≥54.8%) The Charlson Comorbidity Index, an index of comorbid disease severity based on administrative data that has been shown to be associated with 1-year mortality, was used to adjust for the presence chronic diseases that might impact overall health status and opioid use.18


Youth were categorized into three groups based on opioid use in the 18 months following onset of a new episode of the index pain condition as outlined above. Demographic, community level, and clinical variables were examined for youth in each of the three categories. Univariate comparisons were conducted examining for group differences using Chi-square analyses and analyses of variance (ANOVAs) for categorical and continuous variables respectively. Subsequently adjusted logistic regression analyses were conducted examining the association between mental health conditions, number of pain conditions, and chronic and non-chronic opioid use adjusting for potential covariates. Models were constructed separately for three comparisons: chronic users versus non-users, chronic users versus non-chronic users, and non-chronic users versus non-users. Covariates were selected a priori and included: age, gender, community characteristics (race, income, and education level), chronic disease status, and type of index pain condition.


There were 62,560 youth between 13–24 years who met study inclusion criteria. 3,407 youth with a diagnosis of cancer and 76 people with HIV were excluded from the final sample, leaving a final sample of 59,077. Among these, 321 (0.5%) met criteria for new onset chronic opioid use in the 18 months following diagnosis, 16,172 (27.4%) had some opioid use but did not meet criteria for chronic opioid use, and 42,584 (72.1%) had no opioid use.

Chronic use was more common among males, older youth, and youth who lived in communities that were poorer, had more White residents, and had fewer residents who had attended college (Table 2). Among all opioid use categories (no use, non-chronic use and chronic use), the most common qualifying index pain diagnoses, in order of prevalence, were back pain, headache, neck pain, and arthritis. When compared to non-opioid users, youth who had received opioids (chronic or non-chronic users) had significantly higher numbers of pain conditions in the 6-months prior to the index pain condition.

Table 2
Demographic and Health Indicators by Opioid Use Category

Mental health diagnoses were almost twice as common among youth with chronic use when compared to the non-use category and substance use diagnoses were over 5 times more likely amongst youth with chronic use compared to those in the no use category. Overall, 17.1% of youth in the chronic use category had a mental health or substance use diagnosis compared to 10.6% in the non-chronic use category and 8.2% for those in the non-user category. The main difference in prevalence of mental health diagnoses was attributable to higher rates of anxiety disorders and major depressive disorders among chronic users and non-chronic users when compared to non-users (Table 2).

Table 3 shows the results of adjusted analyses of 3 pair-wise comparisons based on a full model. After adjusting for relevant covariates, male sex, older age, and community factors (higher percent white residents and lower rate of college educated residents) remained significant predictors of chronic opioid use. The type of index pain condition and number of pain conditions were not significant predictors of chronic use while presence of a mental health diagnosis continued to be a strong predictor of chronic use. In these adjusted models, youth with a mental health diagnosis were more than 2 times as likely to be a chronic user than a non-user (OR = 2.36, 95% CI – 1.73 – 3.23) and were 1.8 times more likely to be a chronic user than a non-chronic user (OR = 1.83, 95% CI – 1.34 – 2.50).

Table 3
Adjusted Odds ratios for Pair-wise Comparisons based on a Full Model

There was slightly different pattern of factors associated with being a non-chronic user. Like chronic users, youth in the non-chronic use category were significantly older, and were more likely to come from communities with a higher percentage of white residents and a lower percentage of residents who had at least some college education than those in the non-user category. However, unlike chronic users, youth in the non-chronic user category were more likely to be female, had significantly higher Charlson Comorbidity Index scores, and had significantly more pain conditions than non-users. There was a negative association between having neck pain or back pain as a qualifying diagnosis and being in the non-chronic use category. The association between mental health disorders and non-chronic use was of lower magnitude than that seen for chronic user versus non-use (OR = 1.26, 95% CI – 1.18–1.34) but was still significant. Substance use diagnoses were too few to be examined independently in logistic regression analyses and were not included in the final adjusted models.


This is the first analysis to examine the association between mental health disorders and subsequent risk for chronic opioid use among adolescents and young adults with a new episode of chronic pain in a longitudinal manner. We found that, when presenting with a new episode of chronic pain, youth with pre-existing mental health disorders were at greater than 2-fold increased risk of becoming chronic opioid users when compared to no use and were at 1.8-fold increased risk for having some opioid use compared to no use. The risk of chronic use was higher in males and increased with age. We also found that community factors were predictive of the likelihood of chronic and non-chronic use of opioids suggesting that use may vary by community.

The level of association between mental health disorders and opioid use in our study is similar to the results seen in adult populations. Adults with mental health disorders have been found to be at 3-fold increased risk for the initiation of opioid use and greater than 2-fold increased risk for continuing opioids once they were started.9 Mental health disorders have been found to occur in between 54–66% of individuals entering treatment for substance abuse with depression and anxiety as the most common disorders.19 Adults with mental health disorders have also been shown to have an increased likelihood of prescription drug abuse10,20,21 and non-medical use of prescription opioids.22 This finding coupled with the increased likelihood of chronic use in this population has led to the theory that “adverse selection” is occurring, meaning that patients at high risk for poor outcomes of chronic opioid therapy are more likely to receive chronic opioid therapy.14

There are many possible reasons why adolescents and young adults with mental health disorders may be more likely to be prescribed opioids and to become chronic users. First, depression and anxiety have been shown to be associated with an increased number of physical symptoms, including pain complaints. Pain and other aversive physical symptoms associated with chronic medical disorders are rated as more severe in the presence of major depression.23 The higher level of use and chronicity of opioid treatment may be associated with a higher level of physical complaints in this population. Physicians may also be more willing to use chronic opioid therapy for patients with mental health disorders if they attribute symptoms and functional impairment as being related to the pain. Finally, anxiety and depression are more common among parents of youth with mental health disorders which may result in increased treatment seeking on the part of parents.

Few studies have examined the safety or effectiveness of chronic opioid treatment among adolescents and young adults with chronic pain. In one case series of patients from a multidisciplinary pain clinic, chronic opioid use was associated with improved outcomes and no negative consequences.24 This suggests that there may be a role for chronic opioids for the treatment of some patients. However, it is important to note that each of the youth in this case series underwent an intake including psychological assessment and psychological resources were available for follow up assessments. Our finding that chronic opioid use is more common among adolescents and young adults with anxiety and depressive disorders emphasizes the importance of screening for mental health concerns and developing strategies for addressing these concerns in this population.

This study has the following limitations. First, because we used administrative data on pharmacy fills we were unable to examine the actual cause for prescriptions, whether patients were taking medications as prescribed, appropriateness of treatment, response to treatment, or impact on quality of life. Second, the index CNCP conditions with which subjects were included in this study may not represent the full spectrum of chronic painful conditions for which adolescents may be prescribed opioids. Although we tried to identify “new episodes” using a 6-month no diagnosis window it is also possible that some conditions were pre-existing but that youth had not sought care in the prior six months. Additionally, we were unable to assess the role of pain severity or activity interference in the initiation of opioid therapy. Third, we did not have subject-level data on race, ethnicity or income in order to better adjust for these variables. Fourth, mental health and substance use diagnoses are based on coded diagnosis within a 6-month window which may have resulted in misclassification either due to provider reluctance to code diagnoses or lack of treatment episodes for a pre-existing disorder in the 6-month window. Studies have shown that, particularly for adolescents, providers may be reluctant to code these diagnoses.25 Administrative data is likely to pick up youth with the most severe and persistent disorders and may not capture the whole range of mental health or substance use disorders. Studies using administrative data have found that only about 20% to 40% of depression and anxiety disorders among adolescents have been accurately diagnosed by primary care physicians.26,27 Finally, we did not have a large enough sample with substance use diagnoses to examine this as a separate predictor in multivariate analyses.

Despite these limitations, this study has important clinical implications. Adolescents and young adults with chronic pain and comorbid mental health disorders are at increased risk for chronic prescription opioid use. Other studies have suggested that adults with mental health disorders are at increased risk for prescription opioid abuse as well. Our study suggests that, as with adult populations, there may be “adverse selection” of higher risk adolescent and emerging adults into chronic opioid therapy. Providers should screen for mental health disorders prior to starting these medications for chronic pain conditions and should consider referring patients with comorbid depression or anxiety for concurrent mental health treatment either in a multidisciplinary pain clinic setting or in collaboration with a mental health provider in the community when a multidisciplinary clinic is not available.


This work was supported by grants from the Alcohol and Drug Abuse Institute at the University of Washington and from the National Institute on Drug Abuse (NIDA R01 DA022560-01). The authors would like to thank Anna Wallace for her diligent work in constructing the dataset.


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


1. LeResche L, Mancl LA, Drangsholt MT, Saunders K, Korff MV. Relationship of pain and symptoms to pubertal development in adolescents. Pain. 2005 Nov;118(1–2):201–209. [PubMed]
2. Saroyan JM, Cheng WY, Taylor DC, Afzal A, Sonty N, Sullivan MA. Select practice behaviors of clinicians on the use of opioids for adolescents with subacute and chronic nonmalignant pain. J Opioid Manag. 2011 Mar-Apr;7(2):123–134. [PubMed]
3. Caudill-Slosberg MA, Schwartz LM, Woloshin S. Office visits and analgesic prescriptions for musculoskeletal pain in US: 1980 vs. 2000. Pain. 2004 Jun;109(3):514–519. [PubMed]
4. Sullivan MD, Edlund MJ, Fan MY, Devries A, Brennan Braden J, Martin BC. Trends in use of opioids for non-cancer pain conditions 2000–2005 in commercial and Medicaid insurance plans: the TROUP study. Pain. 2008 Aug 31;138(2):440–449. [PMC free article] [PubMed]
5. Sullivan MD, Edlund MJ, Steffick D, Unutzer J. Regular use of prescribed opioids: association with common psychiatric disorders. Pain. 2005 Dec 15;119(1–3):95–103. [PubMed]
6. Richardson LP, Fan MY, McCarty CA, et al. Trends in the Prescription of Opioids for Adolescents with Non-cancer Pain. (under review) [PMC free article] [PubMed]
7. Dunn KM, Saunders KW, Rutter CM, et al. Opioid prescriptions for chronic pain and overdose: a cohort study. Ann Intern Med. Jan 19;152(2):85–92. [PMC free article] [PubMed]
8. Sullivan MD, Edlund MJ, Fan MY, Devries A, Brennan Braden J, Martin BC. Risks for possible and probable opioid misuse among recipients of chronic opioid therapy in commercial and medicaid insurance plans: The TROUP Study. Pain. Aug;150(2):332–339. [PMC free article] [PubMed]
9. Sullivan MD, Edlund MJ, Zhang L, Unutzer J, Wells KB. Association between mental health disorders, problem drug use, regular prescription opioid use. Arch Intern Med. 2006 Oct 23;166(19):2087–2093. [PubMed]
10. Edlund MJ, Martin BC, Fan MY, Devries A, Braden JB, Sullivan MD. Risks for opioid abuse and dependence among recipients of chronic opioid therapy: Results from the TROUP Study. Drug Alcohol Depend. Jul 13; [PMC free article] [PubMed]
11. Schepis TS, Hakes JK. Non-medical prescription use increases the risk for the onset and recurrence of psychopathology: results from the National Epidemiological Survey on Alcohol and Related Conditions. Addiction. 2011 Dec;106(12):2146–2155. [PubMed]
12. Schepis TS, McCabe SE. Exploring age of onset as a causal link between major depression and nonmedical use of prescription medications. Drug Alcohol Depend. 2011 Jul 27; [PMC free article] [PubMed]
13. Martins SS, Keyes KM, Storr CL, Zhu H, Chilcoat HD. Pathways between nonmedical opioid use/dependence and psychiatric disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Drug Alcohol Depend. 2009 Jul 1;103(1–2):16–24. [PMC free article] [PubMed]
14. Sullivan MD. Who gets high-dose opioid therapy for chronic non-cancer pain? Pain. Dec;151(3):567–568. [PubMed]
15. Chou R, Fanciullo GJ, Fine PG, et al. Clinical guidelines for the use of chronic opioid therapy in chronic noncancer pain. J Pain. 2009 Feb;10(2):113–130. [PMC free article] [PubMed]
16. McGorry PD, Purcell R, Goldstone S, Amminger GP. Age of onset and timing of treatment for mental and substance use disorders: implications for preventive intervention strategies and models of care. Curr Opin Psychiatry. 2011 Jul;24(4):301–306. [PubMed]
17. Boudreau D, Von Korff M, Rutter CM, et al. Trends in long-term opioid therapy for chronic non-cancer pain. Pharmacoepidemiol Drug Saf. 2009 Aug 28; [PMC free article] [PubMed]
18. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–383. [PubMed]
19. Cicero TJ, Lynskey M, Todorov A, Inciardi JA, Surratt HL. Co-morbid pain and psychopathology in males and females admitted to treatment for opioid analgesic abuse. Pain. 2008 Sep 30;139(1):127–135. [PubMed]
20. Becker WC, Fiellin DA, Gallagher RM, Barth KS, Ross JT, Oslin DW. The association between chronic pain and prescription drug abuse in Veterans. Pain Med. 2009 Apr;10(3):531–536. [PubMed]
21. Edlund MJ, Steffick D, Hudson T, Harris KM, Sullivan M. Risk factors for clinically recognized opioid abuse and dependence among veterans using opioids for chronic non-cancer pain. Pain. 2007 Jun;129(3):355–362. [PubMed]
22. Becker WC, Sullivan LE, Tetrault JM, Desai RA, Fiellin DA. Non-medical use, abuse and dependence on prescription opioids among U.S. adults: psychiatric, medical and substance use correlates. Drug Alcohol Depend. 2008 Apr 1;94(1–3):38–47. [PubMed]
23. Katon W, Lin EH, Kroenke K. The association of depression and anxiety with medical symptom burden in patients with chronic medical illness. Gen Hosp Psychiatry. 2007 Mar-Apr;29(2):147–155. [PubMed]
24. Slater ME, De Lima J, Campbell K, Lane L, Collins J. Opioids for the management of severe chronic nonmalignant pain in children: a retrospective 1-year practice survey in a children's hospital. Pain Med. Feb;11(2):207–214. [PubMed]
25. Rushton JL, Felt BT, Roberts MW. Coding of pediatric behavioral and mental disorders. Pediatrics. 2002;110(1 Pt 1):e8. [PubMed]
26. Richardson LP, Russo JE, Lozano P, McCauley E, Katon W. Factors associated with detection and receipt of treatment for youth with depression and anxiety disorders. Acad Pediatr. Jan-Feb;10(1):36–40. [PMC free article] [PubMed]
27. Katon WJ, Richardson L, Russo J, Lozano P, McCauley E. Quality of mental health care for youth with asthma and comorbid anxiety and depression. Med Care. 2006 Dec;44(12):1064–1072. [PubMed]