PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
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 November 1.
Published in final edited form as:
PMCID: PMC3479410
NIHMSID: NIHMS356661

Motivational subtypes of nonmedical use of prescription medications: Results from a national study

Abstract

Purpose

Very little research has examined the heterogeneity associated with the nonmedical use of prescription medications in nationally representative samples of adolescents. The main objectives of this study were 1) to identify motivational subtypes of past-year nonmedical use of prescription medications among high school seniors in the United States using a person-centered approach, and 2) examine the associations among motivational subtypes and characteristics of substance abuse (i.e., route of administration, co-ingestion, and subjective high).

Methods

Self-administered questionnaires as part of the Monitoring the Future study were completed by nationally representative samples of high school seniors (modal age 18). The sample consisted of five cohorts (senior years of 2002–2006) made up of 12,431 high school seniors in total, of which 53% were women.

Results

Approximately 75% of past-year nonmedical users of prescription opioids, stimulants, and tranquilizers endorsed more than one motive. Latent class analysis indicated five motivational subtypes associated with nonmedical use of prescription opioids (experiment, relax, get high, pain relief, and affect regulation), four subtypes of prescription stimulants (weight loss/enhance energy, enhance energy/awake/high, experiment, and affect regulation), and five subtypes of prescription tranquilizers (experiment, get high, relax/sleep, relax, affect regulation). Recreational subtypes were positively associated with characteristics of substance abuse while self-treatment subtypes were associated with medical use prior to nonmedical use.

Conclusions

Because multiple motives underlie nonmedical use of prescription medications, identifying subgroups of individuals who endorse combinations of motives, versus a single motive, will better inform intervention efforts to reduce nonmedical prescription medication use.

Keywords: Epidemiology, Adolescents, Prescription drugs, Prescription medications, Latent class analysis, Motivations, Route of administration, Co-Ingestion, Subjective high, High school

Introduction

Several studies have shown that the nonmedical use of prescription medications (NUPM) has significantly increased over the past two decades and is most prevalent among adolescents and young adults [16]. Despite these recent increases, very little research has examined the motivations for NUPM in nationally representative samples [7]. Many existing studies often fail to distinguish between individuals who nonmedically use someone else’s prescription medications to self-treat a medical condition and those who use someone else’s prescription medications recreationally [8]. The findings from at least two regional studies indicate that motivations for NUPM are varied and associated with different adverse consequences among adolescents and young adults [9,10]. These findings warrant additional investigations within national samples of adolescents, because there is a lack of epidemiological research that accurately assesses the motivations and characteristics of those individuals at greatest risk for substance abuse [7,11,12].

At least four studies have examined individual motivations associated with nonmedical use of prescription opioids stimulants, and tranquilizers among secondary students [9,1315]. All of these studies used variable-centered analytic approaches and focused on individual motivations. Because individuals are likely to have multiple motivations underlying their behavior, identifying categories of individuals who endorse combinations of motivations, versus a single motivation, offers a valuable complementary approach to variable-centered approaches. Muthen and Muthen (2000) elaborated the distinction between variable-centered and person-centered statistical methods [16]. Variable-centered approaches focus on the relationships between variables while person-centered approaches focus on the relationships among persons. In contrast to variable-centered approaches, the goal of person-centered approaches “is to group individuals into categories, each one of which contains individuals who are similar to each other and different from individuals in other categories” [16]. Examples of person-centered approaches include cluster analysis and latent class analysis (LCA), each of which can be used to group individuals into similar categories according to their motives for nonmedical use (referred to as motivational classes or subtypes). These methods allow for a more nuanced understanding of why adolescents engage in NUPM.

LCA has been previously utilized in the analysis of illicit drug use [17], alcohol use disorders [18,19], adolescent drinking [20], antisocial personality disorder symptoms among alcohol-dependent subjects [21], adolescent sedative/anxiolytic misuse [22], and the comorbidity of adolescent problem behaviors [23]. For example, an investigation examined the relationships between different patterns of drinking motivations and behaviors in a sample of U.S. 12th grade students using a person-centered approach [20]. The results of this study identified four motivational classes for alcohol use, including Experiment, Thrill-seek, Multi-reason, and Relax. The results also indicated that the riskiest drinking behaviors were related to membership in the Multi-reason class. Despite these recent advances in the alcohol literature, motivational subtypes of NUPM among adolescents have not been examined using person-centered approaches.

The objectives of this study were to 1) identify the motivational subtypes of NUPM within U.S. high school seniors using a person-centered approach, and 2) examine variations in these motivational subtypes of NUPM by characteristics of substance abuse (i.e., route of administration, co-ingestion and subjective high).

Methods

Participants and procedures

The Monitoring the Future study (MTF) annually surveys a cross-sectional, nationally representative sample of high school seniors in approximately 135 public and private schools in the coterminous U.S. MTF uses a multi-stage sampling procedure: In stage 1, geographic areas or primary sampling units are selected; in stage 2, schools within primary sampling units are selected (with probability proportionate to class size); and in stage 3, students within schools are selected. The student response rates for high school seniors ranged from 82% to 83% between 2002 and 2006. Because so many questions are included in the MTF, much of the questionnaire content is divided into six different questionnaire forms which are randomly distributed. This approach results in six virtually identical subsamples. The measures relevant for this study were asked on Form 1, so this study focuses on the subsamples receiving Form 1 within each cohort (for more information on the procedures see [3]).

The total sample size was 12,431 high school seniors. Sample sizes for each year were as follows: 2002 (N=2,256); 2003 (N=2,556); 2004 (N=2,563); 2005 (N=2,581); 2006 (N=2,475). The sample included 53% women, 62% white, 10% Black, and 28% were from other racial groups or did not specify their race. The modal age of the individuals in the sample was 18 years of age. Subsamples for the current study are described in detail below.

Measures

Nonmedical use of prescription medications (NUPM) was assessed with a series of items asking respondents on how many occasions (if any) in the past 12 months they used prescription medications on their own, without a doctor’s orders. There were separate questions for each prescription medication class: (a) Prescription opioids (e.g., Vicodin®, OxyContin®, Percodan®, Percocet®, Demerol®, Dilaudid®, methadone, morphine, codeine); (b) Prescription tranquilizers (e.g., Librium®, Valium®, Xanax®); and (c) Prescription stimulants (e.g., Ritalin®, Dexedrine®). The response scale ranged from 1) no occasions to 7) 40 or more occasions.

Motives for NUPM were assessed by asking respondents who reported past-year NUPM to indicate the most important reasons for NUPM from a check all that apply list of binary items.

Subjective high of NUPM was measured with three items that asked past-year nonmedical users how high they usually get when they use each prescription medication class. The response scale for these items ranged from 1) not at all high to 4) very high.

Co-ingestion of NUPM and other drugs was measured with 9 items focused on the number of times each prescription medication class were used nonmedically at the same time as other drugs so that the effects overlapped (e.g., alcohol, marijuana, LSD, hallucinogens other than LSD, prescription opioids, stimulants, tranquilizers, cocaine, heroin) in the past 12 months. The response scale ranged from 1) not at all to 7) every time.

Routes of administration of NUPM were assessed with five items that asked which methods respondents used for past-year NUPM. The binary items included: 1) intranasal (snorting or sniffing); 2) smoking; 3) injection; 4) orally (by mouth); and 5) other.

Medical use of prescription medications was assessed by asking respondents whether they had ever taken each prescription medication class because a doctor told them to use them. Respondents were informed that prescription medications are sometimes prescribed by doctors and drugstores are not supposed to sell them without a prescription. The response scale included: 1) No medical use; 2) Medical use prior to NUPM; and 3) NUPM prior to medical use.

Data analysis

Questions about motives for NUPM were asked only of those who reported nonmedical use in the past year. Accordingly, analyses focused on participants who reported past-year nonmedical use of opioids (N=768), stimulants (N=782), and/or tranquilizers (N=541) for the years 2002–2006. We used the SAS PROC LCA program [24] to conduct a latent class analysis of the 11 most prevalent motivation items for past-year NUPM of each class, using data from the years 2002 through 2006 (6 items with base rates less than 10% were excluded from prescription opioids and 5 items with base rates less than 12% were excluded from prescription stimulants, and 4 items with base rates less than 10% were excluded from prescription tranquilizers). Beginning with a baseline LCA model, we added classes sequentially and used the likelihood ratio G2 statistic, Akaike’s Information Criterion (AIC) [25], the Bayesian Information Criterion (BIC) [26], model interpretability and parsimony to assess model fit at each step [16,24,27]. Item parameters (i.e., conditional item probabilities that indicate the probability that a given individual in a given latent class endorsed that item) and class probability parameters (i.e., the relative prevalence of each latent class) were estimated [24,28,29]. All LCAs were exploratory, i.e., no a priori constraints were imposed on any of the parameter estimates [28]. After baseline models of motives for each class of prescription medications were identified, we used a LCA with covariates framework to test predictors of latent class membership [24,27].

Results

Sample Representativeness

A binary variable distinguished between participants who reported past-year nonmedical use of at least one prescription medication (n=1,399, 11.3% of the sample) and those who did not (n=11,032, 88.7% of the sample). Past-year NUPM did not vary by sex (females = 11.3%, males = 11.2%); age (<18 years = 11.0%, 18 years or older = 11.4%); or mother’s or father’s education. However, prevalence of past-year NUPM was higher among whites (14.6%) compared to blacks (2.8%), χ2 (1)=128.1, p <. 05; also, nonmedical users were significantly more likely to report low average high school grades, χ2 (8)=103.31, p <. 05.

Prevalence of Motives for NUPM

We estimated the prevalence of NUPM and motives of three prescription medication classes among U.S. high school seniors. The past-year prevalence of NUPM was 8.0% for prescription opioids, 7.2% for stimulants, and 5.3% for tranquilizers. The motives for nonmedical use were different for each prescription medication class (see Table 1). Notably, approximately 75% of nonmedical users for each prescription medication class endorsed more than one motive.

Table 1
Prevalence of Motives for Nonmedical use of Three Prescription Medication Classes among High School Seniors

Latent Classes based on Motives for NUPM

Results from LCA of motives for past-year nonmedical use of prescription opioids indicated that, compared to the baseline model (G2=2018.2, AIC=2040.2, BIC=2091.3), a 5-class solution provided the best fit to the data (G2=894.1, AIC=1012.1, BIC=1286.1) and yielded interpretable classes. Item-response probabilities and latent class membership probabilities for motives of nonmedical use of prescription opioids are presented in Table 2. The most prevalent class was comprised of individuals who had a higher probability of endorsing “to experiment” relative to the other motives, and thus Class 1 was considered the “Experiment” (41.2%) class. Individuals in Class 2, the “Relax” (25.6%) class, were distinguished by relatively high probabilities of endorsing the “to relax” motive combined with low probabilities of endorsing motives related to negative affect reduction. Individuals in Class 3, the “Get High” (14.4%) class, had a relatively high probability of endorsing “to experiment,” but unlike Class 1, the motives “to feel good or get high” and “to have a good time with friends” also had high probabilities. Class 4, the “Pain Relief” (11.4%) class, was characterized by relatively high probabilities of endorsing “pain relief,” combined with low to moderate probabilities for the other motives. Finally, individuals in Class 5, the “Affect Regulation” (7.4%) class were distinguished by relatively high probabilities of endorsing motives related to regulation of positive and negative emotions.

Table 2
Item-Response Probabilities for Latent Class Membership for Nonmedical Prescription Medication Motives

Results from LCA of motives for past-year nonmedical use of prescription stimulants showed that, compared to the baseline model (G2=2536.3, AIC=2558.3, BIC=2609.5), a 4-class solution provided the best fit to the data (G2=1133.0, AIC=1227.0, BIC=1446.1) and yielded interpretable classes. As shown in Table 2, the most prevalent class was comprised of individuals who had moderate probabilities of endorsing “to lose weight” and “to get more energy,” along with relatively low item-response probabilities for other motives, so we labeled Class 1 the “Weight Loss/Enhance Energy” (38.0%) class. Individuals in Class 2, the “Experiment” (25.3%) class, showed a moderate probability of endorsing the “to experiment” motive relative to the other motives. Individuals in Class 3, the “Energy/Awake/High” (24.3%) class, had relatively high probabilities of endorsing “to get high,” “to have a good time with friends,” “to experiment,” “to get more energy,” and “to stay awake.” Finally, individuals in Class 4, the “Affect Regulation” (12.4%) class, had relatively high probabilities of endorsing most of the 11 motives; however, this class was distinguished from the other classes by relatively high probabilities of selecting motives related to affect regulation.

Results from LCA of motives for past-year nonmedical use of prescription tranquilizers indicated that, compared to the baseline model (G2=1691.5, AIC=1713.5, BIC=1760.7), a 5-class solution provided the best fit to the data (G2=806.1, AIC=924.1, BIC=11177.4) and yielded interpretable classes (see Table 2). The most prevalent class was comprised of individuals who had a moderate probability of reporting “to experiment” relative to other motives, and we labeled Class 1 the “Experiment” (30.0%) class. The “Get High” class (23.5%) was characterized by high item-response probabilities for motives to “get high,” and moderate item-response probabilities for “relax/relieve tension,” and “experiment.” By contrast, the “Relax” class (22.1%) was characterized by moderate item-response probabilities for “relax/relieve tension” and “relieve physical pain,” along with low item-response probabilities for all other motives. The next most prevalent class was comprised of individuals who had a relatively high probability of endorsing “to relax/relieve tension” and a moderate item-response probability for “to help sleep,” and we labeled Class 4 the “Relax/Sleep” (16.6%) class. Finally, individuals in Class 5, the “Affect Regulation” (8.2%) class, had moderate to high probabilities of endorsing all 11 motives, especially the affect regulation motives.

Correlates of Latent Class Membership

Having identified adequate baseline models for motives for past-year NUPM, we then used a latent class with covariates framework [24,27] to examine predictors of latent class membership. We focused on the following covariates: sex, subjective high, co-ingestion, route of administration, and history of medical use of each prescription medication. Given the exploratory nature of this research, analyses were limited to testing bivariate associations between each correlate and latent class membership. That is, each correlates were tested individually in separate models. Because all baseline models included more than two latent classes, these analyses took the form of bivariate multinomial regression analyses with latent class membership as the dependent variable. For comparison purposes, the “experiment” class was treated as the reference group across all three medications.

Sex

Results indicated that being female was associated with higher odds of membership in a) the “relax” and “pain relief” classes for opioid motives; b) the “energy/awake/high,” “affect regulation,” and “lose weight” motive classes for stimulant motives; and c) the “relax” and “relax” classes for tranquilizer motives, relative to the “experiment” classes. Sex did not predict membership in the “get high” motive classes for any of the medications.

Subjective High

Subjective high was associated with higher odds of membership in a) the “get high” class for opioid motives; b) the “energy/awake/high” and “affect regulation” motive classes for stimulant motives; and c) the “get high” and “affect regulation” classes for tranquilizer motives, relative to the “experiment” classes. By contrast, subjective high was associated with lower odds of membership in a) the “pain relief” class for opioid motives; b) the “lose weight/energy” motive class for stimulant motives; and c) the “relax” class for tranquilizer motives, relative to the “experiment” classes.

Co-Ingestion

Co-ingestion was associated with a) lower odds of membership in the lose weight/energy” motive class for stimulant motives; b) higher odds of membership in the “get high” and “affect regulation” classes for tranquilizer motives; and c) lower odds of membership in the “relax/sleep” and “relax” classes for tranquilizer motives, relative to the “experiment” classes. Small cell sizes made it impossible to estimate models of co-ingestion as a predictor of class membership for opioid motives.

Route of Administration

Non-oral route of administration was associated with higher odds of membership in the “get high” class for opioid motives. By contrast, non-oral route of administration was associated with lower odds of membership in a) the “relax,” “pain relief,” and “affect regulation” motive classes for opioid motives; and b) the “lose weight/energy” class for stimulant motives, relative to the “experiment” classes.

History of Medical Use

Results for history of medical use were most consistent for opioid motives. Specifically, reporting nonmedical use before medical use of prescription opioids (relative to those who reported no lifetime medical use) was associated with higher odds of membership in all opioid motive classes relative to the “experiment” class. By contrast, reporting medical use first was associated with higher odds of membership in only the “pain relief” class. Similarly, reporting nonmedical use before medical use of tranquilizers was associated with higher odds of membership in all but one of the tranquilizer motive classes relative to the “experiment” class. By contrast, reporting medical use first was associated with higher odds of membership in the “relax/sleep” and “relax” classes. A very different pattern of results was observed for the stimulant motive classes. Specifically, regardless of order, reporting nonmedical and medical use of stimulants was associated with lower odds of membership in the “lose weight/energy” class. By contrast, reporting medical use first was associated with lower odds of membership in the “energy/awake/high” class.

Discussion

The findings of this study revealed a great deal of heterogeneity associated with the nonmedical use of prescription medications (NUPM) in a nationally representative sample of adolescents. The present study found that approximately 75% of past-year nonmedical users of prescription stimulants, tranquilizers, and opioids endorsed more than one motive for NUPM. These findings indicate that most adolescents have multiple motivations underlying their past-year NUPM and suggest that the structure of motives may more closely approximate a categorical versus a dimensional distribution. In this study, we identified four or more motivational subtypes associated with past-year NUPM of each class. While person- and variable-centered approaches are complementary, the results of this person-centered approach identified a greater range of subtypes of nonmedical users than previous studies based on variable-centered analytic approaches [9,10,14,15].

Information regarding motivations can be useful for understanding the consequences associated with NUPM and for designing appropriate prevention and intervention programs [7,10]. Just as research on motives has added to the understanding of consequences associated with drinking behaviors [20,3032], cocaine use [33], and marijuana use [3436], an improved understanding of motives for NUPM will help identify individuals at greatest risk for adverse consequences [912,37]. The findings of the present study provide evidence that motivational subtypes of nonmedical use of prescription stimulants, opioids, and tranquilizers were significantly associated with substance abuse. In particular, we found that the recreational subtypes (e.g., “get high” and “experiment”) had significantly greater odds of feeling moderately or very high when using prescription medications, co-ingesting prescription medications and other drugs, and using non-oral routes of administration.

There were notable sex differences in the subtypes associated with NUPM. In particular, females had greater odds of self-treatment subtypes than males (e.g., “pain relief” and “relax”). These results reinforce the importance of examining differences in subtypes and suggest high school females could be undertreated for some health conditions and disorders such as pain which could contribute to higher prevalence rates of NUPM due to self-treatment. The results of the present study can be used for screening efforts to identify adolescents who may need additional medical attention and the motivational subtypes could be used to match patients to the appropriate intervention. For example, individuals falling into the “weight loss/energy” class for prescription stimulants, “relax” class for prescription tranquilizers, and “pain relief” class for prescription opioids would benefit from a professional assessment by a trained clinician to determine an appropriate diagnosis and treatment. Meanwhile, adolescents in the “affect regulation,” “experiment” and “get high” classes could also benefit from a professional assessment and treatment including more comprehensive substance abuse assessment.

We found that medical use prior to NUPM was significantly greater among self-treatment subtypes for prescription opioids and tranquilizers. Previous work has shown that many nonmedical users obtain prescription medications from a prescription they had previously [3]. For example, 40% of high school seniors who reported nonmedical use of prescription opioids in the past 12 months obtained these medications from their own previous prescription. Taken together, these findings suggest that medically prescribed prescription opioids and tranquilizers should be carefully prescribed and closely monitored to reduce subsequent NUPM of leftover medication.

The present study features several notable strengths such as the inclusion of a large national sample. To date, no large-scale nationally representative studies have examined motivational subtypes for NUPM. There were limitations in the present study that need to be taken into account while considering the implications of the findings. First, the results may not be generalized to other adolescent populations because our sample was drawn from high school seniors and did not include individuals who had dropped out or who were absent from school on the day of survey administration. Also, our results suggested that our sample of past-year nonmedical users were more likely to be white students with poor academic performance. Second, nonresponse may have introduced potential bias in the present study and the data are subject to the potential bias introduced when collecting substance use behaviors via self-reports surveys. The present study attempted to minimize potential biases by implementing conditions that previous research has shown minimizes biases [3839]. It is worth noting that the prevalence rates of NUPM in the present study were comparable to rates reported from other national studies of adolescents [6]. Finally, the prescription stimulant category contained examples that could be perceived as street drugs and was missing commonly misused prescription stimulants such as Adderall® from the list of examples [40].

Despite these limitations, the findings of the present study indicate a wide range of motivational subtypes across the three classes of prescription medications and provide strong support for a great deal of heterogeneity associated with NUPM. Future work is needed to further validate motivational subtypes identified in this study, examine motivational subtypes over time, and incorporate additional characteristics associated with NUPM such as frequency, route of administration and co-ingestion with other substances. Targeted interventions that tailor program content to the distinct motivational subtypes may prove useful for reducing NUPM among adolescents.

Implications and Contribution

High school seniors were grouped into similar categories according to their motivations for using prescription medications nonmedically, allowing for a better understanding of why adolescents engage in such risky behavior. We identified individuals who were at higher risk for substance abuse and such information can be used to guide intervention efforts.

Table 3
Predictors of Latent Class Membership for Opioid Motives (n=768)
Table 4
Correlates of Latent Class Membership for Stimulant Motives (n=782)
Table 5
Correlates of Latent Class Membership for Tranquilizer Motives (n=541)

Acknowledgments

The development of this article was supported by research grants R01DA024678 and R01DA031160 from the National Institute on Drug Abuse, National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Drug Abuse or the National Institutes of Health. The Monitoring the Future data were collected by a research grant DA01411 from the National Institute on Drug Abuse, National Institutes of Health. The authors would like to thank the Substance Abuse and Mental Health Data Archive for providing access to these data.

Footnotes

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.

Drs. Sean Esteban McCabe and James A. Cranford had full access to all the data reported in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

References

1. Blanco C, Alderson D, Ogburn E, et al. Changes in the prevalence of non-medical prescription drug use and drug use disorders in the United States: 1991–1992 and 2001–2002. Drug Alcohol Depend. 2007;90:252–260. [PubMed]
2. Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE. Bethesda, Maryland: National Institute on Drug Abuse; 2007. Monitoring the Future National Survey Results on Drug Use, 1975–2006: Volume II, College Students and Adults Ages 19–45. (NIH Publication No. 07-6206)
3. Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE. Monitoring the Future National Survey Results on Drug Use, 1975–2010. Volume I: Secondary School Students. Ann Arbor, Michigan: University of Michigan Institute for Social Research; 2011.
4. McCabe SE, Cranford JA, Boyd CJ. The relationship between past-year drinking behaviors and nonmedical use of prescription drugs: Prevalence of co-occurrence in a national sample. Drug Alcohol Depend. 2006;84:281–288. [PMC free article] [PubMed]
5. McCabe SE, West B, Wechsler H. Trends and college-level characteristics associated with non-medical use of prescription drugs among US college students from 1993 to 2001. Addiction. 2007;102:455–465. [PMC free article] [PubMed]
6. Substance Abuse and Mental Health Services Administration. Rockville, MD: Office of Applied Studies; 2010. [accessed 9 October 2011]. Results from the 2009 National Survey on Drug Use and Health: Detailed Tables. Available at: http://oas.samhsa.gov/NSDUH/2k9NSDUH/tabs/cover.pdf.
7. Zacny JP, Lichtor SA. Nonmedical use of prescription opioids: motive and ubiquity issues. J Pain. 2008;9:473–486. [PMC free article] [PubMed]
8. Boyd CJ, McCabe SE. Coming to terms with the nonmedical use of prescription medications. Subst Abuse Treat Prev Policy. 2008;3:22. [PMC free article] [PubMed]
9. Boyd CJ, McCabe SE, Cranford JA, et al. Adolescents’ motivations to abuse prescription medications. Pediatrics. 2006;118:2472–2480. [PMC free article] [PubMed]
10. McCabe SE, Cranford JA, Boyd CJ, et al. Motives, diversion and routes of administration associated with nonmedical use of prescription opioids. Addict Behav. 2007;32:562–575. [PMC free article] [PubMed]
11. Compton WM, Volkow ND. Major increases in opioid analgesic abuse in the United States: Concerns and strategies. Drug Alcohol Depend. 2006;81:103–107. [PubMed]
12. Compton WM, Volkow ND. Abuse of prescription drugs and the risk of addiction. Drug Alcohol Depend. 2006;83:S4–S7. [PubMed]
13. Johnston LD, O’Malley PM. Why do the nation’s students use drugs and alcohol? Self-reported reasons from nine national surveys. J Drug Issues. 1986;16:29–66.
14. McCabe SE, Boyd CJ, Cranford JA, et al. Motives for nonmedical use of prescription opioids among high school seniors in the United States: self-treatment and beyond. Arch Pediatr Adolesc Med. 2009;163:739–744. [PMC free article] [PubMed]
15. Terry-McElrath YM, O’Malley PM, Johnston LD. Reasons for drug use among American youth by consumption level, gender, and race-ethnicity: 1976–2005. J Drug Issues. 2009;39:677–714. [PMC free article] [PubMed]
16. Muthen B, Muthen LK. Integrating person-centered and variable-centered analyses: Growth mixture modeling with latent trajectory classes. Alcohol Clin Exp Res. 2000;24:882–891. [PubMed]
17. Lynskey MT, Agrawal A, Bucholz KK, et al. Subtypes of illicit drug users: a latent class analysis of data from an Australian twin sample. Twin Res Hum Genet. 2006;9:523–530. [PubMed]
18. Chung T, Martin CS. Classification and course of alcohol problems among adolescents in addictions treatment programs. Alcohol Clin Exp Res. 2001;25:1734–1742. [PubMed]
19. Thatcher DL, Cornelius JR, Clark DB. Adolescent alcohol use disorders predict adult borderline personality. Addict Behav. 2005;30:1709–1724. [PubMed]
20. Coffman DL, Patrick ME, Palen LA, et al. Why do high school seniors drink? Implications for a targeted approach to intervention. Prev Sci. 2007;8:241–248. [PubMed]
21. Kovac I, Merette C, Legault L, et al. Evidence in an international sample of alcohol-dependent subjects of subgroups with specific symptoms patterns of antisocial personality disorder. Alcohol Clin Exp Res. 2002;26:1088–1096. [PubMed]
22. Hall MT, Howard MO, McCabe SE. Subtypes of adolescent sedative/anxiolytic misusers: a latent profile analysis. Addict Behav. 2010;35:882–889. [PMC free article] [PubMed]
23. Fergusson DM, Horwood LJ, Lynskey MT. The comorbidities of adolescent problem behaviors: a latent class model. J Abnorm Child Psychol. 1994;22:339–354. [PubMed]
24. Lanza ST, Collins LM, Lemmon DR, et al. PROC LCA: A SAS procedure for latent class analysis. Struct Equ Modeling. 2007;14:671–694. [PMC free article] [PubMed]
25. Akaike H. Factor analysis and AIC. Psychometrika. 1987;52:317–332.
26. Schwarz G. Estimating the dimension of a model. Ann Stat. 1978;6:461–464.
27. Collins LM, Lanza ST. Latent Class and Latent Transition Analysis: With Applications in the Social, Behavioral, and Health Sciences. Hoboken, New Jersey: Wiley; 2010.
28. McCutcheon AL. Latent Class Analysis. Newbury Park, California: Sage Publications; 1987.
29. Nylund KL, Asparouhov T, Muthen BO. Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Struct Equ Modeling. 2007;14:535–569.
30. Cooper ML. Motivations of alcohol use among adolescents: development and validation of a four-factor model. Psychol Assess. 1994;6:117–128.
31. Cooper ML, Frone MR, Russell M, et al. Drinking to regulate positive and negative emotions: a motivational model of alcohol use. J Person Soc Psychol. 1995;69:990–1005. [PubMed]
32. Schulenberg JE, Wadsworth KN, O’Malley PM, et al. Adolescent risk factors for binge drinking during the transition to young adulthood: variable- and pattern-centered approaches to change. Dev Psychol. 1996;32:659–674.
33. Weiss RD, Mirin SM. Subtypes of cocaine abusers. Psychiatr Clin North Am. 1986;9:491–501. [PubMed]
34. Schulenberg JE, Merline AC, Johnston LD, et al. Trajectories of marijuana use during the transition to adulthood: the big picture based on national panel data. J Drug Issues. 2005;35:255–279. [PMC free article] [PubMed]
35. Simons J, Correia CJ, Carey KB, et al. Validating a five-factor marijuana motives measure: Relations with use, problems, and alcohol motives. J Couns Psychol. 1998;45:265–273.
36. Zvolensky MJ, Vujanovic AA, Bernstein A, et al. Marijuana use motives: a confirmatory test and evaluation among young adult marijuana users. Addict Behav. 2007;32:3122–3130. [PMC free article] [PubMed]
37. Teter CJ, McCabe SE, Cranford JA, et al. Prevalence and motivations for the illicit use of prescription stimulants in an undergraduate student sample. J Am Coll Health. 2005;53:253–262. [PubMed]
38. Harrison L, Hughes A. Washington: Government Printing Office; 1997. The validity of self-reported drug use: Improving the accuracy of survey estimates. NIH Publication 97-4147, NIDA Research Monograph No. 167. [PubMed]
39. Johnston LD, O’Malley PM. Issues of validity and population coverage in student surveys of drug use. NIDA Res Monogr. 1985;57:31–54. [PubMed]
40. Teter CJ, McCabe SE, LaGrange K, et al. Illicit use of specific prescription stimulants among college students: prevalence, motives, and routes of administration. Pharmacotherapy. 2006;26:1501–1510. [PMC free article] [PubMed]