Abstract
Aims
To examine nonmedical use of prescription opioids (NMUPO) patterns during the transition from adolescence to adulthood, and assess individual characteristics and other substance use behaviors associated with longitudinal patterns of NMUPO.
Design
Nationally representative samples of high school seniors in the United States (wave 1: modal age 18 years) were followed longitudinally across three biennial follow-up waves (waves 2, 3, and 4: modal ages 19/20, 21/22, and 23/24).
Setting
Data were collected via self-administered questionnaires to high school seniors and young adults.
Participants
The longitudinal sample consisted of 27,268 individuals in 30 cohorts (high school senior years 1976–2005) who participated in all four waves.
Measurements
Self-reports of NMUPO and other substance use behaviors.
Findings
Approximately 11.6% (95% CI = 11.2%, 12.0%) of the sample reported past-year NMUPO in at least one of the four waves. Among those who reported past-year NMUPO in at least one wave, 69.0% (67.6%, 70.4%), 20.5% (19.3%, 21.7%), 7.8% (7.1%, 8.6%), and 2.7% (2.3%, 3.1%) reported NMUPO at one, two, three, and four waves, respectively. Several wave 1 variables were associated with greater odds of multiple waves of NMUPO and individuals who reported more waves of NMUPO had greater odds of other substance use behaviors.
Conclusions
Although most nonmedical use of prescription opioids (NMUPO) among 18-year-olds in the United States appears to be non-continuing, about one-third of the sample reporting NMUPO appear to continue use beyond age 18 and have elevated odds of other substance use behaviors at ages 23/24.
Keywords: Longitudinal, Prescription Opioids, Nonmedical Use, Epidemiology, Adolescents, Substance Use
Introduction
The rates of medical and nonmedical use of prescription opioids (NMUPO) have been steadily rising over the past 15–20 years in the United States [1–12]. Although the United States represents less than 5% of the world’s population, Americans consume 80% of the global opioid supply, and 99% of the global hydrocodone supply [10]. While there are several national studies that examine the extent of NMUPO in the United States, there is limited information regarding this public health issue in other developed countries. As a result, cross-national comparisons of NMUPO are difficult to ascertain at this time [12]. In the United States, NMUPO among adolescents and young adults has significantly increased over the past two decades and is most prevalent among individuals 18–25 years of age [1–7]. In 2010, approximately 2 million persons aged 12 or older initiated NMUPO within the past 12 months which was greater than the estimated number of those who initiated use of any other illicit drug with the exception of marijuana [7]. The estimated number of emergency department visits involving NMUPO more than doubled between 2004 and 2008 for patients younger than 21 years of age [8]. Although there have been recent advances in the understanding of NMUPO, considerable gaps in knowledge remain due to limitations in measures and study designs [9–12]. In particular, there are few longitudinal studies, leaving gaps in our understanding of individual patterns of NMUPO over time and of risk factors that accurately identify those individuals at greatest risk for chronic patterns of NMUPO [11].
To date, most large-scale national epidemiological studies have been cross-sectional and have revealed high levels of NMUPO with past-year prevalence rates as high as 10% among secondary school students and 20% among college students in the United States [2,3,13,14]. The age at peak risk for initiating NMUPO is 16 years and the majority of NMUPO is initiated prior to the age of 18 among early and middle-aged adults [2,3,15]. The limited longitudinal evidence suggests that most NMUPO is experimental, but that earlier use sets the stage for later substance use disorders [16]. Based on the National Epidemiologic Survey on Alcohol and Related Conditions, approximately 4 in every 5 adults 18 years or older in the United States who engaged in NMUPO ceased using 3 years later, but that NMUPO among 18–24 year olds was associated with greater odds of developing a substance use disorder three years later [16]. Furthermore, a cross-sectional study found that the initiation of NMUPO before 18 years of age was associated with significantly higher odds of developing prescription opioid abuse and dependence versus those individuals who began using at or after 18 years of age [17]. Indeed, over 40% of individuals in the United States who retrospectively reported initiating NMUPO at age 18 or younger developed prescription drug use disorders in their lifetime [17]. Taken together, these results indicate that NMUPO during the transition from adolescence to adulthood warrants longitudinal examination to distinguish experimental from escalating use.
Much attention has been devoted to the identification of longitudinal patterns of alcohol and marijuana use across the transition to adulthood [18–20]; similar attention has not been given to NMUPO. In addition, although several cross-sectional studies have found strong associations between NMUPO and other substance use behaviors [13,14,21,22], none have considered how NMUPO relates to other substance use over time. To fill these critical gaps in current knowledge, our purpose is to be the first national study to describe longitudinal patterns of NMUPO during the transition to adulthood, to assess demographic and individual characteristics associated with these patterns of NMUPO, and to examine other substance use behaviors associated with patterns of NMUPO. We analyze national panel data from the Monitoring the Future (MTF) study [2,3] and provide a more comprehensive understanding of the heterogeneity associated with NMUPO and its developmental course among adolescents as they move into adulthood.
Methods
The present study used national panel data from the MTF study conducted in the United States [3,23]. Based on a three-stage sampling procedure, MTF has surveyed nationally representative samples of approximately 17,000 high school seniors each year since 1975, using questionnaires administered in classrooms. Stage 1 is the selection of geographic areas; stage 2 is the selection of schools; and stage 3 is the selection of students within each school. Approximately 2,400 high school seniors are selected for biennial follow-ups each year using mailed questionnaires. The biennial follow-up surveys begin one year after high school for one random half of each cohort and two years after high school for the other half. For purposes of these analyses, the two halves were combined (combining modal ages 19/20, 21/22, and 23/24) due to sample size concerns and lack of significant differences across the two halves on substance use measures. Individuals who used illicit drugs in high school were over-sampled for the follow-up surveys. Corrective weighting was used in the analyses to adjust for the unequal probabilities of selection that occurred at any stage of sampling. The project design and sampling methods are described in greater detail elsewhere [3,24].
Sample
The sample for the present study consisted of respondents who were surveyed as high school seniors (wave 1) in 1976 through 2005, and who were surveyed in their first, second, and third biennial follow-up surveys (waves 2, 3 and 4, respectively). Given the aims of the present study, the 30 cohorts were combined and analyses were conducted with 27,268 respondents (weighted number in longitudinal sample) who provided data at all four waves yielding a retention rate of just under 50%. Relaxing this requirement and imputing missing data would have increased our longitudinal sample, but given our focus on individual developmental patterns of NMUPO this full-data inclusion criterion was necessary. This longitudinal sample consisted of 59% females, 82% Whites, 7% Blacks, 5% Hispanics, and 6% other racial/ethnic groups or not specified. As illustrated in Table 1, attrition analyses at wave 1 revealed that those retained in the longitudinal sample differed as compared to those who attrited. For instance, individuals retained in the longitudinal sample were more likely to be female and white, report good grades in high school, have higher parent education, have any college plans and have lower rates of skipping class, evenings out, 2-week binge drinking, and past-year marijuana use. In addition, individuals retained in the longitudinal sample had lower rates of past-year NMUPO and past-year nonmedical use of amphetamines and/or tranquilizers.
Table 1.
Wave 1 sample characteristics for the longitudinal sample and attrition sample
| Longitudinal sample N=27,268 % (95% CI) | Attrition sample N=27,862 % (95% CI) | |
|---|---|---|
| Gender | ||
| Male | 40.7 (40.0–41.4) | 54.6 (53.9–55.3) |
| Female | 59.3 (58.6–60.0) | 45.4 (44.7–46.1) |
| Race/ethnicity | ||
| Black | 7.4 (6.6–8.1) | 16.5 (15.1–17.9) |
| White | 82.1 (81.0–83.1) | 65.4 (63.7–67.1) |
| Hispanic | 4.9 (4.4–5.5) | 10.0 (8.9–11.1) |
| Other | 5.6 (5.2–6.0) | 8.1 (7.7–8.5) |
| Geographical region | ||
| South | 30.2 (28.6–31.7) | 37.2 (35.7–38.8) |
| Northeast | 21.7 (20.3–23.1) | 19.7 (18.5–20.8) |
| Midwest | 31.3 (29.7–32.9) | 23.9 (22.6–25.2) |
| West | 16.8 (15.5–18.1) | 19.2 (17.9–20.6) |
| Urbanicity | ||
| Farm/country | 19.2 (18.1–20.3) | 17.1 (16.2–18.0) |
| Small city or larger | 80.8 (79.7–81.9) | 82.9 (82.0–83.8) |
| Parent education | ||
| Some college | 65.2 (64.1–66.3) | 62.5 (61.5–63.6) |
| High school or less | 34.8 (33.7–35.9) | 37.5 (36.4–38.5) |
| High school GPA | ||
| B- or higher | 82.8 (82.2–83.4) | 70.7 (69.9–71.5) |
| C+ or lower | 17.2 (16.6–17.8) | 29.3 (28.5–30.1) |
| College plans | ||
| Yes college plans | 51.7 (50.6–52.8) | 44.7 (43.7–45.7) |
| No college plans | 48.3 (47.2–49.4) | 55.3 (54.3–56.3) |
| Truancy | ||
| Did not skip any days | 68.4 (67.4–69.3) | 60.8 (59.8–61.8) |
| Skipped any days | 31.6 (30.7–32.6) | 39.2 (38.2–40.2) |
| Work intensity | ||
| No work | 22.5 (21.8–23.1) | 24.9 (24.2–25.6) |
| 1–15 hrs/week | 33.6 (33.0–34.3) | 26.5 (25.9–27.1) |
| 16+ hrs/week | 43.9 (43.1–44.7) | 48.6 (47.8–49.4) |
| Evenings out | ||
| Less than 3 weekly | 52.8 (52.1–53.5) | 48.6 (47.8–49.3) |
| 3 or more weekly | 47.2 (46.5–47.9) | 51.4 (50.7–52.2) |
| Binge drinking | ||
| No, past 2 weeks | 70.5 (69.7–71.3) | 64.7 (63.9–65.5) |
| Yes, past 2 weeks | 29.5 (28.7–30.3) | 35.3 (34.4–36.1) |
| Marijuana use | ||
| No, past 12 months | 68.4 (67.6–69.3) | 60.7 (59.9–61.5) |
| Yes, past 12 months | 31.6 (30.7–32.4) | 39.3 (38.5–40.1) |
| Senior year cohort | ||
| 1976–1991 | 60.0 (57.6–62.5) | 42.2 (39.8–44.7) |
| 1992–2001 | 30.9 (28.6–33.2) | 42.0 (39.5–44.5) |
| 2002–2005 | 9.1 (7.8–10.3) | 15.7 (13.8–17.7) |
Note: All frequencies and percentages are weighted.
Measures
Using six randomly-distributed questionnaire forms, the MTF assesses demographic and psycho-social characteristics and standard measures of substance use.
Nonmedical use of prescription opioids (NMUPO) was assessed at all four waves with an item asking respondents on how many occasions (if any) they used prescription opioids on their own, without a doctor’s orders (e.g., Vicodin®, OxyContin®, Percodan®, Percocet®, Demerol®, Dilaudid®, morphine, methadone, opium, codeine) during the past 12 months. The response scale was (1) no occasions, (2) 1–2 occasions, (3) 3–5 occasions, (4) 6–9 occasions, (5) 10–19 occasions, (6) 20–39 occasions, and (7) 40 or more occasions. It should be noted that the list of examples was updated in 2002 and several “older” medications (e.g., laudanum, Talwin®) were replaced with more “current” medications (e.g., Vicodin®, OxyContin®, Percodan®, Percocet®, Dilaudid®). The addition of these new examples was followed by an uptick in prevalence of NMUPO, likely due in part to the changed question [2,3].
Demographic and lifestyle characteristics were assessed at wave 1 and consisted of student self-reports of the following: gender, race/ethnicity (Black, White, Hispanic, Other), urbanicity (where grew up, farm/country vs. not), parental education (some college vs. high school or less), high school grade point average (B- or higher vs. C+ or lower), college plans (any plans vs. no plans), truancy (did not skip any days in the past four weeks vs. one or more skipped days), work intensity (no work vs. 1–15 hours per week vs. 16 or more hours per week), social evenings out (less than three per week vs. three or more per week), past two-week binge drinking (any vs. none), past-year marijuana use (any vs. none). Geographic region of the country (Northeast, Midwest, South, West) was based on school location information. Senior year cohort was split into three periods based on the high school class survey year (1976–1991, 1992–2001, and 2002–2005). These three cohort periods were selected due to decreases in the prevalence of NMUPO among high school seniors following 1991 and a change in wording for the NMUPO measure in 2002 resulting in an increase in the prevalence of NMUPO [2,3]. Cut-points in the covariates were determined based on sensitivity analyses; categorical covariates were desirable given our analytic and descriptive approaches.
Other substance use behaviors were assessed at wave 4, including two-week binge drinking (any vs. none) and past-year marijuana use (any vs. none). We also included past-year other nonmedical prescription drug use--including amphetamines (e.g., Ritalin®, Dexedrine®) and/or tranquilizers (e.g., Ativan®, Klonopin®, Valium®, Xanax®)-- which were measured with the following questions: “On how many occasions (if any) have you used [specified drug class] during the last 12 months?” The response scale for each substance was the same as for NMUPO. A single dichotomous variable was created for analysis from the separate measures to indicate prevalence of use for one or more of these drug classes.
Data analysis
Given that this was the first national study to systematically examine longitudinal patterns of NMUPO during the transition from adolescence to adulthood, we took a straightforward descriptive approach to addressing our research questions. First, the patterns of past-year NMUPO and the number of waves of past-year NMUPO were described using simple frequency tables. Second, multivariable logistic regression models were used to examine the relationships of wave 1 demographics and lifestyle characteristics with the following three patterns of past-year NMUPO: 1) at one wave only, 2) at two or more waves, and 3) at all four waves. These three patterns of NMUPO were chosen to capture experimental use at one wave only, repeated use at two or more waves, and chronic use at all four waves. Finally, we examined the prevalence of Wave 4 substance use behaviors (i.e., binge drinking in the past two weeks, marijuana use in the past year, and other nonmedical use of prescription drugs in the past year) by the number of waves of NMUPO (i.e., no waves, one, two, three, and four waves). Multivariable logistic regression models were used controlling for wave 1 demographic and lifestyle characteristics to understand the unique contribution of NMUPO patterns to wave 4 substance use behaviors. Analyses incorporated the complex sample design variables and were weighted for follow-up sampling selection using SAS v9.3 SURVEY commands (MEANS, FREQ, LOGISTIC, REG) as appropriate.
Results
Approximately 11.6% (95% CI =11.2%, 12.0%) of the sample reported past-year NMUPO in at least one of the four waves of measurement. The mean level of occasions of past-year NMUPO in the sample held relatively steady over the four waves. The weighted mean frequency of past-year NMUPO (response scale ranged from 1 “no occasions” to 7 “40 or more occasions”) was 1.09 at wave 1 (modal age 18) and 1.08 at waves 2, 3, and 4 (modal ages 19/20, 21/22, and 23/24, respectively). As shown in Table 2, there were 16 possible patterns of past-year NMUPO over time. The majority of high school seniors who reported past-year NMUPO at wave 1 did not engage in this behavior again at waves 2, 3, or 4.
Table 2.
Frequencies and patterns associated with nonmedical use of prescription opioids over four waves
| Past-year nonmedical use of prescription opioids | Number of waves | Frequency (n) | Overall sample (%) (95% CI) | Nonmedical users (n=3,173) (%) (95% CI) |
|---|---|---|---|---|
| No waves | 0 | 24,096 | 88.4 (88.0–88.7) | - |
| Wave 1 only | 1 | 733 | 2.7 (2.5–2.8) | 23.1 (22.0–24.3) |
| Wave 2 only | 1 | 487 | 1.8 (1.6–1.9) | 15.3 (14.2–16.5) |
| Wave 3 only | 1 | 493 | 1.8 (1.7–2.0) | 15.5 (14.4–16.7) |
| Wave 4 only | 1 | 477 | 1.7 (1.6–1.9) | 15.0 (13.8–16.2) |
| Waves 1 and 2 | 2 | 165 | 0.6 (0.5–0.7) | 5.2 (4.6–5.8) |
| Waves 1 and 3 | 2 | 72 | 0.3 (0.2–0.3) | 2.3 (1.9–2.7) |
| Waves 1 and 4 | 2 | 67 | 0.3 (0.2–0.3) | 2.2 (1.8–2.5) |
| Waves 2 and 3 | 2 | 97 | 0.4 (0.3–0.4) | 3.1 (2.5–3.6) |
| Waves 2 and 4 | 2 | 68 | 0.3 (0.2–0.3) | 2.2 (1.7–2.6) |
| Waves 3 and 4 | 2 | 179 | 0.7 (0.6–0.7) | 5.6 (4.9–6.4) |
| Waves 1, 2 and 3 | 3 | 76 | 0.3 (0.2–0.3) | 2.4 (2.0–2.8) |
| Waves 1, 2 and 4 | 3 | 44 | 0.2 (0.1–0.2) | 1.4 (1.1–1.7) |
| Waves 1, 3 and 4 | 3 | 42 | 0.2 (0.1–0.2) | 1.3 (1.0–1.6) |
| Waves 2, 3 and 4 | 3 | 87 | 0.3 (0.3–0.4) | 2.8 (2.2–3.3) |
| All four waves | 4 | 84 | 0.3 (0.3–0.4) | 2.7 (2.3–3.1) |
| Total | -- | 27,268 | ||
| Number of Waves | ||||
| No waves | 24,096 | 88.4 (88.0–88.7) | - | |
| One wave | 2,189 | 8.0 (7.7–8.3) | 69.0 (67.6–70.4) | |
| Two waves | 650 | 2.4 (2.2–2.5) | 20.5 (19.3–21.7) | |
| Three waves | 249 | 0.9 (0.8–1.0) | 7.8 (7.1–8.6) | |
| Four waves | 84 | 0.3 (0.3–0.4) | 2.7 (2.3–3.1) | |
| Total | 27,268 |
Note: All frequencies and percentages are weighted.
In the second phase of the analyses, multivariable logistic regression analyses indicated that several wave 1 variables, including race/ethnicity (White), truancy, binge drinking, marijuana use, and senior year cohorts after 1991 were associated with significantly greater odds of all three patterns of past-year NMUPO (see Table 3). For example, the odds of past-year NMUPO at two or more waves were over 6 times greater among high school seniors who reported past-year marijuana use at wave 1 as compared to those who did not use marijuana. Other covariates were less consistently related to NMUPO patterns. High school seniors who reported “other” race/ethnicity and those who spent three or more evenings out per week had greater odds of past-year NMUPO at one wave only and at two or more waves. High school seniors with at least one college educated parent had greater odds of past-year NMUPO at two or more waves and at all four waves. Those who had lower grade point averages and reported being Hispanic had greater odds of past-year NMUPO at one wave only while high school males had greater odds of past-year NMUPO at two or more waves. High school seniors without plans to attend college had greater odds of past-year NMUPO at all four waves. Finally, urbanicity and work intensity were not associated with any of the three patterns of NMUPO.
Table 3.
Logistic regression results for selected nonmedical use of prescription opioids patterns
| Nonmedical use at one wave only | Nonmedical use at two or more waves | Nonmedical use at all four waves | ||||
|---|---|---|---|---|---|---|
| Wave 1 Characteristics | (%) | Adjusted OR (95% CI) | (%) | Adjusted OR (95% CI) | (%) | Adjusted OR (95% CI) |
| Gender | ||||||
| Male | 8.6 | Referent | 4.4 | Referent | 0.1 | Referent |
| Female | 7.7 | 1.00(0.91,1.09) | 3.1 | 0.79(0.70,0.89)** | 0.2 | 0.75(0.55,1.03) |
| Race/ethnicity | ||||||
| Black | 3.2 | Referent | 1.1 | Referent | 0.0 | Referent |
| White | 8.7 | 2.26(1.73,2.95)*** | 3.9 | 2.46(1.57,3.84)*** | 0.2 | 9.34(1.26,69.19)* |
| Hispanic | 6.0 | 1.59(1.11,2.27)* | 2.4 | 1.34(0.76,2.33) | 0.1 | 3.86(0.45,33.42) |
| Other | 5.8 | 1.78(1.28,2.47)** | 3.0 | 2.07(1.21,3.55)** | 0.0 | 1.96(0.20,19.56) |
| Geographical region | ||||||
| South | 7.4 | Referent | 3.2 | Referent | 0.1 | Referent |
| Northeast | 8.2 | 0.87(0.78,0.98)* | 4.0 | 0.98(0.82,1.17) | 0.2 | 0.72(0.45,1.16) |
| Midwest | 7.9 | 0.88(0.79,0.98)* | 3.4 | 0.90(0.76,1.05) | 0.2 | 0.68(0.45,1.04) |
| West | 9.0 | 1.07(0.93,1.24) | 4.2 | 1.08(0.89,1.33) | 0.2 | 1.06(0.67,1.70) |
| Urbanicity | ||||||
| Farm/country | 6.8 | Referent | 2.8 | Referent | 0.1 | Referent |
| Small city or larger | 8.3 | 1.07(0.95,1.20) | 3.8 | 1.00(0.85,1.17) | 0.2 | 1.29(0.83,2.01) |
| Parental education | ||||||
| Some college | 8.5 | Referent | 4.3 | Referent | 0.2 | Referent |
| High school or less | 7.2 | 0.92(0.83,1.01) | 2.4 | 0.67(0.58,0.78)*** | 0.1 | 0.59(0.38,0.91)* |
| High school GPA | ||||||
| B- or higher | 7.5 | Referent | 3.4 | Referent | 0.2 | Referent |
| C+ or lower | 10.4 | 1.18(1.05,1.32)** | 4.6 | 1.08(0.94,1.25) | 0.2 | 1.17(0.78,1.74) |
| College plans | ||||||
| Yes college plans | 7.6 | Referent | 3.5 | Referent | 0.1 | Referent |
| No college plans | 8.6 | 1.03(0.93,1.14) | 3.8 | 1.13(0.99,1.29) | 0.2 | 1.80(1.27,2.55)** |
| Truancy | ||||||
| Did not skip any days | 6.5 | Referent | 2.3 | Referent | 0.1 | Referent |
| Skipped any days | 11.5 | 1.24(1.13,1.37)*** | 6.4 | 1.64(1.44,1.86)*** | 0.3 | 1.68(1.17,2.40)** |
| Work intensity | ||||||
| No work | 7.0 | Referent | 3.1 | Referent | 0.1 | Referent |
| 1–15 hrs/week | 7.4 | 0.96(0.85,1.09) | 3.2 | 0.91(0.76,1.09) | 0.2 | 1.24(0.80,1.93) |
| 16+ hrs/week | 9.0 | 1.04(0.92,1.17) | 4.1 | 0.98(0.84,1.15) | 0.2 | 1.14(0.74,1.74) |
| Evenings out | ||||||
| Less than 3 weekly | 6.1 | Referent | 2.1 | Referent | 0.1 | Referent |
| 3 or more weekly | 10.1 | 1.17(1.06,1.28)** | 5.3 | 1.44(1.26,1.64)*** | 0.2 | 1.22(0.86,1.74) |
| Binge drinking | ||||||
| No, past 2 weeks | 5.8 | Referent | 2.0 | Referent | 0.1 | Referent |
| Yes, past 2 weeks | 13.6 | 1.32(1.19,1.46)*** | 7.6 | 1.70(1.48,1.95)*** | 0.4 | 2.44(1.66,3.57)*** |
| Marijuana use | ||||||
| No, past 12 months | 4.1 | Referent | 1.1 | Referent | 0.0 | Referent |
| Yes, past 12 months | 16.3 | 3.62(3.23,4.05)*** | 9.1 | 6.30(5.30,7.48)*** | 0.4 | 13.07(6.80,25.12)*** |
| Senior year cohort | ||||||
| 1976–1991 | 7.6 | Referent | 2.7 | Referent | 0.1 | Referent |
| 1992–2001 | 8.0 | 1.29(1.16,1.42)*** | 4.1 | 2.17(1.87,2.53)*** | 0.2 | 3.12(2.05,4.74)*** |
| 2002–2005 | 11.0 | 1.88(1.61,2.21)*** | 8.0 | 4.88(4.06,5.87)*** | 0.4 | 9.71(6.39,14.75)*** |
Referent indicates reference group.
p < 0.05,
p < 0.01,
p < 0.001.
OR: odds ratio; CI: confidence interval.
In the third phase of the analyses, we examined the wave 4 (modal ages 23/24) prevalence rates of binge drinking in the past two weeks, marijuana use in the past year, and other nonmedical use of prescription drugs in the past year as a function of the number of waves of NMUPO and wave 1 covariates from phase two of the analyses. As shown in Table 4, binge drinking was the most common substance use behavior among those who never engaged in past-year NMUPO whereas past-year marijuana use was the most common substance use behavior among those who reported any past-year NMUPO over four waves. Substance use behaviors at wave 4 were more prevalent among those who reported more waves of past-year NMUPO.
Table 4.
Bivariate and logistic regression results for substance use behaviors associated with nonmedical use of prescription opioids
| Wave 4 binge drinking past 2 weeks | Wave 4 past-year marijuana use | Wave 4 past-year other nonmedical prescription drug usea | ||||
|---|---|---|---|---|---|---|
| Number of Waves of Nonmedical Use | % | Adjusted OR (95% CI)b | % | Adjusted OR (95% CI)b | % | Adjusted OR (95% CI)b |
| No waves | 33.1 | Referent | 22.5 | Referent | 6.0 | Referent |
| One wave | 52.9 | 1.99 (1.82,2.17)*** | 55.8 | 3.88 (3.56,4.23)*** | 26.3 | 5.14 (4.62,5.71)*** |
| Two waves | 61.6 | 2.59 (2.24,2.99)*** | 71.1 | 7.71 (6.53,9.11)*** | 45.1 | 12.05 (10.43,13.93)*** |
| Three waves | 72.2 | 3.61 (2.83,4.59)*** | 84.8 | 16.14 (11.99,21.73)*** | 56.4 | 18.51 (14.75,23.22)*** |
| Four waves | 73.3 | 4.10 (2.81,5.99)*** | 85.7 | 17.83 (11.00,28.90)*** | 71.8 | 39.24 (28.04,54.90)*** |
Other nonmedical prescription drug use included nonmedical use of prescription amphetamines and/or tranquilizers.
AOR indicates odds ratios are adjusted for the effects of gender, race/ethnicity, geographical region, parent education, high school grade point average, college plans, truancy, evenings out, and senior year cohort (odds ratios for these variables are not shown).
Referent indicates reference group.
p < 0.001.
OR: odds ratio; CI: confidence interval
As presented in Table 4, multivariable logistic regression analyses confirmed that substance use behaviors at wave 4 were highly associated with number of waves of past-year NMUPO after adjusting for covariates. The odds of substance use behaviors at wave 4 were significantly greater among those who reported more waves of NMUPO as compared to those who never engaged in NMUPO. Notably, individuals who reported NMUPO at all four waves had about 4 times greater odds of binge drinking in the past two weeks, about 17 times greater odds of marijuana use in the past year, and about 39 times greater odds of nonmedical use of other prescription drugs in the past year, than those who had not reported NMUPO at any of the four waves.
Discussion
The findings of the present study extend what is known about the longitudinal patterns of NMUPO among adolescents during the transition to adulthood in several important ways. Sixteen different longitudinal patterns associated with NMUPO were described, and the most prevalent pattern of use included those who reported NMUPO only in their senior year of high school. The majority of individuals who reported NMUPO in their senior year of high school did not engage in this behavior 1–2 years later and even fewer reported NMUPO 3–6 years later. The results of this study were consistent with a previous study that found the majority of adults (18 years or older) who engaged in NMUPO in the United States ceased using 3 years later [16]. Thus, the present study represents an important first step toward understanding the heterogeneity and persistence associated with NMUPO from a longitudinal perspective.
Multivariable logistic regression analyses identified several common wave 1 variables that were associated with significantly greater odds of engaging in multiple waves of past-year NMUPO including race/ethnicity (White), truancy, binge drinking, marijuana use, and more recent high school senior year cohorts (1992–2005). The findings of the present study indicated that White adolescents and those who engaged in multiple problem behaviors while in high school were at increased risk for more chronic patterns of NMUPO during the transition to adulthood. The results also provided additional support for a significant increase in NMUPO among adolescents and young adults over the past two decades [1–3,5,6]. Based on the recent increases in prevalence of NMUPO in the United States, additional prevention and intervention efforts are clearly warranted to reduce NMUPO among adolescents and young adults.
We found significant associations between longitudinal patterns of NMUPO and other substance use behaviors such as binge drinking, marijuana use, and nonmedical use of other prescription drugs. While only about 3% of nonmedical users reported NMUPO at all four waves, nearly all of these individuals reported other substance use behaviors at ages 23/24. The odds of substance use behaviors were significantly greater among individuals who reported more waves of past-year NMUPO and these high rates did not decline at ages 23/24 as they typically do for these substance use behaviors as adolescents transition to adulthood [3,25–27].
High school graduation represents a major developmental transition for many adolescents in the United States. Adolescents and young adults often assume responsibility for their own medication management during the transition to adulthood which may contribute to the high rates of NMUPO found in this age group [2–4,7]. Past work has found that the majority of adolescents and young adults who report NMUPO obtain prescription opioids from their friends for free and more than a third obtain them from their own leftover prescription [2,7,28,29]. The present study found that more than one in every nine individuals in the sample reported past-year NMUPO in at least one of the four waves of measurement. Although there is some evidence based on the present study that the mean frequency of past-year NMUPO held steady for the sample, the prevalence of NMUPO for the sample was lower over time suggesting those reporting NMUPO were using more frequently. The lack of mean level change across the transition to adulthood belies the underlying changes in individual patterns, with the most common NMUPO pattern being to use at only one wave, suggesting strong proclivity toward experimental use only.
The present investigation has several strengths that build upon previous substance abuse research examining NMUPO among adolescents and young adults. First, this study includes nationally representative samples of high school seniors in the United States. Second, multiple cohorts of high school seniors were followed longitudinally across four waves, enabling an assessment of NMUPO over time and historical change. To date, most national studies have been cross-sectional and have not examined patterns of NMUPO by following adolescents from a modal age of 18 to 24 years of age.
The present study also has some limitations that need to be taken into account when considering the implications of the findings. First, the updates to the prescription opioid category in 2002 may have contributed to the associations between patterns of NMUPO and later cohorts (2002–2005). However, no such wording changes were made between 1992 and 2001 and these cohorts were also associated with significantly increased odds of NMUPO which argues for increases of NMUPO in later cohorts. Changes in wording in longitudinal studies always represent a challenge to tracking prescription medications over time, and undeniably, the updates in the 2002 question make it more difficult to interpret the trends. However, similar changes were made to prescription medication questions in other national studies (e.g., National Epidemiologic Survey on Alcohol and Related Conditions, National Survey on Drug Use and Health, College Alcohol Study) and on balance, creating questions that include “current” medications is more important than maintaining items with obsolete wording. An additional limitation is that the survey items did not specify the quantity of prescription opioids that was used on each occasion. Finally, retention over all four waves was approximately 50%; males, non-whites, and those who reported NMUPO and other problem behaviors were less likely to participate in the study over time based on attrition analyses. Thus, it is very likely that individuals who experienced the most serious patterns of and consequences associated with NMUPO are underrepresented in the longitudinal sample, suggesting that our findings represent an underestimation of high use patterns and potential consequences.
Based on increases in NMUPO and prescribing of opioids over the past two decades [30–32], future research should examine the association between prescribing patterns of prescription opioids and NMUPO among adolescents and young adults. Based on the long-term health risks associated with early initiation of NMUPO, the prediction of future NMUPO during childhood and early adolescence represents and important topic for future research [15,17]. Previous research has shown that heavy drinking and other substance use behaviors tend to decline as young adults assume greater responsibilities [3,19,25–27]. An important question for future research, based on the present findings, is whether the typical decline in substance use behaviors holds equally well for NMUPO based on the high abuse potential of prescription opioids [32].
Acknowledgments
The development of this manuscript was supported by research grants R01DA001411, R01DA016575, 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 under research grants R01DA001411 and R01DA016575, and the work of the second through fifth authors on this manuscript was supported by these grants. For the first author, work on this manuscript was supported by research grants R01DA024678 and R01DA031160. The authors would like to thank Dr. Brady West and the anonymous reviewers for their helpful comments on a previous version of this article and the respondents and school personnel for their participation in the study.
Footnotes
Conflict of interest declaration: None
Contributor Information
Sean Esteban McCabe, University of Michigan, Substance Abuse Research Center and Institute for Research on Women and Gender, Ann Arbor, MI, USA 48109
John E. Schulenberg, University of Michigan, Institute for Social Research and Department of Psychology, Ann Arbor, MI, USA 48106-1248
Patrick M. O’Malley, University of Michigan, Institute for Social Research, Ann Arbor, MI, USA 48106-1248
Megan E. Patrick, University of Michigan, Institute for Social Research, Ann Arbor, MI, USA 48106-1248
Deborah D. Kloska, University of Michigan, Institute for Social Research, Ann Arbor, MI, USA 48106-1248
References
- 1.Blanco C, Alderson D, Ogburn E, Grant BF, Nunes EV, Hatzenbuehler ML, 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. doi: 10.1016/j.drugalcdep.2007.04.005. [DOI] [PubMed] [Google Scholar]
- 2.Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE. Monitoring the Future National Survey Results on Drug Use, 1975–2012: Volume I, Secondary School Students. Ann Arbor, MI: University of Michigan Institute for Social Research; 2013. [Google Scholar]
- 3.Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE. Monitoring the Future National Survey Results on Drug Use, 1975–2011: Volume II, College Students and Adults 19–50. Ann Arbor, MI: University of Michigan Institute for Social Research; 2012. [Google Scholar]
- 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. doi: 10.1016/j.drugalcdep.2006.03.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.McCabe SE, West B, Wechsler H. Trends and college-level characteristics associated with the non-medical use of prescription drugs among U.S. college students from 1993 to 2001. Addiction. 2007;102:455–465. doi: 10.1111/j.1360-0443.2006.01733.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.McCabe SE, Cranford JA, West BT. Trends in prescription drug abuse and dependence, co-occurrence with other substance use disorders, and treatment utilization: results from two national surveys. Addict Behav. 2008;33:1297–1305. doi: 10.1016/j.addbeh.2008.06.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Substance Abuse and Mental Health Services Administration. NSDUH Series H-41, HHS Publication No (SMA) 11-4658. Rockville, MD: Office of Applied Studies; 2011. Results from the 2010 National Survey on Drug Use and Health: Summary of National Findings. [Google Scholar]
- 8.Substance Abuse and Mental Health Services Administration. The DAWN Report: Trends in Emergency Department Visits Involving Nonmedical Use of Narcotic Pain Relievers. Rockville, MD: Office of Applied Studies; 2010. [accessed June 22, 2011]. Available at http://oas.samhsa.gov/2k10/dawn016/opioided.htm. [Google Scholar]
- 9.Boyd CJ, McCabe SE. Coming to terms with the nonmedical use of prescription medications. Subst Abuse Treat Prev Policy. 2008;18:3–22. doi: 10.1186/1747-597X-3-22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Manchikanti L, Fellows B, Ailinani H, Pampati V. Therapeutic use, abuse, and nonmedical use of opioids: a ten-year perspective. Pain Physician. 2010;13:401–435. [PubMed] [Google Scholar]
- 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. doi: 10.1016/j.drugalcdep.2005.05.009. [DOI] [PubMed] [Google Scholar]
- 12.Zacny JP, Lichtor SA. Nonmedical use of prescription opioids: motive and ubiquity issues. J Pain. 2008;9:473–486. doi: 10.1016/j.jpain.2007.12.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Boyd CJ, McCabe SE, Teter CJ. Medical and nonmedical use of prescription pain medication by youth in a Detroit-area public school district. Drug Alcohol Depend. 2006;81:37–45. doi: 10.1016/j.drugalcdep.2005.05.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.McCabe SE, Teter CJ, Boyd CJ, Knight JR, Wechsler H. Nonmedical use of prescription opioids among U.S. college students: prevalence and correlates from a national survey. Addict Behav. 2005;30:789–805. doi: 10.1016/j.addbeh.2004.08.024. [DOI] [PubMed] [Google Scholar]
- 15.Meier EA, Troost JP, Anthony JC. Extramedical use of prescription pain relievers by youth aged 12 to 21 years in the United States: national estimates by age and by year. Arch Pediatr Adolesc Med. 2012;166:803–807. doi: 10.1001/archpediatrics.2012.209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Boyd CJ, Teter CJ, West BT, Morales M, McCabe SE. Non-medical use of prescription analgesics: A three-year national longitudinal study. J Addict Dis. 2009;28:232–242. doi: 10.1080/10550880903028452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.McCabe SE, West BT, Morales M, Cranford JA, Boyd CJ. Does early onset of non-medical use of prescription drugs predict subsequent prescription drug abuse and dependence? Results from a national study. Addiction. 2007;102:1920–1930. doi: 10.1111/j.1360-0443.2007.02015.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Schulenberg JE, Maggs JL. A developmental perspective on alcohol use and heavy drinking during adolescence and the transition to young adulthood. J Stud Alcohol. 2002;14:54–70. doi: 10.15288/jsas.2002.s14.54. [DOI] [PubMed] [Google Scholar]
- 19.Schulenberg JE, Merline AC, Johnston LD, O’Malley PM, Bachman JG, Laetz VB. Trajectories of marijuana use during the transition to adulthood: the big picture based on national panel data. J Drug Issues. 2005;35:255–279. doi: 10.1177/002204260503500203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Windle M, Wiesner M. Trajectories of marijuana use from adolescence to young adulthood: Predictors and outcomes. Dev Psychopathol. 2004;16:1007–1027. doi: 10.1017/s0954579404040118. [DOI] [PubMed] [Google Scholar]
- 21.McCabe SE, Boyd CJ, Teter CJ. Illicit use of opioid analgesics by high school seniors. J Subst Abuse Treat. 2005;28:224–229. doi: 10.1016/j.jsat.2004.12.009. [DOI] [PubMed] [Google Scholar]
- 22.McCabe SE, Boyd CJ, Young A. Medical and nonmedical use of prescription drugs among secondary school students. J Adolesc Health. 2007;40:76–83. doi: 10.1016/j.jadohealth.2006.07.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Bachman JG, O’Malley PM, Schulenberg JE, Johnston LD, Bryant AL, Merline AC. The Decline of Substance Use in Young Adulthood: Changes in Social Activities, Roles, and Beliefs. Mahwah, NJ: Lawrence Erlbaum Associates; 2002. [Google Scholar]
- 24.Bachman JG, Johnston LD, O’Malley PM, Schulenberg JE. Monitoring the Future Occasional Paper No 76. Ann Arbor, MI: University of Michigan Institute for Social Research; 2011. The Monitoring the Future Project After Thirty-Seven Years: Design and Procedures. [Google Scholar]
- 25.Bachman JG, Wadsworth KN, O’Malley PM, Johnston LD, Schulenberg JE. Smoking, Drinking and Drug Use in Young Adulthood: The Impacts of New Freedoms and New Responsibilities. Mahwah, NJ: Lawrence Erlbaum Associates; 1997. [Google Scholar]
- 26.Schulenberg J, Maggs JL, Long SW, Sher KJ, Gotham HJ, Baer JS, Kivlahan DR, Marlatt GA, Zucker RA. The problem of college drinking: insights from a developmental perspective. Alcoholism: Clinical and Experimental Research. 2001;25:473–477. [PubMed] [Google Scholar]
- 27.Sher KJ, Gotham HJ. Pathological alcohol involvement: a developmental disorder of young adulthood. Dev Psychopathol. 1999;11:933–956. doi: 10.1017/s0954579499002394. [DOI] [PubMed] [Google Scholar]
- 28.McCabe SE, West BT, Boyd CJ. Leftover prescription opioids and nonmedical use among U.S. high school seniors: A multi-cohort national study. J Adolesc Health. 2013;52:480–485. doi: 10.1016/j.jadohealth.2012.08.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.McCabe SE, Cranford JA, Boyd CJ, Teter CJ. Motives, diversion and routes of administration associated with nonmedical use of prescription opioids. Addict Behav. 2007;32:562–575. doi: 10.1016/j.addbeh.2006.05.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Fortuna RJ, Robbins BW, Caiola E, Joynt M, Halterman JS. Prescribing of controlled medications to adolescents and young adults in the United States. Pediatrics. 2010;126:1108–1116. doi: 10.1542/peds.2010-0791. [DOI] [PubMed] [Google Scholar]
- 31.Thomas CP, Conrad P, Casler R, Goodman E. Trends in the use of psychotropic medications among adolescents, 1994 to 2001. Psychiatr Serv. 2006;57:63–69. doi: 10.1176/appi.ps.57.1.63. [DOI] [PubMed] [Google Scholar]
- 32.Zacny J, Bigelow G, Compton P, Foley K, Iguchi M, Sannerud C. College on Problems of Drug Dependence taskforce on prescription opioid nonmedical use and abuse: position statement. Drug Alcohol Depend. 2003;69:215–232. doi: 10.1016/s0376-8716(03)00003-6. [DOI] [PubMed] [Google Scholar]