DIFFERENCES IN PRESCRIPTION STIMULANT MISUSE MOTIVES ACROSS ADOLESCENTS AND YOUNG ADULTS IN THE UNITED STATES

Abstract

Objective:

While prescription stimulant misuse (PSM) is common in adolescents and young adults (AYAs), PSM motives are poorly understood. We examined a number of PSM motives across the AYA age spectrum using the 2015–18 National Survey on Drug Use and Health.

Methods:

In all, 102,088 AYAs (14–25 years) were included. Individual PSM motives (e.g., to study) and motive categories (i.e., cognitive enhancement only, recreational only, weight loss only, combined motives) were examined by age. Logistic regression models examined links between individual motives or motive categories and educational status, substance use, DSM-IV SUD, and mental health correlates.

Results:

Significant differences were found across AYAs in cognitive enhancement only (14 years=40.4%; 24/25 years=71.2%) and recreational only (14 years=25.8%; 24/25 years=9.8%) or combined PSM motives, (14 years=32.3%; 24/25 years=18.0%); college students and graduates had particularly high rates of cognitive enhancement only (college=78.2%; graduates=74.7%; non- college=63.5%). Recreational only and combined motives were significantly elevated in AYAs with any past-year SUD, especially “to get high” (78–136% higher in those with SUD). While any PSM was associated with higher odds of SUD and mental health outcomes, including suicidal ideation, odds were highest for recreational or combined motives.

Conclusion:

Cognitive enhancement with PSM more often occurs in young adults compared to adolescents; college students endorse more cognitive enhancement than those not in school; and the presence of any PSM in AYAs is linked to a more substance use, suicidal ideation, and other psychopathology. PSM prevention in adolescents as well as screening and intervention among AYA is highly recommended.

INTRODUCTION

Prescription stimulant misuse (PSM), defined as use of another’s controlled stimulant medication or use of one’s own medication in ways not intended by the prescriber, is common in adolescents and young adults (AYAs). The US Monitoring the Future (MTF) survey found that 11.0% of college students engaged in past-year Adderall® misuse, with high rates in non-college young adults (9.1%).¹ While lower in adolescents (<5%), past-year PSM in AYAs exceeds most other drug use prevalence, with the exceptions of alcohol, tobacco, cannabis, and opioid misuse for adolescents.^{1, 2} PSM is associated with concerning correlates in AYAs, including other substance use, psychopathology, and criminal offending.^3–9 Identification of modifiable factors associated with PSM is an important goal for research, and motives, or underlying reasons for PSM, are such a factor. Cannabis use motives covary with changes in use,¹⁰ and alcohol use interventions targeting motives promote significant use reductions.^{11, 12} Understanding PSM motives in AYAs could direct prevention, screening, and treatment to limit PSM.

PSM motives among AYAs have been most studied in local samples of US undergraduate college students. In undergraduates, PSM is primarily motivated by cognitive enhancement,^13–19 typically motives aimed at improving studying, concentration, and alertness. Recreational motives, aimed at enhancing positive affect (e.g., “to get high”), altering other drug effects, and experimentation, are also common, with “to get high” usually the most common recreational motive.^{13, 14} Weight loss motives are more infrequent, though somewhat common in female undergraduates.^{14, 20}

Using MTF data, McCabe and Cranford²¹ found that the most common adolescent PSM motive was “to get more energy”, followed by experimentation, “to get high”, and “to stay awake”, with each near or above 50% endorsement. Weight loss was more common in adolescents (36%). PSM only to study is rare in adolescents (7.4%), and 51.2% endorsed non-study motives,²² suggesting that recreational motives may be more prominent in adolescents.

Thus, preliminary evidence suggests that cognitive enhancement PSM motives are more common in young adults. This corresponds to the increased academic demands among young adults in school-based samples, but exclusive use of school-based samples may bias conclusions across all AYAs. School enrollment/engagement covary with PSM prevalence in AYAs,²³ and motives may as well. The lack of research on PSM motives in AYAs not in school is a major limitation of the literature. AYAs not in school are heterogeneous, including both college graduates and those who dropped out of high school, and these disparate groups have different PSM profiles.^{23, 24} Local samples reflect the idiosyncratic norms of that university’s environment, as PSM varies with university characteristics,²⁵ suggesting a need for studies using nationally representative data. Finally, research examining PSM motive differences through the AYA age spectrum is missing, and such research could direct prevention, screening, and intervention in a developmentally appropriate way.

This research uses four aggregated years of data (2015–18) from the nationally representative US National Survey on Drug Use and Health (NSDUH), aiming to: (1) examine age-based differences in PSM motives and motive categories (i.e., cognitive enhancement only, recreational only, weight loss only, and combined motives); (2) investigate whether motives vary by educational status; (3) link motives to past-year substance use disorder (SUD) prevalence; and,(4) examine the sociodemographic, substance use, and mental health correlates of motive categories.

METHODS

The NSDUH is an annual survey of US residents 12 years and older.^{26, 27} Sampling used an independent, multistage area probability design, weighted for population-based estimates. Sensitive topics were assessed by audio computer-assisted self-interviewing (ACASI) to maximize honesty, with skip-outs and consistency checks to maximize data completeness and accuracy. For 2015–18, the weighted screening response rate ranged from 79.7% to 73.31%, and the weighted interview rate ranged from 69.7% to 66.6%, similar to other nationally representative studies.²⁸ The Research Triangle International IRB approved the NSUDH,²⁹ and the first author’s IRB exempted this research from further oversight.

Participants

For 2015–18, 86,918 AYAs 14–25 years of age composed the analytic sample, of 93,039 possible (93.4%; Table 1). Twelve and thirteen-year-old adolescents were excluded because of very low PSM rates. AYAs were also excluded if they were: (1) homeschooled adolescents (n= 184); (2) young adults in secondary school (n= 2,958); or (3) missing educational or motive data (n= 2,979). Homeschooled adolescents were excluded due to sample size, and young adults in secondary school were excluded because of developmental^{30, 31} and stimulant PDM prevalence²³ differences from other young adults. In addition, young adults in secondary school often have less responsibility for stimulant management than other young adults,³² influencing PSM.

Table 1:

Sociodemographic Characteristics of the Analytic Sample

	Adolescents	Young Adults	Total
Sample Size	35,732	51,186	86,918
	% (95% CI)	% (95% CI)	% (95% CI)
Prevalence of PSM	2.3 (2.1–2.5)	7.4 (7.0–7.7)	5.6 (5.4–5.9)
Male Sex	50.5 (49.8–51.3)	49.5 (48.9–50.1)	49.8 (49.5–50.3)
Age
14 years	24.8 (24.2–25.4)	not applicable	8.5 (8.2–8.7)
15 years	25.0 (24.5–25.6)	not applicable	8.5 (8.3–8.8)
16 years	25.5 (24.9–26.0)	not applicable	8.7 (8.5–8.9)
17 years	24.7 (24.1–25.3)	not applicable	8.4 (8.2–8.7)
18 years	not applicable	8.9 (8.5–9.3)	5.9 (5.6–6.1)
19 years	not applicable	12.2 (11.8–12.7)	8.0 (7.7–8.4)
20 years	not applicable	12.4 (12.0–12.8)	8.2 (7.9–8.5)
21 years	not applicable	12.9 (12.5–13.4)	8.5 (8.2–8.8)
22 years	not applicable	12.9 (12.5–13.3)	8.5 (8.3–8.7)
23 years	not applicable	13.5 (13.1–14.0)	8.9 (8.6–9.2)
24/25 years	not applicable	27.1 (26.4–27.9)	17.9 (17.4–18.3)
Race/Ethnicity
White	53.5 (52.6–54.3)	55.1 (54.2–55.9)	54.5 (53.7–55.2)
Black	13.5 (12.9–14.1)	13.8 (13.3–14.3)	13.7 (13.3–14.2)
Hispanic/Latinx	23.4 (22.7–24.1)	21.5 (20.8–22.3)	22.2 (21.6–22.8)
Asian-American	5.5 (5.1–5.9)	6.2 (5.9–6.6)	6.0 (5.7–6.3)
American Indian	0.5 (0.4–0.6)	0.6 (0.5–0.7)	0.6 (0.5–0.7)
Hawaiian/Pacific Islander	0.4 (0.3–0.6)	0.4 (0.3–0.5)	0.4 (0.4–0.5)
Multiracial	3.2 (3.0–3.4)	2.4 (2.2–2.5)	2.6 (2.5–2.8)
Household Income
< $20,000	15.1 (14.4–15.8)	28.0 (27.1–28.9)	23.6 (22.9–24.3)
$20,000–49,999	27.4 (26.7–28.1)	32.2 (31.6–32.9)	30.6 (30.0–31.2)
$50,000–74,999	14.2 (13.6–14.7)	13.9 (13.4–14.3)	14.0 (13.6–14.3)
≥ $75,000	43.4 (42.3–44.4)	25.9 (25.2–26.7)	31.9 (31.1–32.6)
Population Density
CBSA ≥ 1 million persons	54.9 (53.9–55.8)	53.8 (52.9–54.7)	54.2 (53.5–54.9)
CBSA <1 million persons	39.4 (38.5–40.3)	41.6 (40.7–42.6)	40.9 (40.1–41.6)
Not in a CBSA	5.7 (5.2–6.3)	4.5 (4.2–4.9)	4.9 (4.6–5.4)

Data: 2015–2018 NSDUH

Abbreviations: CBSA = Core-based Statistical Area; PSM = Prescription stimulant misuse; 95% CI = 95% confidence interval of the prevalence estimate

Measures

Participants were asked about lifetime stimulant use (which includes PSM), and those with lifetime use were then asked about lifetime PSM. PSM is stimulant use “in any way a doctor did not direct: using it without a prescription…in greater amounts, more often, or longer than you were told to take it; using it in any other way a doctor did not direct…”. Those with lifetime PSM are also asked about past-year PSM. The PSM assessment includes a variety of trade (e.g., Adderall®, Ritalin®) and generic names (e.g., dextroamphetamine, methylphenidate) and pictures of assessed medications to improve recall.

In those with past-year PSM, motives at the most recent PSM episode are queried via, “What were the reasons you used [specific stimulant] the last time in any way a doctor did not direct you to use it/them?” Participants selected as many motives as applied from: lose weight, concentrate, be alert, study, experiment, get high, alter other drug effects; “because I’m hooked” and other were also included. Answers were dichotomous (yes/no). PSM motives were grouped into four categories: weight loss only, cognitive enhancement only (i.e., concentrate, be alert, and/or study), recreational only (i.e., experiment, get high, alter other drug effects, and/or “because I’m hooked), and combined (i.e., motives from multiple categories). Categories came from past research that categorized motives based on FDA stimulant indications and face validity, with “other” excluded.^{15, 21, 33}

Educational status was also assessed. Adolescent participants were in school and low risk for dropout, in school and at-risk for dropout, or not in school. Dropout risk was based on three characteristics associated with dropout: grades ≤D+ at the last grading period, being at least one year older than typical for grade, and the adolescent stating that s/he “hated going to school”.³⁴ Young adults were in college, college graduates, not in school and high school (HS) graduate, not in school and dropped out of HS.

Sociodemographics were race/ethnicity, sex, grades of C+ or worse at last grading period (only adolescents in school), uninsured status, and past-year offending behavior. Past-year offending behavior was one or more of past-year illegal drug sales, attempted theft of anything worth $50 or more, and/or attacks with intent to seriously harm someone else.

Substance use correlates were past-year prescription opioid misuse, past-year prescription benzodiazepine misuse, past-month binge alcohol use, past-year cannabis use, and past-year substance use disorder (SUD). Past-year SUD is DSM-IV substance abuse or dependence from alcohol, marijuana, cocaine, heroin, hallucinogen, inhalant, methamphetamine, or prescription opioid use, and prescription tranquilizer, sedative, or stimulant misuse. Per NIAAA guidelines,³⁵ past-month binge drinking is four or five alcoholic drinks (for females and males, respectively) in one occasion.

Mental health correlates were (all past-year) major depressive episode, mental health treatment, serious psychological distress (SPD), and suicidal ideation. SPD was from the K6 assessment of non-specific psychological distress,³⁶ and SPD and suicidal ideation were assessed only in young adults.

Data Analyses

Analyses were performed in STATA 16.0 (College Station, TX), incorporating the NSDUH complex survey features. Given use of four years of data, an adjusted person-level weight was used (weight/4), per guidelines.³⁷ Cross-tabulations estimated prevalence and 95% confidence intervals of PSM motives and motive categories by age, educational status, and past-year SUD (the last two separately in adolescents and young adults). For age-based analyses, linearized estimates of difference by year of age were performed, using the margins command. For educational and past-year SUD status, logistic regression models estimated differences between groups, with Bonferroni-corrections for multiple comparisons. Finally, logistic regression and multinomial logistic regression (for race/ethnicity only) models examined the relationship between motive category (excluding weight loss only, due to sample size) and the sociodemographic, substance use, and mental health correlates. All regression models controlled for age, race/ethnicity, sex, population density, and household income.

RESULTS

Overall, 5.6% engaged in past-year PSM; young adult (7.4%) was more common than adolescent PSM (2.3%; Table 1). Motive differences by age are in Table 2. Four PSM motives differed significantly: alert, experiment, get high, and other reasons. While PSM for alertness was 2.2% greater per year (p<0.0001), from 33.6% at 14 years to 53.7% at 24/25 years, the other three motives were more prevalent at younger ages, with smaller prevalence rates associated with aging: experiment by 1.4% (14 years: 19.8%, 24/25 years: 9.4%; p<0.0001), get high by 1.2% (14 years: 27.6%, 24/25 years: 14.1%; p<0.0001), and other reasons by 0.3% (14 years: 10.3%, 24/25 years: 3.5%; p=0.007). Also, study-related motives peaked in the college years, though age-based differences were non-significant. PSM for cognitive enhancement only was 1.8% greater per year (14 years: 40.4%, 24/25 years: 71.2%; p<0.0001), while recreational only (14 years: 25.8%, 24/25 years: 9.8%; p< 0.0001) and combined motives (14 years: 32.3%, 24/25 years: 18.0%; p=0.008) were 1.1% and 0.6% smaller per year, respectively.

Table 2:

Individual Prescription Stimulant Misuse (PSM) and PSM Motive Categories by Age Group, Among Those with Past-Year PSM

	14 years	15 years	16 years	17 years	18 years	19 years	20 years	21 years	22 years	23 years	24/25 years
Sample Size	70	149	263	368	253	421	506	547	546	509	790
	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)
Individual Motives
To Lose Weight	12.2 (5.1–26.3)	3.6 (1.6–8.2)	10.2 (5.2–19.2)	3.3 (1.7–6.3)	8.1 (5.3–12.3)	5.1 (3.1–8.1)	4.6 (2.8–7.5)	5.9 (3.8–8.9)	8.1 (5.6–11.6)	5.9 (3.8–9.1)	5.9 (4.1–8.5)
To Concentrate	59.8 (46.0–72.2)	41.6 (31.3–52.5)	56.6 (48.0–64.9)	59.8 (52.9–66.3)	58.7 (51.9–65.1)	61.3 (55.4–67.0)	61.1 (56.4–65.6)	61.2 (56.1–66.7)	58.1 (52.5–63.5)	55.7 (51.1–60.3)	56.2 (51.7–60.5)
To Be Alert	33.6 (20.6–49.8)	36.5 (27.0–47.1)	30.9 (23.9–39.0)	43.0 (37.5–48.6)	40.7 (34.4–47.4)	39.8 (34.4–45.4)	41.4 (35.5–47.6)	46.1 (41.2–51.1)	46.1 (41.1–51.2)	53.4 (48.4–58.4)	53.7 (48.4–58.9)
To Study	38.0 (25.7–52.1)	38.7 (28.3–50.3)	39.7 (32.5–47.4)	40.9 (34.7–47.3)	47.4 (39.4–55.5)	56.1 (50.1–62.0)	53.4 (49.2–57.7)	57.8 (53.0–62.4)	52.0 (46.0–57.9)	49.3 (44.5–54.1)	38.4 (33.9–43.1)
To Experiment	19.8 (10.7–33.8)	26.7 (19.3–35.8)	27.1 (20.3–35.3)	17.9 (13.5–23.4)	13.3 (9.3–18.5)	9.7 (6.5–14.2)	13.1 (9.7–17.5)	9.3 (6.7–12.8)	9.7 (6.8–13.5)	7.5 (5.5–10.3)	9.4 (7.3–11.9)
To Get High	27.6 (18.0–39.9)	33.8 (23.5–45.9)	23.9 (18.0–30.9)	22.6 (17.6–28.7)	15.8 (11.3–21.5)	13.8 (9.9–18.9)	12.4 (9.5–16.1)	13.7 (10.8–17.1)	10.9 (8.1–14.5)	14.0 (10.6–18.2)	14.1 (11.2–17.7)
To Alter Other Drug Effects	2.0 (0.3–11.1)	3.4 (1.2–9.6)	3.1 (1.5–6.5)	3.2 (1.5–6.6)	1.7 (0.7–3.9)	3.5 (1.7–6.9)	3.1 (1.8–5.4)	3.2 (1.7–5.9)	3.2 (2.0–5.3)	3.7 (2.3–6.1)	3.9 (2.6–6.0)
“Because I’m Hooked”	0.1 (0.01–0.5)	1.9 (0.6–6.0)	1.0 (0.2–3.7)	0.3 (0.1–1.1)	1.6 (0.6–4.5)	0.2 (0.03–1.7)	1.2 (0.5–2.8)	0.2 (0.04–0.9)	0.8 (0.2–2.4)	1.5 (0.5–4.4)	1.6 (0.8–3.2)
Other Reason	10.3 (4.9–20.4)	9.6 (4.8–18.3)	3.5 (1.8–6.9)	7.0 (4.6–10.6)	1.3 (0.5–3.2)	3.7 (1.9–7.4)	2.7 (1.5–4.8)	2.4 (1.2–4.9)	0.9 (0.4–2.0)	2.9 (1.4–6.0)	3.5 (2.2–5.4)
Motive Categoriesa
Weight Loss Only	1.5 (0.3–7.1)	1.8 (0.4–8.1)	1.3 (0.3–4.7)	0.4 (0.1–1.6)	1.3 (0.5–3.4)	1.0 (0.3–3.5)	1.9 (0.8–4.7)	1.4 (0.6–3.2)	1.4 (0.5–3.7)	0.5 (0.1–1.9)	1.0 (0.5–2.0)
Cognitive Enhancement	40.4 (27.2–55.1)	47.5 (38.1–57.0)	53.1 (45.6–60.5)	62.0 (55.7–67.9)	67.6 (60.9–73.7)	73.4 (66.8–79.1)	70.9 (65.3–76.0)	74.7 (69.8–79.1)	71.9 (67.1–76.3)	74.7 (70.1–78.8)	71.2 (66.6–75.3)
Recreational	25.8 (15.9–39.1)	27.5 (20.0–36.7)	22.5 (16.4–30.0)	14.6 (11.0–19.1)	12.6 (8.9–17.5)	8.5 (5.8–12.4)	8.5 (6.2–11.6)	5.7 (3.5–9.0)	8.6 (6.1–12.1)	8.5 (6.3–11.6)	9.8 (7.7–12.4)
Combined	32.3 (20.1–47.4)	23.2 (15.4–33.5)	23.2 (16.4–31.7)	23.0 (17.9–29.2)	18.5 (14.1–23.8)	17.1 (12.4–23.2)	18.6 (14.9–23.0)	18.2 (14.1–23.2)	18.1 (14.3–22.7)	16.3 (13.1–20.1)	18.0 (14.6–22.0)

Data: 2015–18 NSDUH

^aCognitive Enhancement Only is composed of To Concentrate, To Be Alert and To Study; Recreational is To Experiment, To Get High, To Alter Other Drug Effects and “Because I’m Hooked”.

Educational status had no relationship with PSM motives or categories in adolescents (Supplementary Table 1), but motives differed significantly by educational status in young adults (Table 3). Young adults in college had the highest rate of study-related motives (66.6%) and lowest rate of to get high (9.3%); both differed significantly from young adults not in school (ps<0.01). College students also had the highest rate of to concentrate (63.7%) and lowest of weight loss (3.9%), and young adults not in school were significantly more likely to endorse weight loss motives (ps<0.01). Those not completing HS had the lowest rates of to concentrate (47.9%), study (15.4%), and be alert (41.8%), but the highest rates of weight loss (11.8%) and get high (24.8%). College graduates and HS graduates not in school were generally intermediate, with elevated rates of PSM for alertness in both.

Table 3:

Individual Prescription Stimulant Misuse (PSM) and PSM Motive Categories by Educational Characteristic in Young Adults (18–25 years) with Past-Year PSM

	In College (1)a	College Graduates (2)a	Not in School and HS Graduate (3)a	Not in School and dropped out of HS (4)a	Entire Sample
Sample Size	1,686	534	1,179	173	3,572
	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)	% (95% CI)
Individual Motives
To Lose Weight	3.9 (3.0–5.1)3, 4	4.5 (2.8–7.0)3, 4	9.4 (7.3–12.0)1, 2	11.8 (6.2–21.1)1, 2	6.0 (5.1–7.1)
To Concentrate	63.7 (61.0–66.4)2, 3	53.2 (48.5–57.8)1	55.5 (52.3–58.8)1	47.9 (37.5–58.6)	58.8 (57.0–60.6)
To Be Alert	42.2 (39.4–45.0)3	54.7 (49.0–60.2)	52.6 (48.4–56.8)1	41.8 (34.3–49.8)	47.4 (45.1–49.7)
To Study	66.6 (63.6–69.5)3, 4	53.9 (48.9–58.9)3, 4	27.2 (24.2–30.4)1, 2, 4	15.4 (9.7–23.6)1, 2, 3	50.4 (48.2–52.6)
To Experiment	9.5 (7.8–11.6)	7.1 (5.0–9.9)	12.0 (9.3–15.2)	10.1 (6.4–15.5)	9.9 (8.7–11.3)
To Get High	9.3 (7.7–11.3)3, 4	13.4 (9.9–17.9)	18.2 (15.2–21.6)1	24.8 (18.7–32.0)1	13.4 (12.0–14.8)
To Alter Other Drug Effects	3.3 (2.4–4.6)	5.2 (3.3–8.1)	2.8 (1.8–4.2)	2.3 (0.8–6.7)	3.4 (2.8–4.2)
“Because I’m Hooked”	0.8 (0.4–1.5)	1.6 (0.6–3.8)	1.2 (0.6–2.2)	no cases	1.0 (0.6–1.6)
Other Reason	1.3 (0.8–2.2)3, 4	2.5 (1.2–5.4)	4.1 (2.9–5.9)1	6.4 (3.5–11.6)1	2.6 (2.0–3.4)
Motive Categoriesb
Weight Loss Only	0.5 (0.2–1.0)3, 4	0.2 (0.02–1.3)	2.4 (1.4–4.1)1	3.9 (1.5–10.0)1	1.1 (0.8–1.7)
Cognitive Enhancement Only	78.2 (75.1–81.0)	74.7 (69.8–78.9)	64.7 (60.5–68.6)	55.4 (45.6–64.7)	72.6 (70.5–74.6)
Recreational-Only	5.0 (3.8–6.5)3, 4	8.6 (5.8–12.5)	11.9 (9.6–14.6)1	21.4 (14.8–30.0)1	8.4 (7.3–9.5)
Combined	16.4 (13.8–19.3)	16.6 (13.3–20.6)	21.1 (17.7–24.9)	19.3 (13.6–26.7)	17.9 (16.3–19.7)

Data: 2015–18 NSDUH

^aSuperscript numbers denote differences from the group with the letter (i.e., ¹ denotes a significant difference from young adults in college), with all comparisons adjusted for age, race/ethnicity, sex, household income and population density and Bonferroni corrected for multiple comparisons (i.e., p-value for significance is 0.0083, or 0.05/6 comparisons).

^bCognitive Enhancement Only is composed of To Concentrate, To Be Alert and To Study; Recreational is To Experiment, To Get High, To Alter Other Drug Effects and “Because I’m Hooked”. Cognitive Enhancement Only was set as the reference in multinomial logistic regressions.

Significant comparisons for motive categories are set at an a priori p-value of 0.0001 or less, given the large number of comparisons. Abbreviation: 95% CI = 95% confidence interval of the estimate.

Table 4 captures PSM motive and category differences by past-year SUD status, with three similar outcomes across age groups. First, AYAs with past-year SUD had significantly higher rates of PSM to get high (adolescents: 32.2%, young adults: 18.9%) than those without SUD (adolescents: 18.1%, young adults: 8.0%; both ps≤0.001). Second, AYAs with SUD had a higher prevalence of recreational only versus cognitive enhancement motives (adolescents: 23.8%, young adults: 10.5%) than those without SUD (adolescents: 15.9%, young adults: 6.5%; ps≤0.001). Third, AYA with SUD had a higher prevalence of combined motives (adolescents: 30.0%, young adults: 23.5%) than those without SUD (adolescents: 17.3%, young adults: 12.7%; ps≤0.001), versus cognitive enhancement only motives. Also, young adults with SUD were more likely to endorse to lose weight, be alert, experiment, alter other drug effects, and “because I’m hooked” than those without past-year SUD. In contrast, young adults with SUD had lower rates of study- related PSM than those without SUD.

Table 4:

Individual Prescription Stimulant Misuse (PSM) and PSM Motive Categories by Past-Year SUD Status in Adolescents and Young Adults with Past-year PSM

	Adolescents			Young Adults
	Without SUDa	With SUDa	p-valueb	Without SUDa	With SUDa	p-valueb
Sample Size	428	452		1,811	1,761
	% (95% CI)	% (95% CI)		% (95% CI)	% (95% CI)
Individual Motives
To Lose Weight	4.7 (2.7–8.0)	8.9 (5.9–13.1)	0.27	5.1 (3.7–6.8)	7.2 (5.8–8.9)	0.024
To Concentrate	59.2 (53.8–64.3)	50.1 (42.8–57.5)	0.078	59.8 (57.1–62.5)	57.9 (54.9–60.9)	0.35
To Be Alert	34.5 (29.5–39.8)	40.4 (33.7–47.5)	0.43	44.4 (41.6–47.3)	49.9 (46.5–53.3)	0.002
To Study	40.1 (34.8–45.6)	37.9 (31.7–44.7)	0.42	53.8 (50.9–56.6)	46.6 (43.4–50.0)	0.001
To Experiment	19.2 (15.1–24.2)	25.4 (20.6–30.9)	0.12	8.0 (6.6–9.7)	11.9 (10.2–13.9)	0.002
To Get High	18.1 (13.6–23.5)	32.2 (25.8–39.4)	0.001	8.0 (6.6–9.7)	18.9 (16.7–21.4)	< 0.001
To Alter Other Drug Effects	2.7 (1.2–6.4)	3.3 (1.8–6.0)	0.96	1.8 (1.2–2.7)	5.1 (3.9–6.5)	< 0.001
“Because I’m Hooked”	0.5 (0.1–1.8)	0.9 (0.3–2.7)	0.99	0.1 (0.01–0.5)	1.9 (1.2–3.1)	0.002
Other Reason	7.9 (5.4–11.4)	6.6 (4.0–10.7)	0.17	2.2 (1.5–3.2)	3.0 (2.2–4.1)	0.15
Motive Categoriesc
Weight Loss Only	1.7 (0.7–4.2)	1.7 (0.8–3.3)	0.21	0.8 (0.5–1.5)	1.5 (0.9–2.5)	0.022
Cognitive Enhancement Only	65.1 (59.6–70.2)	44.6 (38.6–50.7)	base outcome	80.0 (77.5–82.2)	64.6 (61.3–67.8)	base outcome
Recreational-Only	15.9 (12.4–20.3)	23.8 (18.1–30.6)	0.001	6.5 (5.1–8.2)	10.5 (9.0–12.1)	< 0.001
Combined	17.3 (13.2–22.5)	30.0 (25.1–35.3)	0.001	12.7 (10.9–14.9)	23.5 (20.6–26.6)	< 0.001

Data: 2015–18 NSDUH

^aPast-Year SUD is based on the DSM-IV definition for substance abuse or dependence from alcohol, cannabis, cocaine, heroin, methamphetamine, inhalants, hallucinogens, prescription opioids, prescription tranquilizers, prescriptions sedatives and/or prescription stimulants.

^bp-values are from logistic or multinomial logistic models, controlling for age, race/ethnicity, sex, household income and population density.

^cCognitive Enhancement Only is composed of To Concentrate, To Be Alert and To Study; Recreational is To Experiment, To Get High, To Alter Other Drug Effects and “Because I’m Hooked”.

Abbreviation: 95% CI = 95% confidence interval of the estimate

Finally, all substance use and mental health correlates were more likely in those with any PSM than those without PSM, though odds were generally lowest for cognitive enhancement only (Table 5). To illustrate, the odds ratio for any past-year SUD (versus no PSM) was 6.17 for cognitive enhancement, but 14.89 and 12.83 for recreational or combined motives, respectively. Odds of past-year suicidal ideation were 54%, 202%, and 252% higher in those with cognitive enhancement, recreational, or combined motives (respectively), versus no PSM. PSM across motive categories was associated with adolescent grades of C+ or lower. Differences by race/ethnicity were mainly driven by higher rates of PSM in white and multiracial AYAs, primarily within cognitive enhancement only, given significantly higher rates versus other groups. Past-year offending behavior was higher in those with any PSM, though rates were highest in those with recreational or combined motives. Post hoc sensitivity analyses found no outcome differences when alertness was classified as a recreational motive, instead of cognitive enhancement.

Table 5:

Univariable Outcomes by Prescription Stimulant Misuse (PSM) Motive Category, Among Those with Past-Year PSM

	No Past-Year PSM	Cognitive Enhancement Only PSM	Recreational-Only PSM	Combined PSM
Sample Size	98,438	3,002	542	852
Sociodemographic Outcomes
Race/Ethnicity	RRR (95% CI)	RRR (95% CI)	RRR (95% CI)	RRR (95% CI)
White, non-Hispanic	1.00 (Reference)	1.00 (Reference)	1.00 (Reference)	1.00 (Reference)
Black, non-Hispanic	1.00 (Reference)	0.21 (0.16–0.26)***	0.23 (0.14–0.38)***	0.22 (0.16–0.32)***
Indigenous, non-Hispanic	1.00 (Reference)	0.21 (0.12–0.39)***	0.78 (0.33–1.84)	0.68 (0.59–0.69)
Asian-American, non-Hispanic	1.00 (Reference)	0.40 (0.32–0.51)***	0.64 (0.31–1.35)	0.26 (0.16–0.43)***
Multiracial, non-Hispanic	1.00 (Reference)	0.90 (0.70–1.15)	0.86 (0.55–1.34)	1.04 (0.71–1.54)
Hispanic/Latino	1.00 (Reference)	0.45 (0.39–0.52)***	0.49 (0.34–0.72)***	0.36 (0.25–0.50)***
	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)
Male	1.00 (Reference)	1.33 (1.20–1.46)***	1.19 (0.95–1.49)	1.08 (0.92–1.28)
Current Grades C+ or Lowera	1.00 (Reference)	1.39 (1.04–1.84)*	1.74 (1.13–2.67)*	1.91 (1.27–2.87)**
Uninsured Status	1.00 (Reference)	1.70 (1.44–2.01)***	1.04 (0.77–1.40)	1.45 (1.05–2.00)*
Past-Year Offending Behaviorb	1.00 (Reference)	4.01 (3.57–4.50)***	7.89 (6.29–9.90)***	8.33 (6.71–10.34)***
Substance Use Outcomes	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)
Past-Year Prescription Opioid Misuse	1.00 (Reference)	6.17 (5.55–6.86)***	14.89 (11.90–18.63)***	12.83 (10.88–15.13)***
Past-Year Benzodiazepine Misuse	1.00 (Reference)	11.17 (9.88–12.63)***	20.82 (15.96–27.16)***	17.28 (14.18–21.06)***
Past-Month Binge Alcohol Use	1.00 (Reference)	7.68 (6.73–8.77)***	7.25 (5.39–9.76)***	6.39 (5.18–7.89)***
Past-Year Cannabis Use	1.00 (Reference)	9.94 (8.88–11.11)***	19.33 (13.20–28.31)***	17.74 (13.22–23.80)***
Past-Year Any Substance Use Disorder (SUD)c	1.00 (Reference)	6.29 (5.64–7.01)***	14.77 (11.42–19.11)***	16.08 (13.28–19.47)***
Mental Health Outcomes	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)
Past-Year Major Depressive Episode	1.00 (Reference)	1.77 (1.53–2.05)***	3.23 (2.49–4.19)***	3.87 (3.21–4.67)***
Past-Year Mental Health Treatment	1.00 (Reference)	1.49 (1.35–1.66)***	2.85 (2.19–3.69)***	2.67 (2.22–3.20)***
Past-Year Serious Psychological Distressd	1.00 (Reference)	1.44 (1.28–1.61)***	3.12 (2.40–4.04)***	3.32 (2.62–4.22)***
Past-Year Suicidal Ideationd	1.00 (Reference)	1.54 (1.30–1.83)***	3.02 (2.36–3.87)***	3.52 (2.83–4.38)***

Data: 2015–18 NSDUH

^*denotes p ≤ 0.05;

^**denotes p ≤ 0.01;

^***denotes p ≤ 0.001; All analyses control for age, sex (when applicable), race/ethnicity (when applicable), population density and household income.

^aAnalysis only among adolescents in school (n = 49,049)

^bPast-Year Offending Behavior is one or more of past-year illegal drug sales, attempted theft of anything worth $50 or more, and/or attacks with intent to seriously harm someone else.

^cPast-Year SUD is based on the DSM-IV definition for substance abuse or dependence from alcohol, cannabis, cocaine, heroin, methamphetamine, inhalants, hallucinogens, prescription opioids, prescription tranquilizers, prescriptions sedatives and/or prescription stimulants.

^dAnalyses only in young adults, 18 years and older (n = 55,646 and 55,121, respectively)

Abbreviations: OR = odds ratio; RRR = relative risk ratio; 95% CI = 95% confidence interval of the estimate

DISCUSSION

There were four key findings: one, PSM motives differ significantly over the AYA age range, with lesser endorsement of recreational and greater endorsement of cognitive enhancement motives in young adults, versus adolescents. Two, while motives do not correspond to adolescent educational status, college young adults have greater relative endorsement of cognitive enhancement motives over young adults not in school. Three, recreational and combined motives are strongly linked to past-year any SUD across AYAs, with a striking link between “to get high” and past-year SUD in AYAs – 32.2% of adolescents and 18.9% of young adults with SUD endorsed “to get high”. Put differently, endorsement of “to get high” was 78% and 136% higher in adolescents and young adults, respectively, in those with any past-year SUD. Four, any PSM in AYAs is linked to greater odds of concurrent substance use, suicidal ideation, and other psychopathology, though odds were highest with recreational only or combined PSM motives.

For individual motives, to be alert was more frequently endorsed with age, from 33.6% in 14-year-olds to 53.7% in 24/25-year-olds. Despite this, the prevalence of PSM to concentrate did not differ across the AYA age span, and PSM to study was less common after the tradition college graduation age of 22 or 23 years. The discrepancy between alertness and concentration-related motives warrants further investigation, as do the specific goals motivating young adults to engage in PSM for alertness. Alertness is heterogeneous and may not represent cognitive enhancement, and future studies using latent class models could evaluate this in more detail. In contrast to alertness, experiment and get high became less common with aging, as did recreational only and combined PSM motives.

Young adults in college had a greater prevalence of cognitive enhancement motives than young adults not in school. This was particularly driven by PSM for concentration- and study- related motives. College graduates were intermediate, but they had rates of cognitive enhancement motives generally closer to young adults in college. Cognitive enhancement may underlie the higher PSM prevalence rates in college students and graduates,²³ corresponding to the belief in the academic/cognitive benefits of PSM,³⁸ despite limited evidence of such benefits.³ Alternatively, cognitive enhancement may reflect self-treatment of ADHD symptoms and/or underlying neurocognitive deficits, both elevated in young adults engaged in PSM.³⁹ Young adults not in school endorsed recreational PSM motives much more frequently, particularly PSM to get high among those who dropped out of HS. Given the odds of SUD associated with PSM to get high, this may mark young adults who dropped out of HS as a particularly vulnerable group. Finally, while still infrequent, young adults not in school had higher rates of weight loss motives; this warrants replication and further study.

Notably, recreational motives were associated with any past-year SUD across AYAs. Presence of euphoria seeking may signal elevated SUD in AYAs, providing a helpful tool to guide prevention, screening, and intervention. While study-related motives were less common in young adults with SUD, per Table 5 and Compton et al.,⁴⁰ any PSM is associated with higher SUD rates. Thus, while relative risk is lower than other motives, PSM for study-related motives is still associated with other substance use and SUD, and those engaged in PSM only for cognitive enhancement had over 6 times greater odds of SUD than AYAs without past-year PSM. This is consistent with adolescent research,²² highlighting that all PSM is associated with increased substance use and psychopathology. Finally, the elevations found in suicidal ideation across PSM motives warrant further attention and add to compelling evidence of a link between prescription drug misuse and suicidality across the lifespan.^41–45

Limitations

Given the cross-sectional data, no causal inference can be made. The self-report data and refusal of participation by some potential participants resulted in response and selection bias. Nonetheless, self-report substance use data are likely valid,^{46, 47} and ACASI methods and use of both medication pictures and trade and generic medication names limit self-report bias.⁴⁸ Available variables are limited by the survey, particularly mental health and dropout risk variables here. PSM motives were only captured at the most recent episode, obscuring within-person variance over time. Finally, exclusion of young adults still in secondary school excludes a smaller, but important, group from analyses. This choice was made based on developmental and stimulant-related evidence^{23, 30–32} but remains a limitation.

Summary and Clinical Implications

PSM motives vary significantly over the 14–25 age span, with lesser endorsement of recreational and greater endorsement of cognitive enhancement motives in young adults, relative to adolescents. This may reflect increasing academic or work demands, given the perception of academic benefits from PSM.³⁸ Longitudinal research extending past young adulthood could clarify if recreational motives increase after 25 years; similarly, longitudinal studies could link baseline motives to likelihood of ongoing PSM at follow-up assessment and frequency and consequences of such misuse. Cognitive enhancement only motives may be more likely in white and multiracial AYAs, though future research is needed to examine PSM prevalence and motives in AYAs by race/ethnicity. Future research using latent class analysis is also needed to further validate the motive categories.

PSM rates were significantly higher in young adults, suggesting that prevention^49–51 in adolescents and screening/intervention may be more fruitful in young adults. Furthermore, PSM motive differences across AYAs and by educational level highlight that prevention, screening, and treatment may need to focus on unique motivational profiles at different ages and educational levels. To illustrate, efforts to combat perceptions of academic benefit from PSM and improve academic skills could lower college student PSM. Finally, any PSM at any age is linked to elevated odds of substance use, SUD, and psychopathology, highlighting the significant risk of PSM. PSM to get high can be employed as a screening tool, as it suggests much greater SUD likelihood. AYAs engaged in PSM to get high are likely to need more intensive interventions, and presence of any PSM suggests a need for further screening for concurrent substance use and psychopathology. Prevention, screening, and intervention programs for PSM are needed, especially in college settings, and research on such programs would have great public health impact.

CLINICAL POINTS.

1. Prescription stimulant misuse (PSM) motives differ over the 14–25 years of age span, with increasing cognitive enhancement (e.g., to study or concentrate) and decreasing recreational motives (e.g., to get high); young adults in college had higher rates of cognitive enhancement than those not in school.

2. Any PSM, regardless of motive, was associated with significantly higher odds of other substance use, any substance use disorder, suicidal ideation, and other psychopathology, with the highest odds in those with recreational motives.

3. Clinicians are likely to encounter different PSM motive profiles at different ages and in different educational attainment groups, with prevention and intervention targets varying by age and education.