Academic orientation and cannabis use-findings from a population-based study of Swedish adolescents in upper secondary school.

Abstract

Aim

To examine the association between academic orientation and frequent cannabis use among Swedish adolescents in upper secondary school and include pupils from introductory programs (IPs), a large group of adolescents previously overlooked in research on adolescent cannabis use.

Methods

We used cross-sectional data from two anonymous school surveys carried out in upper secondary school in 2021. The samples consisted of pupils from all academic orientations, and the analysis included 3151 pupils in higher education preparatory programs (HEPs), 1010 pupils in vocational programs (VPs), and 819 pupils in IPs. The association between the exposure academic orientation and the outcome frequent (21+ times) cannabis was analyzed using multi-level mixed-effects Poisson regression.

Results

Estimates from the first model showed a significant (P < 0.05) 2.45 times higher risk of frequent cannabis use among pupils in IPs compared with in HEPs [incidence rate ratio (IRR) 2.45, 95% confidence interval (CI) 1.28–4.66] and 82% higher in VPs (IRR 1.82, 95% CI 1.09–3.04) compared with in HEPs. However, the associations between academic orientation and frequent (21+ times) cannabis use were attenuated and no longer significant when socioeconomic status, truancy, school dissatisfaction, and early onset of substance use were adjusted for.

Conclusions

There was a higher risk of frequent (21+ times) cannabis use among pupils in IPs, and this differential was explained by higher exposure to risk factors in this group. This result is important from a policy perspective as it provides knowledge of a previously neglected risk group for frequent cannabis use.

Article notes

Untitled section

Collection date 2024 Oct.

Author notes

Introduction

Cannabis use is commonly initiated in adolescence [1]. Previous research suggests that cannabis use increases the risk of mental health problems [2, 3] and social consequences such as ending up outside the labor market [4, 5]. The overall picture from prior research is that it is not primarily single occasions of cannabis use that are associated with adverse consequences, but more frequent use [6, 7].

Adolescent lifetime use of cannabis has been stable in Sweden [8] and in other European countries [9] and the USA [10], whereas frequency of use has increased. Between 1989 and 2016, the mean number of times that 16-year-old cannabis users had ever used cannabis tripled from 4.2 to 13.4 times in Sweden [11]. In the USA, the proportion of daily users aged 17–18 years increased from 2% to 6% between 1991 and 2022 [10]. In parallel, the concentration of the main psychoactive component Δ9-tetrahydrocannabinol (THC) of cannabis has increased [7]. A high concentration of THC is associated with increased risk of psychotic disorder [12], mental health problems [13], and addiction [13, 14]. There is also some evidence that young people may be particularly vulnerable to the effects of cannabis use [15].

Improving young people’s well-being and life chances is an important goal for policymakers, and reducing illicit drug use among adolescents and children is an explicit goal in Sweden [16]. From a policy perspective, it is thus important to know if certain groups of adolescents are at risk of more frequent (and possibly more harmful) cannabis use.

Sweden, like many other countries, has a long tradition of monitoring adolescent substance use, including cannabis use, through school surveys [8–10]. However, all previous prevalence studies among older Swedish adolescents are based on national samples of pupils attending year 2 (age 17–18) in the national programs in upper secondary school. Hence, a relatively large group of adolescents has not been included in the national estimates of cannabis use in Sweden. To our knowledge, no previous study has investigated substance use habits in this group of adolescents.

The proportion of pupils not eligible for these programs has increased in the last decade, and in 2020, about 14% of those finishing compulsory school were not qualified to enter the national programs [17]. This situation is not unique to Sweden; a common problem across countries is adolescents leaving school before they are supposed to. The EU average of early leavers from school was 9.7% in 2021 [18]. Thus, many countries conducting school surveys are likely to exclude adolescents who are not able to continue to upper secondary school. This is problematic if this is a risk group for frequent cannabis use.

Swedish children finish compulsory school at age 16. Thereafter, they can attend a 3-year, free-of-charge, upper secondary school education. The upper secondary school has three tracks of academic orientation. There are two types of national programs that lead to a degree: vocational programs (VPs) and higher education preparatory programs (HEPs). Then there is the so-called introductory program (IP), which does not lead to a degree.

In Sweden, most adolescents lacking the qualifications to apply for the national programs in upper secondary school are enrolled in IP. This means that many adolescents who fail compulsory school are still present within the Swedish school system. Pupils in IPs made up approximately 8.5% of all upper secondary pupils in 2021 [19].

Pupils are enrolled in IPs for different reasons, but the common denominator is that they have not completed compulsory school. Previous studies have found that these pupils are characterized by lower motivation for school, early onset of substance use, more often are boys, and report higher levels of truancy and lower levels of school enjoyment compared with pupils in VPs and HEPs [20]. Non-completion of upper secondary school is also more common among adolescents from families with low socioeconomic status (SES) [21].

Previous studies also suggest that academic orientation is associated with substance use [22–24] and that pupils in VPs are at higher risk of binge drinking [24], smoking [23], and lifetime use of cannabis [22] than pupils in HEPs. An obvious shortcoming is that these studies only include pupils in the national programs.

Problems at school [25, 26] and early onset of substance use [25, 27] have been associated with cannabis use. When it comes to SES and cannabis use, previous findings are conflicting [28, 29]. There are results showing that cannabis and other substance use is more common among adolescents from high SES families [30–32], but others have shown that lower childhood SES is associated with more frequent use of cannabis [22]. There is also some evidence that adolescents with low SES have a higher risk of proceeding from experimental to daily cannabis use [33].

Since adolescents in IPs are struggling with multiple disadvantages, including several cannabis-related risk factors, frequent cannabis use may be more prevalent in this group. One possible mechanism behind such an association could be unequal exposure to risk factors [34] between the different academic orientations.

The overall aim of the present study is to examine the association between academic orientation and frequent cannabis use among Swedish adolescents in upper secondary school, including adolescents in IPs. More specifically, we will examine whether there is an increased risk of frequent cannabis use, defined as life-time use of cannabis more than 20 times, among pupils in IPs compared to pupils in HEPs and VPs, and whether the association is attenuated by a higher prevalence of risk factors, such as low SES, truancy, school dissatisfaction, and early onset of substance use among this group of pupils. As pupils in IPs have never been included in previous national estimates of cannabis use among Swedish adolescents, we will also determine if national estimates change if this group is included. Such information is relevant to the international field of school surveys, since it is a common problem that some adolescents drop out of school and therefore are not included in school surveys.

Methods

Participants

We used data from two cross-sectional studies carried out in 2021. Both surveys were anonymous, and the pupils filled out a web-based questionnaire in the classroom. The first dataset (Dataset 1) stems from the Swedish national school survey in year 2 in upper secondary school (age 17–18 years) and comprises about 4000 individuals (with an equal number of boys and girls) representative of pupils enrolled in national programs in year 2 in upper secondary school in Sweden. A two-step, stratified sampling procedure was used to ensure that all regions in Sweden were represented. In the first step, schools were used as the sampling units, and in the second step, one class was selected randomly at each of the selected schools. The sampling of schools was performed by Statistics Sweden (SCB). The response rate among the sampled classes was 74%. At the individual level (i.e. pupils who were present and willing to participate in the survey), the response rate was 81% [8].

In addition, data from a cross-sectional study targeting the IPs were used (Dataset 2). The sampling was performed by SCB using the same method as in the national survey. To avoid too small survey units, which could risk pupils’ anonymity, only schools with at least 20 pupils enrolled in the IP were included in the sampling frame. The survey was offered to all pupils in the randomly selected IPs. The additional survey was designed to generate results comparable to those of the yearly Swedish national school survey, and data were collected during the same time period (March–April 2021). The response rate at the class level was 56%, and at the individual level, it was 67% [20]. The total sample included 5239 pupils (3289 in HEPs, 1060 in VPs, and 890 in IPs). Only pupils who completed all survey items were included in the study (95% of the total sample): 3151 pupils in HEPs, 1010 pupils in VPs, and 819 pupils in IPs (see Table 1).

Table 1.Descriptive statistics—distribution of key variables across academic orientation

		Higher education preparatory programs (HEP)	Vocational programs (VP)	Introduction programs (IP)
Participants	n	3151	1010	819
Gender
Man	%	45	55	60
Woman	%	55	45	40
Lifetime cannabis use	%	12	13	11
Frequent (21+ times) cannabis use	%	2	3	4
Early-onset substance use
Onset before 14 years old	%	12	20	18
Socioeconomic status
At least one parent studied at university/college	%	78	49	47
Truancy
Once/month or more often	%	9	11	22
School dissatisfaction
Dislike school very much/a lot/neither dislike nor like	%	15	18	24

Measures

Outcome

Frequent use of cannabis was based on answers to the question “On how many occasions have you ever used hashish or marijuana?” with the response options 0, 1, 2–4, 5–10, 11–20, 21–50, more than 50 times. Those who reported use 21+ times were defined as frequent users. The cut-off was set to capture users in the higher end of the scale, taking into account the distribution in the sample (18.6% of cannabis users reported use 21+ times) and the fact that cannabis use is relatively rare in Sweden, compared to other countries [9].

Exposure

Information on academic orientation was received from each school before its pupils participated in the survey. In the analysis of Dataset 1, the school classes were coded as either VP or HEP, whereas Dataset 2 included only IPs.

Covariates

Gender was measured using a question where the pupils could choose between “man,” “woman,” and “other gender identity.” Those who reported “other gender identity” or did not answer the question were excluded from the analysis (1%).

SES was measured through the pupils’ reports of their parents’ highest education level. The variable was created based on responses to the questions “Has your father studied at a university or college?” and “Has your mother studied at a university or college?” Those who reported that at least one parent studied at university or college were coded as having at least one parent with higher education. Those who reported that neither parent had studied at university or college, reported no higher education for one parent alongside not knowing, or reported not knowing for both parents were coded as not having any parent with higher education. This was done to avoid losing respondents living with a single parent [31] or who did not know their parents’ education levels. Pupils not answering either of the two questions were excluded (0.5%).

Truancy was measured using a question on how often the pupil was truant. Those with a response ranging from once a month to several times a week were coded as recurrent truants. The responses coded as “no” were “No” and “Yes, once per semester.”

School dissatisfaction was measured using the question “How do you feel about school?” Those who reported being discontent, very discontent, or neither content nor discontent were coded as being dissatisfied with school.

Early onset of substance use was defined as any substance use before 14 years of age and measured using five questions: “How old were you when (if ever) you did the following things for the first time? (1) Drank at least one glass of alcohol? (2) Got drunk on alcohol? (3) Smoked a cigarette? (4) Used moist snuff? (5) Used marijuana or hashish?” Individuals who responded 13 years or younger to any of these questions were defined as having an early onset of substance use.

Statistical analyses

We used multi-level mixed-effects Poisson regression with robust standard errors, including a random intercept for the school/school class level, to analyze data. The pupils in Dataset 1 were clustered by school classes and those in Dataset 2 were clustered by schools. Hence, the assumption of independence between observations was not met. Therefore, data were analyzed with a random intercept for the school/school class level. As shown by Zou (2004), Poisson regression can be used for dichotomous variables and yields easily interpreted results. When applying Poisson regression on a dichotomous outcome, the estimate presented is interpreted as relative risk. Since this method provides conservative results, we used robust standard errors [35].

In the first step, we performed bivariate analyses between each covariate and frequent cannabis use. To examine which explanatory variables were associated with academic orientation, we used the same procedure. In the analyses, HEP was set as the reference category, so the incidence rate ratio (IRR) estimates express how much higher or lower risk of frequent cannabis use pupils in VPs and IPs have, respectively, compared with pupils in HEPs.

All statistical analyses were performed with Stata version 17.0 (StataCorp, College Station, Texas, USA).

To test if national estimates of frequent cannabis use changed when pupils in IPs were included, we calculated a prevalence that also considered the results from pupils in IPs. In order to do this, we weighted the results based on how pupils were distributed across academic orientation in the population.

Results

Descriptive statistics

Table 1 presents descriptive statistics by academic orientation. No substantive differences in lifetime cannabis use were found between pupils with different academic orientations. However, frequent cannabis use was more common among pupils in IPs (4%), compared with those in VPs (3%) or HEPs (2%). A larger proportion of boys was found in IPs (60%) and VPs (55%) compared with in HEPs (45%). Furthermore, truancy and school dissatisfaction were more common in IPs (22% and 24%) than in VPs (11% and 18%) and HEPs (9% and 15%). The proportion of pupils with at least one parent with higher education was higher among pupils in HEPs (78%) compared with pupils in VP and IP (49% and 47%, respectively). Lastly, early onset of substance use was reported by 20% of pupils in VPs, 18% in IPs, and 12% in HEPs (Table 1).

The total prevalence of frequent cannabis use is presented in Table 2. When the prevalence of frequent cannabis use in IPs was combined with the prevalence in the national programs (HEPs and VPs), the total prevalence of frequent cannabis use was 2.24%, compared with 2.09% if pupils in IPs were not included (Table 2).

Table 2.Prevalence of frequent cannabis use in national programs, introductory programs, and weighted total

	% of pupils in population	% frequent (21+ times) cannabis use
National programs	91.5	2.09
Introductory programs	8.5	3.91
Total (weighted)	100	2.24

Regression analysis

The bivariate analyses show that all covariates except SES were associated with both frequent cannabis use and academic orientation (Table 3 and Supplementary Table S1).

Table 3.Bivariate analysis of covariates and frequent (21+ times) cannabis use (multi-level mixed-effects Poisson regression with robust standard errors, including a random intercept for school class level)

n = 4980	IRR	95% CI
Gender	1.75^*	1.18	2.60
Academic orientation
Higher education preparatory (HEP)	ref	ref	ref
Vocational (VP)	1.89^*	1.15	3.10
Introduction (IP)	2.64^*	1.42	4.91
Socioeconomic status (ref = at least one parent studied at university/college)	0.90	0.60	1.33
Early-onset substance use	7.55^*	5.24	10.88
Truancy	2.56^*	1.66	3.96
School dissatisfaction (ref = satisfied)	1.79^*	1.17	2.75

The results of the multi-level mixed-effects Poisson regression are shown in Table 4. The first model (adjusted only for gender) revealed an association between academic orientation and frequent (21+ times) cannabis use. The risk of frequent cannabis use was 2.45 times higher among pupils in IPs compared with those in HEPs [IRR 2.45, 95% confidence interval (CI) 1.28–4.66] and 82% higher in VPs (IRR 1.82, 95% CI 1.09–3.04) compared with in HEPs.

Table 4.The association of academic orientation with frequent (21+ times) cannabis use (multi-level mixed-effects Poisson regression with robust standard errors, including a random intercept for school class level)

	Null model	Model 1 (crude)			Model 2			Model 3			Model 4			Model 5
n = 4980		IRR	95% CI		IRR	95% CI		IRR	95% CI		IRR	95% CI		IRR	95% CI
Gender (ref = woman)		1.65^*	1.09	2.50	1.65^*	1.09	2.49	1.70^*	1.09	2.63	1.66^*	1.11	2.48	1.69^*	1.11	2.59
Higher education preparatory (HEP)		ref			ref			ref			ref			ref
Vocational (VP)		1.82^*	1.09	3.04	1.97^*	1.16	3.34	1.75^*	1.05	2.91	1.35	0.82	2.22	1.48	0.89	2.47
Introduction (IP)		2.45^*	1.28	4.66	2.66^*	1.42	4.98	2.01^*	1.01	3.98	1.80^*	1.02	3.19	1.78	0.98	3.24
Socioeconomic status (ref = at least one parent studied at university/college)					0.75	0.50	1.13							0.72	0.48	1.08
Truancy (ref = no)								2.17^*	1.38	3.42				1.59^*	1.02	2.49
School dissatisfaction (ref = satisfied)								1.55^*	1.01	2.37				1.37	0.90	2.09
Substance onset < 14 years old (ref = no)											7.16^*	5.01	10.24	6.55^*	4.53	9.48
var(_cons)	0.89	0.84	0.48	1.48	0.83	0.46	1.46	0.76	0.43	1.35	0.62	0.32	1.20	0.56	0.28	1.14
Model fitness
df	2	5			6			7			6			9
AIC	1108.1	1096.107			1096.245			1082.193			997.5929			994.3198
BIC	1121.126	1128.673			1135.324			1127.786			1036.672			1052.938

Adjusting for SES did not attenuate the association between academic orientation and frequent cannabis use (Model 2). Adjusting for school-related factors (truancy and school dissatisfaction) attenuated the risk of frequent cannabis use associated with both VPs and IPs compared with HEPs (Model 3). Both truancy and school dissatisfaction were significantly associated with frequent cannabis use. Adjusted for gender, academic orientation, and school dissatisfaction, pupils who were truant once a month or more often had a 2.17 higher risk of frequent cannabis use (IRR 2.17, 95% CI 1.38–3.42). Pupils who were dissatisfied with school had a 55% (IRR 1.55, 95% CI 1.01–2.37) higher risk of frequent cannabis use when we adjusted for gender, academic orientation, and truancy. Adjusting for early onset of substance use also attenuated the risk of frequent cannabis use in both VPs and IPs, and the association was no longer significant for pupils in VPs. When adjusting for early onset of substance use, the risk of frequent cannabis use was 80% (IRR 1.8, 95% CI 1.02–3.19) higher among pupils in Is compared with those in HEPs (Model 4). The analysis also showed that early onset of substance use was significantly associated with a 7.16 (IRR 7.16, 95% CI 5.01–10.24) times higher risk of frequent cannabis use, adjusted for gender and academic orientation.

In the fully adjusted model, the associations between academic orientation and frequent cannabis use were positive, but no longer significant (Model 5), whereas both truancy and early onset of substance use remained significantly associated with frequent cannabis use.

Discussion

The overall aim of the present study was to examine if academic orientation in upper secondary school was associated with frequent cannabis use among adolescents in Sweden. Our analysis included pupils from IPs in addition to pupils in VPs and HEPs. This is a group of adolescents with experiences of severe problems in school that has not been included in previous studies of adolescent cannabis use in Sweden.

We found an increased risk of frequent cannabis use among pupils in both VPs and IPs compared with pupils in HEPs, with the highest risk among those in IPs. Our results are in line with previous findings of an increased risk of negative health behaviors such as smoking, binge drinking, and use of illicit substances among pupils in non-theoretical programs [22–24] and lend support to the idea that academic orientation may serve as an important predictor of negative health behaviors, including frequent cannabis use.

However, when we adjusted for truancy, school dissatisfaction, and early onset of substance use, the association between academic orientation and frequent cannabis use was attenuated and no longer significant. Previous research shows that truancy is associated with both frequent and lifetime cannabis use [22, 25]. In the present study, we found recurrent truancy to increase the risk of frequent cannabis use, net of academic orientation. Early onset of substance use was also strongly associated with an increased risk of frequent cannabis use, which is in line with previous findings [25, 27]. Altogether, this suggests that the increased risk of frequent cannabis use found in VPs and IPs is by and large due to unequal exposure to risk factors for cannabis use.

Our findings that SES neither attenuated the association between academic orientation and frequent cannabis nor was significantly associated with frequent cannabis use were somewhat surprising and differ from previous Swedish findings [22]. One potential explanation may be that previous studies did not include adolescents from IPs.

We found that including a “hard-to-reach” group like pupils in IPs in the sample did not significantly change the national estimates of frequent cannabis use among adolescents in Sweden. The combined total differed by only 0.2 percentage points compared with that for the national programs only. The findings suggest that even though a relatively large group of adolescents dropping out from compulsory school is not included in the national school surveys, this does not dramatically affect the national estimates of cannabis use among adolescents in this age group. Thus, if underreporting of cannabis use is present in school surveys, it is not a result of excluding a group of adolescents that are hard to reach.

Although the inclusion of pupils in IPs did not change the national estimates, our study shows that this group of pupils has an increased risk of frequent cannabis use. In addition to using more cannabis and having a greater number of risk factors for cannabis use, they also run a greater risk of not graduating from upper secondary school [36]. School failure is related to worse future work opportunities and life chances [37, 38], as well as drug abuse [39]. From a social inequality perspective, it is important to reach this group of pupils, who struggle with multiple disadvantages, with prevention efforts. The school setting provides an important and well-established arena for preventive work. Although pupils in IPs are not part of the national programs, they are enrolled in school and can thus be effectively reached with preventive measures.

There are some limitations of the present study that need to be mentioned. Pupils from language introduction were not included in our study. Language introduction is an orientation within IPs which is aimed at newly arrived immigrants who need to learn Swedish before they can be admitted to a national program. Language introduction encompasses 3% of pupils in upper secondary school, and they report lower levels of cannabis use [20]. Hence, our findings cannot be generalized to all students in IP. Moreover, the non-response rate was higher among IP pupils. A study of non-responders found that the pupils who were not present at the time of the survey did report a higher level of substance use, but this did not affect the overall prevalence estimates [40]. However, it cannot be ruled out that we, due to higher non-response rate among IP pupils, underestimate the prevalence of cannabis use in this group.

Furthermore, our analysis relies mainly on self-reported data, with its known risk of underreporting. Among adolescents, overreporting also occurs. To minimize this type of bias, the survey was conducted with a teacher present and with a strong emphasis on each pupil’s anonymity [8]. Information on academic orientation was collected through registers and hence not sensitive to bias associated with self-reporting.

The cross-sectional design makes it difficult to draw conclusions regarding the causal direction of the association between academic orientation and frequent cannabis use. Further research with a longitudinal design is needed to settle this question.

The major strengths of the present study are the large sample size and the design which makes the sample representative of all regions in Sweden. Furthermore, we included a group of adolescents never previously examined, despite having indications of being a risk group for cannabis use. The similar data collection for all three programs in the present study is also a major strength, as it provides a more comprehensive picture of the association between academic orientation and frequent cannabis use.

Conclusions

Frequent cannabis use was more common among pupils in IPs compared with those in HEPs, and this association was explained by higher exposure to risk factors in the IP group. This result is important from a policy perspective as it provides knowledge of a previously disregarded risk group for frequent cannabis use. The inclusion of this hard-to-reach group did not change the national estimate of cannabis use in the age group.

Supplementary Material

Contributor Information

Isabella Gripe, Department of Clinical Neurosciences, Karolinska Institutet, Stockholm, Sweden; The Swedish Council for Information on Alcohol and Other Drugs (CAN), Stockholm, Sweden.

Mats Ramstedt, Department of Clinical Neurosciences, Karolinska Institutet, Stockholm, Sweden; The Swedish Council for Information on Alcohol and Other Drugs (CAN), Stockholm, Sweden.

Patrik Karlsson, Department of Social Work, Stockholm University, Stockholm, Sweden.

Siri Thor, Department of Clinical Neurosciences, Karolinska Institutet, Stockholm, Sweden; The Swedish Council for Information on Alcohol and Other Drugs (CAN), Stockholm, Sweden.

Supplementary data

Supplementary data are available at EURPUB online.

Conflict of interest: The authors declare that they have no competing interests.

Funding

This work was supported by Systembolaget’s Research Council (Grant No. FO2020-0090). The funders had no role in (i) the study design; (ii) the collection, analysis, and interpretation of data; (iii) the writing of the report; or (iv) the decision to submit the manuscript for publication.

Data availability

The data underlying this article cannot be shared publicly as it was created and used under license from the Swedish Council for Information on Alcohol and Other Drugs (CAN). Data and code can be made available upon request for legitimate purposes.

Key points

The present study provides knowledge of a previously disregarded risk group for frequent cannabis use.
Both pupils in VP and IP have higher risk for frequent cannabis use than pupils in HEP.
There is a need to better prevent substance use in these vulnerable groups of adolescents.

References

Untitled section

References

1. Degenhardt L, Stockings E, Patton G et al. The increasing global health priority of substance use in young people. Lancet Psychiatry 2016;3:251–64.
2. Gobbi G, Atkin T, Zytynski T et al. Association of cannabis use in adolescence and risk of depression, anxiety, and suicidality in young adulthood: a systematic review and meta-analysis. JAMA Psychiatry 2019;76:426–34.
3. Kiburi SK, Molebatsi K, Ntlantsana V et al. Cannabis use in adolescence and risk of psychosis: are there factors that moderate this relationship? A systematic review and meta-analysis. Subst Abus 2021;42:527–42.
4. Danielsson AK, Agardh E, Hemmingsson T et al. Cannabis use in adolescence and risk of future disability pension: a 39-year longitudinal cohort study. Drug Alcohol Depend 2014;143:239–43.
5. Danielsson AK, Falkstedt D, Hemmingsson T et al. Cannabis use among Swedish men in adolescence and the risk of adverse life course outcomes: results from a 20 year-follow-up study. Addiction 2015;110:1794–802.
6. Hall W. What has research over the past two decades revealed about the adverse health effects of recreational cannabis use? Addiction 2015;110:19–35.
7. National Academies of Sciences, Engineering, and Medicine. The Health Effects of Cannabis and Cannabinoids: The Current State of Evidence and Recommendations for Research. Washington, DC: The National Academies Press, 2017. Report No. 978-0-309-45304-2.
8. Zetterqvist M, Gripe I, Svensson J et al. CAN:s nationella skolundersökning 2022 [Electronic Resource] [CAN’s National School Survey 2022: Young People’s Experience of Alcohol, Narcotics, Tobacco, Doping and Gambling]. Stockholm: CAN, 2022.
9. ESPAD. ESPAD Report 2019: Results From the European School Survey Project on Alcohol and Other Drugs. Luxembourg: ESPAD, 2020.
10. Johnston LD, Miech RA, Patrick ME et al. Monitoring the Future National Survey Results on Drug Use 1975-2022: Overview, Key Findings on Adolescent Drug Use. Ann Arbor, MI: Institute for Social Research, 2023.
11. Gripe I, Danielsson AK, Ramstedt M. Are changes in drinking related to changes in cannabis use among Swedish adolescents? A time-series analysis for the period 1989-2016. Addiction 2018;40:126–34.
12. Di Forti M, Quattrone D, Freeman TP, EU-GEI WP2 Group et al. The contribution of cannabis use to variation in the incidence of psychotic disorder across Europe (EU-GEI): a multicentre case-control study. Lancet Psychiatry 2019;6:427–36.
13. Hines LA, Freeman TP, Gage SH et al. Association of high-potency cannabis use with mental health and substance use in adolescence. JAMA Psychiatry 2020;77:1044–51.
14. Freeman TP, van der Pol P, Kuijpers W et al. Changes in cannabis potency and first-time admissions to drug treatment: a 16-year study in the Netherlands. Psychol Med 2018;48:2346–52.
15. Kaur N, Bastien G, Gagnon L et al. Variations of cannabis-related adverse mental health and addiction outcomes across adolescence and adulthood: a scoping review. Front Psychiatry 2022;13:973988.
16. SoU25. Socialutskottets betänkande 2021/22: soU25 En samlad strategi för alkohol-, narkotika-, dopnings- och tobakspolitiken samt spel om pengar 2022–2025. [An integrated strategy for alcohol, drug, doping and tobacco policy and gambling 2022–2025.] Stockholm: Sveriges Riksdag 2021. Report No. 2021/22:213.
17. Skolverket. Slutbetyg i Grundskolan Våren 2020 [Final Grades in Compulsory School Spring 2020]. Stockholm: Skolverket/The Swedish National Agency for Education, 2020.
18. Eurostat. Early leavers from education and training. https://ec.europa.eu/eurostat/statisticsexplained/ (25 May 2023, date last accessed).
19. Skolverket. Elever i gymnasieskolan läsåret 2020/21 [Pupils in upper secondary school, school year 2020/21] https://www.skolverket.se/publikationsserier/beskrivande-statistik/2021/elever-i-gymnasieskolan-lasaret-2020-21 (24 June 2024, date last accessed).
20. Gripe I, Thor S. Nya Perspektiv [Elektronisk Resurs] [New Perspectives]. Stockholm: The Swedish Council for Information on Alcohol and Other Drugs, 2022. Report No. 9789172783409.
21. Vaezghasemi M, Mosquera PA, Gustafsson PE et al. Decomposition of income-related inequality in upper secondary school completion in Sweden by mental health, family conditions and contextual characteristics. SSM Popul Health 2020;11:100566.
22. Gripe I, Danielsson AK, Karlsson P et al. Are the well-off youth in Sweden more likely to use cannabis? Drug Alcohol Rev 2021;40:126–34.
23. Hagquist CE. Health inequalities among adolescents: the impact of academic orientation and parents’ education. Eur J Public Health 2007;17:21–6.
24. Thor S, Karlsson P, Landberg J. Social inequalities in harmful drinking and alcohol-related problems among Swedish adolescents. Alcohol Alcohol 2019;54:532–9.
25. Karlsson P, Ekendahl M, Gripe I et al. Individual and school-class correlates of youth cannabis use in Sweden: a multilevel study. Nordisk Alkohol Nark 2018;35:131–46.
26. Townsend L, Flisher AJ, King G. A systematic review of the relationship between high school dropout and substance use. Clin Child Fam Psychol Rev 2007;10:295–317.
27. Millar SR, Mongan D, Smyth BP et al. Relationships between age at first substance use and persistence of cannabis use and cannabis use disorder. BMC Public Health 2021;21:997.
28. Daniel JZ, Hickman M, Macleod J et al. Is socioeconomic status in early life associated with drug use? A systematic review of the evidence. Drug Alcohol Rev 2009;28:142–53.
29. Hanson MD, Chen E. Socioeconomic status and health behaviors in adolescence: a review of the literature. J Behav Med 2007;30:263–85.
30. Charitonidi E, Studer J, Gaume J et al. Socioeconomic status and substance use among Swiss young men: a population-based cross-sectional study. BMC Public Health 2016;16:333.
31. Humensky JL. Are adolescents with high socioeconomic status more likely to engage in alcohol and illicit drug use in early adulthood? Subst Abuse Treat Prev Policy 2010;5:19.
32. Patrick ME, Wightman P, Schoeni RF et al. Socioeconomic status and substance use among young adults: a comparison across constructs and drugs. J Stud Alcohol Drugs 2012;73:772–82.
33. Legleye S, Khlat M, Mayet A et al. From cannabis initiation to daily use: educational inequalities in consumption behaviours over three generations in France. Addiction 2016;111:1856–66.
34. Diderichsen F, Hallqvist J, Whitehead M. Differential vulnerability and susceptibility: how to make use of recent development in our understanding of mediation and interaction to tackle health inequalities. Int J Epidemiol 2019;48:268–74.
35. Zou G. A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004;159:702–6.
36. Skolverket. Introduktionsprogram, eleverna och deras sysselsättning efteråt [Introductory programs, the students and their employment afterwards]. Report No. 2019:484. Stockholm: Skolverket, 2019.
37. Freudenberg N, Ruglis J. Reframing school dropout as a public health issue. Prev Chronic Dis 2007;4:A107.
38. Howieson C, Iannelli C. The effects of low attainment on young people’s outcomes at age 22-23 in Scotland. British Educational Res J 2008;34:269–90.
39. Gauffin K, Vinnerljung B, Fridell M et al. Childhood socio-economic status, school failure and drug abuse: a Swedish national cohort study. Addiction 2013;108:1441–9.
40. Andersson B, Hibell B. Skolelevers Drogvanor: Tabeller Med Sammanfattande Kommentarer. 1993 [Drughabits among Pupils: Tabels and Summary Comments]. Stockholm: Centralförbundet för alkohol-och narkotikaupplysning (CAN; ), 1993.

Associated Data

Supplementary Materials

Data Availability Statement

Key points

The present study provides knowledge of a previously disregarded risk group for frequent cannabis use.
Both pupils in VP and IP have higher risk for frequent cannabis use than pupils in HEP.
There is a need to better prevent substance use in these vulnerable groups of adolescents.

Studies

Conditions

Cannabinoids and active compounds

Related research

Abstract

Aim

Methods

Results

Conclusions

Article notes

Untitled section

Author notes

Introduction

Methods

Participants

Measures

Outcome

Exposure

Covariates

Statistical analyses

Results

Descriptive statistics

Regression analysis

Discussion

Conclusions

Supplementary Material

Contributor Information

Supplementary data

Funding

Data availability

Key points

References

Untitled section

References

Associated Data

Supplementary Materials

Data Availability Statement

Key points

Additional materials