Profiles of drug users in Switzerland and effects of early-onset intensive use of alcohol, tobacco and cannabis on other illicit drug use.

QUESTIONS UNDER STUDY / PRINCIPLES: The main aim of this study was to investigate profiles of drug users, with a particular focus on illicit drugs other than cannabis, and to explore the effect of early-onset intensive use (drunkenness, daily smoking, high on cannabis) on profiles of illicit drug use. METHODS: Baseline data from a representative sample of 5,831 young Swiss men in the ongoing Cohort Study on Substance Use Risk Factors were used. Substance use (alcohol, tobacco, cannabis and 15 types of other illicit drug) and age of onset of intensive use were assessed. The Item Response Theory (IRT) and prevalence rates at different ages of onset were used to reveal different profiles of illicit drug use. RESULTS: In addition to cannabis, there were two profiles of other illicit drug use: (a) "softer" drug users (uppers, hallucinogens and inhaled drugs), among which ecstasy had the highest discriminatory potential (IRT slope = 4.68, standard error (SE) = 0.48; p <0.001); and (b) "harder" drug users (heroin, ketamine, gamma-hydroxybutyrate/gamma-hydroxylactone, research chemicals, crystal meth and spice), among which ketamine had the highest discriminatory potential (slope = 4.05; SE = 0.63; p <0.001). Onset of intensive use at the age of 12 years or younger also discriminated between these two profiles. CONCLUSION: Both the IRT model and the effect of onset of intensive use enabled two groups of illicit drugs to be identified. In particular, very early onset (at 12 years or younger) intensive use of any substance was a marker for later use of the second group of drugs.

L'adolescent consommateur de substances face au réseau de soins [Substance abuse adolescents and the health care network]

Substance user adolescents were asked to report on each contact they had had with any type of care providers since they had begun to use alcohol or illegal drugs regularly. Primary care doctors and social workers represent the main access to the care network. In one out of two contacts substance use was not discussed.

Prevalence and factors associated with use of smokeless tobacco in young Swiss men.

BACKGROUND: Smokeless tobacco is of increasing interest to public health researchers and policy makers. This study aims to measure prevalence of smokeless tobacco use (nasal dry snuff, snus and chewing tobacco) among young Swiss men, and to describe its correlates. METHODS: We invited 13 245 young men to participate in this survey on socio-economic and substance use data. Response rate was 45.2%. We included 5720 participants. Descriptive statistics and multivariable-adjusted logistic regression were performed. RESULTS: Mean age of participants was 19.5 years. Self-reported use once a month or more often was 8% for nasal dry snuff, 3% for snus and negligible for chewing tobacco. In multivariable-adjusted logistic regression, the odds for nasal dry snuff use increased in non daily smokers [odds ratio (OR) 2.41, 95% confidence interval (CI) 1.90-3.05], compared with non smokers, participants reporting risky weekly drinking volume (OR 3.93, 95% CI 1.86-8.32), compared with abstinents, and binge drinking once a month or more often (OR 7.41, 95% CI 4.11-13.38), compared with never binge drinking. Nasal dry snuff use was positively associated with higher BMI, average or above family income and German language, compared with French, and negatively associated with academic higher education, compared with non higher education, and occasional cannabis use, compared with no cannabis use. Correlates of snus were similar to those of nasal dry snuff. CONCLUSION: One in 12 young Swiss men use nasal dry snuff and 3% use snus. Consumption of smokeless tobacco is associated with a cluster of other risky behaviours, especially binge drinking.

Drug use and suicide attempts: the role of personality factors.

This study on suicide attempts is part of a large research project on dependent behavior in adolescents and young adults. 228 subjects aged 14-25 (107 "drug abusers," 121 controls) from the French speaking part of Switzerland were evaluated on the basis of a semi-structured interview (Mini International Neuropsychiatric Interview), enabling a DSM-IV diagnosis, and self-reports (SSS by Zuckermann, MMPI-2, IDI by Hirschfeld). 31.5% of "drug abuser" males and 41.2% of "drug abuser" females committed one or more suicide attempts. The results of a logistic regression show that the significant factor explaining suicide attempts in drug users is emotional reliance for males and experience-seeking for females.

Cannabis use trajectories among Swiss adolescents

Introduction: Swiss data indicate that one fifth of current 16-20 yearold cannabis users do not use tobacco and seem to do better than those smoking both substances. The aim of this research is to assess the substance use trajectories of cannabis users who do not use tobacco and those who use both substances from age 17 to age 23. Methods: Using data from the TREE longitudinal data base, 328 out of 1796 youth 18.3%; 45% females) who smoked cannabis only (Group CAN; N = 46; 36% females) or concurrently with tobacco (Group CANTAB; N = 284; 46% females) at T1 (2001; age 17) were followed at T4 (2004; age 20) and T7 (2007; age 23). Two additional outcome groups were included at T4 and T7: those using only tobacco (Group TOB) and those not using any of these substances (Group NONE). Data were analyzed separately by gender. Results: Females in group CAN at T1 were as likely to be in group TOB (35%) or NONE (35%) at T4 and the percentages increased to 41% and 47%, respectively, at T7. Males in group CAN at T1 were more likely to be in group TOB at T4 (33%) and T7 (61%) than in group NONE (23% and 15%, respectively). Females in group CANTOB at T1 were mainly in group TOB at T4 (52%) and T7 (61%), while males in CANTOB at T1 remained mainly in the same group at T4 (75%) and T7 (61%). Only 10% of females and 5% of males in group CANTOB at T1 were in group NONE at T4 and 15% and 12%, respectively, at T7. Conclusions: Adolescents using only cannabis are globally less likely to continue using cannabis in young adulthood than those using both substances, although a fair percentage (specially males) switch to tobacco use. This result confirms previous research indicating that nicotine dependence and persistent cigarette smoking may be the main public health consequences of cannabis use. A gender difference arises among those using tobacco and cannabis at age 17: while females become mainly tobacco smokers, the majority of males continue to use both substances. Although these results could be explained by a substitution effect, teenagers using both substances seem to have gone beyond the experimentation phase and should be a motive for concern.

Perception of tobacco, cannabis, and alcohol use of others is associated with one's own use.

BACKGROUND: Interventions have been developed to reduce overestimations of substance use among others, especially for alcohol and among students. Nevertheless, there is a lack of knowledge on misperceptions of use for substances other than alcohol. We studied the prevalence of misperceptions of use for tobacco, cannabis, and alcohol and whether the perception of tobacco, cannabis, and alcohol use by others is associated with one's own use. METHODS: Participants (n=5216) in a cohort study from a census of 20-year-old men (N=11,819) estimated the prevalence of tobacco and cannabis use among peers of the same age and sex and the percentage of their peers drinking more alcohol than they did. Using the census data, we determined whether participants overestimated, accurately estimated, or underestimated substance use by others. Regression models were used to compare substance use by those who overestimated or underestimated peer substance with those who accurately estimated peer use. Other variables included in the analyses were the presence of close friends with alcohol or other drug problems and family history of substance use. RESULTS: Tobacco use by others was overestimated by 46.1% and accurately estimated by 37.3% of participants. Cannabis use by others was overestimated by 21.8% and accurately estimated by 31.6% of participants. Alcohol use by others was overestimated by more than half (53.4%) of participants and accurately estimated by 31.0%. In multivariable models, compared with participants who accurately estimated tobacco use by others, those who overestimated it reported smoking more cigarettes per week (incidence rate ratio [IRR] [95% CI], 1.17 [range, 1.05, 1.32]). There was no difference in the number of cigarettes smoked per week between those underestimating and those accurately estimating tobacco use by others (IRR [95% CI], 0.99 [range, 0.84, 1.17]). Compared with participants accurately estimating cannabis use by others, those who overestimated it reported more days of cannabis use per month (IRR [95% CI], 1.43 [range, 1.21, 1.70]), whereas those who underestimated it reported fewer days of cannabis use per month (IRR [95% CI], 0.62 [range, 0.23, 0.75]). Compared with participants accurately estimating alcohol use by others, those who overestimated it reported consuming more drinks per week (IRR [95% CI], 1.57 [range, 1.43, 1.72]), whereas those who underestimated it reported consuming fewer drinks per week (IRR [95% CI], 0.41 [range, 0.34, 0.50]). CONCLUSIONS: Perceptions of substance use by others are associated with one's own use. In particular, overestimating use by others is frequent among young men and is associated with one's own greater consumption. This association is independent of the substance use environment, indicating that, even in the case of proximity to a heavy-usage group, perception of use by others may influence one's own use. If preventive interventions are to be based on normative feedback, and their aim is to reduce overestimations of use by others, then the prevalence of overestimation indicates that they may be of benefit to roughly half the population; or, in the case of cannabis, to as few as 20%. Such interventions should take into account differing strengths of association across substances.

Is cannabis use a significant exposition to nicotine?

Purpose: Young cannabis users are at increased risk for cigarette initiation and later progression to nicotine dependence. The present study assesses to which extent cannabis users are exposed to nicotine through mulling, a widespread process consisting of mixing tobacco to cannabis for its consumption. Methods: Data are issued from an ongoing observational study taking place in Switzerland. A total of 267 eligible participants (mean age 19 years, 46.4% males) completed an anonymous self-administered questionnaire on their tobacco and cannabis intake in the previous 5 days. They also provided a urine sample that was blindly analyzed for cotinine (a key metabolite of nicotine) using liquid-chromatography coupled mass-spectrometry. After the exclusion of cannabis users not having smoked at least one joint/blunt in which tobacco had been mixed (n _ 2), and participants reporting other sources of nicotine exposition than cigarettes or mulling (n _37), four groups were created: cannabis and cigarette abstainers (ABS, n_ 69), cannabis only smokers (CAS; n _ 33), cigarette only smokers (CIS; n _ 62); and cannabis and cigarette smokers (CCS, n _ 64). Cotinine measures of CAS were compared to those of ABS, CIS and CCS. All comparisons were performed using ANCOVA, controlling for age, gender, ethnicity, BMI and environmental exposure to cigarette smoke in the past month (at home, in school/at work, in social settings). The number of mixed joints/blunts smoked in the previous 5 days was additionally taken into account when comparing CAS to CCS. Cotinine values (ng/ml) are reported as means with 95% confidence interval (95% CI). Results: In the previous 5 days, CAS had smoked on average 10 mixed joints/blunts, CIS 30 cigarettes, and CCS 8 mixed joints/ blunts and 41 cigarettes. Cotinine levels of participants considerably differed between groups. The lowest measure was found among ABS (3.2 [0.5-5.9]), followed in growing order by CAS (294.6 [157.1-432.0]), CIS (362.8 [258.4-467.3]), and CCS (649.9 [500.7-799.2]). In the multivariate analysis, cotinine levels of CAS were significantly higher than those of ABS (p _.001), lower than those of CCS (p _ .003), but did not differ from levels of CIS (p _ .384). Conclusions: Our study reveals cannabis users to be significantly exposed to nicotine through mulling, even after controlling for several possible confounders such as environmental exposure to cigarette smoke. Utmost, mixing tobacco to Poster cannabis can result in a substantial nicotine exposition as cotinine levels from cannabis only smokers were as high as those of moderate cigarette smokers. Our findings also suggest that mulling is adding up to the already important nicotine exposition of cigarettes smokers. Because of the addictiveness of nicotine, mulling should be part of a comprehensive assessment of substance use among adolescents and young adults, especially when supporting their cannabis and cigarette quitting attempts. Sources of Support: This study was funded by the Public Health Service of the Canton de Vaud. Dr. BÊlanger's contribution was possible through grants from the Royal College of Physicians and Surgeons of Canada, the CHUQ/CMDP Foundation and the Laval University McLaughlin program, QuÊbec, Canada.

DSM-IV and DSM-5 alcohol use disorder among young Swiss men.

BACKGROUND AND AIMS: Previous studies suggest that the new DSM-5 criteria for alcohol use disorder (AUD) will increase the apparent prevalence of AUD. This study estimates the 12-month prevalence of AUD using both DSM-IV and DSM-5 criteria and compares the characteristics of men in a high risk sample who meet both, only one and neither sets of diagnostic criteria. DESIGN, SETTING AND PARTICIPANTS: 5943 Swiss men aged 18-25 years who participated in the Cohort Study on Substance Use Risk Factors (C-SURF), a population-based cohort study recruited from three of the six military recruitment centres in Switzerland (response rate = 79.2%). MEASUREMENTS: DSM-IV and DSM-5 criteria, alcohol use patterns, and other substance use were assessed. FINDINGS: Approximately 31.7% (30.5-32.8) of individuals met DSM-5 AUD criteria [21.2% mild (20.1-22.2); 10.5% moderate/severe (9.7-11.3)], which was less than the total rate when DSM-IV criteria for alcohol abuse (AA) and alcohol dependence (AD) were combined [36.8% overall (35.5-37.9); 26.6% AA (25.4-27.7); 10.2% AD (9.4-10.9)]. Of 2479 respondents meeting criteria for either diagnoses, 1585 (63.9%) met criteria for both. For those meeting DSM-IV criteria only (n = 598, 24.1%), hazardous use was most prevalent, whereas the criteria larger/longer use than intended and tolerance to alcohol were most prevalent for respondents meeting DSM-5 criteria only (n = 296, 11.9%). Two in five DSM-IV alcohol abuse cases and one-third of DSM-5 mild AUD individuals fulfilled the diagnostic criteria due to the hazardous use criterion. The addition of the craving and excluding of legal criterion, respectively, did not affect estimated AUD prevalence. CONCLUSIONS: In a high-risk sample of young Swiss males, prevalence of alcohol use disorder as diagnosed by DSM-5 was slightly lower than prevalence of DSM-IV diagnosis of dependence plus abuse; 63.9% of those who met either criterion met criteria for both.

Predictors of onset of cannabis and other drug use in male young adults: results from a longitudinal study.

BACKGROUND: The use of cannabis and other illegal drugs is particularly prevalent in male young adults and is associated with severe health problems. This longitudinal study explored variables associated with the onset of cannabis use and the onset of illegal drug use other than cannabis separately in male young adults, including demographics, religion and religiosity, health, social context, substance use, and personality. Furthermore, we explored how far the gateway hypothesis and the common liability to addiction model are in line with the resulting prediction models. METHODS: The data were gathered within the Cohort Study on Substance Use Risk Factors (C-SURF). Young men aged around 20 years provided demographic, social, health, substance use, and personality-related data at baseline. Onset of cannabis and other drug use were assessed at 15-months follow-up. Samples of 2,774 and 4,254 individuals who indicated at baseline that they have not used cannabis and other drugs, respectively, in their life and who provided follow-up data were used for the prediction models. Hierarchical logistic stepwise regressions were conducted, in order to identify predictors of the late onset of cannabis and other drug use separately. RESULTS: Not providing for oneself, having siblings, depressiveness, parental divorce, lower parental knowledge of peers and the whereabouts, peer pressure, very low nicotine dependence, and sensation seeking were positively associated with the onset of cannabis use. Practising religion was negatively associated with the onset of cannabis use. Onset of drug use other than cannabis showed a positive association with depressiveness, antisocial personality disorder, lower parental knowledge of peers and the whereabouts, psychiatric problems of peers, problematic cannabis use, and sensation seeking. CONCLUSIONS: Consideration of the predictor variables identified within this study may help to identify young male adults for whom preventive measures for cannabis or other drug use are most appropriate. The results provide evidence for both the gateway hypothesis and the common liability to addiction model and point to further variables like depressiveness or practising of religion that might influence the onset of drug use.

Patterns of cannabis use and prospective associations with health issues among young males.

BACKGROUND AND AIMS: To test prospective associations between cannabis disorder symptoms/frequency of cannabis use and health issues and to investigate stability versus transience in cannabis use trajectories. DESIGN: Two waves of data collection from the longitudinal Cohort Study on Substance Use Risk Factors (C-SURF). SETTING: A representative sample of young Swiss men in their early 20s from the general population. PARTICIPANTS: A total of 5084 young men (mean age 19.98 ± 1.19 years at time 1). MEASUREMENTS: Cannabis use (life-time use, frequency of use, cannabis disorder symptoms) and self-reported measures of health issues (depression, mental/physical health, health consequences) were assessed. Significant changes in cannabis use were tested using t-test/Wilcoxon's rank test for paired data. Cross-lagged panel models provided evidence regarding longitudinal associations between cannabis use and health issues. FINDINGS: Most of the participants (84.5%) remained in the same use category and cannabis use kept to similar levels at times 1 and 2 (P = 0.114 and P = 0.755; average of 15 ± 2.8 months between times 1 and 2). Cross-lagged panel models showed that cannabis disorder symptoms predicted later health issues (e.g. depression, β = 0.087, P < 0.001; health consequences, β = 0.045, P < 0.05). The reverse paths from health issues to cannabis disorder symptoms and the cross-lagged panel model between frequency of cannabis use and health issues were non-significant. CONCLUSIONS: Patterns of cannabis use showed substantial continuity among young Swiss men in their early 20s. The number of symptoms of cannabis use disorder, rather than the frequency of cannabis use, is a clinically important measure of cannabis use among young Swiss men.

Association between nonmedical prescription drug use and health status among young swiss men.

PURPOSE: To examine the relationship between the nonmedical prescription drug use (NMPDU) of six drug classes and health. METHODS: Data on young adults males (mean age, 19.96 years) from the baseline and follow-up of the Cohort Study on Substance Use Risk Factors (C-SURF) were used (n = 4,958). Two sets of logistic regression models were fitted to examine the associations between NMPDU of opioid analgesics, sedatives or sleeping pills, anxiolytics, antidepressants, beta blockers and stimulants, and health status (assessed using the Medical Outcomes Study 12-Item Short Form Survey Instrument [SF-12 v2]). We first computed odds ratios between NMPDU at baseline and poor mental and physical health at follow-up, adjusting for poor mental or physical health at baseline. We then computed odds ratios between poor mental and physical health at baseline and NMPDU at follow-up, adjusting for NMPDU at baseline. RESULTS: Three key findings regarding mental health were (1) there was a reciprocal risk between poor mental health and sedatives and anxiolytics; (2) poor mental health increased NMPDU of opioid analgesics and antidepressants but not vice versa; and (3) there were no associations with stimulants. Three key findings regarding physical health were (1) poor physical health increased the risk of NMPDU of anxiolytics; (2) the only reciprocal risk was between physical health and NMPDU of opioid analgesics; and (3) there were no associations with stimulants. CONCLUSION: These results, among the first ever on reciprocal effects between NMPDU and mental and physical health status, give unique information concerning the adverse effects of NMPDU on health and vice versa. The study shows that NMPDU is not only a sign of self-medication but may induce health problems.

Psychopathologie chez les enfants de patients souffrant de troubles liés à une substance ou de troubles de l'humeur

L'objectif principal de cette thèse est d'exposer les résultats des premières évaluations psychopathologiques menées, dans le cadre d'une étude familiale, sur une cohorte d'enfants de patients souffrant de troubles liés à des substances ou de troubles de l'humeur. Plus précisément, nous avons examiné des enfants de patients souffrant de dépendance à l'alcool et à l'héroïne, de troubles bipolaires et unipolaires, ainsi qu'un groupe d'enfants témoins, dont les parents étaient des patients contrôles recrutés dans des départements d'orthopédie. Nous nous sommes ainsi proposé d'étudier de façon précise le risque de psychopathologie chez des enfants de parents malades en menant une étude longitudinale prospective au moyen d'évaluations standardisées. Ces enfants ainsi que leurs parents-patients (probands) sont actuellement évalués chaque trois ans. Nous avons aussi effectué une évaluation des co-parents biologiques de ces enfants afin de tenir compte également de l'impact de leurs éventuels troubles psychiatriques sur les manifestations psychopathologiques de leurs enfants. Le volet longitudinal-prospectif de ces familles est actuellement en cours, et les deux articles contenus dans cette thèse, publiés dans des revues à politique éditoriale, exposent les résultats obtenus lors de l'évaluation initiale de ces enfants et de leurs deux parents. Nous résultats montrent que les enfants de patients, pourtant encore très jeunes (moyenne d'âge d'environ 11 ans), présentent déjà de plus hauts taux de manifestations psychopathologiques que les enfants de parents contrôles. Nous avons également pu démontrer le rôle joué par le co-parent, en plus de celui du parent-patient, dans le développement de psychopathologie chez ces enfants. Ces résultats soulignent la nécessité d'une identification et d'une prise en charge très précoce de ces enfants, et indiquent également l'importance de porter un intérêt clinique non seulement au parent-patient, mais également au co-parent de ces enfants. - This thesis presents the baseline results of psychopathological assessments conducted in a prospective longitudinal cohort study of children whose parents have substance use disorders or mood disorders. More specifically, we examined children of alcohol- and heroin- dependent patients, children of patients with bipolar or unipolar disorders, and, as a control group, children whose parents were recruited in orthopedics departments. Using standardized assessments, we conducted this study to accurately assess the risk of psychopathology in children of parents with these substance use or mood disorders. The prospective longitudinal study of these families is currently ongoing, with the children and their parent-patients (probands) being followed up every three years. We also assessed the biological co-parent in order to determine the impact of their possible psychiatric disorders on the risk of mental disorders in offspring. The two articles of this thesis, which have been published in peer-reviewed journals, present the results of baseline assessments of the children and their two parents. Our results show that the children of patients who are still very young at baseline assessment (mean age about 11 years) already have higher rates of psychopathological manifestations compared with children of control parents. We also demonstrate the effect of the co-parents, in addition to that of the parent-patients, on the development of early psychiatric manifestations in these children. These results emphasize the importance of prompt identification and treatment of these children and highlight the need for clinical care not only of the patient, but also of the co- parent, in order to optimize the prevention of mental disorders in their offspring.

Cannabis use and other illicit drug use: Do subjective experiences during first cannabis use increase the probability of using illicit drug use?

Background and aims: Few studies have examined whether subjective experiences during first cannabis use are related to other illicit drug (OID) use. This study investigated this topic. Methods: Baseline data from a representative sample of young Swiss men was obtained from an ongoing Cohort Study on Substance Use Risk Factors (N ¼ 5753). Logistic regressions were performed to examine the relationships between cannabis use and of subjective experiences during first cannabis use with 15 OID. Results: Positive experiences increased the likelihood of using hallucinogens (hallucinogens, salvia divinorum, spice; p50.015), stimulants (speed, ecstasy, cocaine, amphetamines/methamphetamines; p50.006) and also poppers, research chemicals, GHB/GBL, and crystal meth (p50.049). Sniffed drugs (poppers, solvents for sniffing) and ''hard'' drugs (heroin, ketamine, research chemicals, GHB/GBL and crystal meth) were more likely to be used by participants who experienced negative feelings on first use of cannabis (p50.034). Conclusion: Subjective feelings seemed to amplify the association of cannabis with OID. The risk increased for drugs with effects resembling feelings experienced on first cannabis use. Negative experiences should also be a concern, as they were associated with increased risk of using the ''hardest'' illicit drugs.

Is the relationship between major depressive disorder and self-reported alcohol use disorder an artificial one ?

AIMS: Many studies have suggested a close relationship between alcohol use disorder (AUD) and major depressive disorder (MDD). This study aimed to test whether the relationship between self-reported AUD and MDD was artificially strengthened by the diagnosis of MDD. This association was tested comparing relationships between alcohol use and AUD for depressive people and non-depressive people. METHODS: As part of the Cohort Study on Substance Use Risk Factors, 4352 male Swiss alcohol users in their early twenties answered questions concerning their alcohol use, AUD and MDD at two time points. Generalized linear models for cross-sectional and longitudinal associations were calculated. RESULTS: For cross-sectional associations, depressive participants reported a higher number of AUD symptoms (β = 0.743, P < 0.001) than non-depressive participants. Moreover, there was an interaction (β = -0.204, P = 0.001): the relationship between alcohol use and AUD was weaker for depressive participants rather than non-depressive participants. For longitudinal associations, there were almost no significant relationships between MDD at baseline and AUD at follow-up, but the interaction was still significant (β = -0.249, P < 0.001). CONCLUSION: MDD thus appeared to be a confounding variable in the relationship between alcohol use and AUD, and self-reported measures of AUD seemed to be overestimated by depressive people. This result brings into question the accuracy of self-reported measures of substance use disorders. Furthermore, it adds to the emerging debate about the usefulness of substance use disorder as a concept, when heavy substance use itself appears to be a sensitive and reliable indicator.

Prospective pathways between heroin use and nonmedical use of prescription opioids: Trajectories among young Swiss men

BACKGROUND. So far few studies have focused on the last steps of drug-use trajectories. Heroin has been described as a final stage, but the non-medical use of prescription opioids (NMUPOs) is often associated with heroin use. There is, however, no consensus yet about which one precedes the other. AIMS. The objective of this study was to test which of these two substances was likely to be induced by the other using a prospective design. MATERIAL AND METHODS. We used data from the Swiss Longitudinal Cohort Study on Substance Use Risk Factors (C-SURF) to assess exposure to heroin and NMUPO at two times points (N = 5,041). Cross-lagged panel models provided evidence regarding prospective pathways between heroin and NMUPOs. Power analyses provided evidence about significance and clinical relevance. RESULTS. Results showed that heroin use predicted later NMUPO use (? = 1.217, p < 0.001) and that the reverse pathway was non-significant (? = 0.240, p = .233). Heroin use seems to be an important determinant, causing a 150% risk increase for NMUPO use at follow-up, whereas NMUPO use at baseline increases the risk of heroin use at follow-up by a mere non-significant 20%. CONCLUSIONS. Thus, heroin users were more likely to move to NMUPOs than non-heroin users, whereas NMUPO users were not likely to move to heroin use. The pathway of substance use seemed to include first heroin use, then NMUPO use.