0
We're unable to sign you in at this time. Please try again in a few minutes.
Retry
We were able to sign you in, but your subscription(s) could not be found. Please try again in a few minutes.
Retry
There may be a problem with your account. Please contact the AMA Service Center to resolve this issue.
Contact the AMA Service Center:
Telephone: 1 (800) 262-2350 or 1 (312) 670-7827  *   Email: subscriptions@jamanetwork.com
Error Message ......
Article |

Ethnic Differences in Adolescent Substance Initiation Sequences FREE

Lisa M. Guerra, MD, MS; Patrick S. Romano, MD, MPH; Steven J. Samuels, PhD; Philip H. Kass, PhD, DVM
[+] Author Affiliations

From the Department of Pediatrics, University of Southern California, School of Medicine, Los Angeles (Dr Guerra); and Departments of General Medicine and Pediatrics, UC Davis Medical Center, Sacramento, Calif (Dr Romano), Department of Epidemiology and Preventive Medicine, UC Davis School of Medicine (Dr Samuels), and Departments of Population Health and Reproduction and Epidemiology and Preventive Medicine, UC Davis School of Veterinary Medicine and School of Medicine, Davis (Dr Kass).


Arch Pediatr Adolesc Med. 2000;154(11):1089-1095. doi:10.1001/archpedi.154.11.1089.
Text Size: A A A
Published online

Objectives  To evaluate ethnic differences in the initiation sequences of tobacco, alcohol, marijuana, and cocaine use among US high school students and to determine if ethnicity is a predictor of progression from licit to illicit substances or initiation of illicit substances before licit substances.

Design  Cross-sectional analyses of the Centers for Disease Control and Prevention's 1995 Youth Risk Behavior Survey.

Setting  US high schools.

Participants  A total of 8550 high school students randomly selected by cluster design.

Main Outcome Measures  Respondents were categorized based on self-reported sequence of initiating substances as follows: none, licit substances only, licit substances then illicit substances (typical), illicit substances first (reverse), and licit and illicit substances at the same time (concurrent).

Results  Adjusting for age, maternal education, and region, progression from licit to illicit substances was significantly associated with black ethnicity (odds ratio [OR], 1.5; 95% confidence interval [CI], 1.04-2.1) and male sex (OR, 1.4; 95% CI, 1.2-1.6). Black male and Latino female students whose mothers completed at least high school were more likely than white students with similarly educated mothers to initiate illicit substances before licit substances (OR, 3.0; 95% CI, 1.7-5.3; and OR, 5.9; 95% CI, 1.7-20; respectively). Similar trends were noted for the concurrent sequence.

Conclusions  The pattern of initiating tobacco, alcohol, marijuana, and cocaine use differs by ethnicity. Maternal education may be a proxy variable for other significant risk factors.

SUBSTANCE USE can interfere with normal cognitive and psychological development and is associated with poor decision making,1 poor educational and job performance, problematic relationships, and economic instability.2 Many of the behaviors initiated during adolescence continue into adulthood.3,4 As such, examining substance abuse patterns and their correlates in high school students is an important public health priority.

Understanding the differences in prevalence rates of substance initiation patterns may foster the development of more effective interventions. Several studies5 have concluded that definite sequences of substance use exist. At least 4 stages were identified by Kandel et al5 through a longitudinal prospective study of New York high school students: (1) beer or wine, (2) cigarettes or hard liquor, (3) marijuana, and (4) other illicit drugs. Most adolescents went through 2 stages of licit drugs between nonuse and marijuana. Few students in this study progressed to illicit drugs without initiating marijuana use. Seventy-five percent of the high school students who used both licit and illicit substances progressed along the following sequence: nonuse, legal drugs, marijuana, pills (methamphetamine, other amphetamines, barbiturates, or tranquilizers), psychedelics, cocaine, and heroin.5 A longitudinal study6 of subjects from 13 to 24 years of age showed that, in general, alcohol and tobacco use begins at earlier ages than marijuana and cocaine use.

Several studies5,7,8 have demonstrated that tobacco is a gateway drug to marijuana use. However, this appears less likely among African American students, who tend to have a low prevalence of tobacco use and a high prevalence of marijuana use.7,9 Data from the Centers for Disease Control and Prevention's 1995 Youth Risk Behavior Survey (YRBS) demonstrate a difference in the age of drug use initiation by ethnicity. For example, white (25.9%) and Latino (26.6%) students were significantly more likely than African American students (17.2%) to have initiated smoking by 13 years of age.9 Latino students (39.5%) were significantly more likely than white students (30.3%) to have initiated alcohol use before 13 years of age. African American (11.1%) and Latino (12.6%) students were significantly more likely than white students (5.6%) to have initiated marijuana use before 13 years of age. African American (1.3%) and Latino (1.7%) students were more likely than white students (0.9%) to have initiated cocaine use before 13 years of age.9

Substance initiation sequences may be due to the effect of one type of drug on the use of another, the relationships between the use of the substances and the user's demographic, psychosocial, and environmental characteristics, or a combination of both. The period of risk for initiation of cigarettes, alcohol, and marijuana use is completed by the age of 20 years and for most illicit drugs by the age of 21 years.6 The younger the age of onset of legal drug use, the more subjects are at risk of initiating use of an illicit substance.10 Higher prevalence rates of substance use are associated with increasing age.7,11 Sex differences in substance initiation patterns have also been noted. Most men progress along the following sequences: alcohol preceding marijuana and alcohol and marijuana preceding other illicit drugs.12 Most women follow sequences such as alcohol or cigarettes preceding marijuana and alcohol, cigarettes, and marijuana preceding other illicit drugs.12 Parental education may play a role on the adolescent initiation sequence indirectly, since low educational attainment is associated with substance use in adults13 and access to substances in the home is associated with use of tobacco, alcohol, and marijuana among high school students.14 Thus, in examining substance initiation sequences, it would be important to adjust for age, sex, and parental education.

Although attention has been given to substance initiation sequences, few studies have examined differences in these sequences by ethnicity. Most studies5,6,12,15,16 have examined only one ethnic group, compared different countries, or compared whites with nonwhites or African Americans. Given the different prevalence, initiation rates, and risk factors of substance use among African American, Latino, and white high school students,11,17,18 there may be important differences in the substance initiation sequences by ethnicity as well.

We analyzed the YRBS data to answer the following questions about ethnic differences in substance initiation sequences: (1) Among students who start using licit substances, what characteristics are associated with progression to illicit substances? (2) Among students who have used licit and illicit substances, what characteristics are associated with initiation of illicit drugs first?

This study provides a preliminary examination of initiation sequences for the most commonly used licit and illicit substances among high school students: tobacco, alcohol, marijuana, and cocaine. The objective is to identify subgroups at risk for a particular drug sequence to help design more targeted and effective preventive programs.

DATA SOURCE AND INSTRUMENTATION

The YRBS is part of the Youth Risk Behavior Surveillance System initiated by the Centers for Disease Control and Prevention to monitor behaviors associated with significant morbidity and mortality in youth and adulthood. The YRBS consists of 6 demographic questions on sex, age, grade, ethnicity, and parental education and 82 questions on behaviors associated with unintentional and intentional injuries; tobacco, alcohol, and other drug use; sexual behaviors; dietary practices; and physical activity. The instrument was written at the 7th-grade reading level and intended for 9th through 12th graders.

In 1995, the Centers for Disease Control and Prevention conducted a 14-day test-retest reliability study involving students in grades 7 through 12. A κ statistic was calculated for each question and ranged from 14.5% to 91.1%. The questions used in this study received a score of substantial or higher (61%-80%), except for the question pertaining to age of cocaine initiation, which had a κ of 43.6% (moderate).19

SAMPLING SCHEME

The 1995 YRBS used a 3-stage cluster sampling design. Initially, the United States was divided into 1955 primary sampling units based on large counties or groups of smaller, adjacent counties. The primary sampling units were divided into 16 strata based on the degree of urbanization and the relative percentage of African American and Hispanic students. At the first stage, 52 primary sampling units were randomly selected from the 16 strata by probability proportional to school enrollment size. At the second stage, 157 private and public schools were selected with probability proportional to school enrollment size. Schools with large numbers of African American and Hispanic students were sampled at higher rates. At the third stage, 1 or 2 classes of a required subject from grades 9 to 12 were randomly selected from each school.9

ADMINISTRATION

A self-administered, anonymous questionnaire was given to all students in each sampled class after obtaining parental consent. Trained personnel supervised the administration of the questionnaires during regular class time. The questionnaire took approximately 45 minutes to complete. The final sample consisted of 10,904 students from 110 schools. A weighting factor was applied to each student record to correct for nonresponse and oversampling of the African American and Latino populations. The resulting data are representative of 9th through 12th grade students from public and private schools throughout the United States because of the sampling scheme and weighting factors.9 For this reason, we present weighted percentages in all of the tables as others recommend.20

DEFINITIONS

Sequences of tobacco, alcohol, marijuana, and cocaine use were categorized in terms of licit substances defined for adults (tobacco and alcohol) and illicit substances (marijuana and cocaine). We ascertained each respondent's sequence from the answers to the following questions: "How old were you when you . . . smoked a whole cigarette for the first time . . . had your first drink of alcohol other than a few sips . . . tried marijuana for the first time . . . tried any form of cocaine, including powder, crack, or freebase for the first time?" Missing values of each question (n = 1120) were dropped. The allowable responses were as follows: "never, 8 years or younger, 9 or 10 years old, 11 or 12 years old, 13 or 14 years old, 15 or 16 years old, 17 years or older." Based on the individual's responses to these questions, 163 sequences were identified and collapsed into 6 substance-pattern categories: none, licit substances only (licit only sequence), illicit substances only, licit substances then illicit substances (typical sequence), illicit substances then licit substances, and licit and illicit substances at about the same time (concurrent sequence). The illicit only sequence was combined with the illicit to licit sequence, since its numbers were too small for meaningful analysis and both sequence types initiated substance use with illicit drugs. This new sequence was labeled the reverse sequence.

Age at the time of questionnaire administration was categorized as 14 years or younger, 15 years old, 16 years old, 17 years old, and 18 years or older. Ethnicity was obtained from the question, "How do you describe yourself?" The allowable choices included "white–not Hispanic, black–not Hispanic, Hispanic or Latino, Asian or Pacific Islander, American Indian or Alaskan Native, other." The other, Asian or Pacific Islander, and American Indian or Alaskan Native choices were too small for meaningful interpretation and thus not included in the results.

Maternal education was used as an indirect measure of socioeconomic status. Information on paternal education was requested, but 768 more responses were "not sure" compared with the answers for maternal education. Maternal education was grouped into 4 categories, which were then collapsed into "less than high school" and "high school or more."

Because prior literature showed some differences in substance use across geographic regions,21 we controlled for region, with the region variable categorized as Northeast, Midwest, South, and West.

STATISTICAL ANALYSIS

All analyses were performed using the statistical package Stata.22 Stata allowed precise estimation of confidence intervals (CIs) that accounted for the 3-stage cluster-sampling scheme. Subjects with missing data were excluded from the analysis. The demographic characteristics, substance initiation ages, and past month substance use of the excluded subjects were compared with those of included subjects using the F statistic, adjusting for the 3-stage cluster sample design. Significance was defined at an α level of . 05. Tables were constructed to examine the crude association between sequence and ethnicity, age, sex, maternal education, and region. To answer the first research question ("Among students who start using licit substances, what characteristics are associated with progression to illicit substances?"), only the subjects who belonged to either the licit only or typical sequence were analyzed, using logistic regression with a binary outcome (n = 5887). In this model, age had a linear relationship with the log odds of sequence. The other main effects—sex, ethnicity, maternal education, and region—were entered as categorical variables. All 2-way interactions involving ethnicity, sex, age, region, and maternal education were tested using the Wald test. None were statistically significant (P>.05).

To answer the second research question ("Among students who have used licit and illicit substances, what characteristics are associated with initiation of illicit drugs first?), only the subjects who had used both illicit and licit substances were included in the analysis. Because there were more than 2 possible outcomes (typical, reverse, and concurrent sequences), polytomous logistic regression was used to generate a set of logistic models that compared relevant pairs of outcomes. In our case, the relevant outcome pairs were the reverse sequence vs the typical sequence and the concurrent sequence vs the typical sequence. The relationship of the reverse and concurrent sequences to ethnicity controlling for sex, age, maternal education, and region was examined in this manner. The study population included 4473 observations. The outcome variables reference group was the typical sequence. Sex, ethnicity, age, maternal education, and region were all indicator variables with female, white, 18 years or older, high school education or more, and Northeast used as the covariate reference groups, respectively. However, the regions were arbitrarily assigned after the data were collected and therefore the results for region may not offer any meaningful information. Several models were compared using the F statistic. Again, 2-way interactions involving ethnicity, sex, age, region, and maternal education were tested using the Wald test. Significant interaction terms between ethnicity and sex and between ethnicity and maternal education were found.

RESPONSE RATE

The school response rate was 70%, and the student response rate was 86%, making the overall response rate 60% (0.70 × 0.86).

There were 10,904 respondents to the 1995 YRBS. Respondents who did not report their age at first use of tobacco, alcohol, marijuana, or cocaine or any of the other questions involved in the analysis were excluded, leaving a sample of 8550 subjects. Table 1 shows the demographic characteristics of the excluded and included subjects. Among the 8550 participants in the study, 75% were white, 14% were African American, and 11% were Latino. Approximately 49% of the subjects were female, and 86.4% of the students' mothers had at least a high school education. The included and excluded groups were not significantly different in the demographic characteristics (Table 1). However, the included and excluded groups were significantly different in the age they first initiated marijuana use (P = .006) and the use of alcohol and marijuana in the past 30 days (P<.001 and .006, respectively). Included subjects were somewhat more likely than excluded subjects to have reported a history of ever using marijuana (44%; 95% CI, 40%-48%; vs 37%; 95% CI, 32%-42%). Included subjects were more likely than excluded subjects to have started using marijuana at 13 to 14 years of age and 17 years or older, and the excluded group was more likely than the included group to have not used marijuana or alcohol in the past 30 days (data not shown).

Table Graphic Jump LocationTable 1. Weighted Percentages and Frequencies of the Subjects' Demographic Characteristics

Table 2 provides the definition and weighted percentage of each sequence among US high school students in 1995. The most common sequences were the licit only (39.4%) and typical sequences (30.9%).

Table Graphic Jump LocationTable 2. Weighted Percentages and Frequencies of Substance Initiation Sequences

Table 3 and Table 4 summarize the weighted prevalences and weighted unadjusted odds ratios (ORs), respectively, of each sequence by ethnicity, sex, and maternal education. The weighted unadjusted ORs by characteristic were calculated using univariate polytomous logistic regression. The reference group was the typical sequence, since this is the usual drug initiation pattern described in the literature. African Americans were about 1.5 times more likely to exhibit no behaviors, 2.3 times more likely to follow the concurrent sequence, and 3 times as likely to follow the reverse sequence compared with white students. Latinos were also more likely than whites to follow the concurrent or reverse sequences relative to the typical sequence (ORs, 1.4 and 1.8, respectively). High school male students were about 20% less likely to use no substances or to follow the licit only sequence than female students. Although the 95% CI for the ORs includes one for the reverse sequence, there appears to be a trend toward higher odds for this sequence in boys compared with girls. Students whose mothers had less than a high school education were 2.2 times more likely to follow the reverse sequence and 1.7 times more likely to follow the concurrent sequence than students whose mothers had at least a high school education.

Table Graphic Jump LocationTable 3. Frequencies and Weighted Percentages of Substance Initiation Sequences by Demographic Characteristics
Table Graphic Jump LocationTable 4. Weighted Unadjusted Odds Ratios for Specific Substance Initiation Sequences, by Demographic Characteristic, Using Typical Sequence as Reference Group

Table 5 illustrates the factors that were independently associated with proceeding to illicit substance use among students who were already using licit substances. Interaction terms between ethnicity and sex, ethnicity and age, and ethnicity and maternal education were not statistically significant. Age had a linear relationship with the log odds of the typical and licit only sequences. According to this model, progression to illicit substance use (given that a student already initiated licit substance use) was significantly associated with African American ethnicity (OR, 1.5; 95% CI, 1.04-2.1), male sex (OR, 1.4; 95% CI, 1.2-1.6), and age (OR, 1.2; 95% CI, 1.02-1.3). Having a mother with a high school education was not a significant predictor of progression from licit to illicit substances. Latino ethnicity also was not associated with progression from licit to illicit drugs.

Table Graphic Jump LocationTable 5. Factors Associated With Progression to Illicit Substances Once Licit Substances Are Initiated*

The results of the multivariate polytomous logistic regression analysis are presented in Table 6 and Table 7. Table 6 shows the weighted, adjusted, stratum-specific ORs for the concurrent sequence compared with the typical sequence. African American students were more likely than white students to follow the concurrent sequence when their mothers had at least a high school education. This association was observed in both sexes (boys: OR, 2.4; 95% CI, 1.5-3.7; girls: OR, 3.0; 95% CI, 1.7-5.3). Latino girls whose mothers had at least a high school education were more likely to follow the concurrent sequence than white girls whose mothers had the same level of education (OR, 1.9; 95% CI, 1.2-3.1).

Table Graphic Jump LocationTable 6. Adjusted Odds Ratios (95% Confidence Intervals) for the Concurrent Sequence Relative to the Typical Sequence by Ethnicity*
Table Graphic Jump LocationTable 7. Adjusted Odds Ratios (95% Confidence Intervals) for the Reverse Sequence Relative to the Typical Sequence by Ethnicity*

Table 7 shows that African American male students whose mothers had at least a high school education were 3 times more likely than comparable white students to initiate illicit substance use before licit substance use (OR, 3.0; 95% CI, 1.7-5.3). In addition, Latino high school girls whose mothers had at least a high school education were about 6 times more likely to initiate illicit substance use before licit substance use than white high school girls with the same maternal education (OR, 5.9; 95% CI, 1.7-20).

Expressed in another way, white students whose mothers did not graduate from high school were more likely than white students whose mothers completed high school to follow the concurrent and reverse sequences relative to the typical sequence (OR, 2.1; 95% CI, 1.2-3.8; and OR, 3.7; 95% CI, 1.3-10.4, respectively). By contrast, maternal education was not associated with the sequence of substance use initiation among African American (concurrent sequence: OR, 1.0; 95% CI, 0.5-2.0; reverse sequence: OR, 1.1; 95% CI, 0.2-7.4) and Latino students (concurrent sequence: OR, 1.6; 95% CI, 0.4-6.2; reverse sequence: OR, 1.0; 95% CI, 0.4-2.8).

Consistent with prior literature, we found that African American ethnicity, male sex, and increasing age were associated with progression from licit to illicit substance use. We also found, in unadjusted analyses, that African American and Latino students were more likely than white students to follow the concurrent or reverse sequences (relative to the typical sequence). However, closer study revealed that these differences were entirely limited to students whose mothers had graduated from high school. There were no ethnic differences in substance initiation patterns among students whose mothers had less education. This finding has not been previously reported.

Maternal education may be a proxy variable for other environmental and cultural factors and may represent different markers for each ethnicity. Several studies17,2326 have found that more acculturated Latinos have higher prevalences of substance use than recent immigrants. Perhaps more acculturated students have parents with higher education. As such, the findings for Latino female students may indicate that more acculturated female students are at higher risk for initiating illicit substance use than less acculturated female students. Further studies that include acculturation scales, country of birth, and language are needed to clarify the interpretation of the results. It is unlikely that maternal education represents acculturation among white students. In the comparison between African American and white students, maternal education may not represent the same marker either. Maternal education may represent other sociopolitical factors based on the different group experiences and histories in the United States.27,28 Further studies that include the students' experiences, outlook of the future, feelings of disenfranchisement, and parental influences are needed to delineate a better understanding of the difference seen in maternal education among the African American students.29 In addition, the difference between white students and African American and Latino students may represent an increased availability or accessibility of marijuana and cocaine in predominantly Latino and African American communities. Further assessment of this phenomenon when the mothers have the same level of education needs to be obtained.

This study has several important strengths. First, the YRBS is based on a large probability sample. As a result, we can generalize our findings with some caution to the high school students in the United States who attend school. Second, our statistical methods corrected for the design effect that resulted from the 3-stage cluster sampling. Third, the survey instrument was carefully designed by survey research professionals for the purpose of investigating high-risk behaviors and has been repeatedly refined during the past 9 years. Fourth, the instrument has been validated through community-based test-retest research.

In interpreting the results of the study, several limitations must be recognized. First, this is a cross-sectional study. Causal relationships cannot be determined. Second, these data may not reflect the true prevalence of substance use in the study population because of the respondents' desire to give socially acceptable answers and problems with recall. However, we doubt that these factors would differ across ethnic groups, so our ethnic comparisons should be unbiased. Third, the concurrent sequence assumes that the student initiated use of both licit and illicit substances at the same time. However, within this category, there could be students who followed the typical or reverse sequences because of the 2-year age spans in the allowable responses for the substance initiation questions. If each ethnic group accounted for a different proportion of misclassified sequences, the resulting differential misclassification could bias the results. Future studies should attempt to ascertain more precisely when use of each substance was initiated.

The relationship between ethnicity and substance sequence may be different for students who participated in the YRBS and those who did not participate or were excluded because of missing data. The characteristics of the included and excluded groups varied significantly at several levels. In general, the excluded group consisted of significantly more students with younger marijuana use initiation ages and those who never used marijuana. The included group may have fewer never users of alcohol and marijuana in the past 30 days than the excluded group, since the latter group had no sequence of substance use. The differences among the groups may have contributed to biased results. Also, we have no information on substance use among nonrespondents to the questionnaire. The overall response rate was only 60%, which could have contributed to selection bias. The adolescents who were not in the study could have included those who refused consent, had parents who refused consent, were truant, dropped out of school, ran away, or were homeless. These adolescents may be more likely to use substances or may tend to follow different initiation sequences than those at school. As such, the generalizability is limited to students enrolled in and attending high school. In addition, the use of a high school student data set may bias our results. Adolescents who drop out of school may be more likely to have parents who did not graduate from high school. Students who dropped out of school may also be more likely to be involved with substance abuse or associated high-risk behaviors than the students enrolled in school. Thus, our stratum-specific ORs may be biased, because the students who are most at risk and who have mothers with low educational attainment may not be fully represented in the sample. Further studies including adolescents not in school are needed to clarify these issues.

Finally, the data set does not allow for control of likely confounders that might influence the initiation of substance use. Such confounders include socioeconomic status, neighborhood and school characteristics, perceived peer pressure, family characteristics, peer norms, and substance availability. In view of these limitations, future studies should collect longitudinal prospective data on psychosocial, biological, and environmental factors and should include adolescents who are not enrolled in school. In addition, larger subsample sizes to study Asians, American Indians, and subgroups of each ethnicity would contribute to a better understanding of the association between ethnicity and substance initiation sequences.

Our study shows that African American students and Latino female students were more likely than whites to initiate illicit substance use at the same time as or before licit substances when their mothers had graduated from high school. As such, this subset of Latino and African American students may not follow the usual pathway of substance initiation that has been previously described in the literature. Maternal education appears to modify the effect of ethnicity on adolescent substance initiation sequences. Further studies are needed to understand the reasons for this phenomenon and the true markers of maternal education for each ethnic group. Despite the limitations of this study, our results indicate that cultural and family characteristics are important to consider in understanding adolescent substance initiation patterns. Further studies on cultural characteristics of substance use could provide new information to improve recognition of populations at risk in the clinical setting and to improve interventions and prevention programs for adolescents from diverse backgrounds.

Accepted for publication June 22, 2000.

This study was funded by the Hispanic Center of Excellence in the Office of Medical Education at UC Davis School of Medicine.

Presented at the 39th Annual Meeting of the Ambulatory Pediatric Association, San Francisco, Calif, May 3, 1999.

We thank Jann Murray-Garcia, MD, MPH, Robert S. Byrd, MD, MPH, and Jorge A. Garcia, MD, MS, for their invaluable input in the preparation of the manuscript and insightful comments during the editing process.

Reprints: Lisa M. Guerra, MD, MS, LAC + USC, Women's and Children's Hospital, 1240 N Mission Rd, T-11, Los Angeles, CA 90033.

O'Malley  PMJohnston  LDBachman  JG Adolescent substance use: epidemiology, and implications for public policy. Pediatr Clin North Am. 1995;42241- 260
DiClemente  RHansen  WPonton  L Handbook of Adolescent Health Risk Behavior: Issues in Clinical Child Psychology.  New York, NY Plenum Press1996;
Kandel  DBKessler  RCMargulies  RZ Antecedents of adolescent initiation into stages of drug use: a developmental analysis. J Youth Adolesc. 1978;713- 40
Link to Article
Newcomb  MMaddahian  ESkager  RBentler  P Substance abuse and psychosocial risk factors among teenagers: associations with sex, age, ethnicity, and type of school. Am J Drug Alcohol Abuse. 1987;13413- 433
Link to Article
Kandel  DFaust  R Sequence and stages in patterns of adolescent drug use. Arch Gen Psychiatry. 1975;32923- 932
Link to Article
Kandel  DLogan  J Patterns of drug use from adolescence to young adulthood, I: periods of risk for initiation, continued use, and discontinuation. Am J Public Health. 1984;74660- 666
Link to Article
Neamark-Sztainer  DStory  MFrench  SCassuto  NJacobs  DJResnick  M Patterns of health-compromising behaviors among Minnesota adolescents: sociodemographic variations. Am J Public Health. 1996;861599- 1606
Link to Article
Botvin  GBotvin  E Adolescent tobacco, alcohol, and drug abuse: prevention strategies, empirical findings, and assessment issues. J Dev Behav Pediatr. 1992;13290- 301
Link to Article
Kann  LWarren  CWHarris  WA  et al.  Youth risk behavior surveillance—United States, 1995. MMWR Morb Mortal Wkly Rep. 1996;451- 84
Yamaguchi  KKandel  D Patterns of drug use from adolescence to young adulthood, III: predictors of progression. Am J Public Health. 1984;74673- 681
Link to Article
O'Malley  PBachman  JJohnston  L Period, age and cohort effects in substance use among young Americans: a decade of change, 1976-1986. Am J Public Health. 1988;781315- 1321
Link to Article
Yamaguchi  KKandel  D Patterns of drug use from adolescence to young adulthood, II: sequences of progression. Am J Public Health. 1984;74668- 672
Link to Article
Escobedo  LGPeddicord  JP Smoking prevalence in US birth cohorts: the influence of gender and education. Am J Public Health. 1996;86231- 236
Link to Article
Resnick  MDBearman  PSBlum  RW  et al.  Protecting adolescents from harm, findings from the National Longitudinal Study on Adolescent Health. JAMA. 1997;278823- 832
Link to Article
Adler  IKandel  D Cross-cultural perspectives on developmental stages in adolescent drug use. J Stud Alcohol. 1981;42701- 715
Kandel  DAdler  ISudit  M The epidemiology of adolescent drug use in France and Israel. Am J Public Health. 1981;71256- 265
Link to Article
Bachman  JWallace  JJO'Malley  PJohnston  LKurth  CNeighbors  H Racial/ethnic differences in smoking, drinking, and illicit drug use among American high school seniors, 1976-1989. Am J Public Health. 1991;81372- 377
Link to Article
Vega  WZimmerman  RWarheit  GApospori  EGil  A Risk factors for early adolescent drug use in four ethnic and racial groups. Am J Public Health. 1993;83185- 189
Link to Article
Brener  NCollins  JKann  LWarren  CWilliams  B Reliability of the Youth Risk Behavior Survey Questionnaire. Am J Epidemiol. 1995;141575- 580
Korn  ELGraubard  BI Epidemiologic studies utilizing surveys: accounting for the sampling design. Am J Public Health. 1991;811166- 1173
Link to Article
Johnston  LDO'Malley  PMBachman  JG National Survey Results on Drug Use From the Monitoring the Future Study, 1975-1994.  Rockville, Md National Institute on Drug Abuse, US Dept of Health and Human Services, Public Health Service, National Institutes of Health1995-1996;
Corporation  S Stata Statistical Software [computer program]. Release 5.0. College Station, Tex Stata Press1997;
Otero-Sabogal  RSabogal  FPerez-Stable  EJ Psychosocial correlates of smoking among immigrant Latina adolescents. J Natl Cancer Inst Monogr. 1995;1865- 71
Marin  GPerez-Stable  EMarin  B Cigarette smoking among San Francisco Hispanics: the role of acculturation and gender. Am J Public Health. 1989;79196- 198
Link to Article
Palinkas  LPierce  JRosbrook  BPickwell  SJohnson  MBal  D Cigarette smoking behavior and beliefs of Hispanics in California. Am J Prev Med. 1993;9331- 337
Sabogal  F Perceived self-efficacy to avoid cigarette smoking and addiction: differences between Hispanics and non-Hispanics and non-Hispanic whites. Hispanic J Behav Sci. 1989;11136- 147
Link to Article
Murray-Garcia  J African-American youth: essential prevention strategies for every pediatrician. Pediatrics. 1995;96132- 137
Barthwell  AGHewitt  WJilson  I An introduction to ethnic and cultural diversity. Pediatr Clin North Am. 1995;42431- 451
Krieger  NRowley  DHerman  AAvery  BPhillips  M Racism, sexism and social class: implications for studies of health, disease and well-being. Am J Prevent Med. 1993;982- 122

Figures

Tables

Table Graphic Jump LocationTable 1. Weighted Percentages and Frequencies of the Subjects' Demographic Characteristics
Table Graphic Jump LocationTable 2. Weighted Percentages and Frequencies of Substance Initiation Sequences
Table Graphic Jump LocationTable 3. Frequencies and Weighted Percentages of Substance Initiation Sequences by Demographic Characteristics
Table Graphic Jump LocationTable 4. Weighted Unadjusted Odds Ratios for Specific Substance Initiation Sequences, by Demographic Characteristic, Using Typical Sequence as Reference Group
Table Graphic Jump LocationTable 5. Factors Associated With Progression to Illicit Substances Once Licit Substances Are Initiated*
Table Graphic Jump LocationTable 6. Adjusted Odds Ratios (95% Confidence Intervals) for the Concurrent Sequence Relative to the Typical Sequence by Ethnicity*
Table Graphic Jump LocationTable 7. Adjusted Odds Ratios (95% Confidence Intervals) for the Reverse Sequence Relative to the Typical Sequence by Ethnicity*

References

O'Malley  PMJohnston  LDBachman  JG Adolescent substance use: epidemiology, and implications for public policy. Pediatr Clin North Am. 1995;42241- 260
DiClemente  RHansen  WPonton  L Handbook of Adolescent Health Risk Behavior: Issues in Clinical Child Psychology.  New York, NY Plenum Press1996;
Kandel  DBKessler  RCMargulies  RZ Antecedents of adolescent initiation into stages of drug use: a developmental analysis. J Youth Adolesc. 1978;713- 40
Link to Article
Newcomb  MMaddahian  ESkager  RBentler  P Substance abuse and psychosocial risk factors among teenagers: associations with sex, age, ethnicity, and type of school. Am J Drug Alcohol Abuse. 1987;13413- 433
Link to Article
Kandel  DFaust  R Sequence and stages in patterns of adolescent drug use. Arch Gen Psychiatry. 1975;32923- 932
Link to Article
Kandel  DLogan  J Patterns of drug use from adolescence to young adulthood, I: periods of risk for initiation, continued use, and discontinuation. Am J Public Health. 1984;74660- 666
Link to Article
Neamark-Sztainer  DStory  MFrench  SCassuto  NJacobs  DJResnick  M Patterns of health-compromising behaviors among Minnesota adolescents: sociodemographic variations. Am J Public Health. 1996;861599- 1606
Link to Article
Botvin  GBotvin  E Adolescent tobacco, alcohol, and drug abuse: prevention strategies, empirical findings, and assessment issues. J Dev Behav Pediatr. 1992;13290- 301
Link to Article
Kann  LWarren  CWHarris  WA  et al.  Youth risk behavior surveillance—United States, 1995. MMWR Morb Mortal Wkly Rep. 1996;451- 84
Yamaguchi  KKandel  D Patterns of drug use from adolescence to young adulthood, III: predictors of progression. Am J Public Health. 1984;74673- 681
Link to Article
O'Malley  PBachman  JJohnston  L Period, age and cohort effects in substance use among young Americans: a decade of change, 1976-1986. Am J Public Health. 1988;781315- 1321
Link to Article
Yamaguchi  KKandel  D Patterns of drug use from adolescence to young adulthood, II: sequences of progression. Am J Public Health. 1984;74668- 672
Link to Article
Escobedo  LGPeddicord  JP Smoking prevalence in US birth cohorts: the influence of gender and education. Am J Public Health. 1996;86231- 236
Link to Article
Resnick  MDBearman  PSBlum  RW  et al.  Protecting adolescents from harm, findings from the National Longitudinal Study on Adolescent Health. JAMA. 1997;278823- 832
Link to Article
Adler  IKandel  D Cross-cultural perspectives on developmental stages in adolescent drug use. J Stud Alcohol. 1981;42701- 715
Kandel  DAdler  ISudit  M The epidemiology of adolescent drug use in France and Israel. Am J Public Health. 1981;71256- 265
Link to Article
Bachman  JWallace  JJO'Malley  PJohnston  LKurth  CNeighbors  H Racial/ethnic differences in smoking, drinking, and illicit drug use among American high school seniors, 1976-1989. Am J Public Health. 1991;81372- 377
Link to Article
Vega  WZimmerman  RWarheit  GApospori  EGil  A Risk factors for early adolescent drug use in four ethnic and racial groups. Am J Public Health. 1993;83185- 189
Link to Article
Brener  NCollins  JKann  LWarren  CWilliams  B Reliability of the Youth Risk Behavior Survey Questionnaire. Am J Epidemiol. 1995;141575- 580
Korn  ELGraubard  BI Epidemiologic studies utilizing surveys: accounting for the sampling design. Am J Public Health. 1991;811166- 1173
Link to Article
Johnston  LDO'Malley  PMBachman  JG National Survey Results on Drug Use From the Monitoring the Future Study, 1975-1994.  Rockville, Md National Institute on Drug Abuse, US Dept of Health and Human Services, Public Health Service, National Institutes of Health1995-1996;
Corporation  S Stata Statistical Software [computer program]. Release 5.0. College Station, Tex Stata Press1997;
Otero-Sabogal  RSabogal  FPerez-Stable  EJ Psychosocial correlates of smoking among immigrant Latina adolescents. J Natl Cancer Inst Monogr. 1995;1865- 71
Marin  GPerez-Stable  EMarin  B Cigarette smoking among San Francisco Hispanics: the role of acculturation and gender. Am J Public Health. 1989;79196- 198
Link to Article
Palinkas  LPierce  JRosbrook  BPickwell  SJohnson  MBal  D Cigarette smoking behavior and beliefs of Hispanics in California. Am J Prev Med. 1993;9331- 337
Sabogal  F Perceived self-efficacy to avoid cigarette smoking and addiction: differences between Hispanics and non-Hispanics and non-Hispanic whites. Hispanic J Behav Sci. 1989;11136- 147
Link to Article
Murray-Garcia  J African-American youth: essential prevention strategies for every pediatrician. Pediatrics. 1995;96132- 137
Barthwell  AGHewitt  WJilson  I An introduction to ethnic and cultural diversity. Pediatr Clin North Am. 1995;42431- 451
Krieger  NRowley  DHerman  AAvery  BPhillips  M Racism, sexism and social class: implications for studies of health, disease and well-being. Am J Prevent Med. 1993;982- 122

Correspondence

CME
Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.
Submit a Comment

Multimedia

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Articles Related By Topic
Related Collections
PubMed Articles
JAMAevidence.com

Care at the Close of Life EDUCATION GUIDES
Palliative Care for Latino Patients and Their Families