LifeSkills Training (LST)

Study 1

Summary

Botvin et al. (1990, 1995, 2000) used a cluster randomized trial to examine 56 New York State schools with 5,954 seventh-grade student participants. The schools were assigned to two LST groups (one with in-person training and one with video training) and a control group. Students in the schools were followed for 6.5 years to assess self-reported substance use.

Botvin et al. (1990, 1995, 2000) found that, relative to the control group, the two LST groups reported significantly lower

• Cigarette smoking at posttest,
• Cigarette smoking and marijuana use at the three-year follow-up,
• Polydrug use at the 6.5-year follow-up.

Evaluation Methodology

Design: New York State Evaluation: In the spring of 1985, 56 participating schools in three geographic areas of New York State were surveyed to determine the amount of cigarette use of students. The schools were then divided into high, medium or low use schools. The original sample contained 5,954 students who were in 7th grade at the time. From within groups of schools with similar levels of cigarette smoking, schools were randomly assigned to one of the following groups: (1) E1 (n=18 schools): prevention program with a formal 1-day training workshop and implementation feedback by project staff, (2) E2 (n=16 schools): prevention program with training provided by videotape and no implementation feedback, and (3) a treatment as usual control group (n=22 schools). Random assignment successfully ensured equivalent groups at baseline. The program was administered by regular classroom teachers selected by each participating school.

Sample: The sample was approximately half (52%) male and predominantly (91%) white. Sample retention (based on all available students at the pretest) was 93% at the initial posttest (mid-7th grade), 81% at the 16-month follow-up (end of the 8th grade), 75% at the 28-month follow-up (end of the 9th grade), 67% at the 40-month follow-up (end of the 10th grade), and 60% at the end of the 60 month follow-up (end of 12th grade). Retention rates were virtually identical across conditions. Analysis of demographic characteristics of the sample using MANOVA revealed no significant differences between study groups at baseline. ANOVA revealed that substance use at pre-test had a significant effect on attrition rate for the 3-year study, a finding consistent with previous research. Analyses were conducted at the school level to provide for a more conservative test of the intervention.

The sample at the 6-year follow-up studies consisted of 447 individuals who had participated in the original trial and were contacted by mail at the end of the 12th grade (6.5 years from baseline data collection). This sample was 92.3% White and 40% male. Average age was 18.1 years and most (82.5%) lived in two-parent families. At the 6-year study, no significant differences were found between the samples at pre-test, nor were there any differential attrition differences at follow-up, based on demographic characteristics.

Measures: Prior to implementation, breath samples (to measure carbon monoxide levels) were collected from all participants, in order to increase validity of self-report data collected. In order to assess implementation fidelity, researchers observed LST lessons and recorded the amount of lesson information covered per session. A three year cumulative implementation score was computed for each student in the two intervention groups. To assess drug use, students reported frequency of use on three scales (one for each substance: tobacco, alcohol, marijuana). Students were also asked to report on the amount of tobacco and alcohol used on two scales. An additional scale was used to determine how often individuals drank to the point of intoxication. From these scales, measures to assess "heavy" use and polydrug use were created by partitioning collected measures into binary variables.

At the 6-year follow-up, participants were asked about frequency of illicit drug use, using 13 different categories of drugs that were based on the Monitoring the Future study. Six composite scores were created that reflected the sum of the individual drug items (marijuana, cocaine, inhalants, nonmedical pill use, heroin and other narcotics, and hallucinogens), and a second summary score that represented the sum of all "illicit drug use other than marijuana."

Analysis: The school was used as the unit of analysis, although 9th grade analyses were also supplemented with individual level analyses. Analyses were based on means for each drug use and polydrug use variable. Ordinary least-squares (OLS) regression was conducted, with the school as the unit of analysis. Separate regressions were performed for the full sample and the high fidelity sample. Inferences were based on one-tailed significant tests.

The 6 year follow-up on illicit drug use (Botvin et al., 2000) analyzed data using GLM ANCOVAs, adjusting for relevant covariates. Generalized estimating equations (GEE) were also conducted to control for intracluster correlations among students within schools. P-values for both analyses represented two-tailed significance levels.

Outcomes

Post-test: (Botvin, 1989)
The only behavioral effects found during the first 2 years of this study (grade 7 and grade 8) were for cigarette smoking. Prevention effects were evident for interpersonal skills knowledge, domain-specific knowledge and normative expectations concerning tobacco, alcohol, and marijuana use.

Long-term: 3-year - End of 9th grade (Botvin, Baker, Dusenbury, Tortu, & Botvin, 1990)
Results were restricted to the sample of students who had received a minimum of 60% of the prevention program. While the mean implementation fidelity score was 68%, researchers lowered the standard in order to establish a minimum standard of acceptable program implementation while retaining as much of the sample as possible. After inclusion criteria were applied, 75% of the sample remained. There were no significant differences found between the full sample and the analyzed sub-sample. Significant treatment effects were found using MANCOVA for cigarette smoking and marijuana use, with lower rates in both E1 and E2. Although there were no significant effects found for drinking frequency or amount, the frequency of getting drunk was significantly less in E2 than among controls.

6-year (End of 12th Grade): (Botvin, Baker, Dusenbury, Botvin, & Diaz, 1995)
Full Sample: The prevalence of weekly and monthly cigarette smoking was significantly lower for both intervention groups than the control group (monthly = .27, .26, .33; and weekly = .23, .21, .27, for E1, E2, and control groups respectively). Heavy smoking was significantly lower for the E2 intervention group than the control group (.09 vs .12). The prevalence of heavy drinking was significantly lower for the intervention groups (.34, .33, .40, for E1, E2, and control group). There were no significant differences for the monthly, weekly, or 3 drinks or more per occasion rates, nor were there significant differences for marijuana use. On polydrug use measures, prevalence of weekly use of all three drugs was significantly lower among intervention youth (both groups) than for the control participants. Other polydrug measures with two drugs also showed significantly lower prevalence rates among both treatment groups than in the control group, with the exception of prevalence of weekly cigarette smoking and alcohol use among the E1 group (approached significance).

6.5 year illicit drug use results - six months after 12th grade (Botvin, Griffin, Diaz, Scheier, Williams, & Epstein, 2000)
Overall rates of illicit drug use were relatively low overall at follow-up, with a few exceptions. Of the 13 drugs evaluated, three showed rates of use across the sample of over 10% or more: marijuana (48.5%), LSD or other psychedelics (15.1%), and amphetamines (10.6%). The raw proportion of students using each illicit drug was higher in the control group than in the experimental group for all drugs. The GLM analysis showed significant differences in means at follow-up favoring the treatment over the control groups on four of the six composite scores: marijuana, inhalants, heroin and other narcotics, and hallucinogens. P-values were also significant for total illicit substance use and total illicits other than marijuana. After GEE analysis, marijuana became marginally significant and inhalants was nonsignificant. Lifetime rates of illicit drug use among treatment youth were 25% lower than those from the control group (22.5 vs. 30.1); rates of hallucinogen use were 38% lower (13.0 vs. 21.0); rates of narcotics use were 56% lower (3.4 vs. 7.7).

High Fidelity Sample: The results were stronger for the high-fidelity sample. This group received at least 60% of the intervention (n=2,752). No differences were found between this group and the full sample in terms of demographic characteristics. The experimental groups were significantly different from the control group for all measures of cigarette use, weekly alcohol use, 3 drinks or more per occasion, drunk, and weekly marijuana use. Monthly marijuana use was significantly lower for the E1 group compared to the control group. Monthly alcohol use was significantly lower for the E2 group compared to the controls. Polydrug use results yielded similar findings. Treatment youth had significantly lower rates of polydrug use than control youth for nearly all combinations of the three drugs (both monthly and weekly levels of use). Both intervention groups reported 66% fewer adolescents who used all three drugs at least weekly.

Summary of Sustained Effects: Reductions in alcohol, tobacco, and marijuana use endure to the end of high school in the high-fidelity sample. In the full sample, effects are sustained for tobacco and heavy alcohol use.

Study 2

Summary

Griffin et al. (2006) used the sample and assignment from the cluster randomized trial in Study 1 but examined measures of sexual risk behaviors at a 10-year follow-up.

Griffin et al. (2006) found that, relative to the control group, the LST group had significantly

• Lower HIV risk at the 10-year follow-up.

Evaluation Methodology

This evaluation examined the long-term impact of the Life Skills Training program on drug use and sexual behaviors that put one at elevated risk for HIV infection. The data for this study were collected as part of a larger long-term follow-up study of a randomized drug abuse prevention trial.

Design: In the original school-based prevention trial, a randomized block design was used. Please see above (Study 1) for a detailed description of the study design. At the baseline assessment in 1985, surveys were completed by 5,569 participants prior to the start of the intervention, of which 3,815 (68.5%) received the prevention program. For the follow-up assessment during young adulthood, attempts were made to confirm contact information for all participants who completed the baseline survey. A total of 3,108 (56%) addresses for the original sample were confirmed by telephone records or mail requests for change of address information, while 1,519 (27%) home addresses were obtained but not confirmed by either telephone or mail. Contact information was not obtained for 17% of the original population. The follow-up survey was mailed to those for whom home addresses had been confirmed or unconfirmed and participants were offered $20.00 as an incentive to complete the survey. Of the 2,042 youth (37% of the original baseline sample) who participated in the 10-year follow-up, 1,360 (66.6%) received the prevention program and 682 were in the control group of the original prevention study. The final follow-up sample consisted of 1,080 girls and 962 boys. There were no significant pre-test differences in the young adult follow-up sample between the experimental and control groups in terms of any of the demographic variables, or in terms of rates of substance use or grades received in school in the 7th grade. At the follow-up assessment, there were no significant differences across conditions in terms of percent married or cohabitating, percent of college graduates, or percent with incomes of $15,000 per year or less. Overall attrition from the baseline to the young adult follow-up assessment was 63.3% and was similar across conditions. Those who reported smoking, drinking, or marijuana use at baseline were more likely to drop out of the study relative to those who did not report using these substances. However, the rate of attrition of substance users did not differ across experimental conditions. Males and minorities dropped out of the study at a higher rate compared to females and non-minorities, but this did not differ across experimental conditions.

Sample: Participants were primarily from middle-class suburban and rural areas of New York State, and 77.6% lived in two-parent families during junior high school. The sample was 52.8% female, and the vast majority was Caucasian (91.2%). Almost half (49.6%) were college graduates, 39.6% were married or cohabitating, and the median age was 24.6 years.

Measures:
School-based surveys: Frequency of alcohol intoxication was measured by responses to the question "How often (if ever) do you get drunk?" with response options ranging from "Never" to "More than once a day." Frequency of marijuana use was measured by the question "How often (if ever) do you usually smoke marijuana?" with response options including "Never," "Tried it but don't use it now," up to "More than once a day."

Young adult follow-up survey: HIV risk behavior outcomes included several high-risk sexual and substance use behaviors. Questions related to risky sexual behavior inquired about the number of sexual partners in the past year, the frequency of engaging in sexual intercourse while drunk or very high, and condom use. Substance use was assessed by asking participants how often (if ever) they had used any of 13 different illicit drug categories based on those used in the Monitoring the Future study, including marijuana, cocaine, amphetamines, Quaaludes, barbiturates, tranquilizers, heroin, narcotics other than heroin, inhalants, amyl or butyl nitrites, LSD PCP, and MDMA (ecstasy), with response options on a seven-point Likert-type scale. Alcohol and marijuana intoxication were assessed with items that asked about the frequency of drinking until drunk and smoking until high on a nine-item Likert-type scale. Participants were identified as high-risk substance users if they reported alcohol or marijuana intoxication or using any other illicit substance in the past month, and were considered to be engaging in high-risk behavior for HIV infection if they reported (1) having multiple sex partners, (2) having intercourse when drunk or very high, and (3) recent (past month) high-risk substance use.

Analysis: A logistic regression analysis was conducted to examine the effect of the intervention on the HIV risk index with the dependent variable being the dichotomous HIV risk index score, and demographic covariates included gender and minority status, along with lifetime smoking, lifetime alcohol use, lifetime marijuana use, percent living in two-parent families, percent receiving grades in school of C or less, and (at the young adult follow-up assessment) percent married, percent cohabitating, percent of college graduates, and percent with incomes of $15,000 per year or less. Growth modeling procedures were used to examine potential mechanisms of intervention effects.

Outcomes

Long-term: Findings indicated that the intervention had a protective effect on the HIV risk index. Covariates that were predictive of being at high HIV risk included being male and protective factors for being at high HIV risk included being married or cohabitating with a partner. This protective intervention effect remained significant after controlling for clustering within schools.

Mediating mechanisms: Prior to testing for intervention effects, a confirmatory factor analysis was conducted to examine the measurement properties of the HIV Risk Behavior latent factor, which was designed to capture covariation among three dichotomous indicator items assessing whether the participant had multiple sex partners, had sex when drunk or high, and reported high risk substance use. There were no differences across experimental conditions at the follow-up assessment in terms of the use of condoms, therefore this behavior was not included in the analyses. The percentage of the treatment group engaging in each behavior was lower than the percentage of the control group in all three cases. The confirmatory factor analysis showed that the factor loadings for these three indicators on the HIV Risk Behavior latent factor were all statistically significant. Next, prior to looking at mediating mechanisms, a direct effect model of the intervention on HIV Risk Behavior within a SEM framework was tested. Findings indicated, however, that the path from experimental condition to HIV Risk Behavior was not significant for the entire sample. Additional analyses were conducted on a subgroup of participants who had received at least 60% of the intervention during the three intervention years (n = 1,487; n = 690 men and n = 797 women). The demographic characteristics of this sample were virtually identical to those of the full follow-up sample. Among this high-fidelity follow-up subsample, the path from experimental condition to HIV Risk Behavior was significant, indicating that there was a protective effect of the intervention on HIV Risk Behavior in young adulthood, with those assigned to the intervention condition reporting less HIV risk behavior at the end of follow-up compared to control participants. Growth in serious levels of substance use involvement was measured by Alcohol and Marijuana Intoxication (AMI) during the 7th through 12th grades. The model included an AMI Slope factor to estimate growth over time and an AMI Intercept factor to estimate individual differences in alcohol and marijuana intoxication at baseline. The correlation between slope and intercept factors were estimated in the model to account for the possibility that differences in initial levels of substance use may affect rate of growth over time. A path from experimental condition to the AMI Intercept factor was also included to control for pre-test differences. One-tailed tests of significance revealed that the rate of growth in alcohol and marijuana intoxication was lower in the intervention group relative to controls. There was also a significant direct effect from the AMI slope factor to the HIV Risk Behavior factor, indicating that more growth in alcohol and marijuana intoxication during junior and senior high school was associated with greater HIV risk behavior in adulthood. In the final model, the correlation between the AMI Intercept and Slope factors was not significant, nor was the path from experimental condition to the AMI Intercept factor. Furthermore, the direct effect from experimental condition to HIV Risk Behavior dropped to nonsignificance, suggesting that the effect of the intervention on HIV risk during young adulthood was partially mediated by reduced growth in alcohol and marijuana intoxication over the course of adolescence.

Study 3

Summary

Griffin et al. (2004) used the sample and assignment from the cluster randomized trial in Study 1 but examined driving records from the Department of Motor Vehicles at the end of a six-year period. The outcome measures of risky driving included traffic violations and driver's license points.

Griffin et al. (2004) found that, relative to the control group, the LST group had

• Fewer indicators of risky driving at the six-year follow-up.

Evaluation Methodology

Design: Please see Botvin et al., 1995 above for a detailed description of the randomized block design used in the school-based drug prevention program. In order to obtain data on risky driving for the current evaluation, a list of names and addresses of students in the prevention program was provided to the New York State Department of Motor Vehicles (DMV). For students whose name and address could be matched to the DMV database, the DMV provided information on traffic violations on students' driving records. Of the over 3,500 students that participated in the long-term follow-up study, the DMV was able to provide a match for 2,042 (58%) students. The length of the follow-up period between the initial baseline data collection for the LST and the DMV data was approximately six years. The final sample of 2,042 students included 1,360 students from the treatment group and 682 control students from the original study. There were no baseline differences between experimental conditions at baseline in terms of gender or the alcohol use index.

Sample: The sample was 53% male, 91% were Caucasian, and the median age was 18.1 years. Participants were primarily from middle-class suburban and rural areas of New York State, and 86% lived in two-parent families.

Measures: Data on demographic factors, self-reported alcohol use, and experimental condition in the seventh grade were used in the present analysis. Follow-up data on antidrinking attitudes in the 10th grade and self-reported alcohol use in the 12th grade from the school-based survey were also used. Antidrinking attitudes were assessed with 10 items such as "Drinking alcohol makes you look cool" and "Drinking alcohol makes you look more grown-up," with higher scores indicating greater disagreement. Alcohol consumption was measured using three items reflecting the frequency of alcohol use, the quantity of use per drinking occasion, and the frequency of drunkenness. Participants were designated as regular alcohol users if they (1) drank alcohol in the past week, (2) reported having three or more drinks per occasion, or (3) got drunk in the past month. Data on risky driving was obtained through the state department of motor vehicles (DMV). In addition to the number of traffic violations, the number of points on students' driver's licenses was also used as outcome variables.

Analysis: A series of logistic regression analyses were conducted to examine the effect of the intervention on risky driving during high school. In each analysis, gender and alcohol use in the 12th grade were included as covariates.

Outcomes

Post-test: As this is a longitudinal follow-up, no post-test data was analyzed.

Long term: In the 12th grade, 27% of students reported drinking in the last week, 56% reported typically taking three or more drinks per occasion, and 35% reported getting drunk in the past month. In terms of risky driving, 77% of the sample had no violations and 79% had no points on their DMV record. For those with violations on their record, the mode was 3 violations (range 1-9); for those with points on their licenses, the mode was 4 points (range 2-12). Due to the skewed nature of the driving outcomes, two dichotomous scores (one indicating the presence of any violations on one's driving record and the other indicating the presence of any points) as the main outcomes in the subsequent analyses.

Results of the logistic regression analyses indicated that the intervention had a significant protective effect on risky driving. Specifically, LST had a protective effect in terms of the presence of violations on one's DMV record, controlling for gender and alcohol use. In this analysis, being male was associated with increased likelihood of violations, as was regular alcohol use. Results were similar using the presence of points as the outcome variable: LST had a protective effect in terms of the presence of points on one's license, controlling for gender and alcohol use. Being male was associated with an increased likelihood of points on one's license, as was regular alcohol use. The protective effects of the intervention on violations and points remained significant when alcohol use was not included as a covariate. Additional analyses were conducted to control for intracluster correlations (ICCs) among students within schools. When the ICCs were taken into account using the generalized estimating equations (GEE) method, the prevention effects remained statistically significant for both number of violations and number of points. Thus, findings indicate that those who received the intervention were less likely to have indicators of risky driving on their DMV records as compared to those in the control group, and these findings remained significant when school-level clustering was taken into account.

Mediational analyses on a subgroup of participants that completed survey data in the 10th grade were examined to identify potential mediators of program effects on risky driving. Findings indicated that those in the intervention group had higher antidrinking attitudes in the 10th grade compared to controls; higher antidrinking attitudes predicted significantly fewer total violations in the 12th grade; and the direct effect of the intervention on the total violations became nonsignificant with antidrinking attitudes included in the model. These findings indicate that the program effects on total violations were mediated in part by increased antidrinking attitudes among those that received the prevention program. However, the mediational model was not significant for total number of points on participants' licenses.

Study 4

Summary

Botvin et al. (2001a, 2001b) and Griffin et al. (2003) used a cluster randomized trial to examine 29 New York City public schools with 5,222 seventh-grade student participants. The schools were assigned to an LST group or a control group that received the standard curriculum. Students in the schools were followed for two years to assess self-reported substance use.

Botvin et al. (2001a, 2001b) and Griffin et al. (2003) found that, relative to the control group, the LST group reported significantly lower

• Alcohol use and polydrug use at posttest,
• Tobacco use, alcohol use, and inhalant use at the one-year follow-up,
• Binge drinking at the two-year follow-up.

Evaluation Methodology

Design: Twenty-nine New York City public schools participated in the study. Schools were surveyed to determine the amount of cigarette use of students. The schools were then divided into high, medium or low use schools and randomized to either receive the LST intervention (16 schools) or be in the control group (13 schools). Students in the intervention condition received the 15 session LST curriculum in the 7th grade, and the 10 session booster curriculum in the 8th grade. Modifications to the standard LST curriculum were made to make the program more appropriate for the targeted population. These modifications included the inclusion of pictures of minority youth, appropriate language and behavioral rehearsal scenarios, and adjustment of the reading level. No changes were made that would affect the underlying prevention strategy of the lessons. The sessions were taught by the regular classroom teacher. Control youth received the standard curriculum in place in NYC schools.

Teachers who taught the program had attended a 1-day teacher training workshop. Trained observers randomly attended classes and completed observational forms to assess implementation fidelity in each year.

Sample: The sample consisted of a total of 5,222 seventh grade students. Demographic characteristics of the sample included approximately half (47%) male and predominantly minority (61% African-American, 22% Hispanic, 6% Asian, 6% White, and 5% other or mixed background). Sixty-two percent of participants were eligible for the free lunch program. Approximately half (53%) of students lived in dual-parent households and 36% lived in mother-only households. After the schools were randomized into groups, 69% of the sample (n=3,621) were in the treatment condition.

Measures: Participants were surveyed prior to treatment (pretest), 3 months after the first year of the intervention, and at one-year follow-up after the initial posttest at the end of the 8th grade. Students completed questionnaires of self-reported drug use behavior. Questionnaires were administered by a team of data collectors who were members of the same ethnic groups as participating students. Prior to implementation, breath samples (to measure carbon monoxide levels) were collected from all participants, in order to increase validity of self-report data collected. Two measures of cigarette smoking were conducted: frequency (9-item scale) and quantity (11-item scale). Two measures of alcohol consumption were conducted: frequency (9-item scale) and amount consumed per drinking occasion (6-item scale). Frequencies of smoking marijuana, getting "high" from marijuana, and using inhalants were also measured by using a 9-item scale. Polydrug use measures were based on the responses to single drug use frequencies and examined in terms of lifetime use and current use. Also included in the analysis of this study were assessments of behavioral intentions, drug attitudes and knowledge, and social and personal competence.

Analysis: Generalized linear models (GLM) ANCOVA were used to compare means of drug use between the conditions at posttest and follow-up. Since the intervention was administered at the school level, researchers also controlled for intra-cluster correlations (ICCs) among students within the schools by using generalized estimating equations (GEE) to adjust the estimated standard error. Regression analyses were conducted to determine the effects of mediating factors (knowledge, intention, normative expectations). Effects on binge drinking at the one- and two-year follow-up assessments were tested.

Outcomes

Baseline Equivalence and Differential Attrition: Pretest analyses indicated no significant differences between conditions for any of the substance-use variables. There were a few demographic differences between the groups in terms of race and proportion of students who received free lunch. These variables were controlled for in the regression analyses for program effects. Analyses conducted to determine differential attrition at posttest revealed that participants who reported substance use at pretest were more likely to not be included in the posttest measures. This resulted in a more conservative test of the program.

Fidelity Monitoring: The mean level of implementation fidelity was 48%, which suggests a lower level of fidelity than had been found in previous research.

Posttest: Posttest data were collected three months after intervention. Analysis revealed significant effects on each of the alcohol use measures (frequency, drunkenness, and drinking quantity) as well as lifetime polydrug use, compared with the control condition. When the ICCs were included in the more conservative GEE analysis, the p-value for drinking frequency became nonsignificant and polydrug use approached significance. Prevention effects included increased drinking knowledge, and normative expectations for smoking and drinking.

One-year: GLM analysis indicates significant effects after one year on measures of all drugs, tobacco (frequency and quantity), alcohol (frequency, getting drunk, quantity), and marijuana (frequency, getting high), as well as inhalant use, as compared to the control condition. When GEE analysis was conducted, p-values for both marijuana variables became nonsignificant, while all other variables retained significant effects.

One- and two-year effects on binge drinking: The prevention program had protective effects in terms of binge drinking at the 1-year (8th grade) and 2-year (9th grade) follow-up assessments. The proportion of binge drinkers was over 50% lower in the intervention group relative to the control group at the follow-up assessments. There were also several significant program effects on proximal drinking variables, including drinking knowledge, pro-drinking attitudes, and peer drinking norms.

High-Risk Subsample: Griffin et al. (2003) examined 21% of the original sample that was classified as high risk for substance use initiation based on having poor grades and peers who used substances (n = 802). Tests at one-year follow-up for the subsample controlled for pretest scores and used generalized estimating equations to adjust for school clustering. The text reported one-tailed probability levels but provided information to infer two-tailed significance levels. The (two-tailed) results showed that the LST group had significantly lower substance use means than the control group for smoking (d = .22), drinking (d = .22), and polydrug use (d = .21). In addition, the study reported no significant baseline differences between conditions and no differences across conditions in the determinants of attrition.

Study 5

Summary

Zollinger et al. (2003) used a quasi-experimental design to examine 16 Indiana middle schools, 12 of which self-selected to implement LST. Sixth-grade students (n = 1,598) were followed through eighth grade to assess cigarette use and attitudes and knowledge about cigarette use.

Zollinger et al. (2003) found that, relative to the control group, the LST group reported significantly lower

• Smoking prevalence
• Smoking knowledge and self-efficacy.

Evaluation Methodology

Design: Sixteen middle schools from the Indianapolis Public School (IPS) system were included in the study. The Life Skills Training curriculum was implemented in 12 of the 16 schools. Intervention schools were not randomly chosen, but the specifics surrounding their selection were not discussed in the study. Self-report surveys were administered by the IPS system in December of each study year (1997-2000) for all middle school students. The survey used a repeated panel design including those who were and those who were not exposed to the program. Survey administrators were instructed on guidelines and protocols for administering the survey, including issues regarding confidentiality, consistency, and logistics. Baseline data were collected in 1997 on sixth-grade students. In 1998, sixth- and seventh-grade students were surveyed, and in 1999 and 2000, sixth-, seventh-, and eighth-grade students in middle school were surveyed. Student identification numbers assigned by IPS were used to match the surveys completed by students in the sixth, seventh, and eighth grades. Two cohort groups were identified: those who participated in all of the 1997, 1998, and 1999 surveys, and those who participated in all of the 1998, 1999, and 2000 surveys. For this analysis, the two cohorts were combined (n = 1,598).

Responses to tobacco use items were used to classify students as frequent smokers, current smokers, those who tried smoking cigarettes, and non-smokers. Current smokers had smoked in the past 30 days. Non-smoking students had never smoked a cigarette, not even a puff or two. All other students were classified as having tried smoking.

Sample: Students enrolled in the IPS system from 1997-2000 (n = 27,865) were included in the study. About one-fourth (28.9%, n = 8,048) of students declined to participate or were not available when the survey was administered. A total of 610 surveys (2.2%) were excluded because students did not answer at least half of the questions, or staffing judged students did not complete the survey with true or serious responses. Average response rate used for analysis was 68.9% for the four surveys. A final total of 1,598 eighth-grade students completed the Youth Tobacco Survey while in the sixth, seventh, and eighth grades. Approximately 56% of the participants were female, 59% were African American, and 31% were White. Although the intervention schools were not randomly chosen, tobacco-related behavior and attitudes of these students at baseline did not differ significantly from nonintervention schools.

Measures: A self-administered survey collected data about middle school students' knowledge, attitudes, beliefs, self-efficacy, decision-making ability, and behavior toward tobacco use and related issues. Items for the survey instrument were derived from published instruments including the CDC Youth Risk Behavior Survey Questionnaire, 1993, CDC Behavioral Risk Factor Surveillance System Questionnaire, 1997, Health Survey for England-the Booklet for 13-15 year-olds, 1996, Alcohol and Other Drug Use Survey, Indiana Prevention Resource Center, 1995, Maryland Adolescent Survey, 1994, and Statewide Survey of Drug and Alcohol Use Among California Students, grades 7, 9, and 11, 1986. Other items were developed specifically for the survey.

Analysis: Completed surveys were compiled and verified. Responses were compared using the z -test for proportions to determine statistical significance. Results were presented for students with no exposure to the LST program (26.9%), exposure during one school year (32.9%), and exposure during two school years (40.2%).

Outcomes

Although survey data were collected annually from 1997-2000, no post-test analyses were conducted immediately after program implementation. Data was analyzed after the completion of the final round of data collection.

Smoking behaviors:

Current smokers: Overall, 12.5% of the participants were currently smoking, and 39.4% had tried smoking in the past. Roughly one-half (48.2%) had never tried smoking, not even a puff or two. Significantly fewer current smokers existed among those who completed the LST curriculum once (one year) or twice (two years) (10.5% and 10.3%) compared to those with no exposure (18.1%). There were significantly more non-smokers in the group exposed to LST at least one time. No differences existed between one and two exposures to LST in any of the smoking behavior categories. Significantly fewer White students exposed twice to the LST curriculum were currently smoking, compared to those not exposed. Significantly more students of both genders and racial groups exposed to LST indicated they did not hang out with friends who smoke cigarettes.

Tried smoking: No significant impact on students who had tried smoking.

Non-smokers: There were significantly more non-smokers with one or two years exposure to the program compared to those with no exposure. This was also true in the subgroups of males, females, and Whites.

Intentions to smoke: When non-smokers were asked about their intention to try smoking in the next 12 months, 83.7% indicated they would not do so. Significantly more males compared to females and more African American students compared to White students indicated they would not try smoking in the next 12 months. Significantly larger proportions of female students, White students, and all students exposed twice to the program indicated they did not think they would try smoking, compared to the no-exposure groups.

The program had no significant impact on quit-attempt rates among smokers.

Self-efficacy: Significantly more students exposed to LST indicated it would not be difficult to refuse an offer of a cigarette. Significantly more female students twice exposed to the curriculum reported it would not be difficult to say "no." Significantly fewer female students with more than one exposure to the program reported that their decision to smoke was affected by friends' smoking behaviors compared to students with no exposure. Significantly fewer African American students with exposure to the program were affected by their friends' smoking.

Attitudes: Significantly more students exposed to LST once or twice thought it was a good idea to pass laws against smoking in schools and other public buildings, compared to those with no exposure. More students in each gender group exposed to the curriculum thought it was a good idea to pass laws restricting smoking.

Knowledge: Although the vast majority (90%) of students knew cigarette smoking caused damage to the lungs, makes teeth look bad, and causes lung cancer and bad breath, significantly more students exposed to LST one or two times knew smoking caused damage to the heart, eyes, unborn babies, cancer of the mouth and lungs, strokes to the brain, and damage to the ears. These results were consistent across gender and racial variables.

Summary of effects: The LST curriculum positively impacted tobacco use and attitudes of IPS middle school students. Exposure to LST one or two times was associated with a reduction in the prevalence of youth smoking as well as positive shifts in self-efficacy, attitudes and knowledge. Most improvements occurred with one exposure, although some required two exposures.

Study 6

Summary

Botvin et al. (1997) used a cluster randomized trial to examine seven junior high schools in New York City with 833 student participants. The schools were assigned to an LST group or a control group that received the standard curriculum. Students in the schools were assessed on substance use at baseline and posttest.

Botvin et al. (1997) found that, relative to the control group, the LST group reported significantly lower

• Cigarette smoking at posttest
• Drinking at posttest
• Marijuana use at posttest
• Polydrug use at posttest.

Evaluation Methodology

Design: Seven junior high schools in New York City participated in the study. Assignment was at the school-level to either the Life Skills Training treatment condition or the standard care control condition. Treatment condition students received the 15-session Life Skills Training program. The program was revised to be more appropriate for the targeted population. Modifications included adjusting the reading level, illustrative examples, and suggested situations for behavioral exercises. Treatment teachers attended a one-day training workshop.

Sample Characteristics: There were 833 participating students at pretest. Of these, 721 (87%) also completed posttest measures. The majority of students were girls (53%) and the mean age of the students was 12.6 years. The ethnic-racial composition of the sample was 25.8% African-American, 69.6% Hispanic, .7% White, 1.4% Asian, 1.5% Native American, and 1.0% Other. Most of the sample lived with their mother-only (37.3%) or both parents (35%). The majority of students (78.6%) qualified for free or reduced lunch.

Measures: Measures were collected at pretest and at post-intervention (about three months after pretest). Students completed two measures: a questionnaire and a carbon monoxide (CO) breath sample to enhance the validity of the self-reported data. Students completed the questionnaire during class and answered questions about current drug use (5 behavioral measures: smoking, drinking, drinking amount, drunkenness, marijuana use; 2 multiple substance measures: ever use and current use) and intentions for drug use in the future. Also assessed were behavioral intention (for drug use), normative expectations, attitudes towards drug use, and social competence (decision making, advertising influences, anxiety reduction, and communication).

Analysis: General Linear Modeling was used for the analysis, and used only data provided by students who completed both the pre- and posttest. One-tailed significance tests were used. To examine the impact of mediating variables, ANCOVAs were used on measures of attitudes, normative expectations, and skills use.

Outcomes

Baseline Equivalence: Crosstabs were performed to determine pretest equivalence of the demographic variables by condition. There were no significant differences between conditions for gender, free lunch, or family structure. There were differences between conditions on race/ethnicity, with a lower proportion of Hispanic students and a higher proportion of African-American students in the control condition compared to the treatment condition. However, race/ethnicity was not related to any of the pretest drug use variables, meaning conditions were comparable at pretest.

Posttest: There were significant treatment effects found on all five individual drug use behavior variables and both multiple drug use measures, indicating that students in the treatment condition reported using all measured substances less often than students in the control condition. Significant treatment effects were also found on intentions to use for three of six measures (cigarettes, beer/wine, and marijuana). On the mediating variables, significant differences were found on all but one of the normative expectations variables, indicating that the intervention resulted in lower normative expectations for treatment students concerning various drugs, compared to students in the control group. Refusal assertiveness (under skills use measures) was also found to be a significant mediating variable.

Study 7

Summary

Botvin et al. (2006) used a cluster randomized trial to examine 41 New York City public and parochial schools with 4,858 sixth-grade student participants. The schools were randomly assigned to an LST group or a control group that received the standard curriculum. Students in the schools were assessed on measures of verbal and physical aggression, fighting, and delinquency at pretest and posttest.

Botvin et al. (2006) found that, relative to the control group, the LST group reported significantly lower

• Delinquency at posttest
• Frequent fighting at posttest.

Evaluation Methodology

Design: The sample consisted of 4,858 sixth grade students from 41 randomly assigned NYC public and parochial schools. There were 20 LST schools (n=2,374) and 21 control schools (n=2,484) who received the standard health eduction curriculum. Pretest assessment was in the sixth grade and posttest approximately three months later, after the LST students had received the first year of the curriculum.

Sample: The sample was 51% boys and 49% girls; 39% African American, 33% Hispanic, 10% White, 6% Asian, 2% Native American, and 10% of other or mixed ethnicity.

Measures: Measures assessed verbal and physical aggression, fighting, and delinquency.

Analysis: Analyses were run using generalized estimating equations to account for the within-cluster correlation. Pretest values of each outcome, plus gender, percent black, percent Hispanic, grades, and implementation score were used as covariates in the models.

Outcomes

Baseline Equivalence: The two conditions did not differ at baseline on any of the violence or delinquency scales, or on gender or academic performance. The intervention group had more Hispanic students (36.7%) than controls (30.2%), while the control group had more black students (43.8%) than the intervention group (33.1%).

Posttest Results:

In the full sample, the intervention reduced "any delinquency in the past year," "frequent fighting in the past year," and "frequent delinquency in the past year." Five other variables were not significantly different including past month and high frequency verbal and physical aggression, as well as fighting in the past year.

In the high fidelity sample, with students who received at least half of the LST intervention, there were significant prevention effects on Physical Aggression in the Past Month, Violence in the Past Year, and Delinquency in the Past Year. There were also significant effects for the high frequency of outcome behaviors (top quartile), including Frequent Verbal Aggression in the Past Month, Frequent Physical Aggression in the Past Month, Frequent Fighting in the Past Year, and Frequent Delinquency in the Past Year.

Study 8

Summary

Mackillop et al. (2006) used a pre-post design without a control group to examine two school districts in upstate New York that implemented LST.

Mackillop et al. (2006) examined change in schools implementing LST without a control group and found no significant improvement in substance use.

Evaluation Methodology

Design: An experimental pre- post-test design was used in this evaluation. Two school districts in suburban towns in Upstate New York were selected in an unspecified manner to implement LST to groups of 6th graders (approximately 11 years of age). The towns from which the school districts were drawn maintained populations of approximately 40,000 and 17,000, respectively, and were primarily Caucasian (92% in School District One and 96% in School District Two) with small percentages of ethnic minorities present. Prior to the study, parents of children enrolled in the sixth grade in participating districts were informed by letter that a new substance use prevention program was being implemented as part of the educational curriculum and would be evaluated on two occasions. Parents were given the option of refusing to allow their child to complete the assessments, but all children in classes where LST was implemented would receive the program as a part of the standard curriculum. No parents withdrew consent for their children to participate in the evaluation. The LST curriculum was delivered once per week for 15 weeks in School District One, while School District Two delivered the curriculum once per day in 15 consecutive class periods. In School District One, six students (7%) provided data for only one time point and were not included in the study; in School District Two, seven students (4%) provided data for only one time point and were not included in the study.

Sample: The sample for School District One was 55% male, 89% Caucasian, 4% African American, 0% Hispanic, 4% Native American, 1% Asian, and 0% Other. The sample for School District Two was 54% male, 83% Caucasian, 2% African American, 2% Hispanic, 9% Native American, 2% Asian, and 1% Other. An unusually large percentage of respondents indicated they were of Native American heritage, which is possible, but may also indicate a misunderstanding of the category Native American meaning "born in America."

Measures: The Life Skills Training Questionnaire (LSTQ) was used in both school districts. The LSTQ assesses a number of domains related to the LST program. The 90-item, 7 subscale Alcohol Expectancy Questionnaire-Adolescent Version (AEQ-A) was used to evaluate adolescents' outcome expectancies for drinking alcohol. Due to school district administrators' concerns, subscale 7, Sexual Enhancement, was not administered. The 36-item Self-Perception Profile for Children (SPPC) was used to measure perceived self-competence in children. Due to class period time constrictions, only one additional measure could be included with the LSTQ; therefore, students in School District One were administered the LSTQ and AEQ-A, while the students at School District Two were administered the LSTQ and the SPPC. Fidelity was assessed using two approaches. First, at the end of each lesson, teachers completed a Life Skills Training Implementation Checklist (LST-IC), a checklist for teachers and independent observers to assess two aspects of LST lesson adherence: objectives and topics/activities. Second, on 16 occasions independent observers rated fidelity using the LST-IC.

Analysis: Individual within subjects analyses of variance (ANOVAs) were conducted for each school district. In order to reduce skewness and kurtosis, in School District One inverse transformations were used on the pro-attitude toward substance use subscales and perceived peer substance use subscale, and a square root transformation was used on the second drug refusal skills subscale. In School District Two transformations were used in the substance use and intention to use scales. In both school districts, the substance use and intention to use scales were severely skewed due to low rates of substance use behavior, and transformations did not improve skewness. As a result, the data were recoded dichotomously and the McNemar test was used for these variables. For all variables, analyses included participants who provided valid pre-intervention and post-intervention data. Potential treatment-by-gender interactions were examined using 2 (male/female) X 2 (pre-test/post-test) mixed ANOVAs.

Outcomes

Implementation Fidelity: In School District One, the teacher presented the 15 LST lessons to six classes for a total of 90 class periods. A total of 48 of 90 (53%) LST-ICs were completed, which revealed that the mean proportion of objectives completed was 95% and that the mean topics/activities completed was also 95%. Six lessons were independently observed with the mean proportion of objectives completed rated at 99% and the mean proportion of topics/activities completed was rated at 100% by the independent observer. Mean student daily attendance was 94%.

In School District Two, the teacher also implemented LST for six classes and completed LST-ICs following all 90 lessons with a mean proportion of objectives completed of 93% and a mean proportion of topics/activities of 80%. Independent observation of 12 lessons resulted in a mean rating of 99% of objectives completed and 73% of topics/activities completed. Mean student daily attendance was 93%.

LST Outcomes:

School District One: At post-test there were statistically significant changes in the Overall Knowledge, Life Skills Knowledge, Drug Knowledge, Perceived Adult Substance Use, Pro-Smoking Attitudes, Pro-Drinking Attitudes, Assertiveness Skills, and Anxiety Reduction Skills subscales. All of these changes were in the hypothesized direction, reflecting positive prevention effects. The McNemar test revealed no changes in Use or Intention to Use from pre- to post-test for either specific substances or aggregate estimates.

School District Two: At post-test there were statistically significant changes on the Overall Knowledge, Life Skills Knowledge, Drug Knowledge, Assertiveness Skills, Self-control Skills, Perceived Adult Substance Use, and Perceived Peer Substance Use subscales. All changes were in the hypothesized direction, with two exceptions: A significant effect on the Perceived Peer Substance Use subscale reflected an increase in perceived prevalence and a significant effect on the Drug Refusal Skills II subscale reflected a decrease in self-reported drug refusal skills. As was the case for School District One, the McNemar tests detected no changes from pre- to post-test on the substance use or intention to use for either specific substances or aggregate estimates.

Effects by Gender: In School District One, no main effects or interactions were evident between gender and LST, with one exception: A main effect for anxiety reduction skills indicated that females generally reported greater anxiety reduction skills than males, regardless of the intervention. In School District Two, gender interaction effects were found on the Drug Knowledge subscale, with females learning more about drugs than males, and on the Anxiety Reduction Skills subscale, with females again exhibiting an improvement in anxiety reduction skills while males actually reported a decrease in self-reported anxiety skills. The analyses also revealed main effects of gender on the Overall Knowledge subscale and Life Skills Knowledge subscale, with both cases reflecting poorer performance in males.

Study 9

Summary

Spoth et al. (2002, 2006, 2008, 2008b, 2014) and Trudeau et al. (2003) used a cluster randomized trial that assigned 36 middle schools with 1,664 seventh-grade students to three conditions: LST, LST plus Strengthening Families 10-14, or a control group. Assessments of substance use continued through age 22.

Spoth et al. (2002, 2006, 2008, 2008b, 2014) and Trudeau et al. (2003) found that, relative to the control group, the LST group reported significantly

• Lower substance use initiation at 1.5 years after baseline
• Lower cigarette initiation at 5.5 years after baseline
• Slower growth rates in cigarette initiation and drunkenness at 5.5 years after baseline.

Evaluation Methodology

Design: Participants in the study were seventh graders enrolled in 36 randomly selected rural schools in 22 contiguous counties in a Midwestern state. Criteria for the selection of the initial pool of schools were: 20% or more of households in the school district within 185% of the federal poverty level; community size (school district enrollment under 1,200); and all middle-school grades (6-8) taught at one location. A randomized block design guided the assignment of the 36 schools to one of three experimental conditions: 1) a combined Life Skills Training (LST) and Strengthening Families Program 10-14 (SFP 10-14) group, 2) an LST only group, and 3) a no-treatment control group.

Students in the combined LST and SFP 10-14 group received both curricula, including booster sessions (n = 4 booster sessions for the SFP 10-14 and n = 5 booster sessions for LST), while students in the LST-only group received the LST curriculum including 5 booster sessions. For a detailed description of the SFP 10-14 program, see the complete write-up. After schools were matched and randomly assigned to conditions, school officials were contacted and informed of the experimental condition to which their school had been assigned. All seventh grade students in participating schools were recruited for participation. On average, 46 students in each school completed the pre-test (n = 1,664 total), with 549 in the combined LST and SFP 10-14 group, 621 in the LST-only group, and 494 in the control group. A total of 1,563 students completed the post-test (n = 517 in the combined LST and SFP 10-14 group, n = 583 in the LST- only group, and n = 463 in the control group), while 1,372 students completed the long-term follow-up (n = 453 in the LST and SFP 10-14 group, n = 503 in the LST only group, and n = 416 in the control group).

Trudeau et al. (2003) compared trends in outcomes between the LST-only and control groups. For this analysis, only students with no missing data and who had not changed schools were included. This resulted in a sample of 847 students in 24 schools and completion rates of 86-90% at the last assessment.

Data collection in the form of student surveys was completed in classrooms at pre-test, post-test (one month after completion of the intervention), and at the long-term follow-up (one year after completion of the intervention). In addition, a bogus pipeline procedure was performed in order to promote honesty in answering smoking related questions. The sample was analyzed for pre-test equivalence on sociodemographic and outcome measures; the only significant difference discovered was that the control group contained more dual-parent families than the two intervention groups. This variable was included as a control variable in the subsequent outcome analyses. Analysis of differential attrition revealed no significant dropout by condition interactions from pre- to post-test or from post-test to follow-up for any outcome or sociodemographic variable.

Spoth et al. (2014) presented results for 11 assessments overall, with four of the assessments coming after high school, at ages 19, 20, 21, and 22. The authors noted that their study eliminated 18 students "who changed conditions (i.e., moved from a school district in one condition into one in a different condition)." According to Figure 1, about 72% of the baseline students provided data at age 22. Of students eligible for the young adult follow-ups - those who participated in the 11th or 12th grade assessments - 84% provided data at age 22.  Overall, those who remained in the study had a lower level of substance use at pretest than those who dropped out. However, the authors reported only one instance of differential attrition: a lower rate of attrition among control group participants from dual biological parent families at the 19-year-old assessment point. "No other significant pretest or differential attrition effects were found" (p. 951).

Sample: Participants included all seventh grade students at these schools, who were recruited to participate. On average, 46 students per school participated in the pre-test, slightly over half (53%) were male, and the majority of participants (96%) were Caucasian. Analysis of the demographic characteristics of the groups at baseline revealed one difference: the control group contained more youth with dual-parent families, thus lowering their level of risk. This variable was included in the outcome analysis as a control variable. At post-test and follow-up, analyses were conducted to rule out differential attrition in the sample by examining Condition X Dropout Status interactions. No significant interactions were found at either follow-up, for any outcome or sociodemographic measure.

In a longitudinal analysis (Trudeau et al., 2003), the sample was limited to two conditions and students with no missing data across three time-points and who had not changed schools. The resulting sample was 52% male and 97% Caucasian. 77% were living with biological parents, and 21% qualified for free or reduced-price lunch.

Measures: Self-reported use of alcohol, cigarettes, marijuana, and medications not prescribed to the participant was obtained from the classroom-administered questionnaire. Individual items included (a) "Have you ever had a drink of alcohol?", (b) "Have you ever smoked a cigarette?", (c) "Have you ever smoked marijuana or hashish?", (d) "Have you ever used drugs or medications that were prescribed by a doctor to someone else?" All four items were answered using a yes/no format. Inconsistent reports in lifetime substance use were corrected. Lifetime use measures were adjusted to control for baseline use, with these adjusted lifetime use measures (new-user rates) indicating whether use was initiated since baseline. Three lifetime use items were individually examined and summed to form the substance initiation index (SII). Prior to implementation, breath samples (to measure carbon monoxide levels) were collected from all participants, in order to increase validity of self-report data collected.

Trudeau et al. (2003) also examined trends in three outcomes of interest. 1) Substance use expectancies were defined as negative attitudes towards substance use due to personal and social consequences. The variable was created by averaging five Likert-style attitudinal items (alpha = .85). 2) Refusal intentions measured students' expectation of refusing substances. The Likert-style items used for this variable came from two scales and had an alpha of .80. 3) Substance initiation measured whether students had already used alcohol, tobacco, and/or marijuana. Three dichotomous yes/no variables were summed to create an initiation scale. Since data were collected at three time-points, some answers were corrected for consistency. In other words, a student could not respond "no" if they had previously responded "yes."

Analysis: A multilevel (mixed model) analysis of covariance (ANCOVA) using SAS Proc Mixed with restricted maximum likelihood estimation and listwise deletion of missing data was used to test for intervention effects on the SII. Because assignment to treatment conditions was made at the school level, school was incorporated as a random effect in the analyses. In addition, new-user analyses for specific substances were conducted at the school level, based on the proportions of new users in each school. For the prescription drug misuse outcome (Spoth et al., 2008b), Fisher's exact test was used due to the small numbers of participants responding in the affirmative and lack of within-school dependence. The tests did not adjust for randomization of schools.

Trudeau et al. (2003) used latent growth and structural equation models to examine differences in trends between the LST-only and control conditions. Due to between-school variability and the need to adjust for within-school clustering, corrected standard errors and chi-square tests of model fit were calculated using restricted maximum likelihood estimation. The growth models incorporated baseline outcomes in the intercept and slope estimates. Consistent with an intent-to-treat analysis, the study used all students with data, although no attempt was made to follow students moving to other schools.

Spoth et al. (2014) used latent growth models with adjustments for clustering within schools and full information maximum likelihood estimation to include all available data. Rather than examining condition differences in outcomes during young adulthood, the models estimated the indirect effects of the interventions on substance use outcomes via adolescent initiation. That is, the intervention affected adolescent initiation of substance use, which in turn affected young adult substance use.

Outcomes

1.5 Years After Baseline (Spoth, Redmond, Trudeau, and Shin, 2002)

Because only initiation measures were applied in the outcome analysis, the post-test was considered to be the baseline time point (since the analyses examined differences in substance initiation after delivery of the interventions), and the pretest was delivered several months before the intervention was delivered. Results are reported at one year after the intervention posttest (1-1/2 years after baseline).

The substance initiation index (SII) score was lowest for the LST and SFP 10-14 combined condition, while the LST-only group had the next lowest SII score and the control group had the highest SII score. Adjusting for the one-tailed tests, the LST and SFP 10-14 combined group scored significantly lower on the SII than the control group, but the difference between the combined group and the LST only group was non-significant. The LST-only group was marginally significantly lower on the SII than the control group.

New User Rates: The LST and SFP 10-14 combined condition demonstrated the lowest new user rate for alcohol and marijuana compared to the LST only and control groups. The relative reduction rate (the percentage difference in the proportion of new users in the intervention group relative to the control group) for the combined condition was 30% for alcohol initiation, while the same rate for the LST only condition was 4.1%. There were no significant findings associated with cigarette initiation. With regard to the contrast of LST and the control group, marijuana new users was marginally significantly lower in the LST group, but the contrasts with new users of alcohol and cigarettes were not significant.

1.5 Year Trends (Trudeau, Spoth, Lillehoj, Redmond, & Wickrama, 2003)

Using data from three time-points (baseline, the end of the first year, and the end of the second year following booster lessons), researchers compared the LST-only and control groups in terms of their trends in three outcomes of interest. 1) The decline over time in negative expectancies towards substance was smaller for the treatment group than the control group, but the difference was marginally significant at p < 0.07. 2) Intention to refuse substances was also expected to decrease over time and the decrease was significantly (p < 0.01) smaller for the treatment than the control. 3) Substance use initiation was expected to increase over time, and the researchers found that this increase was significantly (p < 0.01) smaller for the treatment group than the control. Additional moderation tests found stronger intervention effects for girls than boys.

5.5 Years After Baseline (12th Grade Outcomes):

Spoth, Randall, Trudeau, Shin, Redmond, 2008

Using multilevel analysis of covariance (HLM with students nested within schools) with 428 12th grade LST-only students and 347 Control students, adjusting for the one-tailed tests, the index of substance use initiation (which includes alcohol, marijuana, and cigarette use) was significantly lower for LST-only vs. Controls. The growth trajectory was marginally significant. Examining the individual initiation measures showed that the LST-only group had significantly lower mean levels of cigarette initiation and marginally significant lower levels of marijuana initiation. The LST-only group also showed a significantly slower rate of increase across time for cigarette initiation and drunkenness initiation.

There were no significant effects found for either the 12th grade mean levels or for the growth trajectories for any of the more serious substance use outcomes (frequency of use, monthly poly-substance use, and advanced poly-substance use index). However, there were significant effects on all measures, with the exception of drunkenness frequency, for a higher-risk subgroup defined as those students who reported use of at least two of three substances (alcohol, cigarettes, and marijuana) at pretest.

11th and 12th Grade Methamphetamine Use (Spoth, Clair, Shin, Redmond, 2006): There was no significant difference in methamphetamine use between LST-only and control conditions at the 11th grade follow-up. At 5.5 years (12th grade), there was a significant difference between the LST-only and control groups in lifetime methamphetamine use. The LST + SFP group was lower in lifetime methamphetamine use at both 4.5 and 5.5 years, and in past year methamphetamine use at 4.5 years.

11th and 12th Grade Prescription Drug Misuse (Spoth, Trudeau, Shin, Redmond, 2008): There was no significant difference in prescription drug misuse between LST-only and control conditions at either the 11th or 12th grade follow-ups. The LST + SFP group was significantly lower in reported prescription drug misuse in the 11th grade, and this difference was marginally significant in the 12th grade.

9.5 Years After Baseline (Outcomes at Ages 19, 20, 21, and 22)

Spoth et al. (2014)

The latent growth models examined the indirect effects of LST versus the control group through adolescent initiation of substance use. The indirect effects produced lower levels in the LST group than the control group on four outcomes at ages 19-22: drunkenness frequency, alcohol-related problems during the past year, cigarette use frequency, and illicit drug use frequency. Only one of the outcomes, alcohol-related problems, showed an indirect effect of LST on the slope or rate of change, suggesting that the LST advantage declined over the young adult years. With controls for the indirect effects, LST had little direct influence on the outcomes. The results support arguments that the program influences adult substance use by delaying adolescent imitation, which then moderates substance use in young adulthood.

Study 10

Summary

Botvin & Eng (1980) and Botvin et al. (1980) used a cluster randomized trial to examine 281 students in grades 8-10 who were attending two New York City schools. The two schools were randomly assigned to an LST or control group. Student participants were assessed on smoking, smoking knowledge, and social relations at baseline, posttest, and three-month follow-up.

Botvin & Eng (1980) and Botvin et al. (1980) found that, relative to the control group, the LST group reported significantly

• Fewer smokers
• Higher smoking knowledge.

Evaluation Methodology

Design: A sample of 281 students was drawn from a population of eighth, ninth, and tenth grade science and health education students in two suburban New York City schools in an unspecified manner. Both schools (School A and School B) were generally comparable with respect to socioeconomic status and the prevalence of cigarette smoking and were predominantly middle class. The two schools were randomly assigned to either the experimental (n = 121) or the control (n = 160) condition. All participants were pre- and post-tested by questionnaire with respect to self-reported smoking status, smoking knowledge, psychosocial knowledge, locus of control, self-image, social anxiety, influenceability, and the need for group acceptance. Following the pre-test, students in the experimental condition participated in a 10-session smoking prevention program. Although sessions were conducted weekly, the time interval between the beginning and the end of the program was 12 weeks due to school holidays. The program was administered by an outside specialist and utilized a combination of group discussion and special skills training. Sessions included content on self-image, decision making, advertising techniques, coping with anxiety, communication skills, social skills, and assertiveness training. In addition to the material covered in each session, students were given outside assignments either to prepare them for specific sessions or to reinforce material already covered. All students participated in a Self-Improvement Project in which they worked over the course of the 10-week program toward improving some skill or toward changing some specific personal behavior. Self-improvement goals were broken down into a series of weekly subgoals in order to enable students to gradually shape their own behavior and to chart their weekly progress. Students in both groups completed two post-tests. The first post-test was administered at the completion of the smoking prevention program (12 weeks after the pre-test), and the second post-test was administered approximately three months later. Data for the three month post-test was collected on roughly 77% of the immediate post-test sample (80% for the experimental group and 74% for the control group).

Sample: No specific information regarding the gender or racial composition of the sample was provided.

Measures: The questionnaire consisted of 58-items (excluding basic demographic data) and was divided into 3 sections: questions relating to smoking behavior (10 items), knowledge questions (20 items), and questions designed to tap various psychological variables (28 items).

Analysis: All pre-test smokers were eliminated from the analysis of smoking status, permitting the comparison of the experimental and control groups in terms of the number of new smokers. A Chi-square analysis was performed to test between group differences in the number of new smokers in the experimental and control groups. A two-way analysis of variance (sex X treatment condition) was used to compare the between-group differences in knowledge and personality scores from pre- to post-test. These between-group comparisons were performed both for the total sample and for each of the three grade levels within the total sample.

Outcomes

Post-test: Botvin, Eng, and Williams, 1980
Significantly fewer students in the experimental group began smoking during the course of the study when compared to students in the control group. The LST smoking prevention program was not equally effective for all grade levels, however. LST was most effective (100%) in preventing the onset of smoking among eighth graders, less effective (75%) among the ninth graders, and the least effective (44%) among the tenth graders. Overall, the experimental group had a significantly greater increase in smoking knowledge than the control group, and there was a significant two-way interaction between sex and treatment condition for social anxiety, with the males in the experimental condition showing the greatest decrease in social anxiety.

Long-term: Botvin and Eng, 1980
At the three-month follow-up, there were still significantly fewer new smokers in the experimental group compared to the control group, although the percentage of students beginning to experiment with cigarettes increased between the immediate and three-month post-tests in both groups. For the eighth graders, there was a significantly greater decrease in the need for group acceptance among the students in the experimental group compared to the control group, with males in the experimental group showing the greatest decrease in social anxiety. For the ninth graders, there was a significantly greater increase in smoking knowledge among students in the experimental group than among the controls as well as a significantly greater decrease in the need for group acceptance. Among tenth graders, as with the eighth graders, there was a significant interaction between sex and treatment condition for social anxiety, with the males showing the greatest decrease.

Overall, the experimental group had a significantly greater increase in smoking knowledge between the immediate and three-month posttest than did the control group, with males in the experimental condition again showing the greatest decrease in social anxiety. Girls in the experimental group maintained a significantly greater decrease in identification with their peers. Among the eighth graders, the students in the experimental group had a significantly greater decrease in social anxiety compared to the control group. Similarly, there was a significantly greater decrease in the need for group acceptance and peer identification. For the ninth graders, the only significant difference between the experimental and control groups was where students in the experimental group had a greater decrease in the influenceability between the immediate and the three-month follow-up compared to the control group. Finally, among the tenth graders there was a significant two-way interaction between sex and condition for smoking knowledge.

Study 11

Summary

Botvin et al. (1992) used a cluster randomized trial to examine 47 New York City schools that were randomly assigned to an LST group (n = 1,795 students) or control group (n = 1,358 students). Student participants were assessed on smoking, smoking knowledge, and psychological well-being at baseline and posttest.

Botvin et al. (1992) found that, relative to the control group, the LST group reported significantly

• Lower past month smoking and smoking onset at posttest
• Higher smoking knowledge at posttest
• Lower normative expectations for smoking at posttest.

Evaluation Methodology

Design: Forty-seven schools in four boroughs of New York City participated in the study, with 3,153 students (90% of the available 3,518 seventh graders) providing pre- and post-test data. Schools were first blocked by school type (public or parochial) and percentage of Hispanic students per school (25-49%, 50-74%, 75-100%) and then randomized into either a treatment or control condition. Blocking occurred due to expected differences in smoking or smoking risk. The study sample consisted of 25 schools (19 parochial, 6 public; 1,795 students) in the treatment condition and 22 schools (17 parochial, 5 public; 1,358 students) in the control condition.

Treatment students received the 15-session prevention program. This version of the curriculum only addressed cigarette smoking (and not alcohol and marijuana). In order to make the curriculum more appropriate for the targeted population (urban minority), a few modifications were made, including adjusting the reading level, examples used to illustrate program content, and suggested situations for behavioral rehearsal exercises. Teachers who taught the program had attended a 1-day teacher training workshop. Trained observers randomly attended classes and completed observational forms to assess implementation fidelity.

Sample Characteristics: The majority of students (n=1,836) attended one of the 11 participating public schools (there were 1,364 parochial school students in 36 schools). Students were 51% female and 49% male and students had a mean age of 12 years, 10 months. Most schools (83%) had students with average income levels at or below 150% of the federal poverty level. The majority of students (56%) were Hispanic, followed by Black (19%), White (14%), and Other (12%).

Measures: Measures were collected at baseline and post-intervention. Students completed a questionnaire during class that collected information about smoking status (self and significant others), as well as cognitive, attitudinal, and psychological characteristics hypothesized to be related to (mediate) smoking initiation (smoking knowledge, skills knowledge, attitudes and normative beliefs, skills use, skills confidence, skills efficacy, self-efficacy, and psychological well-being). Reliability estimates on measures ranged from .69 to .82. Students also submitted carbon monoxide (CO) breath samples to enhance the validity of the self-reported data.

Analysis: Analysis was conducted only on individuals who provided both pre- and posttest data (n=3,153). Individual-level data was aggregated for each school for data analysis. A general linear model procedure was used using pretest scores as covariates. Results were presented as overal mean differences, as well as by school type and ethnic composition (percent Hispanic). Mediating effects were also analyzed using a structural modeling approach.

Outcomes

Baseline Equivalence and Differential Attrition: T-tests were performed to determine baseline equivalence, and showed no significant differences between treatment and control groups prior to program implementation.

Fidelity Monitoring: The mean level of implementation fidelity was 59.8%, with distribution of implementation scores indicating that at least half the treatment participants received at least 60% of the program. There was a trend of lower implementation fidelity among the public schools, but this trend was not significant.

Posttest: On the measure of smoking behavior, results indicated significant program impact on the percentage of treatment students reporting past month smoking and smoking onset, compared to control condition students. The reductions in onset of smoking rates compared to the control condition were almost 30% lower. There were no significant differences between groups on current smoking, past week smoking, or behavioral intention. On the psychosocial variables assessed, there were significant program effects on knowledge and normative expectation measures (promximal variables), but not on attitudes. Students who participated in the program had significantly higher posttest knowledge scores and significantly lower normative expectation scores than students who did not receive the program. Causal modeling analysis also demonstrated that the impact of the intervention on cigarette smoking was mediated by these variables.

Study 12

Summary

Velasco et al. (2017) used a quasi-experimental design that examined 55 high schools in Italy that had already implemented or not implemented the LST program. The LST schools had 1,350 students, while the control schools with similar characteristics had 1,014 students. Student assessments of substance use came at baseline and posttest.

Velasco et al. (2017) found that, relative to the control group, the LST group reported significantly

• Lower rates of smoking initiation at posttest
• Greater substance use knowledge at posttest
• Less positive attitudes toward smoking at posttest
• Lower normative expectations about smoking and drinking at posttest.

Evaluation Methodology

Design:

Velasco et al. (2017) conducted a quasi-experimental design with crude matching that involved 55 schools in the Lombardy region of Italy. Data were collected before (baseline) and after (posttest) implementation, and again one- and two-years following program completion.

Assignment:

All students who participated in the study were enrolled in one of 55 high schools distributed across Italy (n=3,048 students, 138 classrooms). Thirty-one intervention schools (n = 1,350 students) were randomly selected from the set of schools participating in a larger implementation of the program in Italy after being stratified by geographic area. For the comparison group, schools with similar characteristics (e.g., municipality size, school size, and students' demographics) as the intervention schools were selected, resulting in twenty-four control schools (n=1,014 students). Within each treatment and control school, 3 classes (if available) were randomly selected to participate in the study. During the first year of middle school, students in the intervention schools received 15 sessions of the treatment followed by 10 booster sessions implemented during the second year of middle school and nine during the third year. All comparison schools were not involved in other drug prevention interventions.

Attrition:

A total of 1,586 students from 106 classes and 48 schools completed a posttest after the final year of program implementation (i.e., at the two-year follow-up after implementation of the 19 booster sessions), for an overall attrition rate of 48%, 23% and 13%, respectively.

Sample:

Students were 51% female for the intervention and comparison and students had a mean age of 11 years. The majority of students in the intervention (92%) and comparison (90%) were Italian.

Measures:

Three forms of substance use were assessed, smoking, alcohol use, and drunkenness, each of which were measured using a 9-point frequency response scale anchored by 1 (never) to 9 (more than once a day).

Thirteen measures assessing risk and protective factors were collected, including:

• Life skills - assertiveness (sample α = .73); social skills (sample α = .68); decision-making (sample α = .80); advertising resistance skills (sample α = .69); and anxiety reduction skills (sample α = .67).
• Psychosocial outcomes - students' distress (sample α = .81); well-being (sample α = .81); and risk-taking (sample α = .68).
• Beliefs about substance use - knowledge of the physiological effects of substance use (no alpha reported); beliefs about myths/misconceptions about drugs (no alpha reported); attitudes about smoking (sample α = .78) and alcohol (sample α = .74); normative expectations related to tobacco and alcohol use in terms of perceived prevalence of drug use among adults (no sample alpha).

Analysis:

Analyses reported here examined the effects of the intervention at the 2-year follow-up (i.e., after the booster sessions). While the text states that covariates included pretest scores, gender and age, Table 4 (which reports results after the two-year follow up when the full program was completed) states that the covariates included just gender and age. Thus, it was not clear whether analyses controlled for baseline pretests. Analyses used GEE to account for the clustering of students within schools. Scores on the behavioral measures (smoking, alcohol use, and drunkenness) were recorded to identify students who initiated alcohol or tobacco use (never vs. more than once in lifetime), and those who transitioned from less than weekly use to weekly use in order to verify the effectiveness of the program in preventing the initiation or the regular use of these substances.

Intent-to-Treat:

It appeared that students were analyzed according to the condition in which they were assigned, which is in line with intent-to-treat protocol. Students with missing data at the posttest and 2-year follow up were dropped from the analysis.

Outcomes

Implementation Fidelity:

Not reported.

Baseline Equivalence:

There were no significant baseline differences between groups in terms of school characteristics (municipality size and school size) and school-level demographics (gender, age, and nationality). There were also no school-level baseline differences in substance use (tobacco, alcohol or drunkenness). Baseline equivalence tests, however, were not conducted at the classroom- or student-level.

Differential Attrition:

Differential attrition tests (i.e., assessing the baseline and demographic differences between completers and attritors) were not conducted. There were no significant differences in attrition found between groups with regards to baseline outcomes (i.e., well-being or substance use), but these attrition-by-condition tests did not assess demographic variables.

Posttest:

At the two-year follow-up (when the program, including all 19 booster sessions, was completed), results showed that compared to control, treatment students self-reported:

• Lower rates of initiating smoking (behavioral outcome)
• Higher rates of skills awareness and anxiety reduction skills (R&P factor)
• Greater knowledge of the physiological effects of substance use (R&P factor)
• More accurate believes and fewer misconceptions about drugs (R&P factor)
• Less positive attitudes toward smoking (R&P factor)
• Lower normative expectations about adults' smoking and drinking (R&P factors).

Out of the 13 R&P measures, however there was one negative effect (i.e., at the posttest, lower assertive skills were reported in the treatment group compared to the control group).

Long-Term:

Not conducted.

Study 13

Summary

Crowley et al. (2014) used a quasi-experimental design with propensity score matching that examined seventh-grade students in 28 school districts across Iowa and Pennsylvania. Schools in the treatment districts but not the control districts implemented LST (or one of several other programs). Use of nonmedical prescription opioid use served as the outcome.

Crowley et al. (2014) found that, relative to students in the control group, students in the LST group reported significantly

• Lower use of nonmedical prescription opioids.

Evaluation Methodology

Design:

The National Institutes of Health funded an evaluation of the PROSPER model designed to disseminate a menu of evidence-based prevention interventions (EPBIs) that are both family- and school-based. Included in the menu of EPBIs were the Strengthening Families Program offered in 6^th grade; and All Stars, Life Skills Training and/or Project Alert school-based programs offered in 7^th grade. For the PROSPER evaluation, communities were randomly assigned to treatment (receiving the menu of EBPI's) or control (no menu of EBPI's). Crowley et al. (2014) conducted a multi-cohort quasi-experimental design utilizing propensity score matching to evaluate outcomes of students who received individual programs or some combination of school-based programs within the menu of EPBIs (treatment) compared to control. Self-report data assessing use of prescription opioids for nonmedical purposes were collected in 6^th grade (pretest) and at the end of each year through 12^th grade.

Recruitment:

From the 68 available school districts in Iowa and Pennsylvania where the PROSPER evaluation took place, 40 were excluded for not meeting the following eligibility requirements: (1) school district enrollment between 1301 and 5200 students, (2) at least 15% of families eligible for reduced cost lunch, (3) maximum of 50% of the adult population employed at or attending a college or university, and (4) the community could not be involved in other university-affiliated, youth-focused prevention initiatives. For the PROSPER evaluation, 28 school districts were matched by geographic location and size, and each pair of districts were randomized to condition (14 in the treatment and 14 in the control). The total number of individuals or families was not reported (just mean cluster sizes and cluster size ranges were reported). Within the treatment group, teams led by local cooperative extension agents and school officials selected a universal family and school program from one of the following evidenced based preventative interventions: 1) All Stars, 2) Life Skills, and 3) Project Alert. In addition, all families in the intervention group were offered the Strengthening Families Program for families with students in the 6th grade, but only some enrolled.

Assignment:

Which school-based evidence-based prevention interventions selected was determined within each of the 14 districts assigned to treatment, and families within the treatment group chose whether to attend the evening family program. Though not explicitly reported, it appeared that all 14 districts were involved in the quasi-experimental design reported by Crowley et al. (2014). In addition, a total of 5,026 students received one of the 3 school-based programs in 7^th grade, but it was not reported how many students (overall or by condition) were assigned to the treatment or control group. The propensity score model used to match control with treatment included 43 covariates covering four levels: individual, family, school, and research team. Examples of variables included: 18 student-level variables (prescription opioid use, gender, alcohol use, ever been intoxicated, level of alcohol use, inhalant use, hard drug use, tobacco use, youth substance use expectations, school attitude, problem solving capacity, school adjustment, school attendance, refusal efficacy, refusal intentions, stress management, substance use norms, and future use), 9 school-level variables (school uses a structured curriculum, percentage of free lunch, parent outreach, community pressure, teacher resistance, involvement of agency, school attitude towards prevention, district attitude towards prevention, and number of teachers in school), and 5 variables related to the research team (extension reputation, team size, time for parent recruitment, schools' prevention attitudes, and success of community coalition). These variables appeared to have been collected by surveys administered as part of the PROSPER evaluation.

Attrition:

No districts dropped from the sample, and n's at the student level were not reported (either at assignment or follow-up).

Sample:

50% of the students were female for the intervention and 51% for the treatment, and students had a mean age of 11.8 years overall. Students in the intervention and control groups had an average family income of $50,174 and $52,704, respectively. Around half of the sample in both treatment (50%) and control (53%) were from dual parent households.

Measures:

To evaluate youth nonmedical prescription opioid use, each participant was asked whether they had ever used prescription opioids for nonmedical purposes at the 6th grade pre-test and at the end of each school year through 12th grade.

Analysis:

A multi-step analytic framework was employed that included: (1) estimation of participants' propensity to receive different programs, (2) fitting marginal structural models to estimate the impact of receiving different programs on ever using prescription opioids for non-medical purposes, and (3) multi-level logistic models to account for clustering.

Intent-to-Treat:

It appeared that students were analyzed according to the condition in which they were assigned, which is in line with intent-to-treat protocol. Multiple imputation was utilized for missing data.

Outcomes

Implementation Fidelity:

Not reported.

Baseline Equivalence:

Not reported.

Differential Attrition:

Not reported.

Posttest:

There was no posttest result reported. Rather, results reflected findings across six years of data collection and showed that receipt of the Life Skills Training Program led to a significantly reduced probability of youth having ever used prescription opioids for nonmedical purposes by grade 12 compared to the control condition. No significant differences were observed between the All Stars and Project Alert Programs compared to control. Receipt of the Life Skills and family programs together as well as receipt of the All Stars and family programs together revealed a significant difference from the control condition in favor of the treatment group.

Long-Term:

Not conducted.

Study 14

Summary

Smith et al. (2004) and Vicary et al. (2004, 2006) used a cluster randomized trial to examine 732 seventh-grade students in nine Pennsylvania middle schools. The schools were randomly assigned to LST, infused LST that delivered the program as part of the regular curriculum, or a control group. Assessments through the end of ninth grade examined substance use outcomes.

Smith et al. (2004) found no effects of LST compared to the control group for males and found effects in grade seven for females that disappeared by grade eight. Vicary et al. (2004, 2006) found no effects on substance use in grades eight or nine.

Evaluation Methodology

Design:

Recruitment: The sample came from nine middle schools in nine rural Pennsylvania school districts. The schools were of low socioeconomic status (one third of the student body in the school district qualifying to receive free or reduced lunch) and had relatively small enrollment (less than 1,000). A total of 732 consented youths who began grade seven in 1999-2000 participated in the study.

Assignment: The cluster randomized design randomly assigned the nine middle schools to three conditions, each with three schools: LST (n = 234 students), infused LST that delivered the program as part of the regular curriculum (n = 297 students), and a control group that received neither program (n = 201 students).

Assessments/Attrition: The four assessments occurred at baseline (the beginning of grade seven) and the end of grades seven, eight, and nine. Students received the main program in grades seven and eight and received booster sessions in grade nine. Thus, all three of the post-baseline assessments occurred while the three-year LST program was ongoing. Smith et al. (2004) reported retention rates of 97% at baseline, 96% at the grade seven assessment, and 90% at the grade eight assessment. Vicary et al. (2006) reported that 79% of the students completed all four assessments.

Sample:

The sample was 54% male and 97% white.

Measures:

The six substance use outcome measures came from student self-reports and included frequency of cigarette use, alcohol use, drunkenness, binge drinking, marijuana use, and inhalant use. To maintain confidentiality, ID codes were used in place of names, and teachers were not present. However, 19 students were removed at the third assessment for suspicious or inconsistent answering patterns. Vicary et al. (2004) used log or double log transformations of these substance use variables to better approximate a normal distribution.

Vicary et al. (2004, 2006) examined numerous other self-reported risk and protective measures related to substance use, including attitudes, normative beliefs, knowledge, decision-making, communication skills, refusal skills, media awareness, assertiveness, and coping with anxiety. The scales all had good reliabilities.

Analysis:

The analysis used regression models with fixed effects for schools, controls for baseline outcomes and other covariates, and robust standard errors for non-normally distributed outcomes. There were too few schools to use multilevel models, but checks found "negligible" ICCs. The analysis also used multiple imputation with academic performance and religious participation serving as auxiliary predictors in the imputation model. The analysis did not examine main effects, as the authors stated that they made the decision a priori to analyze the outcomes separately for males and females.

Intent-to-Treat: The analysis used multiple imputation to include all participants.

Outcomes

Implementation Fidelity:

For the LST condition, the teachers delivered 90% of the lessons, and student attendance averaged 93%. The authors noted informally that program teachers would have benefitted from refresher training.

Baseline Equivalence:

Smith et al. (2004) noted condition differences for three variables (free lunch eligibility, substance use, and problem behavior) but gave no other details on the number of tests or the size of the differences. The models controlled for these three baseline measures. Vicary et al. (2004) stated that "Pre-intervention comparisons of the sample revealed no significant differences by treatment condition for any of the substance use variables or the variables used to construct risk status."

Differential Attrition:

Vicary et al. (2004) stated that completion rates for students were similar across the conditions (68% for LST versus 72% for the control group). Vicary et al. (2006) reported that there was "no differential attrition across the three conditions" but provided no details.

Posttest:

Seventh- and Eighth-Grade Results

Smith et al. (2004) presented separate results for males and females. For males, none of the 12 outcome tests (six at the seventh-grade assessment and six at the eighth-grade assessment) showed a significant difference in substance use between the LST group and the control group. For females, four of the outcomes at the end of seventh grade (alcohol use, binge drinking, marijuana use, and inhalant use) were significantly lower for the LST group than the control group, but all four differences declined to non-significance by the end of eighth grade.

Vicary et al. (2004) examined females only and presented separate results for low-risk and high-risk groups. Although the program significantly reduced substance use among both low-risk and high-risk girls at the end of grade seven, all the effects fell to non-significance by the end of grade eight. Tests for effects on the risk and protective factors showed a significant program effect at the end of grade eight only for knowledge among low-risk girls and assertiveness skills among high-risk girls. Overall, tests for moderation by risk status showed stronger treatment effects for the high-risk females than low-risk females.

Ninth-Grade Results

Vicary et al. (2006) presented separate analyses for females and males. They found no significant effects of LST relative to the control group for any of the substance use outcomes at the end of grade nine. For the risk and protective factors, they found that the LST groups did better than the control group on coping and communication for girls but found no effects for boys.

Long-Term:

Not examined.

Study 15

Summary

Aviles (2019) used a quasi-experimental design with propensity score matching to examine sixth-grade students in 13 schools that had implemented LST and 10 schools that had not. Self-reported substance use served as the outcome.

Aviles (2019) found that, relative to the control group, the LST group had significantly

• Fewer smokers.

Evaluation Methodology

Design:

Recruitment: The study examined sixth-grade students in 13 schools located in the mid-Atlantic region that had implemented LST.

Assignment: The quasi-experimental design matched the 13 LST schools (n = 1,052 students) to 10 control schools not utilizing the treatment (n = 494 students). The propensity scores used to match the schools resulted from the following predictors: gender distribution, socioeconomic status (i.e., percentage of low income students in the school as reflected by the percentage of youth receiving free or reduced-price lunch), locality (e.g., urban, suburban, or rural), aggregated community risk factor scale scores for the school (e.g., perceived availability of drugs), aggregated family risk factor scale scores for the school (e.g., poor family management, parental attitudes favorable to substance use), and aggregated school-based risk factor scale scores for the school (e.g., poor academic performance).

Assessments/Attrition: The cross-sectional design did not include baseline data. It instead measured the outcomes in the fall of sixth grade, presumably while the program was ongoing.

Sample:

The student sample was 89% Caucasian with 52% females.

Measures:

Data came from the Pennsylvania Youth Survey, administered in the fall every two years to students in grade six. The survey included 11 self-reported substance use measures and 12 measures of risk and protective factors. To maintain reliability of the substance use measures, students who provided inconsistent answers or reported using a fake drug were treated as having missing data. Reliabilities for the scales were as low as .42 and .61.

Analysis:

The analysis used two-level random effects models with linear, Poisson, inflated Poisson, binomial, and negative binomial link functions. Low ICC values of .002 to .014 indicated little clustering.

Intent-to-Treat: The use of Full Information Maximum Likelihood estimation enabled all participants to be included in the analysis except those with missing data on the gender, race, or age covariates.

Outcomes

Implementation Fidelity:

Not examined.

Baseline Equivalence:

Not examined.

Differential Attrition:

Not examined.

Posttest:

Of the 11 substance use measures, two showed a significant effect. The program partly impacted 30-day cigarette use by significantly reducing whether a person smoked but not the amount a person smoked. The program also had a significant iatrogenic effect on lifetime prescription drug use. There were no significant effects on the 12 risk and protective measures. Tests for moderation suggested stronger effects for boys than girls.

Long-Term:

Not examined.