Study
Haggerty and colleagues (2015) conducted a randomized controlled trial to test whether SCT effects would sustain from adolescence into early adulthood (ages 18 to 25). The study used the same sample and eligibility criteria as the 2005 study by Haggerty and colleagues (Study 1), which recruited eighth-grade students from the Seattle Public Schools. The study authors attempted to contact all 331 students who were included in the 2005 study. The study authors conducted a 72-month follow up (when students were about age 20) to detect lasting effects of the program.
The 331 contacted students ranged in age 18 to 22 years. About 90.1 percent were re-contacted, completed the self-administered survey, and provided a urine sample for drug screening (n = 301). Young adults remained in their original stratified group conditions based on race and gender. There were 118 in the PA format intervention group, 107 in the SA format intervention group, and 106 in the control condition (the CrimeSolutions review of this study focused on intervention effects for the PA format condition versus the control condition).
Of the 301 young adults, 67 were Black males, 73 were Black females, 82 were white males, and 79 were white females. The majority were enrolled in school (57.8 percent), employed at the time of the study (45.6 percent), or were neither employed nor attending school regularly (18 percent). The demographic characteristics continued to show a statistically significant difference by race. Black young adults had a higher likelihood of coming from single-parent families with larger households, lower per capita income, and lower level of parent education.
The main outcome of interest was drug use frequency at age 20. The drug use frequency index consisted of four indicators of substance use frequency: 1) heavy episodic drinking in the past 2 weeks (five drinks in a row for males, four drinks in a row for females); ranging from none to six or more times; 2) marijuana use in the past month, ranging from none to 20 or more times; 3) cigarette use in the past month, ranging from none to more than one pack per day; and 4) any other illegal drug use in the past year, ranging from none to six or more occasions. All measures were self-reported items through surveys constructed for the study. The four data collection/assessment periods were as follows: 1) at posttest, 2) during a 1-year follow up (that was conducted in the ninth grade), 3) at a 2-year follow up (that was conducted in the 10th grade), and 4) at a 72-month follow up (when students were about age 20). A latent construct approach was then used to assess intervention effects on drug use frequency at age 20. The study authors conducted subgroup analyses to determine whether there were differences between white and Black young adults in the PA and the SA intervention groups.
Study
Haggerty and colleagues (2007) used an intent-to-treat, randomized controlled trial to test the efficacy of two different formats for administering Staying Connected with Your Teen (SCT). The two formats were 1) the traditional parent and adolescent group–administered (PA) format, and 2) a self-administered (SA) format with weekly telephone support from program personnel (the CrimeSolutions review of this study focused on the PA format of the SCT intervention versus the control condition). The study was conducted in Seattle, Washington, and eligible participants were recruited through the Seattle Public Schools. Families were eligible to participate if they had a white or Black eighth grader living at home, spoke English as their primary language, and planned on living in the Seattle area for the next 6 months. The study authors conducted a 24-month (2-year) follow up to detect any program effects.
Families that consented were stratified on race and gender and randomly assigned to one of the treatment conditions (participating in the PA or SA format of the SCT program) or the control condition (no treatment or programming). The treatment group had 225 families; 118 were assigned to the PA format condition, and 107 were assigned to the SA format condition. There were 106 families in the control condition. Families in the control condition did not participate in the SCT program or receive assistance from program personnel.
The study sample showed statistically significant differences by race for most of the parental demographic characteristics. Black adolescents tended to come from single-parent families with larger households, lower per capita income, and lower parent education. Black adolescents also self-reported significantly less alcohol use and significantly more sexual intercourse at baseline than white adolescents. Baseline levels were entered as covariates in all analyses to reduce the potential for baseline differences to affect the program’s outcomes.
All outcome measures consisted of teen self-reports on survey items constructed for this study. The main outcomes measured were 1) perceived harm of substance abuse, 2) favorable attitudes toward substance abuse, 3) delinquent behavior, and 4) violent behavior. Substance abuse items began as, “It hurts people to…” or “Do you think it’s OK for someone your age to…” followed by “smoke cigarettes, drink alcohol, smoke marijuana, or use other illegal drugs.” Delinquent and violent behaviors were measured by student-reported frequency of behavior in the past 30 days. There were four data collection/assessments: 1) at baseline, 2) at a posttest at the end of eighth grade, 3) at a 12-month follow up in ninth grade, and 4) at a 24-month follow up in 10th grade. Repeated measures mixed-model regressions were used to assess change in the outcome measures. The study authors conducted subgroup analyses to determine whether there were differences in outcomes between white and Black students in the PA and SA intervention groups and the control group.