Preventing drug abuse among adolescent girls: outcome data from an internet-based intervention
Effectiveness bank home page. Opens new windowResearch analysis

This entry is our analysis of a study considered particularly relevant to improving outcomes from drug or alcohol interventions in the UK. The original study was not published by Findings; click Title to order a copy. Free reprints may be available from the authors – click prepared e-mail. The summary conveys the findings and views expressed in the study. Below is a commentary from Drug and Alcohol Findings.

Links to other documents. Hover over for notes. Click to highlight passage referred to. Unfold extra text Unfold supplementary text
Copy title and link | Comment/query |

Preventing drug abuse among adolescent girls: outcome data from an internet-based intervention.

Schwinn T.M., Schinke S.P., Di Noia J.
Prevention Science: 2010, 11, p. 24–32.
Unable to obtain a copy by clicking title? Try asking the author for a reprint by adapting this prepared e-mail or by writing to Dr Schwinn at You could also try this alternative source.

In this US study a substance use prevention programme for adolescent girls accessed over the internet from home had effects comparable to school-based drug education, yet occupied no classroom or teacher time and could inexpensively be replicated across the internet-linked population. Also described are later reports from similar studies.

Summary Computer-delivered prevention programmes for young people enable them to privately disclose and develop their skills in relation to sensitive issues at their own pace, and to benefit from content which can be standardised in accordance with theory and research findings. Such programmes are also inexpensive to deliver, easy to use, and allow for interactivity and data storage and retrieval.

The featured study's research team developed one such programme called RealTeen. It drew on an earlier skills-based programme delivered via a CD-ROM, but was accessed over the internet and tailored for young teenage girls. RealTeen's 12 roughly 25-minute sessions sought to develop general personal and social skills (self-efficacy, communication, assertiveness) and skills specific to dealing with situations involving possible substance use. Within each session steps were taken to generate interactivity and peer-to-peer communication. Across sessions an older female animated character guided girls through the content and practice exercises.

To test RealTeen's effectiveness the researchers e-mailed 13–14-year-old female registered users a youth-oriented social web site asking them to join the study. About 450 were interested of whom 236 received verified parental permission and joined the study. All were from the USA or Canada.

The girls were randomly assigned to have access to the RealTeen site or to a control A group of people, households, organisations, communities or other units who do not participate in the intervention(s) being evaluated. Instead, they receive no intervention or none relevant to the outcomes being assessed, carry on as usual, or receive an alternative intervention (for the latter the term comparison group may be preferable). Outcome measures taken from the controls form the benchmark against which changes in the intervention group(s) are compared to determine whether the intervention had an impact and whether this is statistically significant. Comparability between control and intervention groups is essential. Normally this is best achieved by randomly allocating research participants to the different groups. Alternatives include sequentially selecting participants for one then the other group(s), or deliberately selecting similar set of participants for each group. group who were asked just to complete the same questionnaires over the internet assessing their substance use and other variables such as self-efficacy which might affect and be affected by the programme. Assessments were made immediately before the RealTeen group accessed the programme (the baseline), immediately after they finished it (post-programme – averaging a month later) or at roughly the same time for the control group, and then six months later. All but 7% and 9% respectively of the girls completed post-programme and six-month assessments. Girls allocated to RealTeen were rewarded for completing its sessions with points which could be exchanged for small gifts.

Main findings

All but 10 of the 118 girls allocated to RealTeen completed all 12 sessions. The 10 who did not also did not complete follow-up assessments, meaning that the results of the programme were based entirely on completers.

Assessed over the past 30 days and as changes from the baseline, post-programme levels of substance use among RealTeen completers did not significantly differ from those assessed at roughly the same time among the control group. However, six months later there were statistically significant differences indicating relatively lower substance use among girls allocated to RealTeen.

At all assessments the girls were asked how many times (from 0 to 40) in the past month they had used alcohol, cigarettes, cannabis, cocaine, inhalants, methamphetamines, or ecstasy. Relative to their starting use levels, at the final follow-up the RealTeen girls had on average drunk alcohol significantly less often over the past 30 days (0.45 times versus 1.08). Results were similar for using cannabis (0.04 times versus 0.42) but there were no statistically significant differences in frequency of smoking.

The researchers also amalgamated all the substance use measures to arrive at figure for the number of different types of drugs (out of seven) each girl had used over the past 30 days. This too was on average significantly lower among RealTeen than control girls (0.33 substances versus 0.68), as was a similar measure which excluded alcohol (0.13 substances versus 0.38 out of six). In each case the magnitude of the reduction expressed as an effect size A standard way of expressing the magnitude of a difference (eg, between outcomes in control and intervention groups) applicable to most quantitative data. Enables different measures taken in different studies to be compared or (in meta-analyses) combined. Based on expressing the difference in the average outcomes between control and experimental groups as a proportion of how much the outcome varies across both groups. The most common statistic used to quantify this difference is called Cohen's d. Conventionally this is considered to indicate a small effect when no greater than 0.2, a medium effect when around 0.5, and a large effect when at least 0.8. was about 0.2, a small effect.

The questionnaires also assessed some psychological variables which might account for the programme's impact on substance use. Of these, after completing the programme the RealTeen girls reported fewer substance users and/or less acceptance of substance among their peers and friends, but this effect did not persist to the six-month follow-up. The reverse was the case for confidence in one's ability to resist pressure from a boyfriend (or a boy one likes) to use substances; after the programme RealTeen girls were no more confident than control girls, but they were more confident six months later. Other variables were apparently unaffected by the programme including decision-making skills, the ease with which girls felt they could refuse an offer to use drugs from a best friend, ability to manage stress, goal-setting, social skills, self-esteem, and satisfaction with one's looks.

The authors' conclusions

Findings at the six-month follow-up suggest that the study's female-specific, internet-based programme did help prevent substance use, and lend credence to the use of the internet to deliver skills-based interactive interventions. Lack of impact as assessed immediately after the programme may reflect the fact that the assessed period included days when little of the programme would have been completed, and that time and real-life opportunities to apply acquired skills are required for these to impact on substance use.

The magnitude of the impacts from the intervention are comparable to those achieved by similar classroom-based face-to-face programmes and considerably better than those recorded after older, non-interactive, knowledge-based programmes. These results were achieved without labour-intensive and costly staff training and classroom lessons, and without eating in to the time the school has available for other lessons. The programme lends itself to home rather than public delivery, but could be accessed at private computer terminals at community-based agencies, schools, or libraries.

It should however be remembered that the study would have attracted girls who were computer literate and motivated and who lived in what were then the minority of North American homes with broadband internet access. Its effects may not have been the same among a wider or different population.

Findings logo commentary This is one of at least three studies from the same research stable to have demonstrated the potential of computer-based prevention programming among adolescents from different sectors of US society and recruited in different ways. While its results were statistically significant, only a longer term study could determine whether the small changes (typical of online prevention) in on average very low levels of substance use would prove significant in prevention and health promotion terms. For example, the impact on drinking amounted to a reduction from drinking about one day out of 30 to about one day out of 60, with no information about the quantity drunk on each occasion.

In all the studies it seems likely that the children and parents who volunteered for and completed the studies were particularly keen on developing the skills on offer in the programmes and on controlling the substance use of the children. In the featured study in particular an unusually high proportion of girls completed all 12 sessions, a testament perhaps to the programme's attractions and the gifts on offer, but also perhaps a reflection of the type of girls who joined the study. If, like school-based drug education, the programmes were implemented routinely across the adolescent population, it seems possible that completion rates would be lower and results might differ. Offered instead on a voluntary basis, take-up of online health promotion is low especially among high-risk groups, and concentrated among the highly educated population.

The small effects seen in the featured study must be considered vulnerable to methodological influences, such as girls who had worked their way through the programme being less willing to admit to substance use. Perhaps too the 10 who did not complete the programme and were also missing from the follow-up assessments were particularly resistant to prevention efforts; had their substance use been able to be assessed, it might have tipped the balance against the programme. Generally outcomes can more securely be attributed to an intervention if the links in between are also visible – if it can be seen not just that it worked, but that it worked via the intended mechanisms. Of 18 possibilities for the programme to demonstrate that it had worked via theorised psychological changes, in just two cases did results reach statistical significance, and in no case was the same change evident at both follow-ups. With so many possibilities, the few positives might have been due to chance.

Among the same team's other studies was one of an interactive CD-ROM tested among a very different population – primarily black or Hispanic 10–12-year olds. A three-year follow-up showed that the programme had helped prevent use of alcohol, tobacco or cannabis. Most of the same sample were also followed up when aged 17–18, revealing long-term impacts on drinking, heavy drinking, smoking and alcohol-related adverse consequences, but not on cannabis use. When the children were a year older another analysis from the same study focused on their drinking. It found that the children allocated to the CD-ROM on average nearly eight years before were still less frequently drinking or drinking heavily and also smoking far less often. In this study there were no lasting benefits from adding parental involvement A video and newsletters explaining the project and attempting to engage parents in strengthening the family structure and helping their children assimilate the CD-ROM. elements to the programme.

Like the featured study, girls were the target of another study. Again, instead of being primarily white the girls were black and Hispanic US adolescents aged 10–13, who each joined the study with their mothers. The pairs were randomly allocated to a control group, or to work together through a 10-session computerised programme. It was intended to improve communication between them and general family dynamics, to encourage parents to monitor their children and implement rules on substance use, and to help the girls manage their emotions and difficult situations, including refusing peer pressure. They also learned about the true prevalence of smoking, drinking, and drug use among their peers. As assessed at the end of what was intended to be 10 weekly sessions, girls allocated to the programme had drank on average fewer drinks over the past 30 days and most of the psychological and family process variables thought to underpin the programme's effect on substance use had changed in the expected ways. In this study it seems that only mother-daughter pairs who completed the intervention programme were directed to the online follow-up measures, though the study records few drop-outs between baseline and these measures being taken. Nevertheless the possibility remains that the requirement to complete – one which could not be replicated among the controls – biased outcomes by excluding non-completers among the intervention pairs. Alternatively the low rate of drop-out suggests an extraordinarily high completion rate unlikely to be found in a less self-selected population.

Thanks for their comments on this entry in draft to research author Traci M. Schwinn of Columbia University in the USA and Andrew Brown of Mentor based in London, England. Commentators bear no responsibility for the text including the interpretations and any remaining errors.

Last revised 12 May 2019. First uploaded 31 July 2013

Open Effectiveness Bank home page

Top 10 most closely related documents on this site. For more try a subject or free text search

STUDY 2010 The effectiveness of a school-based substance abuse prevention program: 18-month follow-up of the EU-Dap cluster randomized controlled trial

NASTY SURPRISES 2004 Confident kids ... like to party

STUDY 2003 Secondary school DARE ineffective without interactive extensions

STUDY 2000 Education's uncertain saviour

STUDY 2011 Effects of a school-based prevention program on European adolescents' patterns of alcohol use

REVIEW 2015 Prevention of addictive behaviours

STUDY 2002 Growth in youth drinking curbed by correcting 'normative' beliefs

STUDY 2009 Evaluating mediators of the impact of the Linking the Interests of Families and Teachers (LIFT) multimodal preventive intervention on substance use initiation

STUDY 2008 Harnessing peer interaction in school-based prevention can backfire

STUDY 2000 Everyone is NOT doing it - important prevention message for early teens