Bodin M.C., Strandberg A.K.
Addiction: 2011, in press.
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 Bodin at email@example.com;firstname.lastname@example.org.
When its developers tested it, in Sweden routine parent-school meetings including presentations encouraging parents to take a strong stand against underage drinking had a remarkable impact on adolescent drunkenness; why then did this Swedish trial by other researchers fail to replicate the original findings?
Summary Rather than through classroom lessons, the Örebro Prevention Programme works via the parents, building on the fact that Swedish schools start each term with a parent information meeting. In the original trial conducted by the programme's developers, this low cost intervention was found to halve the increase in the frequency of drunkenness between ages 13 and 16 both among pupils in general and among high risk pupils who had already been drunk at age 13. Though the intervention had focused on drinking, there were also statistically significant and medium to large reductions in criminal or antisocial behaviour. As expected, the programme seemed to have worked by maintaining the parents' strict anti-drinking norms.
The featured trial aimed to test whether these findings could be replicated in a study conducted by independent researchers not involved in the programme's development, and using the current (in Sweden) widely disseminated version of the programme presented by experienced Örebro presenters. This version includes six 20-minute standardised PowerPoint presentations given to parents of 13–16 year olds during termly regular school-based parent-teacher meetings. During these parents are advised to maintain a zero-tolerance stance towards youth drinking and to communicate clear rules to their children. This is reinforced by inviting attending parents to sign agreements on how they will prevent their children from drinking. A summary of the event and the class agreement is mailed to all parents including those who had not been at the meeting. The original trial also used the mailings to promote organised leisure activities. However, there was no evidence that involvement in adult-led organised group activities had in fact been enhanced by the intervention, so this element was dropped from the featured study.
Of 716 schools told about the study, 40 were both eligible and volunteered to participate. They were allocated at random to implement the Örebro Prevention Programme in two or three classes or to carry on as usual (the 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. classes) without implementing the programme. The initial sample of 1752 seventh-grade pupils (about age 13) surveyed before the intervention started were re-surveyed at follow-ups 12 and 30 months later when around 90% responded. Parents too were surveyed and around 70% responded.
The programme was fairly fully implemented. Of the intended six presentations, the study averaged just under five per school after researchers made extensive efforts to facilitate arrangements. Presentation summaries and class agreements were posted by the research assistant to all parents after each presentation. Three quarters of the 46 Örebro classes made written class agreements during the first or second presentations and another 15% came to verbal agreements.
At the start of the study at age 13, around 1 in 8 of the pupils said they had already been drunk, though very few (around 1%) were currently drunk frequently and few too drank at least weekly. Whether the programme had any impact on them was tested by analysing the responses of just the pupils who completed the relevant follow-up survey, by assuming good outcomes among the remainder, by assuming bad outcomes among the remainder, and by estimating how missing pupils would have responded based on the available data. In respect of frequent (at least twice in the past four weeks) drunkenness, only under one of these assumptions (assuming bad outcomes) was there a statistically significant difference between pupils in Örebro classes versus those in control classes and then only at the first follow-up. There were no statistically significant differences at all in the proportions of pupils who had ever been drunk or who were drinking weekly.
However, there were differences in parental attitudes and behaviour. Consistently at the last follow-up, pupils in Örebro classes were less likely to say they had been allowed to drink at home and also at this time their parents who responded to the survey reported more strict and restrictive attitudes to youth drinking than the parents of children in control classes.
For the authors their results did not support the case that the programme prevented youth drunkenness. The sole statistically significant difference in drunkenness was probably due to differential drop-out rather than any real impact of the programme, and the apparent impacts on parental attitudes and behaviour may have been due to parents and children exposed to the Örebro programme being more likely to give the responses 'approved' by the programme.
Though a contrast to the original Swedish study, these findings are consistent with a trial in the Netherlands of a Dutch version of the Örebro programme. This found it ineffective on its own in curbing drinking as were alcohol use prevention lessons, but the combination – targeting parents and their children at the same time – did substantially curb the expected increase in weekly drinking and drinking heavily each week over the next 34 months.
The difference between the findings of the featured study and the original Swedish trial may have been due to several nationwide media campaigns encouraging parents to be stricter about youth drinking, which perhaps left too little room for improvement. It is also the case that programme developers
have been found
Editor's note: A reference is being made here to the 'researcher allegiance' effect. In several social research areas,1 programme developers and other researchers with an interest in the programme's success have been found to record more positive findings than fully independent researchers. Such overlaps between developers and researchers are endemic2 in drug prevention research.
1. See articles at the following web addresses:
2. See article at the following web address:
http://dx.doi.org/10.1016/j.evalprogplan.2007.06.004 to record more positive findings than fully independent researchers.
commentary With this study the Örebro programme joins the growing ranks of prevention programmes found effective when first tested by their developers but which later fail in independent trials. However, the original Swedish trial cannot yet be dismissed. It constituted a convincing demonstration of the potential power of harnessing the parent involvement mechanisms and influence of the school to reinforce parental responsibility in respect of their children's drinking – at least, in a culture conducive to such messages, and perhaps too when the schools and the presenters are taking part in a trailblazing project led by the programme's developers, who can be expected to be both expert on its implementation and enthusiastic about its potential.
Efforts to involve parents have generally been more elaborate but than the one trialled in the featured study and usually of at best modest effectiveness. A meta-analysis A study which uses recognised procedures to summarise quantitative results from several studies of the same or similar interventions to arrive at composite outcome scores. Usually undertaken to allow the intervention's effectiveness to be assessed with greater confidence than on the basis of the studies taken individually. combining findings from randomised studies of parent-focused substance use prevention programmes found modest effects in the form of fewer adolescent children starting to drink and a lower frequency of drinking. This was particularly the case when whole schools were engaged in the intervention, offering an opportunity for pupils and parents who participated in the programme to influence those who did not. However, the findings were undermined by a general failure to account for families which were unable to be followed up.
A common practical problem is getting parents to participate in face-to-face substance use prevention programmes. Typically in Britain (see for example 1 2 3) and elsewhere in Europe, attendance is very low, especially among parents most in need of parenting support and with lenient attitudes to substance use. Generally in these studies the attempt was to encourage attendance at special add-on events. On this count the featured study's strategy of incorporating prevention in to the school's core parent involvement programme has a distinct advantage. The downside is that at these events schools have a limited time in which communicate with parents; educational and other social issues (such as knife-carrying, guns, bullying, illegal drugs, teenage pregnancy) are likely to be seen as higher priorities both by the school and by the parents. Other solutions tried in Australia and the USA involve mailings to parents from the school or parent-child homework assignments.
Thanks for their comments on this entry in draft to Maria Bodin of the Centre for Psychiatry Research in Stockholm. Commentators bear no responsibility for the text including the interpretations and any remaining errors.
Last revised 12 August 2011
STUDY 2012 Differential impact of a Dutch alcohol prevention program targeting adolescents and parents separately and simultaneously: low self-control and lenient parenting at baseline predict effectiveness