Koning I.M., MacKinnon D., Maric M. et al.
Journal of Consulting and Clinical Psychology: 2015, 83(4), p. 719–727.
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 Koning at firstname.lastname@example.org.
First get the parents to set and communicate strict limits on their children’s drinking was the implication of this analysis of how in the Netherlands a combined adolescent education and parenting programme exerted unusually strong impacts on later drinking.
Summary The featured report derives from a previously analysed evaluation conducted in the Netherlands of an intervention aimed at reducing drinking in adolescents by educating pupils and prompting parents to set and communicate explicit limits to their children’s drinking. First this account offers a résumé of the study and findings on the effectiveness of the interventions, before turning to the featured report’s findings on how these results were achieved.
In the Netherlands combined parent limit-setting on drinking and adolescent education – but not either component alone – substantially restrained growth of the childrens’ drinking.
The combination worked at least partly by inducing parents to communicate stricter limits to their child’s drinking which directly and by bolstering the child’s self-control, led to less drinking.
Implies that programmes of this kind should start by prompting parents to explicitly control their child’s drinking, creating a context within which children react well to later educational efforts.
The study had randomly assigned participating schools to receive either just the parent intervention, just the adolescent education intervention, both interventions, or to act as control schools which simply carried on with the normal alcohol education.
The programme’s parental limit-setting component was based on the Örebro programme developed and tested in Sweden. It entailed a brief presentation from an alcohol expert at the first parents’ meeting in each school year on the adverse effects of youth drinking and of permissive parental attitudes to drinking. After this parents of children from the same class were meant to meet to agree rules about their children’s drinking. The other component was classroom-based education providing children alcohol-related information and skills-training.
At the start of the study the children were 12–13-years-old. Nearly three years later the study assessed how many had started drinking at least weekly or routinely started drinking heavily each weekend. To measure adolescent self-control and parental limit-setting, adolescents were given statements like, “I have trouble saying no,” or, “I am allowed to have one glass of alcohol when my parents are at home,” and asked to rate how far these applied to them.
When – and only when – parental and child components were combined did the programme restrain the adolescents’ drinking, effects several times greater and more consistent than those typical of education-based alcohol prevention programmes; at the final 34-month follow-up, for every four pupils allocated to parenting plus alcohol education, one was prevented from drinking weekly and also one from drinking heavily each week at age 15.
The featured report focused on the amount of drinking in a typical week at the final follow-up 34 months after pre-intervention baseline assessments. As with the main drinking outcomes, among combined-intervention pupils this was significantly lower among than among control pupils, not the case among pupils allocated to either of the component interventions. What led combined-intervention pupils to drink less was investigated by looking at the degree of self-control and parental limit-setting pupils had reported in the preceding follow-ups 10 and 22 months after baseline, satisfying a key requirement for establishing one thing caused another: cause must come before effect.
Relative to usual-education controls, by the 10-month follow-up the combined intervention had significantly bolstered the strictness of parental limit-setting about drinking, not an effect seen for the non-combined interventions, though the parent component came close. Stricter limit-setting at 10 months led directly and via greater pupil self-control at 22 months to less drinking at the final 34-month follow-up. The combined intervention also led to stricter limit-setting at 22 months, which led directly to less drinking at the final 34-month follow-up. A scenario which reversed the causal chain – greater adolescent self-control leading later to stricter parental limit-setting – was not supported by the data.
Unexpectedly, relative to usual-education controls, on their own the parent and adolescent components seemed to diminish self-control at the 22-month follow-up.
Compared to other children, the greater increase in adolescents’ self-control after the combined intervention was due to their parents previously communicating stricter limits to the child’s drinking. Together these processes led to relatively less drinking at the final follow-up. Regardless of any impacts on the child’s self-control, stricter parental limit-setting also directly reduced later drinking, perhaps because it affected other influences such as whether the child’s friends were drinkers. These findings confirm other research indicative of the pivotal role of parenting and imply that such interventions should first target parents in order to directly influence their children’s drinking and set the conditions for the children themselves to develop control over their drinking. However, these processes accounted for only around a third of the combined programme’s impact on drinking, indicating that it worked not just in these ways but in others yet to be identified.
commentary The analysis strongly suggests the combined intervention worked at least partly by inducing parents to communicate stricter limits to their child’s drinking. Directly and by bolstering the child’s confidence that they could control their drinking, this restrained the growth of drinking over the following years. Another implication of the findings is that explicit and strict parental limit-setting sets the scene for children to react as desired to later educational efforts intended to bolster self-control over their drinking. This mechanism is not only supported by the data, but accords with broader understandings of parenting as setting the context within which the child reacts to other influences and develops their own strengths and vulnerabilities.
Another analysis of results from the study found the combined intervention reduced the proportion of weekly drinkers only among the half of the children with the weakest self-control at the start of the study, or the half whose parents were most lenient about drinking – characteristics which offered the intervention the greatest scope to exert its effects. These characteristics did not, however, affect impacts on heavy weekly drinking.
Though the results were striking and study and analysis methodologically sound, they derived from just 20 the 80 schools asked to join the study; 60 refused. If schools participated because they were unusually committed to alcohol prevention, their outcomes might not be generally replicated. The study’s authors say most schools refused because they already hosted other research projects. Measures of parental limit-setting were based on the child’s perceptions, subject to influence by factors other than whether the parent really was strict. These concerns are mitigated by the fact that an intervention intended to bolster parental limit-setting had the desired effects, suggesting the parents’ stance and communications really were active ingredients. The first implementation of the parent intervention came about six months before the first alcohol lessons in school as part of the combined intervention; had the order been reversed, perhaps too the causal chain might have been reversed. However, initial implementations of both components were completed several months before the first follow-up; there was time for limit-setting and self-control to have been affected, but still the order was limit-setting leading only later to greater self-control.
Thanks for their comments on this entry to UK drug education expert Andrew Brown. Commentators bear no responsibility for the text including the interpretations and any remaining errors.
Last revised 09 September 2015. First uploaded 15 August 2015
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
STUDY 2000 Education's uncertain saviour
STUDY 2003 Family programme improves on school lessons