New meta-review highlights the best approaches for preventing violence in young people

Young man with shot gun

The WISQARS Leading Causes of Death website is a mine of information for researchers and bloggers alike.

I searched it to find out what the leading causes of death are for people (all races, both sexes) in the US aged 10-24 years.

The top three in 2010 were:

  1. Unintentional injury (e.g. road traffic accidents, poisoning, drowning) (total deaths = 13,226)
  2. Suicide (total deaths = 4,867)
  3. Homicide (total deaths = 4,828)

Just to put this into perspective, you can combine the remaining top 10 causes of death (cancer, heart disease, congenital anomalies, cerebrovascular, chronic low respiratory disease, influenza, pneumonia and diabetes) and even together they don’t total as much as homicides.

We no doubt face a long wait for US Congress to make wholesale changes to gun laws, but in the meantime there is significant interest in what can be done to improve family, school and community settings so that future violence can be prevented.

A new meta-review published in the journal Aggression and Violent Behavior, evaluates youth violence prevention programs and brings together a huge amount of research that can help inform this work.

Methods

The authors carried out a really comprehensive search and found 37 meta-analyses and 15 systematic reviews to include in their meta-review.

The studies covered a wide range of youth violence including physical assault, carrying a weapon, criminal activity, aggression and antisocial behaviour.

The majority of studies targeted the individual or the family, but many also focused on school settings and peers.

Clearly with a review of this size, it’s important to categorise the evidence into different types. The authors did this with the 52 included studies, grouping them as follows:

  • General youth violence prevention programs (n=9)
  • Family programs (n=15)
  • Treatment-specific programs (n=8)
  • School-based programs (n=15)
  • Community-based programs (n=5)

Most included studies scored in the moderate range of the AMSTAR Rating Scale, meaning that they were of moderate quality.

Strong program effects were found for cognitive behavioural therapy, behavioural parent training, peer mediation and other school-based interventions

Strong program effects were found for cognitive behavioural therapy, behavioural parent training, peer mediation and other school-based interventions

Results

  • Most of the meta-analyses and systematic reviews were found to demonstrate moderate program effects
  • Strong program effects were found for cognitive behavioural therapy, behavioural parent training, peer mediation and other school-based interventions
  • For family programs, the strongest effects on aggression and delinquency were found for preschool and elementary-age youth (rather than early childhood)
  • Interestingly, one study found that participants in “Scared Straight” or similar programs were 1.5–1.96 times more likely to commit a crime and/or be delinquent at first follow-up than those in the control group (p<.01). A clear sign of this ‘treatment’ causing more harm than good

Conclusions

The authors concluded:

Meta-reviews are important for the field in order to summarize “what works” to prevent youth violence perpetration. They are a complement to meta-analyses and systematic reviews, which tend to focus on specific types of program evaluations (e.g., treatment-specific approaches).

In recent years, there has been an emphasis placed on utilizing evidence-based programs to maximize the likelihood that a program will reduce youth violence perpetration. Information gleaned from this meta-review can inform prevention efforts and can guide the field in determining next steps in program development and evaluation.

Link

Matjasko, JL et al. A systematic meta-review of evaluations of youth violence prevention programs: common and divergent findings from 25 years of meta-analyses and systematic reviews. Aggression and Violent Behavior, 17(6), Nov/Dec 2012, pp.540-552. [ScienceDirect abstract]

Share on Facebook Tweet this on Twitter Share on LinkedIn Share on Google+