Two meta-analyses reports have reviewed SFBT
outcomes in the general population across a wide
range of studies. Stams et al. (2006) conducted a
meta-analysis of 21 studies investigating the effects
of SFBT, using Cohen’s d to measure effect sizes.
This meta-analysis found an overall small effect size
for SFBT (Cohen’s d = 0.37; 95% confidence interval:
0.19 < d < 0.55, P < 0.001). Studies that compared
SFBT with ‘no treatment’ (n = 4) yielded a
medium effect size of Cohen’s d (d = 0.57;
P < 0.01). Studies that compared SFBT with other
treatments (n = 7) yielded a small effect size that
was not statistically significant of Cohen’s d
(d = 0.16; not significant). Kim (2008) conducted a
second meta-analysis examining the effectiveness of
SFBT (22 comparison group studies) for different
types of outcomes: externalising behaviour problems,
internalising behaviour problems and family
or relationship problems. This meta-analysis found
small but positive treatment effects favouring the
SFBT groups. However, only the overall weighted
mean effect size for internalising problems, such as
depression, anxiety, self-concept and self-esteem,
was statistically significant at the P < 0.05 level, indicating
that the treatment effect of the SFBT groups
was better than that of the control groups. SFBT
appeared to be less effective with externalising
behaviour problems such as hyperactivity, conduct
problems, aggression, and family and relationship
problems. In a review of SFBT outcome research
Gingerich et al. (2012) stated: ‘SFBT is as good or
slightly better than other accepted treatments, but it
is clearly better than no treatment at all’ (p. 106).