This study has some limitations concerning the
choice and type of outcomes, the length of the follow-up period, and the research design. The
first issue to consider is that any choice of standardised
instruments automatically implies restrictions.
During SFBT, each individual formulated
his or her own goal. It is possible that the chosen
goal did not sufficiently match the measuring pretention
of the instruments used. This does not
apply to the SQP, because this measurement
adjusts itself to the individual’s goal. However, it
does hold true for the IDQOL and the POS, as
the quality of life domains and the resilience
domains within these instruments were broad and
could differ from what people with MID considered
to be relevant outcomes. The fact that SFBT
had minimal effect on, for example, social optimism,
may confirm this thought. Second, it is dif-
ficult to conclude from this study whether the
improvements attributed to SFBT can hold over
time. Although gains were made through the interventions,
it remains uncertain whether these
improvements will last over time (e.g. longer than
1 year). Another consideration is that the choice
and allocation of participants may be subject to
discussion. All SFBT clients were referred by staff
and not randomly allocated to both conditions. It
is possible that the selected clients tended to be
more co-operative in therapy and the outcomes
could be more favourable to SFBT compared with
a random selection. In future research, the intent
will be to recruit more potential participants in a
relatively shorter period of time (e.g. by collaboration
with other service providers). This would
enable researchers the ability to random allocation
of participants to the SFBT or CAU group.
Change process research can identify how clients
can benefit from any particular intervention
(McKeel 2012). For example in SFBT, goal setting
is an important issue. In future research, it may be
of interest to study to what extent this goal setting
accounts for the effects, rather than really working
on the goals. Additionally, measurements (in both
the SBT and the CAU condition) were also
administered directly by the therapists/researchers.
This meant that the participants and therapists/
researchers were not blinded to the treatment condition
or the treatment results. Biases due to the
non-random allocation of groups and non-blind
assessors may have influenced the results and
cannot be ruled out.