Interventions for Suicide Survivors: A Review of the Literature

340 Interventions for Suicide Survivors
qualitative and quantitative (meta-analytic) re- interventions with children than with adults.
view of bereavement intervention studies. The authors also evaluated seven secondary
Their review of 13 studies yielded an overall prevention studies that focused on bereaved
effect size of .052, .272, and .095 for the re- persons defined as being at high risk for deduction
of depressive, somatic, and all other veloping bereavement related problems. Re-
psychological symptoms, respectively. The view of these studies revealed a modest amount
authors also computed a global effect size of of support for intervention efficacy. Finally,
.114 across all outcome measures, and con- Schut et al. (2001) examined seven tertiary
cluded from these findings “psychological in- intervention studies for people who had al-
terventions for bereavement are not effective ready developed a complicated mourning
interventions” (p. 293). No statistical exami- response, including populations who were
nation of moderator variables was reported suffering from clinical levels of depression,
in this review. anxiety, and other bereavement induced dis-
In another meta-analytic review, Fororders at the time of entry into the studies.
tner and Neimeyer (in Neimeyer, 2000) re- Despite some methodological limitations, the
viewed only investigations that met their cri- authors found that tertiary intervention was
teria of random assignment to treatment and generally successful, and concluded that:
control groups. They included studies in- “The general pattern emerging from this revolving
interventions for both children and view is that the more complicated the grief
adults across all types of losses. The authors process appears to be, the better the chances
found an overall effect size of .13 across the of interventions leading to positive results”
total sample of 23 studies. Using a novel mea- (p. 731).
sure of “treatment induced deterioration” What conclusions can be drawn about
(Neimeyer, 2000, p. 544), they also found the effectiveness of general bereavement inthat
approximately 38% of participants would terventions from this brief summary of re-
have had a better outcome had they been ascent reviews, and what implications do they
signed to the control, rather than the treat- have for interventions with suicide survivors?
ment condition. In an examination of moder- In general, the literature suggests that the efator
variables, the researchers also found that fect size of formal interventions for the be-
a greater length of time since the death, reaved is quite low, with the greatest reported
younger age of the subject, and higher levels effects being less than half those of typical
of risk (sudden violent death or evidence of psychotherapy outcome studies. While the
chronic grief) were related to increased effect reasons for this rather surprising finding are
size for the interventions (Neimeyer, 2000). not clear, one likely explanation is that most
No effects were found for other variables uncomplicated grief is naturally self-limiting
such as length of therapy, credentials of the for the majority of mourners (Raphael, Min-
therapist (professional vs. nonprofessional), kov, & Dobson, 2001; Stroebe, Hansson,
modality of treatment (individual vs. group), Stroebe, & Schut, 2001). This is evidenced
or theoretical approach used by the therapist. by the fact that in many studies, participants
In the most recent qualitative review in the control groups tended to improve
of bereavement intervention research, Schut, without any intervention, thus washing out
Stroebe, van den Bout, and Terheggen (2001) differences between control and treatment
provided a qualitative summary of three cate- groups. This same phenomenon may also be
gories of studies: primary, secondary, and ter- true for some suicide survivors, raising the
tiary interventions after loss that included in- important possibility that some survivors may
dividual, group, and family modalities. Their be neither at risk nor in need of formal sup-
evaluation of 16 primary intervention studies port services (McIntosh, 1999).
revealed little data to support the effective- A second factor that may contribute to
ness of such programs, although there was the apparent ineffectiveness of grief counsel-
marginally more support for the efficacy of
ing is the possible differential response of